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{-mcrc}] [@b{-mno-crc}]
426 [@b{-minsn32}] [@b{-mno-insn32}]
427 [@b{-mfix7000}] [@b{-mno-fix7000}]
428 [@b{-mfix-rm7000}] [@b{-mno-fix-rm7000}]
429 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
430 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
431 [@b{-mdebug}] [@b{-no-mdebug}]
432 [@b{-mpdr}] [@b{-mno-pdr}]
436 @emph{Target MMIX options:}
437 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
438 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
439 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
440 [@b{--linker-allocated-gregs}]
444 @emph{Target Nios II options:}
445 [@b{-relax-all}] [@b{-relax-section}] [@b{-no-relax}]
450 @emph{Target NDS32 options:}
451 [@b{-EL}] [@b{-EB}] [@b{-O}] [@b{-Os}] [@b{-mcpu=@var{cpu}}]
452 [@b{-misa=@var{isa}}] [@b{-mabi=@var{abi}}] [@b{-mall-ext}]
453 [@b{-m[no-]16-bit}] [@b{-m[no-]perf-ext}] [@b{-m[no-]perf2-ext}]
454 [@b{-m[no-]string-ext}] [@b{-m[no-]dsp-ext}] [@b{-m[no-]mac}] [@b{-m[no-]div}]
455 [@b{-m[no-]audio-isa-ext}] [@b{-m[no-]fpu-sp-ext}] [@b{-m[no-]fpu-dp-ext}]
456 [@b{-m[no-]fpu-fma}] [@b{-mfpu-freg=@var{FREG}}] [@b{-mreduced-regs}]
457 [@b{-mfull-regs}] [@b{-m[no-]dx-regs}] [@b{-mpic}] [@b{-mno-relax}]
462 @emph{Target PDP11 options:}
463 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
464 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
465 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
469 @emph{Target picoJava options:}
474 @emph{Target PowerPC options:}
476 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|@b{-m403}|@b{-m405}|
477 @b{-m440}|@b{-m464}|@b{-m476}|@b{-m7400}|@b{-m7410}|@b{-m7450}|@b{-m7455}|@b{-m750cl}|@b{-mppc64}|
478 @b{-m620}|@b{-me500}|@b{-e500x2}|@b{-me500mc}|@b{-me500mc64}|@b{-me5500}|@b{-me6500}|@b{-mppc64bridge}|
479 @b{-mbooke}|@b{-mpower4}|@b{-mpwr4}|@b{-mpower5}|@b{-mpwr5}|@b{-mpwr5x}|@b{-mpower6}|@b{-mpwr6}|
480 @b{-mpower7}|@b{-mpwr7}|@b{-mpower8}|@b{-mpwr8}|@b{-mpower9}|@b{-mpwr9}@b{-ma2}|
481 @b{-mcell}|@b{-mspe}|@b{-mspe2}|@b{-mtitan}|@b{-me300}|@b{-mcom}]
482 [@b{-many}] [@b{-maltivec}|@b{-mvsx}|@b{-mhtm}|@b{-mvle}]
483 [@b{-mregnames}|@b{-mno-regnames}]
484 [@b{-mrelocatable}|@b{-mrelocatable-lib}|@b{-K PIC}] [@b{-memb}]
485 [@b{-mlittle}|@b{-mlittle-endian}|@b{-le}|@b{-mbig}|@b{-mbig-endian}|@b{-be}]
486 [@b{-msolaris}|@b{-mno-solaris}]
487 [@b{-nops=@var{count}}]
491 @emph{Target PRU options:}
494 [@b{-mno-warn-regname-label}]
498 @emph{Target RISC-V options:}
499 [@b{-fpic}|@b{-fPIC}|@b{-fno-pic}]
500 [@b{-march}=@var{ISA}]
501 [@b{-mabi}=@var{ABI}]
505 @emph{Target RL78 options:}
507 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
511 @emph{Target RX options:}
512 [@b{-mlittle-endian}|@b{-mbig-endian}]
513 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
514 [@b{-muse-conventional-section-names}]
515 [@b{-msmall-data-limit}]
518 [@b{-mint-register=@var{number}}]
519 [@b{-mgcc-abi}|@b{-mrx-abi}]
523 @emph{Target s390 options:}
524 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
525 [@b{-mregnames}|@b{-mno-regnames}]
526 [@b{-mwarn-areg-zero}]
530 @emph{Target SCORE options:}
531 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
532 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
533 [@b{-march=score7}][@b{-march=score3}]
534 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
538 @emph{Target SPARC options:}
539 @c The order here is important. See c-sparc.texi.
540 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Aleon}|@b{-Asparclet}|@b{-Asparclite}
541 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av8plusb}|@b{-Av8plusc}|@b{-Av8plusd}
542 @b{-Av8plusv}|@b{-Av8plusm}|@b{-Av9}|@b{-Av9a}|@b{-Av9b}|@b{-Av9c}
543 @b{-Av9d}|@b{-Av9e}|@b{-Av9v}|@b{-Av9m}|@b{-Asparc}|@b{-Asparcvis}
544 @b{-Asparcvis2}|@b{-Asparcfmaf}|@b{-Asparcima}|@b{-Asparcvis3}
545 @b{-Asparcvisr}|@b{-Asparc5}]
546 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}]|@b{-xarch=v8plusb}|@b{-xarch=v8plusc}
547 @b{-xarch=v8plusd}|@b{-xarch=v8plusv}|@b{-xarch=v8plusm}|@b{-xarch=v9}
548 @b{-xarch=v9a}|@b{-xarch=v9b}|@b{-xarch=v9c}|@b{-xarch=v9d}|@b{-xarch=v9e}
549 @b{-xarch=v9v}|@b{-xarch=v9m}|@b{-xarch=sparc}|@b{-xarch=sparcvis}
550 @b{-xarch=sparcvis2}|@b{-xarch=sparcfmaf}|@b{-xarch=sparcima}
551 @b{-xarch=sparcvis3}|@b{-xarch=sparcvisr}|@b{-xarch=sparc5}
554 [@b{--enforce-aligned-data}][@b{--dcti-couples-detect}]
558 @emph{Target TIC54X options:}
559 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
560 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
564 @emph{Target TIC6X options:}
565 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
566 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
567 [@b{-mpic}|@b{-mno-pic}]
571 @emph{Target TILE-Gx options:}
572 [@b{-m32}|@b{-m64}][@b{-EB}][@b{-EL}]
575 @c TILEPro has no machine-dependent assembler options
579 @emph{Target Visium options:}
580 [@b{-mtune=@var{arch}}]
584 @emph{Target Xtensa options:}
585 [@b{--[no-]text-section-literals}] [@b{--[no-]auto-litpools}]
586 [@b{--[no-]absolute-literals}]
587 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
588 [@b{--[no-]transform}]
589 [@b{--rename-section} @var{oldname}=@var{newname}]
590 [@b{--[no-]trampolines}]
594 @emph{Target Z80 options:}
595 [@b{-z80}] [@b{-r800}]
596 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
597 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
598 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
599 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
600 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
601 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
605 @c Z8000 has no machine-dependent assembler options
614 @include at-file.texi
617 Turn on listings, in any of a variety of ways:
621 omit false conditionals
624 omit debugging directives
627 include general information, like @value{AS} version and options passed
630 include high-level source
636 include macro expansions
639 omit forms processing
645 set the name of the listing file
648 You may combine these options; for example, use @samp{-aln} for assembly
649 listing without forms processing. The @samp{=file} option, if used, must be
650 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
653 Begin in alternate macro mode.
655 @xref{Altmacro,,@code{.altmacro}}.
658 @item --compress-debug-sections
659 Compress DWARF debug sections using zlib with SHF_COMPRESSED from the
660 ELF ABI. The resulting object file may not be compatible with older
661 linkers and object file utilities. Note if compression would make a
662 given section @emph{larger} then it is not compressed.
665 @cindex @samp{--compress-debug-sections=} option
666 @item --compress-debug-sections=none
667 @itemx --compress-debug-sections=zlib
668 @itemx --compress-debug-sections=zlib-gnu
669 @itemx --compress-debug-sections=zlib-gabi
670 These options control how DWARF debug sections are compressed.
671 @option{--compress-debug-sections=none} is equivalent to
672 @option{--nocompress-debug-sections}.
673 @option{--compress-debug-sections=zlib} and
674 @option{--compress-debug-sections=zlib-gabi} are equivalent to
675 @option{--compress-debug-sections}.
676 @option{--compress-debug-sections=zlib-gnu} compresses DWARF debug
677 sections using zlib. The debug sections are renamed to begin with
678 @samp{.zdebug}. Note if compression would make a given section
679 @emph{larger} then it is not compressed nor renamed.
683 @item --nocompress-debug-sections
684 Do not compress DWARF debug sections. This is usually the default for all
685 targets except the x86/x86_64, but a configure time option can be used to
689 Ignored. This option is accepted for script compatibility with calls to
692 @item --debug-prefix-map @var{old}=@var{new}
693 When assembling files in directory @file{@var{old}}, record debugging
694 information describing them as in @file{@var{new}} instead.
696 @item --defsym @var{sym}=@var{value}
697 Define the symbol @var{sym} to be @var{value} before assembling the input file.
698 @var{value} must be an integer constant. As in C, a leading @samp{0x}
699 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
700 value. The value of the symbol can be overridden inside a source file via the
701 use of a @code{.set} pseudo-op.
704 ``fast''---skip whitespace and comment preprocessing (assume source is
709 Generate debugging information for each assembler source line using whichever
710 debug format is preferred by the target. This currently means either STABS,
714 Generate stabs debugging information for each assembler line. This
715 may help debugging assembler code, if the debugger can handle it.
718 Generate stabs debugging information for each assembler line, with GNU
719 extensions that probably only gdb can handle, and that could make other
720 debuggers crash or refuse to read your program. This
721 may help debugging assembler code. Currently the only GNU extension is
722 the location of the current working directory at assembling time.
725 Generate DWARF2 debugging information for each assembler line. This
726 may help debugging assembler code, if the debugger can handle it. Note---this
727 option is only supported by some targets, not all of them.
729 @item --gdwarf-sections
730 Instead of creating a .debug_line section, create a series of
731 .debug_line.@var{foo} sections where @var{foo} is the name of the
732 corresponding code section. For example a code section called @var{.text.func}
733 will have its dwarf line number information placed into a section called
734 @var{.debug_line.text.func}. If the code section is just called @var{.text}
735 then debug line section will still be called just @var{.debug_line} without any
739 @item --size-check=error
740 @itemx --size-check=warning
741 Issue an error or warning for invalid ELF .size directive.
743 @item --elf-stt-common=no
744 @itemx --elf-stt-common=yes
745 These options control whether the ELF assembler should generate common
746 symbols with the @code{STT_COMMON} type. The default can be controlled
747 by a configure option @option{--enable-elf-stt-common}.
749 @item --generate-missing-build-notes=yes
750 @itemx --generate-missing-build-notes=no
751 These options control whether the ELF assembler should generate GNU Build
752 attribute notes if none are present in the input sources.
753 The default can be controlled by the @option{--enable-generate-build-notes}
759 Print a summary of the command line options and exit.
762 Print a summary of all target specific options and exit.
765 Add directory @var{dir} to the search list for @code{.include} directives.
768 Don't warn about signed overflow.
771 @ifclear DIFF-TBL-KLUGE
772 This option is accepted but has no effect on the @value{TARGET} family.
774 @ifset DIFF-TBL-KLUGE
775 Issue warnings when difference tables altered for long displacements.
780 Keep (in the symbol table) local symbols. These symbols start with
781 system-specific local label prefixes, typically @samp{.L} for ELF systems
782 or @samp{L} for traditional a.out systems.
787 @item --listing-lhs-width=@var{number}
788 Set the maximum width, in words, of the output data column for an assembler
789 listing to @var{number}.
791 @item --listing-lhs-width2=@var{number}
792 Set the maximum width, in words, of the output data column for continuation
793 lines in an assembler listing to @var{number}.
795 @item --listing-rhs-width=@var{number}
796 Set the maximum width of an input source line, as displayed in a listing, to
799 @item --listing-cont-lines=@var{number}
800 Set the maximum number of lines printed in a listing for a single line of input
803 @item --no-pad-sections
804 Stop the assembler for padding the ends of output sections to the alignment
805 of that section. The default is to pad the sections, but this can waste space
806 which might be needed on targets which have tight memory constraints.
808 @item -o @var{objfile}
809 Name the object-file output from @command{@value{AS}} @var{objfile}.
812 Fold the data section into the text section.
814 @item --hash-size=@var{number}
815 Set the default size of GAS's hash tables to a prime number close to
816 @var{number}. Increasing this value can reduce the length of time it takes the
817 assembler to perform its tasks, at the expense of increasing the assembler's
818 memory requirements. Similarly reducing this value can reduce the memory
819 requirements at the expense of speed.
821 @item --reduce-memory-overheads
822 This option reduces GAS's memory requirements, at the expense of making the
823 assembly processes slower. Currently this switch is a synonym for
824 @samp{--hash-size=4051}, but in the future it may have other effects as well.
827 @item --sectname-subst
828 Honor substitution sequences in section names.
830 @xref{Section Name Substitutions,,@code{.section @var{name}}}.
835 Print the maximum space (in bytes) and total time (in seconds) used by
838 @item --strip-local-absolute
839 Remove local absolute symbols from the outgoing symbol table.
843 Print the @command{as} version.
846 Print the @command{as} version and exit.
850 Suppress warning messages.
852 @item --fatal-warnings
853 Treat warnings as errors.
856 Don't suppress warning messages or treat them as errors.
865 Generate an object file even after errors.
867 @item -- | @var{files} @dots{}
868 Standard input, or source files to assemble.
876 @xref{AArch64 Options}, for the options available when @value{AS} is configured
877 for the 64-bit mode of the ARM Architecture (AArch64).
882 The following options are available when @value{AS} is configured for the
883 64-bit mode of the ARM Architecture (AArch64).
886 @include c-aarch64.texi
887 @c ended inside the included file
895 @xref{Alpha Options}, for the options available when @value{AS} is configured
896 for an Alpha processor.
901 The following options are available when @value{AS} is configured for an Alpha
905 @include c-alpha.texi
906 @c ended inside the included file
913 The following options are available when @value{AS} is configured for an ARC
917 @item -mcpu=@var{cpu}
918 This option selects the core processor variant.
920 Select either big-endian (-EB) or little-endian (-EL) output.
922 Enable Code Density extenssion instructions.
927 The following options are available when @value{AS} is configured for the ARM
931 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
932 Specify which ARM processor variant is the target.
933 @item -march=@var{architecture}[+@var{extension}@dots{}]
934 Specify which ARM architecture variant is used by the target.
935 @item -mfpu=@var{floating-point-format}
936 Select which Floating Point architecture is the target.
937 @item -mfloat-abi=@var{abi}
938 Select which floating point ABI is in use.
940 Enable Thumb only instruction decoding.
941 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
942 Select which procedure calling convention is in use.
944 Select either big-endian (-EB) or little-endian (-EL) output.
945 @item -mthumb-interwork
946 Specify that the code has been generated with interworking between Thumb and
949 Turns on CodeComposer Studio assembly syntax compatibility mode.
951 Specify that PIC code has been generated.
959 @xref{Blackfin Options}, for the options available when @value{AS} is
960 configured for the Blackfin processor family.
965 The following options are available when @value{AS} is configured for
966 the Blackfin processor family.
970 @c ended inside the included file
977 See the info pages for documentation of the CRIS-specific options.
981 The following options are available when @value{AS} is configured for
984 @cindex D10V optimization
985 @cindex optimization, D10V
987 Optimize output by parallelizing instructions.
992 The following options are available when @value{AS} is configured for a D30V
995 @cindex D30V optimization
996 @cindex optimization, D30V
998 Optimize output by parallelizing instructions.
1002 Warn when nops are generated.
1004 @cindex D30V nops after 32-bit multiply
1006 Warn when a nop after a 32-bit multiply instruction is generated.
1012 The following options are available when @value{AS} is configured for the
1013 Adapteva EPIPHANY series.
1016 @xref{Epiphany Options}, for the options available when @value{AS} is
1017 configured for an Epiphany processor.
1021 @c man begin OPTIONS
1022 The following options are available when @value{AS} is configured for
1023 an Epiphany processor.
1025 @c man begin INCLUDE
1026 @include c-epiphany.texi
1027 @c ended inside the included file
1035 @xref{H8/300 Options}, for the options available when @value{AS} is configured
1036 for an H8/300 processor.
1040 @c man begin OPTIONS
1041 The following options are available when @value{AS} is configured for an H8/300
1044 @c man begin INCLUDE
1045 @include c-h8300.texi
1046 @c ended inside the included file
1054 @xref{i386-Options}, for the options available when @value{AS} is
1055 configured for an i386 processor.
1059 @c man begin OPTIONS
1060 The following options are available when @value{AS} is configured for
1063 @c man begin INCLUDE
1064 @include c-i386.texi
1065 @c ended inside the included file
1070 @c man begin OPTIONS
1072 The following options are available when @value{AS} is configured for the
1078 Specifies that the extended IP2022 instructions are allowed.
1081 Restores the default behaviour, which restricts the permitted instructions to
1082 just the basic IP2022 ones.
1088 The following options are available when @value{AS} is configured for the
1089 Renesas M32C and M16C processors.
1094 Assemble M32C instructions.
1097 Assemble M16C instructions (the default).
1100 Enable support for link-time relaxations.
1103 Support H'00 style hex constants in addition to 0x00 style.
1109 The following options are available when @value{AS} is configured for the
1110 Renesas M32R (formerly Mitsubishi M32R) series.
1115 Specify which processor in the M32R family is the target. The default
1116 is normally the M32R, but this option changes it to the M32RX.
1118 @item --warn-explicit-parallel-conflicts or --Wp
1119 Produce warning messages when questionable parallel constructs are
1122 @item --no-warn-explicit-parallel-conflicts or --Wnp
1123 Do not produce warning messages when questionable parallel constructs are
1130 The following options are available when @value{AS} is configured for the
1131 Motorola 68000 series.
1136 Shorten references to undefined symbols, to one word instead of two.
1138 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1139 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1140 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1141 Specify what processor in the 68000 family is the target. The default
1142 is normally the 68020, but this can be changed at configuration time.
1144 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1145 The target machine does (or does not) have a floating-point coprocessor.
1146 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1147 the basic 68000 is not compatible with the 68881, a combination of the
1148 two can be specified, since it's possible to do emulation of the
1149 coprocessor instructions with the main processor.
1151 @item -m68851 | -mno-68851
1152 The target machine does (or does not) have a memory-management
1153 unit coprocessor. The default is to assume an MMU for 68020 and up.
1161 @xref{Nios II Options}, for the options available when @value{AS} is configured
1162 for an Altera Nios II processor.
1166 @c man begin OPTIONS
1167 The following options are available when @value{AS} is configured for an
1168 Altera Nios II processor.
1170 @c man begin INCLUDE
1171 @include c-nios2.texi
1172 @c ended inside the included file
1178 For details about the PDP-11 machine dependent features options,
1179 see @ref{PDP-11-Options}.
1182 @item -mpic | -mno-pic
1183 Generate position-independent (or position-dependent) code. The
1184 default is @option{-mpic}.
1187 @itemx -mall-extensions
1188 Enable all instruction set extensions. This is the default.
1190 @item -mno-extensions
1191 Disable all instruction set extensions.
1193 @item -m@var{extension} | -mno-@var{extension}
1194 Enable (or disable) a particular instruction set extension.
1197 Enable the instruction set extensions supported by a particular CPU, and
1198 disable all other extensions.
1200 @item -m@var{machine}
1201 Enable the instruction set extensions supported by a particular machine
1202 model, and disable all other extensions.
1208 The following options are available when @value{AS} is configured for
1209 a picoJava processor.
1213 @cindex PJ endianness
1214 @cindex endianness, PJ
1215 @cindex big endian output, PJ
1217 Generate ``big endian'' format output.
1219 @cindex little endian output, PJ
1221 Generate ``little endian'' format output.
1229 @xref{PRU Options}, for the options available when @value{AS} is configured
1230 for a PRU processor.
1234 @c man begin OPTIONS
1235 The following options are available when @value{AS} is configured for a
1238 @c man begin INCLUDE
1240 @c ended inside the included file
1245 The following options are available when @value{AS} is configured for the
1246 Motorola 68HC11 or 68HC12 series.
1250 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1251 Specify what processor is the target. The default is
1252 defined by the configuration option when building the assembler.
1254 @item --xgate-ramoffset
1255 Instruct the linker to offset RAM addresses from S12X address space into
1256 XGATE address space.
1259 Specify to use the 16-bit integer ABI.
1262 Specify to use the 32-bit integer ABI.
1264 @item -mshort-double
1265 Specify to use the 32-bit double ABI.
1268 Specify to use the 64-bit double ABI.
1270 @item --force-long-branches
1271 Relative branches are turned into absolute ones. This concerns
1272 conditional branches, unconditional branches and branches to a
1275 @item -S | --short-branches
1276 Do not turn relative branches into absolute ones
1277 when the offset is out of range.
1279 @item --strict-direct-mode
1280 Do not turn the direct addressing mode into extended addressing mode
1281 when the instruction does not support direct addressing mode.
1283 @item --print-insn-syntax
1284 Print the syntax of instruction in case of error.
1286 @item --print-opcodes
1287 Print the list of instructions with syntax and then exit.
1289 @item --generate-example
1290 Print an example of instruction for each possible instruction and then exit.
1291 This option is only useful for testing @command{@value{AS}}.
1297 The following options are available when @command{@value{AS}} is configured
1298 for the SPARC architecture:
1301 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1302 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1303 Explicitly select a variant of the SPARC architecture.
1305 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1306 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1308 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1309 UltraSPARC extensions.
1311 @item -xarch=v8plus | -xarch=v8plusa
1312 For compatibility with the Solaris v9 assembler. These options are
1313 equivalent to -Av8plus and -Av8plusa, respectively.
1316 Warn when the assembler switches to another architecture.
1321 The following options are available when @value{AS} is configured for the 'c54x
1326 Enable extended addressing mode. All addresses and relocations will assume
1327 extended addressing (usually 23 bits).
1328 @item -mcpu=@var{CPU_VERSION}
1329 Sets the CPU version being compiled for.
1330 @item -merrors-to-file @var{FILENAME}
1331 Redirect error output to a file, for broken systems which don't support such
1332 behaviour in the shell.
1337 @c man begin OPTIONS
1338 The following options are available when @value{AS} is configured for
1343 This option sets the largest size of an object that can be referenced
1344 implicitly with the @code{gp} register. It is only accepted for targets that
1345 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1347 @cindex MIPS endianness
1348 @cindex endianness, MIPS
1349 @cindex big endian output, MIPS
1351 Generate ``big endian'' format output.
1353 @cindex little endian output, MIPS
1355 Generate ``little endian'' format output.
1373 Generate code for a particular MIPS Instruction Set Architecture level.
1374 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1375 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1376 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1377 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1378 @samp{-mips32r5}, @samp{-mips32r6}, @samp{-mips64}, @samp{-mips64r2},
1379 @samp{-mips64r3}, @samp{-mips64r5}, and @samp{-mips64r6} correspond to generic
1380 MIPS V, MIPS32, MIPS32 Release 2, MIPS32 Release 3, MIPS32 Release 5, MIPS32
1381 Release 6, MIPS64, MIPS64 Release 2, MIPS64 Release 3, MIPS64 Release 5, and
1382 MIPS64 Release 6 ISA processors, respectively.
1384 @item -march=@var{cpu}
1385 Generate code for a particular MIPS CPU.
1387 @item -mtune=@var{cpu}
1388 Schedule and tune for a particular MIPS CPU.
1392 Cause nops to be inserted if the read of the destination register
1393 of an mfhi or mflo instruction occurs in the following two instructions.
1396 @itemx -mno-fix-rm7000
1397 Cause nops to be inserted if a dmult or dmultu instruction is
1398 followed by a load instruction.
1402 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1403 section instead of the standard ELF .stabs sections.
1407 Control generation of @code{.pdr} sections.
1411 The register sizes are normally inferred from the ISA and ABI, but these
1412 flags force a certain group of registers to be treated as 32 bits wide at
1413 all times. @samp{-mgp32} controls the size of general-purpose registers
1414 and @samp{-mfp32} controls the size of floating-point registers.
1418 The register sizes are normally inferred from the ISA and ABI, but these
1419 flags force a certain group of registers to be treated as 64 bits wide at
1420 all times. @samp{-mgp64} controls the size of general-purpose registers
1421 and @samp{-mfp64} controls the size of floating-point registers.
1424 The register sizes are normally inferred from the ISA and ABI, but using
1425 this flag in combination with @samp{-mabi=32} enables an ABI variant
1426 which will operate correctly with floating-point registers which are
1430 @itemx -mno-odd-spreg
1431 Enable use of floating-point operations on odd-numbered single-precision
1432 registers when supported by the ISA. @samp{-mfpxx} implies
1433 @samp{-mno-odd-spreg}, otherwise the default is @samp{-modd-spreg}.
1437 Generate code for the MIPS 16 processor. This is equivalent to putting
1438 @code{.module mips16} at the start of the assembly file. @samp{-no-mips16}
1439 turns off this option.
1442 @itemx -mno-mips16e2
1443 Enable the use of MIPS16e2 instructions in MIPS16 mode. This is equivalent
1444 to putting @code{.module mips16e2} at the start of the assembly file.
1445 @samp{-mno-mips16e2} turns off this option.
1448 @itemx -mno-micromips
1449 Generate code for the microMIPS processor. This is equivalent to putting
1450 @code{.module micromips} at the start of the assembly file.
1451 @samp{-mno-micromips} turns off this option. This is equivalent to putting
1452 @code{.module nomicromips} at the start of the assembly file.
1455 @itemx -mno-smartmips
1456 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1457 equivalent to putting @code{.module smartmips} at the start of the assembly
1458 file. @samp{-mno-smartmips} turns off this option.
1462 Generate code for the MIPS-3D Application Specific Extension.
1463 This tells the assembler to accept MIPS-3D instructions.
1464 @samp{-no-mips3d} turns off this option.
1468 Generate code for the MDMX Application Specific Extension.
1469 This tells the assembler to accept MDMX instructions.
1470 @samp{-no-mdmx} turns off this option.
1474 Generate code for the DSP Release 1 Application Specific Extension.
1475 This tells the assembler to accept DSP Release 1 instructions.
1476 @samp{-mno-dsp} turns off this option.
1480 Generate code for the DSP Release 2 Application Specific Extension.
1481 This option implies @samp{-mdsp}.
1482 This tells the assembler to accept DSP Release 2 instructions.
1483 @samp{-mno-dspr2} turns off this option.
1487 Generate code for the DSP Release 3 Application Specific Extension.
1488 This option implies @samp{-mdsp} and @samp{-mdspr2}.
1489 This tells the assembler to accept DSP Release 3 instructions.
1490 @samp{-mno-dspr3} turns off this option.
1494 Generate code for the MIPS SIMD Architecture Extension.
1495 This tells the assembler to accept MSA instructions.
1496 @samp{-mno-msa} turns off this option.
1500 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1501 This tells the assembler to accept XPA instructions.
1502 @samp{-mno-xpa} turns off this option.
1506 Generate code for the MT Application Specific Extension.
1507 This tells the assembler to accept MT instructions.
1508 @samp{-mno-mt} turns off this option.
1512 Generate code for the MCU Application Specific Extension.
1513 This tells the assembler to accept MCU instructions.
1514 @samp{-mno-mcu} turns off this option.
1518 Generate code for the MIPS cyclic redundancy check (CRC) Application
1519 Specific Extension. This tells the assembler to accept CRC instructions.
1520 @samp{-mno-crc} turns off this option.
1524 Only use 32-bit instruction encodings when generating code for the
1525 microMIPS processor. This option inhibits the use of any 16-bit
1526 instructions. This is equivalent to putting @code{.set insn32} at
1527 the start of the assembly file. @samp{-mno-insn32} turns off this
1528 option. This is equivalent to putting @code{.set noinsn32} at the
1529 start of the assembly file. By default @samp{-mno-insn32} is
1530 selected, allowing all instructions to be used.
1532 @item --construct-floats
1533 @itemx --no-construct-floats
1534 The @samp{--no-construct-floats} option disables the construction of
1535 double width floating point constants by loading the two halves of the
1536 value into the two single width floating point registers that make up
1537 the double width register. By default @samp{--construct-floats} is
1538 selected, allowing construction of these floating point constants.
1540 @item --relax-branch
1541 @itemx --no-relax-branch
1542 The @samp{--relax-branch} option enables the relaxation of out-of-range
1543 branches. By default @samp{--no-relax-branch} is selected, causing any
1544 out-of-range branches to produce an error.
1546 @item -mignore-branch-isa
1547 @itemx -mno-ignore-branch-isa
1548 Ignore branch checks for invalid transitions between ISA modes. The
1549 semantics of branches does not provide for an ISA mode switch, so in
1550 most cases the ISA mode a branch has been encoded for has to be the
1551 same as the ISA mode of the branch's target label. Therefore GAS has
1552 checks implemented that verify in branch assembly that the two ISA
1553 modes match. @samp{-mignore-branch-isa} disables these checks. By
1554 default @samp{-mno-ignore-branch-isa} is selected, causing any invalid
1555 branch requiring a transition between ISA modes to produce an error.
1557 @item -mnan=@var{encoding}
1558 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1559 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1562 @item --emulation=@var{name}
1563 This option was formerly used to switch between ELF and ECOFF output
1564 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1565 removed in GAS 2.24, so the option now serves little purpose.
1566 It is retained for backwards compatibility.
1568 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1569 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1570 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1571 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1572 preferred options instead.
1575 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1582 Control how to deal with multiplication overflow and division by zero.
1583 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1584 (and only work for Instruction Set Architecture level 2 and higher);
1585 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1589 When this option is used, @command{@value{AS}} will issue a warning every
1590 time it generates a nop instruction from a macro.
1596 The following options are available when @value{AS} is configured for
1602 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1603 The command line option @samp{-nojsri2bsr} can be used to disable it.
1607 Enable or disable the silicon filter behaviour. By default this is disabled.
1608 The default can be overridden by the @samp{-sifilter} command line option.
1611 Alter jump instructions for long displacements.
1613 @item -mcpu=[210|340]
1614 Select the cpu type on the target hardware. This controls which instructions
1618 Assemble for a big endian target.
1621 Assemble for a little endian target.
1630 @xref{Meta Options}, for the options available when @value{AS} is configured
1631 for a Meta processor.
1635 @c man begin OPTIONS
1636 The following options are available when @value{AS} is configured for a
1639 @c man begin INCLUDE
1640 @include c-metag.texi
1641 @c ended inside the included file
1646 @c man begin OPTIONS
1648 See the info pages for documentation of the MMIX-specific options.
1654 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1655 for a NDS32 processor.
1657 @c ended inside the included file
1661 @c man begin OPTIONS
1662 The following options are available when @value{AS} is configured for a
1665 @c man begin INCLUDE
1666 @include c-nds32.texi
1667 @c ended inside the included file
1674 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1675 for a PowerPC processor.
1679 @c man begin OPTIONS
1680 The following options are available when @value{AS} is configured for a
1683 @c man begin INCLUDE
1685 @c ended inside the included file
1693 @xref{RISC-V-Options}, for the options available when @value{AS} is configured
1694 for a RISC-V processor.
1698 @c man begin OPTIONS
1699 The following options are available when @value{AS} is configured for a
1702 @c man begin INCLUDE
1703 @include c-riscv.texi
1704 @c ended inside the included file
1709 @c man begin OPTIONS
1711 See the info pages for documentation of the RX-specific options.
1715 The following options are available when @value{AS} is configured for the s390
1721 Select the word size, either 31/32 bits or 64 bits.
1724 Select the architecture mode, either the Enterprise System
1725 Architecture (esa) or the z/Architecture mode (zarch).
1726 @item -march=@var{processor}
1727 Specify which s390 processor variant is the target, @samp{g5} (or
1728 @samp{arch3}), @samp{g6}, @samp{z900} (or @samp{arch5}), @samp{z990} (or
1729 @samp{arch6}), @samp{z9-109}, @samp{z9-ec} (or @samp{arch7}), @samp{z10} (or
1730 @samp{arch8}), @samp{z196} (or @samp{arch9}), @samp{zEC12} (or @samp{arch10}),
1731 @samp{z13} (or @samp{arch11}), or @samp{z14} (or @samp{arch12}).
1733 @itemx -mno-regnames
1734 Allow or disallow symbolic names for registers.
1735 @item -mwarn-areg-zero
1736 Warn whenever the operand for a base or index register has been specified
1737 but evaluates to zero.
1745 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1746 for a TMS320C6000 processor.
1750 @c man begin OPTIONS
1751 The following options are available when @value{AS} is configured for a
1752 TMS320C6000 processor.
1754 @c man begin INCLUDE
1755 @include c-tic6x.texi
1756 @c ended inside the included file
1764 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1765 for a TILE-Gx processor.
1769 @c man begin OPTIONS
1770 The following options are available when @value{AS} is configured for a TILE-Gx
1773 @c man begin INCLUDE
1774 @include c-tilegx.texi
1775 @c ended inside the included file
1783 @xref{Visium Options}, for the options available when @value{AS} is configured
1784 for a Visium processor.
1788 @c man begin OPTIONS
1789 The following option is available when @value{AS} is configured for a Visium
1792 @c man begin INCLUDE
1793 @include c-visium.texi
1794 @c ended inside the included file
1802 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1803 for an Xtensa processor.
1807 @c man begin OPTIONS
1808 The following options are available when @value{AS} is configured for an
1811 @c man begin INCLUDE
1812 @include c-xtensa.texi
1813 @c ended inside the included file
1818 @c man begin OPTIONS
1821 The following options are available when @value{AS} is configured for
1822 a Z80 family processor.
1825 Assemble for Z80 processor.
1827 Assemble for R800 processor.
1828 @item -ignore-undocumented-instructions
1830 Assemble undocumented Z80 instructions that also work on R800 without warning.
1831 @item -ignore-unportable-instructions
1833 Assemble all undocumented Z80 instructions without warning.
1834 @item -warn-undocumented-instructions
1836 Issue a warning for undocumented Z80 instructions that also work on R800.
1837 @item -warn-unportable-instructions
1839 Issue a warning for undocumented Z80 instructions that do not work on R800.
1840 @item -forbid-undocumented-instructions
1842 Treat all undocumented instructions as errors.
1843 @item -forbid-unportable-instructions
1845 Treat undocumented Z80 instructions that do not work on R800 as errors.
1852 * Manual:: Structure of this Manual
1853 * GNU Assembler:: The GNU Assembler
1854 * Object Formats:: Object File Formats
1855 * Command Line:: Command Line
1856 * Input Files:: Input Files
1857 * Object:: Output (Object) File
1858 * Errors:: Error and Warning Messages
1862 @section Structure of this Manual
1864 @cindex manual, structure and purpose
1865 This manual is intended to describe what you need to know to use
1866 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1867 notation for symbols, constants, and expressions; the directives that
1868 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1871 We also cover special features in the @value{TARGET}
1872 configuration of @command{@value{AS}}, including assembler directives.
1875 This manual also describes some of the machine-dependent features of
1876 various flavors of the assembler.
1879 @cindex machine instructions (not covered)
1880 On the other hand, this manual is @emph{not} intended as an introduction
1881 to programming in assembly language---let alone programming in general!
1882 In a similar vein, we make no attempt to introduce the machine
1883 architecture; we do @emph{not} describe the instruction set, standard
1884 mnemonics, registers or addressing modes that are standard to a
1885 particular architecture.
1887 You may want to consult the manufacturer's
1888 machine architecture manual for this information.
1892 For information on the H8/300 machine instruction set, see @cite{H8/300
1893 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1894 Programming Manual} (Renesas).
1897 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1898 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1899 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1900 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1903 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1907 @c I think this is premature---doc@cygnus.com, 17jan1991
1909 Throughout this manual, we assume that you are running @dfn{GNU},
1910 the portable operating system from the @dfn{Free Software
1911 Foundation, Inc.}. This restricts our attention to certain kinds of
1912 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1913 once this assumption is granted examples and definitions need less
1916 @command{@value{AS}} is part of a team of programs that turn a high-level
1917 human-readable series of instructions into a low-level
1918 computer-readable series of instructions. Different versions of
1919 @command{@value{AS}} are used for different kinds of computer.
1922 @c There used to be a section "Terminology" here, which defined
1923 @c "contents", "byte", "word", and "long". Defining "word" to any
1924 @c particular size is confusing when the .word directive may generate 16
1925 @c bits on one machine and 32 bits on another; in general, for the user
1926 @c version of this manual, none of these terms seem essential to define.
1927 @c They were used very little even in the former draft of the manual;
1928 @c this draft makes an effort to avoid them (except in names of
1932 @section The GNU Assembler
1934 @c man begin DESCRIPTION
1936 @sc{gnu} @command{as} is really a family of assemblers.
1938 This manual describes @command{@value{AS}}, a member of that family which is
1939 configured for the @value{TARGET} architectures.
1941 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1942 should find a fairly similar environment when you use it on another
1943 architecture. Each version has much in common with the others,
1944 including object file formats, most assembler directives (often called
1945 @dfn{pseudo-ops}) and assembler syntax.@refill
1947 @cindex purpose of @sc{gnu} assembler
1948 @command{@value{AS}} is primarily intended to assemble the output of the
1949 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1950 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1951 assemble correctly everything that other assemblers for the same
1952 machine would assemble.
1954 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1957 @c This remark should appear in generic version of manual; assumption
1958 @c here is that generic version sets M680x0.
1959 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1960 assembler for the same architecture; for example, we know of several
1961 incompatible versions of 680x0 assembly language syntax.
1966 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1967 program in one pass of the source file. This has a subtle impact on the
1968 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1970 @node Object Formats
1971 @section Object File Formats
1973 @cindex object file format
1974 The @sc{gnu} assembler can be configured to produce several alternative
1975 object file formats. For the most part, this does not affect how you
1976 write assembly language programs; but directives for debugging symbols
1977 are typically different in different file formats. @xref{Symbol
1978 Attributes,,Symbol Attributes}.
1981 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1982 @value{OBJ-NAME} format object files.
1984 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1986 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1987 SOM or ELF format object files.
1992 @section Command Line
1994 @cindex command line conventions
1996 After the program name @command{@value{AS}}, the command line may contain
1997 options and file names. Options may appear in any order, and may be
1998 before, after, or between file names. The order of file names is
2001 @cindex standard input, as input file
2003 @file{--} (two hyphens) by itself names the standard input file
2004 explicitly, as one of the files for @command{@value{AS}} to assemble.
2006 @cindex options, command line
2007 Except for @samp{--} any command line argument that begins with a
2008 hyphen (@samp{-}) is an option. Each option changes the behavior of
2009 @command{@value{AS}}. No option changes the way another option works. An
2010 option is a @samp{-} followed by one or more letters; the case of
2011 the letter is important. All options are optional.
2013 Some options expect exactly one file name to follow them. The file
2014 name may either immediately follow the option's letter (compatible
2015 with older assemblers) or it may be the next command argument (@sc{gnu}
2016 standard). These two command lines are equivalent:
2019 @value{AS} -o my-object-file.o mumble.s
2020 @value{AS} -omy-object-file.o mumble.s
2024 @section Input Files
2027 @cindex source program
2028 @cindex files, input
2029 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
2030 describe the program input to one run of @command{@value{AS}}. The program may
2031 be in one or more files; how the source is partitioned into files
2032 doesn't change the meaning of the source.
2034 @c I added "con" prefix to "catenation" just to prove I can overcome my
2035 @c APL training... doc@cygnus.com
2036 The source program is a concatenation of the text in all the files, in the
2039 @c man begin DESCRIPTION
2040 Each time you run @command{@value{AS}} it assembles exactly one source
2041 program. The source program is made up of one or more files.
2042 (The standard input is also a file.)
2044 You give @command{@value{AS}} a command line that has zero or more input file
2045 names. The input files are read (from left file name to right). A
2046 command line argument (in any position) that has no special meaning
2047 is taken to be an input file name.
2049 If you give @command{@value{AS}} no file names it attempts to read one input file
2050 from the @command{@value{AS}} standard input, which is normally your terminal. You
2051 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
2054 Use @samp{--} if you need to explicitly name the standard input file
2055 in your command line.
2057 If the source is empty, @command{@value{AS}} produces a small, empty object
2062 @subheading Filenames and Line-numbers
2064 @cindex input file linenumbers
2065 @cindex line numbers, in input files
2066 There are two ways of locating a line in the input file (or files) and
2067 either may be used in reporting error messages. One way refers to a line
2068 number in a physical file; the other refers to a line number in a
2069 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
2071 @dfn{Physical files} are those files named in the command line given
2072 to @command{@value{AS}}.
2074 @dfn{Logical files} are simply names declared explicitly by assembler
2075 directives; they bear no relation to physical files. Logical file names help
2076 error messages reflect the original source file, when @command{@value{AS}} source
2077 is itself synthesized from other files. @command{@value{AS}} understands the
2078 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
2079 @ref{File,,@code{.file}}.
2082 @section Output (Object) File
2088 Every time you run @command{@value{AS}} it produces an output file, which is
2089 your assembly language program translated into numbers. This file
2090 is the object file. Its default name is @code{a.out}.
2091 You can give it another name by using the @option{-o} option. Conventionally,
2092 object file names end with @file{.o}. The default name is used for historical
2093 reasons: older assemblers were capable of assembling self-contained programs
2094 directly into a runnable program. (For some formats, this isn't currently
2095 possible, but it can be done for the @code{a.out} format.)
2099 The object file is meant for input to the linker @code{@value{LD}}. It contains
2100 assembled program code, information to help @code{@value{LD}} integrate
2101 the assembled program into a runnable file, and (optionally) symbolic
2102 information for the debugger.
2104 @c link above to some info file(s) like the description of a.out.
2105 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
2108 @section Error and Warning Messages
2110 @c man begin DESCRIPTION
2112 @cindex error messages
2113 @cindex warning messages
2114 @cindex messages from assembler
2115 @command{@value{AS}} may write warnings and error messages to the standard error
2116 file (usually your terminal). This should not happen when a compiler
2117 runs @command{@value{AS}} automatically. Warnings report an assumption made so
2118 that @command{@value{AS}} could keep assembling a flawed program; errors report a
2119 grave problem that stops the assembly.
2123 @cindex format of warning messages
2124 Warning messages have the format
2127 file_name:@b{NNN}:Warning Message Text
2131 @cindex file names and line numbers, in warnings/errors
2132 (where @b{NNN} is a line number). If both a logical file name
2133 (@pxref{File,,@code{.file}}) and a logical line number
2135 (@pxref{Line,,@code{.line}})
2137 have been given then they will be used, otherwise the file name and line number
2138 in the current assembler source file will be used. The message text is
2139 intended to be self explanatory (in the grand Unix tradition).
2141 Note the file name must be set via the logical version of the @code{.file}
2142 directive, not the DWARF2 version of the @code{.file} directive. For example:
2146 error_assembler_source
2152 produces this output:
2156 asm.s:2: Error: no such instruction: `error_assembler_source'
2157 foo.c:31: Error: no such instruction: `error_c_source'
2160 @cindex format of error messages
2161 Error messages have the format
2164 file_name:@b{NNN}:FATAL:Error Message Text
2167 The file name and line number are derived as for warning
2168 messages. The actual message text may be rather less explanatory
2169 because many of them aren't supposed to happen.
2172 @chapter Command-Line Options
2174 @cindex options, all versions of assembler
2175 This chapter describes command-line options available in @emph{all}
2176 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
2177 for options specific
2179 to the @value{TARGET} target.
2182 to particular machine architectures.
2185 @c man begin DESCRIPTION
2187 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2188 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2189 The assembler arguments must be separated from each other (and the @samp{-Wa})
2190 by commas. For example:
2193 gcc -c -g -O -Wa,-alh,-L file.c
2197 This passes two options to the assembler: @samp{-alh} (emit a listing to
2198 standard output with high-level and assembly source) and @samp{-L} (retain
2199 local symbols in the symbol table).
2201 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2202 command-line options are automatically passed to the assembler by the compiler.
2203 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2204 precisely what options it passes to each compilation pass, including the
2210 * a:: -a[cdghlns] enable listings
2211 * alternate:: --alternate enable alternate macro syntax
2212 * D:: -D for compatibility
2213 * f:: -f to work faster
2214 * I:: -I for .include search path
2215 @ifclear DIFF-TBL-KLUGE
2216 * K:: -K for compatibility
2218 @ifset DIFF-TBL-KLUGE
2219 * K:: -K for difference tables
2222 * L:: -L to retain local symbols
2223 * listing:: --listing-XXX to configure listing output
2224 * M:: -M or --mri to assemble in MRI compatibility mode
2225 * MD:: --MD for dependency tracking
2226 * no-pad-sections:: --no-pad-sections to stop section padding
2227 * o:: -o to name the object file
2228 * R:: -R to join data and text sections
2229 * statistics:: --statistics to see statistics about assembly
2230 * traditional-format:: --traditional-format for compatible output
2231 * v:: -v to announce version
2232 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2233 * Z:: -Z to make object file even after errors
2237 @section Enable Listings: @option{-a[cdghlns]}
2247 @cindex listings, enabling
2248 @cindex assembly listings, enabling
2250 These options enable listing output from the assembler. By itself,
2251 @samp{-a} requests high-level, assembly, and symbols listing.
2252 You can use other letters to select specific options for the list:
2253 @samp{-ah} requests a high-level language listing,
2254 @samp{-al} requests an output-program assembly listing, and
2255 @samp{-as} requests a symbol table listing.
2256 High-level listings require that a compiler debugging option like
2257 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2260 Use the @samp{-ag} option to print a first section with general assembly
2261 information, like @value{AS} version, switches passed, or time stamp.
2263 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2264 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2265 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2266 omitted from the listing.
2268 Use the @samp{-ad} option to omit debugging directives from the
2271 Once you have specified one of these options, you can further control
2272 listing output and its appearance using the directives @code{.list},
2273 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2275 The @samp{-an} option turns off all forms processing.
2276 If you do not request listing output with one of the @samp{-a} options, the
2277 listing-control directives have no effect.
2279 The letters after @samp{-a} may be combined into one option,
2280 @emph{e.g.}, @samp{-aln}.
2282 Note if the assembler source is coming from the standard input (e.g.,
2284 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
2285 is being used) then the listing will not contain any comments or preprocessor
2286 directives. This is because the listing code buffers input source lines from
2287 stdin only after they have been preprocessed by the assembler. This reduces
2288 memory usage and makes the code more efficient.
2291 @section @option{--alternate}
2294 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2297 @section @option{-D}
2300 This option has no effect whatsoever, but it is accepted to make it more
2301 likely that scripts written for other assemblers also work with
2302 @command{@value{AS}}.
2305 @section Work Faster: @option{-f}
2308 @cindex trusted compiler
2309 @cindex faster processing (@option{-f})
2310 @samp{-f} should only be used when assembling programs written by a
2311 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2312 and comment preprocessing on
2313 the input file(s) before assembling them. @xref{Preprocessing,
2317 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2318 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2323 @section @code{.include} Search Path: @option{-I} @var{path}
2325 @kindex -I @var{path}
2326 @cindex paths for @code{.include}
2327 @cindex search path for @code{.include}
2328 @cindex @code{include} directive search path
2329 Use this option to add a @var{path} to the list of directories
2330 @command{@value{AS}} searches for files specified in @code{.include}
2331 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2332 many times as necessary to include a variety of paths. The current
2333 working directory is always searched first; after that, @command{@value{AS}}
2334 searches any @samp{-I} directories in the same order as they were
2335 specified (left to right) on the command line.
2338 @section Difference Tables: @option{-K}
2341 @ifclear DIFF-TBL-KLUGE
2342 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2343 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2344 where it can be used to warn when the assembler alters the machine code
2345 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2346 family does not have the addressing limitations that sometimes lead to this
2347 alteration on other platforms.
2350 @ifset DIFF-TBL-KLUGE
2351 @cindex difference tables, warning
2352 @cindex warning for altered difference tables
2353 @command{@value{AS}} sometimes alters the code emitted for directives of the
2354 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2355 You can use the @samp{-K} option if you want a warning issued when this
2360 @section Include Local Symbols: @option{-L}
2363 @cindex local symbols, retaining in output
2364 Symbols beginning with system-specific local label prefixes, typically
2365 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2366 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2367 such symbols when debugging, because they are intended for the use of
2368 programs (like compilers) that compose assembler programs, not for your
2369 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2370 such symbols, so you do not normally debug with them.
2372 This option tells @command{@value{AS}} to retain those local symbols
2373 in the object file. Usually if you do this you also tell the linker
2374 @code{@value{LD}} to preserve those symbols.
2377 @section Configuring listing output: @option{--listing}
2379 The listing feature of the assembler can be enabled via the command line switch
2380 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2381 hex dump of the corresponding locations in the output object file, and displays
2382 them as a listing file. The format of this listing can be controlled by
2383 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2384 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2385 @code{.psize} (@pxref{Psize}), and
2386 @code{.eject} (@pxref{Eject}) and also by the following switches:
2389 @item --listing-lhs-width=@samp{number}
2390 @kindex --listing-lhs-width
2391 @cindex Width of first line disassembly output
2392 Sets the maximum width, in words, of the first line of the hex byte dump. This
2393 dump appears on the left hand side of the listing output.
2395 @item --listing-lhs-width2=@samp{number}
2396 @kindex --listing-lhs-width2
2397 @cindex Width of continuation lines of disassembly output
2398 Sets the maximum width, in words, of any further lines of the hex byte dump for
2399 a given input source line. If this value is not specified, it defaults to being
2400 the same as the value specified for @samp{--listing-lhs-width}. If neither
2401 switch is used the default is to one.
2403 @item --listing-rhs-width=@samp{number}
2404 @kindex --listing-rhs-width
2405 @cindex Width of source line output
2406 Sets the maximum width, in characters, of the source line that is displayed
2407 alongside the hex dump. The default value for this parameter is 100. The
2408 source line is displayed on the right hand side of the listing output.
2410 @item --listing-cont-lines=@samp{number}
2411 @kindex --listing-cont-lines
2412 @cindex Maximum number of continuation lines
2413 Sets the maximum number of continuation lines of hex dump that will be
2414 displayed for a given single line of source input. The default value is 4.
2418 @section Assemble in MRI Compatibility Mode: @option{-M}
2421 @cindex MRI compatibility mode
2422 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2423 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2424 compatible with the @code{ASM68K} assembler from Microtec Research.
2425 The exact nature of the
2426 MRI syntax will not be documented here; see the MRI manuals for more
2427 information. Note in particular that the handling of macros and macro
2428 arguments is somewhat different. The purpose of this option is to permit
2429 assembling existing MRI assembler code using @command{@value{AS}}.
2431 The MRI compatibility is not complete. Certain operations of the MRI assembler
2432 depend upon its object file format, and can not be supported using other object
2433 file formats. Supporting these would require enhancing each object file format
2434 individually. These are:
2437 @item global symbols in common section
2439 The m68k MRI assembler supports common sections which are merged by the linker.
2440 Other object file formats do not support this. @command{@value{AS}} handles
2441 common sections by treating them as a single common symbol. It permits local
2442 symbols to be defined within a common section, but it can not support global
2443 symbols, since it has no way to describe them.
2445 @item complex relocations
2447 The MRI assemblers support relocations against a negated section address, and
2448 relocations which combine the start addresses of two or more sections. These
2449 are not support by other object file formats.
2451 @item @code{END} pseudo-op specifying start address
2453 The MRI @code{END} pseudo-op permits the specification of a start address.
2454 This is not supported by other object file formats. The start address may
2455 instead be specified using the @option{-e} option to the linker, or in a linker
2458 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2460 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2461 name to the output file. This is not supported by other object file formats.
2463 @item @code{ORG} pseudo-op
2465 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2466 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2467 which changes the location within the current section. Absolute sections are
2468 not supported by other object file formats. The address of a section may be
2469 assigned within a linker script.
2472 There are some other features of the MRI assembler which are not supported by
2473 @command{@value{AS}}, typically either because they are difficult or because they
2474 seem of little consequence. Some of these may be supported in future releases.
2478 @item EBCDIC strings
2480 EBCDIC strings are not supported.
2482 @item packed binary coded decimal
2484 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2485 and @code{DCB.P} pseudo-ops are not supported.
2487 @item @code{FEQU} pseudo-op
2489 The m68k @code{FEQU} pseudo-op is not supported.
2491 @item @code{NOOBJ} pseudo-op
2493 The m68k @code{NOOBJ} pseudo-op is not supported.
2495 @item @code{OPT} branch control options
2497 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2498 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2499 relaxes all branches, whether forward or backward, to an appropriate size, so
2500 these options serve no purpose.
2502 @item @code{OPT} list control options
2504 The following m68k @code{OPT} list control options are ignored: @code{C},
2505 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2506 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2508 @item other @code{OPT} options
2510 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2511 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2513 @item @code{OPT} @code{D} option is default
2515 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2516 @code{OPT NOD} may be used to turn it off.
2518 @item @code{XREF} pseudo-op.
2520 The m68k @code{XREF} pseudo-op is ignored.
2525 @section Dependency Tracking: @option{--MD}
2528 @cindex dependency tracking
2531 @command{@value{AS}} can generate a dependency file for the file it creates. This
2532 file consists of a single rule suitable for @code{make} describing the
2533 dependencies of the main source file.
2535 The rule is written to the file named in its argument.
2537 This feature is used in the automatic updating of makefiles.
2539 @node no-pad-sections
2540 @section Output Section Padding
2541 @kindex --no-pad-sections
2542 @cindex output section padding
2543 Normally the assembler will pad the end of each output section up to its
2544 alignment boundary. But this can waste space, which can be significant on
2545 memory constrained targets. So the @option{--no-pad-sections} option will
2546 disable this behaviour.
2549 @section Name the Object File: @option{-o}
2552 @cindex naming object file
2553 @cindex object file name
2554 There is always one object file output when you run @command{@value{AS}}. By
2555 default it has the name @file{a.out}.
2556 You use this option (which takes exactly one filename) to give the
2557 object file a different name.
2559 Whatever the object file is called, @command{@value{AS}} overwrites any
2560 existing file of the same name.
2563 @section Join Data and Text Sections: @option{-R}
2566 @cindex data and text sections, joining
2567 @cindex text and data sections, joining
2568 @cindex joining text and data sections
2569 @cindex merging text and data sections
2570 @option{-R} tells @command{@value{AS}} to write the object file as if all
2571 data-section data lives in the text section. This is only done at
2572 the very last moment: your binary data are the same, but data
2573 section parts are relocated differently. The data section part of
2574 your object file is zero bytes long because all its bytes are
2575 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2577 When you specify @option{-R} it would be possible to generate shorter
2578 address displacements (because we do not have to cross between text and
2579 data section). We refrain from doing this simply for compatibility with
2580 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2583 When @command{@value{AS}} is configured for COFF or ELF output,
2584 this option is only useful if you use sections named @samp{.text} and
2589 @option{-R} is not supported for any of the HPPA targets. Using
2590 @option{-R} generates a warning from @command{@value{AS}}.
2594 @section Display Assembly Statistics: @option{--statistics}
2596 @kindex --statistics
2597 @cindex statistics, about assembly
2598 @cindex time, total for assembly
2599 @cindex space used, maximum for assembly
2600 Use @samp{--statistics} to display two statistics about the resources used by
2601 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2602 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2605 @node traditional-format
2606 @section Compatible Output: @option{--traditional-format}
2608 @kindex --traditional-format
2609 For some targets, the output of @command{@value{AS}} is different in some ways
2610 from the output of some existing assembler. This switch requests
2611 @command{@value{AS}} to use the traditional format instead.
2613 For example, it disables the exception frame optimizations which
2614 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2617 @section Announce Version: @option{-v}
2621 @cindex assembler version
2622 @cindex version of assembler
2623 You can find out what version of as is running by including the
2624 option @samp{-v} (which you can also spell as @samp{-version}) on the
2628 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2630 @command{@value{AS}} should never give a warning or error message when
2631 assembling compiler output. But programs written by people often
2632 cause @command{@value{AS}} to give a warning that a particular assumption was
2633 made. All such warnings are directed to the standard error file.
2637 @cindex suppressing warnings
2638 @cindex warnings, suppressing
2639 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2640 This only affects the warning messages: it does not change any particular of
2641 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2644 @kindex --fatal-warnings
2645 @cindex errors, caused by warnings
2646 @cindex warnings, causing error
2647 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2648 files that generate warnings to be in error.
2651 @cindex warnings, switching on
2652 You can switch these options off again by specifying @option{--warn}, which
2653 causes warnings to be output as usual.
2656 @section Generate Object File in Spite of Errors: @option{-Z}
2657 @cindex object file, after errors
2658 @cindex errors, continuing after
2659 After an error message, @command{@value{AS}} normally produces no output. If for
2660 some reason you are interested in object file output even after
2661 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2662 option. If there are any errors, @command{@value{AS}} continues anyways, and
2663 writes an object file after a final warning message of the form @samp{@var{n}
2664 errors, @var{m} warnings, generating bad object file.}
2669 @cindex machine-independent syntax
2670 @cindex syntax, machine-independent
2671 This chapter describes the machine-independent syntax allowed in a
2672 source file. @command{@value{AS}} syntax is similar to what many other
2673 assemblers use; it is inspired by the BSD 4.2
2678 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2682 * Preprocessing:: Preprocessing
2683 * Whitespace:: Whitespace
2684 * Comments:: Comments
2685 * Symbol Intro:: Symbols
2686 * Statements:: Statements
2687 * Constants:: Constants
2691 @section Preprocessing
2693 @cindex preprocessing
2694 The @command{@value{AS}} internal preprocessor:
2696 @cindex whitespace, removed by preprocessor
2698 adjusts and removes extra whitespace. It leaves one space or tab before
2699 the keywords on a line, and turns any other whitespace on the line into
2702 @cindex comments, removed by preprocessor
2704 removes all comments, replacing them with a single space, or an
2705 appropriate number of newlines.
2707 @cindex constants, converted by preprocessor
2709 converts character constants into the appropriate numeric values.
2712 It does not do macro processing, include file handling, or
2713 anything else you may get from your C compiler's preprocessor. You can
2714 do include file processing with the @code{.include} directive
2715 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2716 to get other ``CPP'' style preprocessing by giving the input file a
2717 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2718 Output, gcc info, Using GNU CC}.
2720 Excess whitespace, comments, and character constants
2721 cannot be used in the portions of the input text that are not
2724 @cindex turning preprocessing on and off
2725 @cindex preprocessing, turning on and off
2728 If the first line of an input file is @code{#NO_APP} or if you use the
2729 @samp{-f} option, whitespace and comments are not removed from the input file.
2730 Within an input file, you can ask for whitespace and comment removal in
2731 specific portions of the by putting a line that says @code{#APP} before the
2732 text that may contain whitespace or comments, and putting a line that says
2733 @code{#NO_APP} after this text. This feature is mainly intend to support
2734 @code{asm} statements in compilers whose output is otherwise free of comments
2741 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2742 Whitespace is used to separate symbols, and to make programs neater for
2743 people to read. Unless within character constants
2744 (@pxref{Characters,,Character Constants}), any whitespace means the same
2745 as exactly one space.
2751 There are two ways of rendering comments to @command{@value{AS}}. In both
2752 cases the comment is equivalent to one space.
2754 Anything from @samp{/*} through the next @samp{*/} is a comment.
2755 This means you may not nest these comments.
2759 The only way to include a newline ('\n') in a comment
2760 is to use this sort of comment.
2763 /* This sort of comment does not nest. */
2766 @cindex line comment character
2767 Anything from a @dfn{line comment} character up to the next newline is
2768 considered a comment and is ignored. The line comment character is target
2769 specific, and some targets multiple comment characters. Some targets also have
2770 line comment characters that only work if they are the first character on a
2771 line. Some targets use a sequence of two characters to introduce a line
2772 comment. Some targets can also change their line comment characters depending
2773 upon command line options that have been used. For more details see the
2774 @emph{Syntax} section in the documentation for individual targets.
2776 If the line comment character is the hash sign (@samp{#}) then it still has the
2777 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2778 to specify logical line numbers:
2781 @cindex lines starting with @code{#}
2782 @cindex logical line numbers
2783 To be compatible with past assemblers, lines that begin with @samp{#} have a
2784 special interpretation. Following the @samp{#} should be an absolute
2785 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2786 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2787 new logical file name. The rest of the line, if any, should be whitespace.
2789 If the first non-whitespace characters on the line are not numeric,
2790 the line is ignored. (Just like a comment.)
2793 # This is an ordinary comment.
2794 # 42-6 "new_file_name" # New logical file name
2795 # This is logical line # 36.
2797 This feature is deprecated, and may disappear from future versions
2798 of @command{@value{AS}}.
2803 @cindex characters used in symbols
2804 @ifclear SPECIAL-SYMS
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
2812 A @dfn{symbol} is one or more characters chosen from the set of all
2813 letters (both upper and lower case), digits and the three characters
2814 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2820 On most machines, you can also use @code{$} in symbol names; exceptions
2821 are noted in @ref{Machine Dependencies}.
2823 No symbol may begin with a digit. Case is significant.
2824 There is no length limit; all characters are significant. Multibyte characters
2825 are supported. Symbols are delimited by characters not in that set, or by the
2826 beginning of a file (since the source program must end with a newline, the end
2827 of a file is not a possible symbol delimiter). @xref{Symbols}.
2829 Symbol names may also be enclosed in double quote @code{"} characters. In such
2830 cases any characters are allowed, except for the NUL character. If a double
2831 quote character is to be included in the symbol name it must be preceeded by a
2832 backslash @code{\} character.
2833 @cindex length of symbols
2838 @cindex statements, structure of
2839 @cindex line separator character
2840 @cindex statement separator character
2842 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2843 @dfn{line separator character}. The line separator character is target
2844 specific and described in the @emph{Syntax} section of each
2845 target's documentation. Not all targets support a line separator character.
2846 The newline or line separator character is considered to be part of the
2847 preceding statement. Newlines and separators within character constants are an
2848 exception: they do not end statements.
2850 @cindex newline, required at file end
2851 @cindex EOF, newline must precede
2852 It is an error to end any statement with end-of-file: the last
2853 character of any input file should be a newline.@refill
2855 An empty statement is allowed, and may include whitespace. It is ignored.
2857 @cindex instructions and directives
2858 @cindex directives and instructions
2859 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2860 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2862 A statement begins with zero or more labels, optionally followed by a
2863 key symbol which determines what kind of statement it is. The key
2864 symbol determines the syntax of the rest of the statement. If the
2865 symbol begins with a dot @samp{.} then the statement is an assembler
2866 directive: typically valid for any computer. If the symbol begins with
2867 a letter the statement is an assembly language @dfn{instruction}: it
2868 assembles into a machine language instruction.
2870 Different versions of @command{@value{AS}} for different computers
2871 recognize different instructions. In fact, the same symbol may
2872 represent a different instruction in a different computer's assembly
2876 @cindex @code{:} (label)
2877 @cindex label (@code{:})
2878 A label is a symbol immediately followed by a colon (@code{:}).
2879 Whitespace before a label or after a colon is permitted, but you may not
2880 have whitespace between a label's symbol and its colon. @xref{Labels}.
2883 For HPPA targets, labels need not be immediately followed by a colon, but
2884 the definition of a label must begin in column zero. This also implies that
2885 only one label may be defined on each line.
2889 label: .directive followed by something
2890 another_label: # This is an empty statement.
2891 instruction operand_1, operand_2, @dots{}
2898 A constant is a number, written so that its value is known by
2899 inspection, without knowing any context. Like this:
2902 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2903 .ascii "Ring the bell\7" # A string constant.
2904 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2905 .float 0f-314159265358979323846264338327\
2906 95028841971.693993751E-40 # - pi, a flonum.
2911 * Characters:: Character Constants
2912 * Numbers:: Number Constants
2916 @subsection Character Constants
2918 @cindex character constants
2919 @cindex constants, character
2920 There are two kinds of character constants. A @dfn{character} stands
2921 for one character in one byte and its value may be used in
2922 numeric expressions. String constants (properly called string
2923 @emph{literals}) are potentially many bytes and their values may not be
2924 used in arithmetic expressions.
2928 * Chars:: Characters
2932 @subsubsection Strings
2934 @cindex string constants
2935 @cindex constants, string
2936 A @dfn{string} is written between double-quotes. It may contain
2937 double-quotes or null characters. The way to get special characters
2938 into a string is to @dfn{escape} these characters: precede them with
2939 a backslash @samp{\} character. For example @samp{\\} represents
2940 one backslash: the first @code{\} is an escape which tells
2941 @command{@value{AS}} to interpret the second character literally as a backslash
2942 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2943 escape character). The complete list of escapes follows.
2945 @cindex escape codes, character
2946 @cindex character escape codes
2947 @c NOTE: Cindex entries must not start with a backlash character.
2948 @c NOTE: This confuses the pdf2texi script when it is creating the
2949 @c NOTE: index based upon the first character and so it generates:
2950 @c NOTE: \initial {\\}
2951 @c NOTE: which then results in the error message:
2952 @c NOTE: Argument of \\ has an extra }.
2953 @c NOTE: So in the index entries below a space character has been
2954 @c NOTE: prepended to avoid this problem.
2957 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2959 @cindex @code{ \b} (backspace character)
2960 @cindex backspace (@code{\b})
2962 Mnemonic for backspace; for ASCII this is octal code 010.
2965 @c Mnemonic for EOText; for ASCII this is octal code 004.
2967 @cindex @code{ \f} (formfeed character)
2968 @cindex formfeed (@code{\f})
2970 Mnemonic for FormFeed; for ASCII this is octal code 014.
2972 @cindex @code{ \n} (newline character)
2973 @cindex newline (@code{\n})
2975 Mnemonic for newline; for ASCII this is octal code 012.
2978 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2980 @cindex @code{ \r} (carriage return character)
2981 @cindex carriage return (@code{backslash-r})
2983 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2986 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2987 @c other assemblers.
2989 @cindex @code{ \t} (tab)
2990 @cindex tab (@code{\t})
2992 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2995 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2996 @c @item \x @var{digit} @var{digit} @var{digit}
2997 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2999 @cindex @code{ \@var{ddd}} (octal character code)
3000 @cindex octal character code (@code{\@var{ddd}})
3001 @item \ @var{digit} @var{digit} @var{digit}
3002 An octal character code. The numeric code is 3 octal digits.
3003 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
3004 for example, @code{\008} has the value 010, and @code{\009} the value 011.
3006 @cindex @code{ \@var{xd...}} (hex character code)
3007 @cindex hex character code (@code{\@var{xd...}})
3008 @item \@code{x} @var{hex-digits...}
3009 A hex character code. All trailing hex digits are combined. Either upper or
3010 lower case @code{x} works.
3012 @cindex @code{ \\} (@samp{\} character)
3013 @cindex backslash (@code{\\})
3015 Represents one @samp{\} character.
3018 @c Represents one @samp{'} (accent acute) character.
3019 @c This is needed in single character literals
3020 @c (@xref{Characters,,Character Constants}.) to represent
3023 @cindex @code{ \"} (doublequote character)
3024 @cindex doublequote (@code{\"})
3026 Represents one @samp{"} character. Needed in strings to represent
3027 this character, because an unescaped @samp{"} would end the string.
3029 @item \ @var{anything-else}
3030 Any other character when escaped by @kbd{\} gives a warning, but
3031 assembles as if the @samp{\} was not present. The idea is that if
3032 you used an escape sequence you clearly didn't want the literal
3033 interpretation of the following character. However @command{@value{AS}} has no
3034 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
3035 code and warns you of the fact.
3038 Which characters are escapable, and what those escapes represent,
3039 varies widely among assemblers. The current set is what we think
3040 the BSD 4.2 assembler recognizes, and is a subset of what most C
3041 compilers recognize. If you are in doubt, do not use an escape
3045 @subsubsection Characters
3047 @cindex single character constant
3048 @cindex character, single
3049 @cindex constant, single character
3050 A single character may be written as a single quote immediately followed by
3051 that character. Some backslash escapes apply to characters, @code{\b},
3052 @code{\f}, @code{\n}, @code{\r}, @code{\t}, and @code{\"} with the same meaning
3053 as for strings, plus @code{\'} for a single quote. So if you want to write the
3054 character backslash, you must write @kbd{'\\} where the first @code{\} escapes
3055 the second @code{\}. As you can see, the quote is an acute accent, not a grave
3058 @ifclear abnormal-separator
3059 (or semicolon @samp{;})
3061 @ifset abnormal-separator
3063 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
3068 immediately following an acute accent is taken as a literal character
3069 and does not count as the end of a statement. The value of a character
3070 constant in a numeric expression is the machine's byte-wide code for
3071 that character. @command{@value{AS}} assumes your character code is ASCII:
3072 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
3075 @subsection Number Constants
3077 @cindex constants, number
3078 @cindex number constants
3079 @command{@value{AS}} distinguishes three kinds of numbers according to how they
3080 are stored in the target machine. @emph{Integers} are numbers that
3081 would fit into an @code{int} in the C language. @emph{Bignums} are
3082 integers, but they are stored in more than 32 bits. @emph{Flonums}
3083 are floating point numbers, described below.
3086 * Integers:: Integers
3094 @subsubsection Integers
3096 @cindex constants, integer
3098 @cindex binary integers
3099 @cindex integers, binary
3100 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
3101 the binary digits @samp{01}.
3103 @cindex octal integers
3104 @cindex integers, octal
3105 An octal integer is @samp{0} followed by zero or more of the octal
3106 digits (@samp{01234567}).
3108 @cindex decimal integers
3109 @cindex integers, decimal
3110 A decimal integer starts with a non-zero digit followed by zero or
3111 more digits (@samp{0123456789}).
3113 @cindex hexadecimal integers
3114 @cindex integers, hexadecimal
3115 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
3116 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
3118 Integers have the usual values. To denote a negative integer, use
3119 the prefix operator @samp{-} discussed under expressions
3120 (@pxref{Prefix Ops,,Prefix Operators}).
3123 @subsubsection Bignums
3126 @cindex constants, bignum
3127 A @dfn{bignum} has the same syntax and semantics as an integer
3128 except that the number (or its negative) takes more than 32 bits to
3129 represent in binary. The distinction is made because in some places
3130 integers are permitted while bignums are not.
3133 @subsubsection Flonums
3135 @cindex floating point numbers
3136 @cindex constants, floating point
3138 @cindex precision, floating point
3139 A @dfn{flonum} represents a floating point number. The translation is
3140 indirect: a decimal floating point number from the text is converted by
3141 @command{@value{AS}} to a generic binary floating point number of more than
3142 sufficient precision. This generic floating point number is converted
3143 to a particular computer's floating point format (or formats) by a
3144 portion of @command{@value{AS}} specialized to that computer.
3146 A flonum is written by writing (in order)
3151 (@samp{0} is optional on the HPPA.)
3155 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
3157 @kbd{e} is recommended. Case is not important.
3159 @c FIXME: verify if flonum syntax really this vague for most cases
3160 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
3161 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
3164 On the H8/300 and Renesas / SuperH SH architectures, the letter must be
3165 one of the letters @samp{DFPRSX} (in upper or lower case).
3167 On the ARC, the letter must be one of the letters @samp{DFRS}
3168 (in upper or lower case).
3170 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3174 One of the letters @samp{DFRS} (in upper or lower case).
3177 One of the letters @samp{DFPRSX} (in upper or lower case).
3180 The letter @samp{E} (upper case only).
3185 An optional sign: either @samp{+} or @samp{-}.
3188 An optional @dfn{integer part}: zero or more decimal digits.
3191 An optional @dfn{fractional part}: @samp{.} followed by zero
3192 or more decimal digits.
3195 An optional exponent, consisting of:
3199 An @samp{E} or @samp{e}.
3200 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3201 @c principle this can perfectly well be different on different targets.
3203 Optional sign: either @samp{+} or @samp{-}.
3205 One or more decimal digits.
3210 At least one of the integer part or the fractional part must be
3211 present. The floating point number has the usual base-10 value.
3213 @command{@value{AS}} does all processing using integers. Flonums are computed
3214 independently of any floating point hardware in the computer running
3215 @command{@value{AS}}.
3218 @chapter Sections and Relocation
3223 * Secs Background:: Background
3224 * Ld Sections:: Linker Sections
3225 * As Sections:: Assembler Internal Sections
3226 * Sub-Sections:: Sub-Sections
3230 @node Secs Background
3233 Roughly, a section is a range of addresses, with no gaps; all data
3234 ``in'' those addresses is treated the same for some particular purpose.
3235 For example there may be a ``read only'' section.
3237 @cindex linker, and assembler
3238 @cindex assembler, and linker
3239 The linker @code{@value{LD}} reads many object files (partial programs) and
3240 combines their contents to form a runnable program. When @command{@value{AS}}
3241 emits an object file, the partial program is assumed to start at address 0.
3242 @code{@value{LD}} assigns the final addresses for the partial program, so that
3243 different partial programs do not overlap. This is actually an
3244 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3247 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3248 addresses. These blocks slide to their run-time addresses as rigid
3249 units; their length does not change and neither does the order of bytes
3250 within them. Such a rigid unit is called a @emph{section}. Assigning
3251 run-time addresses to sections is called @dfn{relocation}. It includes
3252 the task of adjusting mentions of object-file addresses so they refer to
3253 the proper run-time addresses.
3255 For the H8/300, and for the Renesas / SuperH SH,
3256 @command{@value{AS}} pads sections if needed to
3257 ensure they end on a word (sixteen bit) boundary.
3260 @cindex standard assembler sections
3261 An object file written by @command{@value{AS}} has at least three sections, any
3262 of which may be empty. These are named @dfn{text}, @dfn{data} and
3267 When it generates COFF or ELF output,
3269 @command{@value{AS}} can also generate whatever other named sections you specify
3270 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3271 If you do not use any directives that place output in the @samp{.text}
3272 or @samp{.data} sections, these sections still exist, but are empty.
3277 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3279 @command{@value{AS}} can also generate whatever other named sections you
3280 specify using the @samp{.space} and @samp{.subspace} directives. See
3281 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3282 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3283 assembler directives.
3286 Additionally, @command{@value{AS}} uses different names for the standard
3287 text, data, and bss sections when generating SOM output. Program text
3288 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3289 BSS into @samp{$BSS$}.
3293 Within the object file, the text section starts at address @code{0}, the
3294 data section follows, and the bss section follows the data section.
3297 When generating either SOM or ELF output files on the HPPA, the text
3298 section starts at address @code{0}, the data section at address
3299 @code{0x4000000}, and the bss section follows the data section.
3302 To let @code{@value{LD}} know which data changes when the sections are
3303 relocated, and how to change that data, @command{@value{AS}} also writes to the
3304 object file details of the relocation needed. To perform relocation
3305 @code{@value{LD}} must know, each time an address in the object
3309 Where in the object file is the beginning of this reference to
3312 How long (in bytes) is this reference?
3314 Which section does the address refer to? What is the numeric value of
3316 (@var{address}) @minus{} (@var{start-address of section})?
3319 Is the reference to an address ``Program-Counter relative''?
3322 @cindex addresses, format of
3323 @cindex section-relative addressing
3324 In fact, every address @command{@value{AS}} ever uses is expressed as
3326 (@var{section}) + (@var{offset into section})
3329 Further, most expressions @command{@value{AS}} computes have this section-relative
3332 (For some object formats, such as SOM for the HPPA, some expressions are
3333 symbol-relative instead.)
3336 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3337 @var{N} into section @var{secname}.''
3339 Apart from text, data and bss sections you need to know about the
3340 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3341 addresses in the absolute section remain unchanged. For example, address
3342 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3343 @code{@value{LD}}. Although the linker never arranges two partial programs'
3344 data sections with overlapping addresses after linking, @emph{by definition}
3345 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3346 part of a program is always the same address when the program is running as
3347 address @code{@{absolute@ 239@}} in any other part of the program.
3349 The idea of sections is extended to the @dfn{undefined} section. Any
3350 address whose section is unknown at assembly time is by definition
3351 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3352 Since numbers are always defined, the only way to generate an undefined
3353 address is to mention an undefined symbol. A reference to a named
3354 common block would be such a symbol: its value is unknown at assembly
3355 time so it has section @emph{undefined}.
3357 By analogy the word @emph{section} is used to describe groups of sections in
3358 the linked program. @code{@value{LD}} puts all partial programs' text
3359 sections in contiguous addresses in the linked program. It is
3360 customary to refer to the @emph{text section} of a program, meaning all
3361 the addresses of all partial programs' text sections. Likewise for
3362 data and bss sections.
3364 Some sections are manipulated by @code{@value{LD}}; others are invented for
3365 use of @command{@value{AS}} and have no meaning except during assembly.
3368 @section Linker Sections
3369 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3374 @cindex named sections
3375 @cindex sections, named
3376 @item named sections
3379 @cindex text section
3380 @cindex data section
3384 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3385 separate but equal sections. Anything you can say of one section is
3388 When the program is running, however, it is
3389 customary for the text section to be unalterable. The
3390 text section is often shared among processes: it contains
3391 instructions, constants and the like. The data section of a running
3392 program is usually alterable: for example, C variables would be stored
3393 in the data section.
3398 This section contains zeroed bytes when your program begins running. It
3399 is used to hold uninitialized variables or common storage. The length of
3400 each partial program's bss section is important, but because it starts
3401 out containing zeroed bytes there is no need to store explicit zero
3402 bytes in the object file. The bss section was invented to eliminate
3403 those explicit zeros from object files.
3405 @cindex absolute section
3406 @item absolute section
3407 Address 0 of this section is always ``relocated'' to runtime address 0.
3408 This is useful if you want to refer to an address that @code{@value{LD}} must
3409 not change when relocating. In this sense we speak of absolute
3410 addresses being ``unrelocatable'': they do not change during relocation.
3412 @cindex undefined section
3413 @item undefined section
3414 This ``section'' is a catch-all for address references to objects not in
3415 the preceding sections.
3416 @c FIXME: ref to some other doc on obj-file formats could go here.
3419 @cindex relocation example
3420 An idealized example of three relocatable sections follows.
3422 The example uses the traditional section names @samp{.text} and @samp{.data}.
3424 Memory addresses are on the horizontal axis.
3428 @c END TEXI2ROFF-KILL
3431 partial program # 1: |ttttt|dddd|00|
3438 partial program # 2: |TTT|DDD|000|
3441 +--+---+-----+--+----+---+-----+~~
3442 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3443 +--+---+-----+--+----+---+-----+~~
3445 addresses: 0 @dots{}
3452 \line{\it Partial program \#1: \hfil}
3453 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3454 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3456 \line{\it Partial program \#2: \hfil}
3457 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3458 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3460 \line{\it linked program: \hfil}
3461 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3462 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3463 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3464 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3466 \line{\it addresses: \hfil}
3470 @c END TEXI2ROFF-KILL
3473 @section Assembler Internal Sections
3475 @cindex internal assembler sections
3476 @cindex sections in messages, internal
3477 These sections are meant only for the internal use of @command{@value{AS}}. They
3478 have no meaning at run-time. You do not really need to know about these
3479 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3480 warning messages, so it might be helpful to have an idea of their
3481 meanings to @command{@value{AS}}. These sections are used to permit the
3482 value of every expression in your assembly language program to be a
3483 section-relative address.
3486 @cindex assembler internal logic error
3487 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3488 An internal assembler logic error has been found. This means there is a
3489 bug in the assembler.
3491 @cindex expr (internal section)
3493 The assembler stores complex expression internally as combinations of
3494 symbols. When it needs to represent an expression as a symbol, it puts
3495 it in the expr section.
3497 @c FIXME item transfer[t] vector preload
3498 @c FIXME item transfer[t] vector postload
3499 @c FIXME item register
3503 @section Sub-Sections
3505 @cindex numbered subsections
3506 @cindex grouping data
3512 fall into two sections: text and data.
3514 You may have separate groups of
3516 data in named sections
3520 data in named sections
3526 that you want to end up near to each other in the object file, even though they
3527 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3528 use @dfn{subsections} for this purpose. Within each section, there can be
3529 numbered subsections with values from 0 to 8192. Objects assembled into the
3530 same subsection go into the object file together with other objects in the same
3531 subsection. For example, a compiler might want to store constants in the text
3532 section, but might not want to have them interspersed with the program being
3533 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3534 section of code being output, and a @samp{.text 1} before each group of
3535 constants being output.
3537 Subsections are optional. If you do not use subsections, everything
3538 goes in subsection number zero.
3541 Each subsection is zero-padded up to a multiple of four bytes.
3542 (Subsections may be padded a different amount on different flavors
3543 of @command{@value{AS}}.)
3547 On the H8/300 platform, each subsection is zero-padded to a word
3548 boundary (two bytes).
3549 The same is true on the Renesas SH.
3553 Subsections appear in your object file in numeric order, lowest numbered
3554 to highest. (All this to be compatible with other people's assemblers.)
3555 The object file contains no representation of subsections; @code{@value{LD}} and
3556 other programs that manipulate object files see no trace of them.
3557 They just see all your text subsections as a text section, and all your
3558 data subsections as a data section.
3560 To specify which subsection you want subsequent statements assembled
3561 into, use a numeric argument to specify it, in a @samp{.text
3562 @var{expression}} or a @samp{.data @var{expression}} statement.
3565 When generating COFF output, you
3570 can also use an extra subsection
3571 argument with arbitrary named sections: @samp{.section @var{name},
3576 When generating ELF output, you
3581 can also use the @code{.subsection} directive (@pxref{SubSection})
3582 to specify a subsection: @samp{.subsection @var{expression}}.
3584 @var{Expression} should be an absolute expression
3585 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3586 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3587 begins in @code{text 0}. For instance:
3589 .text 0 # The default subsection is text 0 anyway.
3590 .ascii "This lives in the first text subsection. *"
3592 .ascii "But this lives in the second text subsection."
3594 .ascii "This lives in the data section,"
3595 .ascii "in the first data subsection."
3597 .ascii "This lives in the first text section,"
3598 .ascii "immediately following the asterisk (*)."
3601 Each section has a @dfn{location counter} incremented by one for every byte
3602 assembled into that section. Because subsections are merely a convenience
3603 restricted to @command{@value{AS}} there is no concept of a subsection location
3604 counter. There is no way to directly manipulate a location counter---but the
3605 @code{.align} directive changes it, and any label definition captures its
3606 current value. The location counter of the section where statements are being
3607 assembled is said to be the @dfn{active} location counter.
3610 @section bss Section
3613 @cindex common variable storage
3614 The bss section is used for local common variable storage.
3615 You may allocate address space in the bss section, but you may
3616 not dictate data to load into it before your program executes. When
3617 your program starts running, all the contents of the bss
3618 section are zeroed bytes.
3620 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3621 @ref{Lcomm,,@code{.lcomm}}.
3623 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3624 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3627 When assembling for a target which supports multiple sections, such as ELF or
3628 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3629 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3630 section. Typically the section will only contain symbol definitions and
3631 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3638 Symbols are a central concept: the programmer uses symbols to name
3639 things, the linker uses symbols to link, and the debugger uses symbols
3643 @cindex debuggers, and symbol order
3644 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3645 the same order they were declared. This may break some debuggers.
3650 * Setting Symbols:: Giving Symbols Other Values
3651 * Symbol Names:: Symbol Names
3652 * Dot:: The Special Dot Symbol
3653 * Symbol Attributes:: Symbol Attributes
3660 A @dfn{label} is written as a symbol immediately followed by a colon
3661 @samp{:}. The symbol then represents the current value of the
3662 active location counter, and is, for example, a suitable instruction
3663 operand. You are warned if you use the same symbol to represent two
3664 different locations: the first definition overrides any other
3668 On the HPPA, the usual form for a label need not be immediately followed by a
3669 colon, but instead must start in column zero. Only one label may be defined on
3670 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3671 provides a special directive @code{.label} for defining labels more flexibly.
3674 @node Setting Symbols
3675 @section Giving Symbols Other Values
3677 @cindex assigning values to symbols
3678 @cindex symbol values, assigning
3679 A symbol can be given an arbitrary value by writing a symbol, followed
3680 by an equals sign @samp{=}, followed by an expression
3681 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3682 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3683 equals sign @samp{=}@samp{=} here represents an equivalent of the
3684 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3687 Blackfin does not support symbol assignment with @samp{=}.
3691 @section Symbol Names
3693 @cindex symbol names
3694 @cindex names, symbol
3695 @ifclear SPECIAL-SYMS
3696 Symbol names begin with a letter or with one of @samp{._}. On most
3697 machines, you can also use @code{$} in symbol names; exceptions are
3698 noted in @ref{Machine Dependencies}. That character may be followed by any
3699 string of digits, letters, dollar signs (unless otherwise noted for a
3700 particular target machine), and underscores.
3704 Symbol names begin with a letter or with one of @samp{._}. On the
3705 Renesas SH you can also use @code{$} in symbol names. That
3706 character may be followed by any string of digits, letters, dollar signs (save
3707 on the H8/300), and underscores.
3711 Case of letters is significant: @code{foo} is a different symbol name
3714 Symbol names do not start with a digit. An exception to this rule is made for
3715 Local Labels. See below.
3717 Multibyte characters are supported. To generate a symbol name containing
3718 multibyte characters enclose it within double quotes and use escape codes. cf
3719 @xref{Strings}. Generating a multibyte symbol name from a label is not
3720 currently supported.
3722 Each symbol has exactly one name. Each name in an assembly language program
3723 refers to exactly one symbol. You may use that symbol name any number of times
3726 @subheading Local Symbol Names
3728 @cindex local symbol names
3729 @cindex symbol names, local
3730 A local symbol is any symbol beginning with certain local label prefixes.
3731 By default, the local label prefix is @samp{.L} for ELF systems or
3732 @samp{L} for traditional a.out systems, but each target may have its own
3733 set of local label prefixes.
3735 On the HPPA local symbols begin with @samp{L$}.
3738 Local symbols are defined and used within the assembler, but they are
3739 normally not saved in object files. Thus, they are not visible when debugging.
3740 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols})
3741 to retain the local symbols in the object files.
3743 @subheading Local Labels
3745 @cindex local labels
3746 @cindex temporary symbol names
3747 @cindex symbol names, temporary
3748 Local labels are different from local symbols. Local labels help compilers and
3749 programmers use names temporarily. They create symbols which are guaranteed to
3750 be unique over the entire scope of the input source code and which can be
3751 referred to by a simple notation. To define a local label, write a label of
3752 the form @samp{@b{N}:} (where @b{N} represents any non-negative integer).
3753 To refer to the most recent previous definition of that label write
3754 @samp{@b{N}b}, using the same number as when you defined the label. To refer
3755 to the next definition of a local label, write @samp{@b{N}f}. The @samp{b}
3756 stands for ``backwards'' and the @samp{f} stands for ``forwards''.
3758 There is no restriction on how you can use these labels, and you can reuse them
3759 too. So that it is possible to repeatedly define the same local label (using
3760 the same number @samp{@b{N}}), although you can only refer to the most recently
3761 defined local label of that number (for a backwards reference) or the next
3762 definition of a specific local label for a forward reference. It is also worth
3763 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3764 implemented in a slightly more efficient manner than the others.
3775 Which is the equivalent of:
3778 label_1: branch label_3
3779 label_2: branch label_1
3780 label_3: branch label_4
3781 label_4: branch label_3
3784 Local label names are only a notational device. They are immediately
3785 transformed into more conventional symbol names before the assembler uses them.
3786 The symbol names are stored in the symbol table, appear in error messages, and
3787 are optionally emitted to the object file. The names are constructed using
3791 @item @emph{local label prefix}
3792 All local symbols begin with the system-specific local label prefix.
3793 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3794 that start with the local label prefix. These labels are
3795 used for symbols you are never intended to see. If you use the
3796 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3797 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3798 you may use them in debugging.
3801 This is the number that was used in the local label definition. So if the
3802 label is written @samp{55:} then the number is @samp{55}.
3805 This unusual character is included so you do not accidentally invent a symbol
3806 of the same name. The character has ASCII value of @samp{\002} (control-B).
3808 @item @emph{ordinal number}
3809 This is a serial number to keep the labels distinct. The first definition of
3810 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3811 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3812 the number @samp{1} and its 15th definition gets @samp{15} as well.
3815 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3816 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3818 @subheading Dollar Local Labels
3819 @cindex dollar local symbols
3821 On some targets @code{@value{AS}} also supports an even more local form of
3822 local labels called dollar labels. These labels go out of scope (i.e., they
3823 become undefined) as soon as a non-local label is defined. Thus they remain
3824 valid for only a small region of the input source code. Normal local labels,
3825 by contrast, remain in scope for the entire file, or until they are redefined
3826 by another occurrence of the same local label.
3828 Dollar labels are defined in exactly the same way as ordinary local labels,
3829 except that they have a dollar sign suffix to their numeric value, e.g.,
3832 They can also be distinguished from ordinary local labels by their transformed
3833 names which use ASCII character @samp{\001} (control-A) as the magic character
3834 to distinguish them from ordinary labels. For example, the fifth definition of
3835 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3838 @section The Special Dot Symbol
3840 @cindex dot (symbol)
3841 @cindex @code{.} (symbol)
3842 @cindex current address
3843 @cindex location counter
3844 The special symbol @samp{.} refers to the current address that
3845 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3846 .long .} defines @code{melvin} to contain its own address.
3847 Assigning a value to @code{.} is treated the same as a @code{.org}
3849 @ifclear no-space-dir
3850 Thus, the expression @samp{.=.+4} is the same as saying
3854 @node Symbol Attributes
3855 @section Symbol Attributes
3857 @cindex symbol attributes
3858 @cindex attributes, symbol
3859 Every symbol has, as well as its name, the attributes ``Value'' and
3860 ``Type''. Depending on output format, symbols can also have auxiliary
3863 The detailed definitions are in @file{a.out.h}.
3866 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3867 all these attributes, and probably won't warn you. This makes the
3868 symbol an externally defined symbol, which is generally what you
3872 * Symbol Value:: Value
3873 * Symbol Type:: Type
3875 * a.out Symbols:: Symbol Attributes: @code{a.out}
3878 * COFF Symbols:: Symbol Attributes for COFF
3881 * SOM Symbols:: Symbol Attributes for SOM
3888 @cindex value of a symbol
3889 @cindex symbol value
3890 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3891 location in the text, data, bss or absolute sections the value is the
3892 number of addresses from the start of that section to the label.
3893 Naturally for text, data and bss sections the value of a symbol changes
3894 as @code{@value{LD}} changes section base addresses during linking. Absolute
3895 symbols' values do not change during linking: that is why they are
3898 The value of an undefined symbol is treated in a special way. If it is
3899 0 then the symbol is not defined in this assembler source file, and
3900 @code{@value{LD}} tries to determine its value from other files linked into the
3901 same program. You make this kind of symbol simply by mentioning a symbol
3902 name without defining it. A non-zero value represents a @code{.comm}
3903 common declaration. The value is how much common storage to reserve, in
3904 bytes (addresses). The symbol refers to the first address of the
3910 @cindex type of a symbol
3912 The type attribute of a symbol contains relocation (section)
3913 information, any flag settings indicating that a symbol is external, and
3914 (optionally), other information for linkers and debuggers. The exact
3915 format depends on the object-code output format in use.
3919 @subsection Symbol Attributes: @code{a.out}
3921 @cindex @code{a.out} symbol attributes
3922 @cindex symbol attributes, @code{a.out}
3925 * Symbol Desc:: Descriptor
3926 * Symbol Other:: Other
3930 @subsubsection Descriptor
3932 @cindex descriptor, of @code{a.out} symbol
3933 This is an arbitrary 16-bit value. You may establish a symbol's
3934 descriptor value by using a @code{.desc} statement
3935 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3936 @command{@value{AS}}.
3939 @subsubsection Other
3941 @cindex other attribute, of @code{a.out} symbol
3942 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3947 @subsection Symbol Attributes for COFF
3949 @cindex COFF symbol attributes
3950 @cindex symbol attributes, COFF
3952 The COFF format supports a multitude of auxiliary symbol attributes;
3953 like the primary symbol attributes, they are set between @code{.def} and
3954 @code{.endef} directives.
3956 @subsubsection Primary Attributes
3958 @cindex primary attributes, COFF symbols
3959 The symbol name is set with @code{.def}; the value and type,
3960 respectively, with @code{.val} and @code{.type}.
3962 @subsubsection Auxiliary Attributes
3964 @cindex auxiliary attributes, COFF symbols
3965 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3966 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3967 table information for COFF.
3972 @subsection Symbol Attributes for SOM
3974 @cindex SOM symbol attributes
3975 @cindex symbol attributes, SOM
3977 The SOM format for the HPPA supports a multitude of symbol attributes set with
3978 the @code{.EXPORT} and @code{.IMPORT} directives.
3980 The attributes are described in @cite{HP9000 Series 800 Assembly
3981 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3982 @code{EXPORT} assembler directive documentation.
3986 @chapter Expressions
3990 @cindex numeric values
3991 An @dfn{expression} specifies an address or numeric value.
3992 Whitespace may precede and/or follow an expression.
3994 The result of an expression must be an absolute number, or else an offset into
3995 a particular section. If an expression is not absolute, and there is not
3996 enough information when @command{@value{AS}} sees the expression to know its
3997 section, a second pass over the source program might be necessary to interpret
3998 the expression---but the second pass is currently not implemented.
3999 @command{@value{AS}} aborts with an error message in this situation.
4002 * Empty Exprs:: Empty Expressions
4003 * Integer Exprs:: Integer Expressions
4007 @section Empty Expressions
4009 @cindex empty expressions
4010 @cindex expressions, empty
4011 An empty expression has no value: it is just whitespace or null.
4012 Wherever an absolute expression is required, you may omit the
4013 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
4014 is compatible with other assemblers.
4017 @section Integer Expressions
4019 @cindex integer expressions
4020 @cindex expressions, integer
4021 An @dfn{integer expression} is one or more @emph{arguments} delimited
4022 by @emph{operators}.
4025 * Arguments:: Arguments
4026 * Operators:: Operators
4027 * Prefix Ops:: Prefix Operators
4028 * Infix Ops:: Infix Operators
4032 @subsection Arguments
4034 @cindex expression arguments
4035 @cindex arguments in expressions
4036 @cindex operands in expressions
4037 @cindex arithmetic operands
4038 @dfn{Arguments} are symbols, numbers or subexpressions. In other
4039 contexts arguments are sometimes called ``arithmetic operands''. In
4040 this manual, to avoid confusing them with the ``instruction operands'' of
4041 the machine language, we use the term ``argument'' to refer to parts of
4042 expressions only, reserving the word ``operand'' to refer only to machine
4043 instruction operands.
4045 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
4046 @var{section} is one of text, data, bss, absolute,
4047 or undefined. @var{NNN} is a signed, 2's complement 32 bit
4050 Numbers are usually integers.
4052 A number can be a flonum or bignum. In this case, you are warned
4053 that only the low order 32 bits are used, and @command{@value{AS}} pretends
4054 these 32 bits are an integer. You may write integer-manipulating
4055 instructions that act on exotic constants, compatible with other
4058 @cindex subexpressions
4059 Subexpressions are a left parenthesis @samp{(} followed by an integer
4060 expression, followed by a right parenthesis @samp{)}; or a prefix
4061 operator followed by an argument.
4064 @subsection Operators
4066 @cindex operators, in expressions
4067 @cindex arithmetic functions
4068 @cindex functions, in expressions
4069 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
4070 operators are followed by an argument. Infix operators appear
4071 between their arguments. Operators may be preceded and/or followed by
4075 @subsection Prefix Operator
4077 @cindex prefix operators
4078 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
4079 one argument, which must be absolute.
4081 @c the tex/end tex stuff surrounding this small table is meant to make
4082 @c it align, on the printed page, with the similar table in the next
4083 @c section (which is inside an enumerate).
4085 \global\advance\leftskip by \itemindent
4090 @dfn{Negation}. Two's complement negation.
4092 @dfn{Complementation}. Bitwise not.
4096 \global\advance\leftskip by -\itemindent
4100 @subsection Infix Operators
4102 @cindex infix operators
4103 @cindex operators, permitted arguments
4104 @dfn{Infix operators} take two arguments, one on either side. Operators
4105 have precedence, but operations with equal precedence are performed left
4106 to right. Apart from @code{+} or @option{-}, both arguments must be
4107 absolute, and the result is absolute.
4110 @cindex operator precedence
4111 @cindex precedence of operators
4118 @dfn{Multiplication}.
4121 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4127 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4130 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4134 Intermediate precedence
4139 @dfn{Bitwise Inclusive Or}.
4145 @dfn{Bitwise Exclusive Or}.
4148 @dfn{Bitwise Or Not}.
4155 @cindex addition, permitted arguments
4156 @cindex plus, permitted arguments
4157 @cindex arguments for addition
4159 @dfn{Addition}. If either argument is absolute, the result has the section of
4160 the other argument. You may not add together arguments from different
4163 @cindex subtraction, permitted arguments
4164 @cindex minus, permitted arguments
4165 @cindex arguments for subtraction
4167 @dfn{Subtraction}. If the right argument is absolute, the
4168 result has the section of the left argument.
4169 If both arguments are in the same section, the result is absolute.
4170 You may not subtract arguments from different sections.
4171 @c FIXME is there still something useful to say about undefined - undefined ?
4173 @cindex comparison expressions
4174 @cindex expressions, comparison
4179 @dfn{Is Not Equal To}
4183 @dfn{Is Greater Than}
4185 @dfn{Is Greater Than Or Equal To}
4187 @dfn{Is Less Than Or Equal To}
4189 The comparison operators can be used as infix operators. A true results has a
4190 value of -1 whereas a false result has a value of 0. Note, these operators
4191 perform signed comparisons.
4194 @item Lowest Precedence
4203 These two logical operations can be used to combine the results of sub
4204 expressions. Note, unlike the comparison operators a true result returns a
4205 value of 1 but a false results does still return 0. Also note that the logical
4206 or operator has a slightly lower precedence than logical and.
4211 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4212 address; you can only have a defined section in one of the two arguments.
4215 @chapter Assembler Directives
4217 @cindex directives, machine independent
4218 @cindex pseudo-ops, machine independent
4219 @cindex machine independent directives
4220 All assembler directives have names that begin with a period (@samp{.}).
4221 The names are case insensitive for most targets, and usually written
4224 This chapter discusses directives that are available regardless of the
4225 target machine configuration for the @sc{gnu} assembler.
4227 Some machine configurations provide additional directives.
4228 @xref{Machine Dependencies}.
4231 @ifset machine-directives
4232 @xref{Machine Dependencies}, for additional directives.
4237 * Abort:: @code{.abort}
4239 * ABORT (COFF):: @code{.ABORT}
4242 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4243 * Altmacro:: @code{.altmacro}
4244 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4245 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4246 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4247 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, etc
4248 * Byte:: @code{.byte @var{expressions}}
4249 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4250 * Comm:: @code{.comm @var{symbol} , @var{length} }
4251 * Data:: @code{.data @var{subsection}}
4252 * Dc:: @code{.dc[@var{size}] @var{expressions}}
4253 * Dcb:: @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
4254 * Ds:: @code{.ds[@var{size}] @var{number} [,@var{fill}]}
4256 * Def:: @code{.def @var{name}}
4259 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4265 * Double:: @code{.double @var{flonums}}
4266 * Eject:: @code{.eject}
4267 * Else:: @code{.else}
4268 * Elseif:: @code{.elseif}
4271 * Endef:: @code{.endef}
4274 * Endfunc:: @code{.endfunc}
4275 * Endif:: @code{.endif}
4276 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4277 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4278 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4280 * Error:: @code{.error @var{string}}
4281 * Exitm:: @code{.exitm}
4282 * Extern:: @code{.extern}
4283 * Fail:: @code{.fail}
4284 * File:: @code{.file}
4285 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4286 * Float:: @code{.float @var{flonums}}
4287 * Func:: @code{.func}
4288 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4290 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4291 * Hidden:: @code{.hidden @var{names}}
4294 * hword:: @code{.hword @var{expressions}}
4295 * Ident:: @code{.ident}
4296 * If:: @code{.if @var{absolute expression}}
4297 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4298 * Include:: @code{.include "@var{file}"}
4299 * Int:: @code{.int @var{expressions}}
4301 * Internal:: @code{.internal @var{names}}
4304 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4305 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4306 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4307 * Lflags:: @code{.lflags}
4308 @ifclear no-line-dir
4309 * Line:: @code{.line @var{line-number}}
4312 * Linkonce:: @code{.linkonce [@var{type}]}
4313 * List:: @code{.list}
4314 * Ln:: @code{.ln @var{line-number}}
4315 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4316 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4318 * Local:: @code{.local @var{names}}
4321 * Long:: @code{.long @var{expressions}}
4323 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4326 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4327 * MRI:: @code{.mri @var{val}}
4328 * Noaltmacro:: @code{.noaltmacro}
4329 * Nolist:: @code{.nolist}
4330 * Nops:: @code{.nops @var{size}[, @var{control}]}
4331 * Octa:: @code{.octa @var{bignums}}
4332 * Offset:: @code{.offset @var{loc}}
4333 * Org:: @code{.org @var{new-lc}, @var{fill}}
4334 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4336 * PopSection:: @code{.popsection}
4337 * Previous:: @code{.previous}
4340 * Print:: @code{.print @var{string}}
4342 * Protected:: @code{.protected @var{names}}
4345 * Psize:: @code{.psize @var{lines}, @var{columns}}
4346 * Purgem:: @code{.purgem @var{name}}
4348 * PushSection:: @code{.pushsection @var{name}}
4351 * Quad:: @code{.quad @var{bignums}}
4352 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4353 * Rept:: @code{.rept @var{count}}
4354 * Sbttl:: @code{.sbttl "@var{subheading}"}
4356 * Scl:: @code{.scl @var{class}}
4359 * Section:: @code{.section @var{name}[, @var{flags}]}
4362 * Set:: @code{.set @var{symbol}, @var{expression}}
4363 * Short:: @code{.short @var{expressions}}
4364 * Single:: @code{.single @var{flonums}}
4366 * Size:: @code{.size [@var{name} , @var{expression}]}
4368 @ifclear no-space-dir
4369 * Skip:: @code{.skip @var{size} [,@var{fill}]}
4372 * Sleb128:: @code{.sleb128 @var{expressions}}
4373 @ifclear no-space-dir
4374 * Space:: @code{.space @var{size} [,@var{fill}]}
4377 * Stab:: @code{.stabd, .stabn, .stabs}
4380 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4381 * Struct:: @code{.struct @var{expression}}
4383 * SubSection:: @code{.subsection}
4384 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4388 * Tag:: @code{.tag @var{structname}}
4391 * Text:: @code{.text @var{subsection}}
4392 * Title:: @code{.title "@var{heading}"}
4394 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4397 * Uleb128:: @code{.uleb128 @var{expressions}}
4399 * Val:: @code{.val @var{addr}}
4403 * Version:: @code{.version "@var{string}"}
4404 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4405 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4408 * Warning:: @code{.warning @var{string}}
4409 * Weak:: @code{.weak @var{names}}
4410 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4411 * Word:: @code{.word @var{expressions}}
4412 @ifclear no-space-dir
4413 * Zero:: @code{.zero @var{size}}
4416 * 2byte:: @code{.2byte @var{expressions}}
4417 * 4byte:: @code{.4byte @var{expressions}}
4418 * 8byte:: @code{.8byte @var{bignums}}
4420 * Deprecated:: Deprecated Directives
4424 @section @code{.abort}
4426 @cindex @code{abort} directive
4427 @cindex stopping the assembly
4428 This directive stops the assembly immediately. It is for
4429 compatibility with other assemblers. The original idea was that the
4430 assembly language source would be piped into the assembler. If the sender
4431 of the source quit, it could use this directive tells @command{@value{AS}} to
4432 quit also. One day @code{.abort} will not be supported.
4436 @section @code{.ABORT} (COFF)
4438 @cindex @code{ABORT} directive
4439 When producing COFF output, @command{@value{AS}} accepts this directive as a
4440 synonym for @samp{.abort}.
4445 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4447 @cindex padding the location counter
4448 @cindex @code{align} directive
4449 Pad the location counter (in the current subsection) to a particular storage
4450 boundary. The first expression (which must be absolute) is the alignment
4451 required, as described below.
4453 The second expression (also absolute) gives the fill value to be stored in the
4454 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4455 padding bytes are normally zero. However, on most systems, if the section is
4456 marked as containing code and the fill value is omitted, the space is filled
4457 with no-op instructions.
4459 The third expression is also absolute, and is also optional. If it is present,
4460 it is the maximum number of bytes that should be skipped by this alignment
4461 directive. If doing the alignment would require skipping more bytes than the
4462 specified maximum, then the alignment is not done at all. You can omit the
4463 fill value (the second argument) entirely by simply using two commas after the
4464 required alignment; this can be useful if you want the alignment to be filled
4465 with no-op instructions when appropriate.
4467 The way the required alignment is specified varies from system to system.
4468 For the arc, hppa, i386 using ELF, iq2000, m68k, or1k,
4469 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4470 alignment request in bytes. For example @samp{.align 8} advances
4471 the location counter until it is a multiple of 8. If the location counter
4472 is already a multiple of 8, no change is needed. For the tic54x, the
4473 first expression is the alignment request in words.
4475 For other systems, including ppc, i386 using a.out format, arm and
4476 strongarm, it is the
4477 number of low-order zero bits the location counter must have after
4478 advancement. For example @samp{.align 3} advances the location
4479 counter until it a multiple of 8. If the location counter is already a
4480 multiple of 8, no change is needed.
4482 This inconsistency is due to the different behaviors of the various
4483 native assemblers for these systems which GAS must emulate.
4484 GAS also provides @code{.balign} and @code{.p2align} directives,
4485 described later, which have a consistent behavior across all
4486 architectures (but are specific to GAS).
4489 @section @code{.altmacro}
4490 Enable alternate macro mode, enabling:
4493 @item LOCAL @var{name} [ , @dots{} ]
4494 One additional directive, @code{LOCAL}, is available. It is used to
4495 generate a string replacement for each of the @var{name} arguments, and
4496 replace any instances of @var{name} in each macro expansion. The
4497 replacement string is unique in the assembly, and different for each
4498 separate macro expansion. @code{LOCAL} allows you to write macros that
4499 define symbols, without fear of conflict between separate macro expansions.
4501 @item String delimiters
4502 You can write strings delimited in these other ways besides
4503 @code{"@var{string}"}:
4506 @item '@var{string}'
4507 You can delimit strings with single-quote characters.
4509 @item <@var{string}>
4510 You can delimit strings with matching angle brackets.
4513 @item single-character string escape
4514 To include any single character literally in a string (even if the
4515 character would otherwise have some special meaning), you can prefix the
4516 character with @samp{!} (an exclamation mark). For example, you can
4517 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4519 @item Expression results as strings
4520 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4521 and use the result as a string.
4525 @section @code{.ascii "@var{string}"}@dots{}
4527 @cindex @code{ascii} directive
4528 @cindex string literals
4529 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4530 separated by commas. It assembles each string (with no automatic
4531 trailing zero byte) into consecutive addresses.
4534 @section @code{.asciz "@var{string}"}@dots{}
4536 @cindex @code{asciz} directive
4537 @cindex zero-terminated strings
4538 @cindex null-terminated strings
4539 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4540 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4543 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4545 @cindex padding the location counter given number of bytes
4546 @cindex @code{balign} directive
4547 Pad the location counter (in the current subsection) to a particular
4548 storage boundary. The first expression (which must be absolute) is the
4549 alignment request in bytes. For example @samp{.balign 8} advances
4550 the location counter until it is a multiple of 8. If the location counter
4551 is already a multiple of 8, no change is needed.
4553 The second expression (also absolute) gives the fill value to be stored in the
4554 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4555 padding bytes are normally zero. However, on most systems, if the section is
4556 marked as containing code and the fill value is omitted, the space is filled
4557 with no-op instructions.
4559 The third expression is also absolute, and is also optional. If it is present,
4560 it is the maximum number of bytes that should be skipped by this alignment
4561 directive. If doing the alignment would require skipping more bytes than the
4562 specified maximum, then the alignment is not done at all. You can omit the
4563 fill value (the second argument) entirely by simply using two commas after the
4564 required alignment; this can be useful if you want the alignment to be filled
4565 with no-op instructions when appropriate.
4567 @cindex @code{balignw} directive
4568 @cindex @code{balignl} directive
4569 The @code{.balignw} and @code{.balignl} directives are variants of the
4570 @code{.balign} directive. The @code{.balignw} directive treats the fill
4571 pattern as a two byte word value. The @code{.balignl} directives treats the
4572 fill pattern as a four byte longword value. For example, @code{.balignw
4573 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4574 filled in with the value 0x368d (the exact placement of the bytes depends upon
4575 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4578 @node Bundle directives
4579 @section Bundle directives
4580 @subsection @code{.bundle_align_mode @var{abs-expr}}
4581 @cindex @code{bundle_align_mode} directive
4583 @cindex instruction bundle
4584 @cindex aligned instruction bundle
4585 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4586 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4587 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4588 disabled (which is the default state). If the argument it not zero, it
4589 gives the size of an instruction bundle as a power of two (as for the
4590 @code{.p2align} directive, @pxref{P2align}).
4592 For some targets, it's an ABI requirement that no instruction may span a
4593 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4594 instructions that starts on an aligned boundary. For example, if
4595 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4596 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4597 effect, no single instruction may span a boundary between bundles. If an
4598 instruction would start too close to the end of a bundle for the length of
4599 that particular instruction to fit within the bundle, then the space at the
4600 end of that bundle is filled with no-op instructions so the instruction
4601 starts in the next bundle. As a corollary, it's an error if any single
4602 instruction's encoding is longer than the bundle size.
4604 @subsection @code{.bundle_lock} and @code{.bundle_unlock}
4605 @cindex @code{bundle_lock} directive
4606 @cindex @code{bundle_unlock} directive
4607 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4608 allow explicit control over instruction bundle padding. These directives
4609 are only valid when @code{.bundle_align_mode} has been used to enable
4610 aligned instruction bundle mode. It's an error if they appear when
4611 @code{.bundle_align_mode} has not been used at all, or when the last
4612 directive was @w{@code{.bundle_align_mode 0}}.
4614 @cindex bundle-locked
4615 For some targets, it's an ABI requirement that certain instructions may
4616 appear only as part of specified permissible sequences of multiple
4617 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4618 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4619 instruction sequence. For purposes of aligned instruction bundle mode, a
4620 sequence starting with @code{.bundle_lock} and ending with
4621 @code{.bundle_unlock} is treated as a single instruction. That is, the
4622 entire sequence must fit into a single bundle and may not span a bundle
4623 boundary. If necessary, no-op instructions will be inserted before the
4624 first instruction of the sequence so that the whole sequence starts on an
4625 aligned bundle boundary. It's an error if the sequence is longer than the
4628 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4629 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4630 nested. That is, a second @code{.bundle_lock} directive before the next
4631 @code{.bundle_unlock} directive has no effect except that it must be
4632 matched by another closing @code{.bundle_unlock} so that there is the
4633 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4636 @section @code{.byte @var{expressions}}
4638 @cindex @code{byte} directive
4639 @cindex integers, one byte
4640 @code{.byte} expects zero or more expressions, separated by commas.
4641 Each expression is assembled into the next byte.
4643 @node CFI directives
4644 @section CFI directives
4645 @subsection @code{.cfi_sections @var{section_list}}
4646 @cindex @code{cfi_sections} directive
4647 @code{.cfi_sections} may be used to specify whether CFI directives
4648 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4649 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4650 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4651 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4652 directive is not used is @code{.cfi_sections .eh_frame}.
4654 On targets that support compact unwinding tables these can be generated
4655 by specifying @code{.eh_frame_entry} instead of @code{.eh_frame}.
4657 Some targets may support an additional name, such as @code{.c6xabi.exidx}
4658 which is used by the @value{TIC6X} target.
4660 The @code{.cfi_sections} directive can be repeated, with the same or different
4661 arguments, provided that CFI generation has not yet started. Once CFI
4662 generation has started however the section list is fixed and any attempts to
4663 redefine it will result in an error.
4665 @subsection @code{.cfi_startproc [simple]}
4666 @cindex @code{cfi_startproc} directive
4667 @code{.cfi_startproc} is used at the beginning of each function that
4668 should have an entry in @code{.eh_frame}. It initializes some internal
4669 data structures. Don't forget to close the function by
4670 @code{.cfi_endproc}.
4672 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4673 it also emits some architecture dependent initial CFI instructions.
4675 @subsection @code{.cfi_endproc}
4676 @cindex @code{cfi_endproc} directive
4677 @code{.cfi_endproc} is used at the end of a function where it closes its
4678 unwind entry previously opened by
4679 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4681 @subsection @code{.cfi_personality @var{encoding} [, @var{exp}]}
4682 @cindex @code{cfi_personality} directive
4683 @code{.cfi_personality} defines personality routine and its encoding.
4684 @var{encoding} must be a constant determining how the personality
4685 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4686 argument is not present, otherwise second argument should be
4687 a constant or a symbol name. When using indirect encodings,
4688 the symbol provided should be the location where personality
4689 can be loaded from, not the personality routine itself.
4690 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4691 no personality routine.
4693 @subsection @code{.cfi_personality_id @var{id}}
4694 @cindex @code{cfi_personality_id} directive
4695 @code{cfi_personality_id} defines a personality routine by its index as
4696 defined in a compact unwinding format.
4697 Only valid when generating compact EH frames (i.e.
4698 with @code{.cfi_sections eh_frame_entry}.
4700 @subsection @code{.cfi_fde_data [@var{opcode1} [, @dots{}]]}
4701 @cindex @code{cfi_fde_data} directive
4702 @code{cfi_fde_data} is used to describe the compact unwind opcodes to be
4703 used for the current function. These are emitted inline in the
4704 @code{.eh_frame_entry} section if small enough and there is no LSDA, or
4705 in the @code{.gnu.extab} section otherwise.
4706 Only valid when generating compact EH frames (i.e.
4707 with @code{.cfi_sections eh_frame_entry}.
4709 @subsection @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4710 @code{.cfi_lsda} defines LSDA and its encoding.
4711 @var{encoding} must be a constant determining how the LSDA
4712 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), the second
4713 argument is not present, otherwise the second argument should be a constant
4714 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4715 meaning that no LSDA is present.
4717 @subsection @code{.cfi_inline_lsda} [@var{align}]
4718 @code{.cfi_inline_lsda} marks the start of a LSDA data section and
4719 switches to the corresponding @code{.gnu.extab} section.
4720 Must be preceded by a CFI block containing a @code{.cfi_lsda} directive.
4721 Only valid when generating compact EH frames (i.e.
4722 with @code{.cfi_sections eh_frame_entry}.
4724 The table header and unwinding opcodes will be generated at this point,
4725 so that they are immediately followed by the LSDA data. The symbol
4726 referenced by the @code{.cfi_lsda} directive should still be defined
4727 in case a fallback FDE based encoding is used. The LSDA data is terminated
4728 by a section directive.
4730 The optional @var{align} argument specifies the alignment required.
4731 The alignment is specified as a power of two, as with the
4732 @code{.p2align} directive.
4734 @subsection @code{.cfi_def_cfa @var{register}, @var{offset}}
4735 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4736 address from @var{register} and add @var{offset} to it}.
4738 @subsection @code{.cfi_def_cfa_register @var{register}}
4739 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4740 now on @var{register} will be used instead of the old one. Offset
4743 @subsection @code{.cfi_def_cfa_offset @var{offset}}
4744 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4745 remains the same, but @var{offset} is new. Note that it is the
4746 absolute offset that will be added to a defined register to compute
4749 @subsection @code{.cfi_adjust_cfa_offset @var{offset}}
4750 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4751 value that is added/subtracted from the previous offset.
4753 @subsection @code{.cfi_offset @var{register}, @var{offset}}
4754 Previous value of @var{register} is saved at offset @var{offset} from
4757 @subsection @code{.cfi_val_offset @var{register}, @var{offset}}
4758 Previous value of @var{register} is CFA + @var{offset}.
4760 @subsection @code{.cfi_rel_offset @var{register}, @var{offset}}
4761 Previous value of @var{register} is saved at offset @var{offset} from
4762 the current CFA register. This is transformed to @code{.cfi_offset}
4763 using the known displacement of the CFA register from the CFA.
4764 This is often easier to use, because the number will match the
4765 code it's annotating.
4767 @subsection @code{.cfi_register @var{register1}, @var{register2}}
4768 Previous value of @var{register1} is saved in register @var{register2}.
4770 @subsection @code{.cfi_restore @var{register}}
4771 @code{.cfi_restore} says that the rule for @var{register} is now the
4772 same as it was at the beginning of the function, after all initial
4773 instruction added by @code{.cfi_startproc} were executed.
4775 @subsection @code{.cfi_undefined @var{register}}
4776 From now on the previous value of @var{register} can't be restored anymore.
4778 @subsection @code{.cfi_same_value @var{register}}
4779 Current value of @var{register} is the same like in the previous frame,
4780 i.e. no restoration needed.
4782 @subsection @code{.cfi_remember_state} and @code{.cfi_restore_state}
4783 @code{.cfi_remember_state} pushes the set of rules for every register onto an
4784 implicit stack, while @code{.cfi_restore_state} pops them off the stack and
4785 places them in the current row. This is useful for situations where you have
4786 multiple @code{.cfi_*} directives that need to be undone due to the control
4787 flow of the program. For example, we could have something like this (assuming
4788 the CFA is the value of @code{rbp}):
4798 .cfi_def_cfa %rsp, 8
4801 /* Do something else */
4804 Here, we want the @code{.cfi} directives to affect only the rows corresponding
4805 to the instructions before @code{label}. This means we'd have to add multiple
4806 @code{.cfi} directives after @code{label} to recreate the original save
4807 locations of the registers, as well as setting the CFA back to the value of
4808 @code{rbp}. This would be clumsy, and result in a larger binary size. Instead,
4820 .cfi_def_cfa %rsp, 8
4824 /* Do something else */
4827 That way, the rules for the instructions after @code{label} will be the same
4828 as before the first @code{.cfi_restore} without having to use multiple
4829 @code{.cfi} directives.
4831 @subsection @code{.cfi_return_column @var{register}}
4832 Change return column @var{register}, i.e. the return address is either
4833 directly in @var{register} or can be accessed by rules for @var{register}.
4835 @subsection @code{.cfi_signal_frame}
4836 Mark current function as signal trampoline.
4838 @subsection @code{.cfi_window_save}
4839 SPARC register window has been saved.
4841 @subsection @code{.cfi_escape} @var{expression}[, @dots{}]
4842 Allows the user to add arbitrary bytes to the unwind info. One
4843 might use this to add OS-specific CFI opcodes, or generic CFI
4844 opcodes that GAS does not yet support.
4846 @subsection @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4847 The current value of @var{register} is @var{label}. The value of @var{label}
4848 will be encoded in the output file according to @var{encoding}; see the
4849 description of @code{.cfi_personality} for details on this encoding.
4851 The usefulness of equating a register to a fixed label is probably
4852 limited to the return address register. Here, it can be useful to
4853 mark a code segment that has only one return address which is reached
4854 by a direct branch and no copy of the return address exists in memory
4855 or another register.
4858 @section @code{.comm @var{symbol} , @var{length} }
4860 @cindex @code{comm} directive
4861 @cindex symbol, common
4862 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4863 common symbol in one object file may be merged with a defined or common symbol
4864 of the same name in another object file. If @code{@value{LD}} does not see a
4865 definition for the symbol--just one or more common symbols--then it will
4866 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4867 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4868 the same name, and they do not all have the same size, it will allocate space
4869 using the largest size.
4872 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4873 an optional third argument. This is the desired alignment of the symbol,
4874 specified for ELF as a byte boundary (for example, an alignment of 16 means
4875 that the least significant 4 bits of the address should be zero), and for PE
4876 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4877 boundary). The alignment must be an absolute expression, and it must be a
4878 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4879 common symbol, it will use the alignment when placing the symbol. If no
4880 alignment is specified, @command{@value{AS}} will set the alignment to the
4881 largest power of two less than or equal to the size of the symbol, up to a
4882 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4883 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4884 @samp{--section-alignment} option; image file sections in PE are aligned to
4885 multiples of 4096, which is far too large an alignment for ordinary variables.
4886 It is rather the default alignment for (non-debug) sections within object
4887 (@samp{*.o}) files, which are less strictly aligned.}.
4891 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4892 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4896 @section @code{.data @var{subsection}}
4897 @cindex @code{data} directive
4899 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4900 end of the data subsection numbered @var{subsection} (which is an
4901 absolute expression). If @var{subsection} is omitted, it defaults
4905 @section @code{.dc[@var{size}] @var{expressions}}
4906 @cindex @code{dc} directive
4908 The @code{.dc} directive expects zero or more @var{expressions} separated by
4909 commas. These expressions are evaluated and their values inserted into the
4910 current section. The size of the emitted value depends upon the suffix to the
4911 @code{.dc} directive:
4915 Emits N-bit values, where N is the size of an address on the target system.
4919 Emits double precision floating-point values.
4921 Emits 32-bit values.
4923 Emits single precision floating-point values.
4925 Emits 16-bit values.
4926 Note - this is true even on targets where the @code{.word} directive would emit
4929 Emits long double precision floating-point values.
4932 If no suffix is used then @samp{.w} is assumed.
4934 The byte ordering is target dependent, as is the size and format of floating
4938 @section @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
4939 @cindex @code{dcb} directive
4940 This directive emits @var{number} copies of @var{fill}, each of @var{size}
4941 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
4942 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
4943 @var{size} suffix, if present, must be one of:
4947 Emits single byte values.
4949 Emits double-precision floating point values.
4951 Emits 4-byte values.
4953 Emits single-precision floating point values.
4955 Emits 2-byte values.
4957 Emits long double-precision floating point values.
4960 If the @var{size} suffix is omitted then @samp{.w} is assumed.
4962 The byte ordering is target dependent, as is the size and format of floating
4966 @section @code{.ds[@var{size}] @var{number} [,@var{fill}]}
4967 @cindex @code{ds} directive
4968 This directive emits @var{number} copies of @var{fill}, each of @var{size}
4969 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
4970 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
4971 @var{size} suffix, if present, must be one of:
4975 Emits single byte values.
4977 Emits 8-byte values.
4979 Emits 4-byte values.
4981 Emits 12-byte values.
4983 Emits 4-byte values.
4985 Emits 2-byte values.
4987 Emits 12-byte values.
4990 Note - unlike the @code{.dcb} directive the @samp{.d}, @samp{.s} and @samp{.x}
4991 suffixes do not indicate that floating-point values are to be inserted.
4993 If the @var{size} suffix is omitted then @samp{.w} is assumed.
4995 The byte ordering is target dependent.
5000 @section @code{.def @var{name}}
5002 @cindex @code{def} directive
5003 @cindex COFF symbols, debugging
5004 @cindex debugging COFF symbols
5005 Begin defining debugging information for a symbol @var{name}; the
5006 definition extends until the @code{.endef} directive is encountered.
5011 @section @code{.desc @var{symbol}, @var{abs-expression}}
5013 @cindex @code{desc} directive
5014 @cindex COFF symbol descriptor
5015 @cindex symbol descriptor, COFF
5016 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
5017 to the low 16 bits of an absolute expression.
5020 The @samp{.desc} directive is not available when @command{@value{AS}} is
5021 configured for COFF output; it is only for @code{a.out} or @code{b.out}
5022 object format. For the sake of compatibility, @command{@value{AS}} accepts
5023 it, but produces no output, when configured for COFF.
5029 @section @code{.dim}
5031 @cindex @code{dim} directive
5032 @cindex COFF auxiliary symbol information
5033 @cindex auxiliary symbol information, COFF
5034 This directive is generated by compilers to include auxiliary debugging
5035 information in the symbol table. It is only permitted inside
5036 @code{.def}/@code{.endef} pairs.
5040 @section @code{.double @var{flonums}}
5042 @cindex @code{double} directive
5043 @cindex floating point numbers (double)
5044 @code{.double} expects zero or more flonums, separated by commas. It
5045 assembles floating point numbers.
5047 The exact kind of floating point numbers emitted depends on how
5048 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5052 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
5053 in @sc{ieee} format.
5058 @section @code{.eject}
5060 @cindex @code{eject} directive
5061 @cindex new page, in listings
5062 @cindex page, in listings
5063 @cindex listing control: new page
5064 Force a page break at this point, when generating assembly listings.
5067 @section @code{.else}
5069 @cindex @code{else} directive
5070 @code{.else} is part of the @command{@value{AS}} support for conditional
5071 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
5072 of code to be assembled if the condition for the preceding @code{.if}
5076 @section @code{.elseif}
5078 @cindex @code{elseif} directive
5079 @code{.elseif} is part of the @command{@value{AS}} support for conditional
5080 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
5081 @code{.if} block that would otherwise fill the entire @code{.else} section.
5084 @section @code{.end}
5086 @cindex @code{end} directive
5087 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
5088 process anything in the file past the @code{.end} directive.
5092 @section @code{.endef}
5094 @cindex @code{endef} directive
5095 This directive flags the end of a symbol definition begun with
5100 @section @code{.endfunc}
5101 @cindex @code{endfunc} directive
5102 @code{.endfunc} marks the end of a function specified with @code{.func}.
5105 @section @code{.endif}
5107 @cindex @code{endif} directive
5108 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
5109 it marks the end of a block of code that is only assembled
5110 conditionally. @xref{If,,@code{.if}}.
5113 @section @code{.equ @var{symbol}, @var{expression}}
5115 @cindex @code{equ} directive
5116 @cindex assigning values to symbols
5117 @cindex symbols, assigning values to
5118 This directive sets the value of @var{symbol} to @var{expression}.
5119 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
5122 The syntax for @code{equ} on the HPPA is
5123 @samp{@var{symbol} .equ @var{expression}}.
5127 The syntax for @code{equ} on the Z80 is
5128 @samp{@var{symbol} equ @var{expression}}.
5129 On the Z80 it is an error if @var{symbol} is already defined,
5130 but the symbol is not protected from later redefinition.
5131 Compare @ref{Equiv}.
5135 @section @code{.equiv @var{symbol}, @var{expression}}
5136 @cindex @code{equiv} directive
5137 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
5138 the assembler will signal an error if @var{symbol} is already defined. Note a
5139 symbol which has been referenced but not actually defined is considered to be
5142 Except for the contents of the error message, this is roughly equivalent to
5149 plus it protects the symbol from later redefinition.
5152 @section @code{.eqv @var{symbol}, @var{expression}}
5153 @cindex @code{eqv} directive
5154 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
5155 evaluate the expression or any part of it immediately. Instead each time
5156 the resulting symbol is used in an expression, a snapshot of its current
5160 @section @code{.err}
5161 @cindex @code{err} directive
5162 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
5163 message and, unless the @option{-Z} option was used, it will not generate an
5164 object file. This can be used to signal an error in conditionally compiled code.
5167 @section @code{.error "@var{string}"}
5168 @cindex error directive
5170 Similarly to @code{.err}, this directive emits an error, but you can specify a
5171 string that will be emitted as the error message. If you don't specify the
5172 message, it defaults to @code{".error directive invoked in source file"}.
5173 @xref{Errors, ,Error and Warning Messages}.
5176 .error "This code has not been assembled and tested."
5180 @section @code{.exitm}
5181 Exit early from the current macro definition. @xref{Macro}.
5184 @section @code{.extern}
5186 @cindex @code{extern} directive
5187 @code{.extern} is accepted in the source program---for compatibility
5188 with other assemblers---but it is ignored. @command{@value{AS}} treats
5189 all undefined symbols as external.
5192 @section @code{.fail @var{expression}}
5194 @cindex @code{fail} directive
5195 Generates an error or a warning. If the value of the @var{expression} is 500
5196 or more, @command{@value{AS}} will print a warning message. If the value is less
5197 than 500, @command{@value{AS}} will print an error message. The message will
5198 include the value of @var{expression}. This can occasionally be useful inside
5199 complex nested macros or conditional assembly.
5202 @section @code{.file}
5203 @cindex @code{file} directive
5205 @ifclear no-file-dir
5206 There are two different versions of the @code{.file} directive. Targets
5207 that support DWARF2 line number information use the DWARF2 version of
5208 @code{.file}. Other targets use the default version.
5210 @subheading Default Version
5212 @cindex logical file name
5213 @cindex file name, logical
5214 This version of the @code{.file} directive tells @command{@value{AS}} that we
5215 are about to start a new logical file. The syntax is:
5221 @var{string} is the new file name. In general, the filename is
5222 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
5223 to specify an empty file name, you must give the quotes--@code{""}. This
5224 statement may go away in future: it is only recognized to be compatible with
5225 old @command{@value{AS}} programs.
5227 @subheading DWARF2 Version
5230 When emitting DWARF2 line number information, @code{.file} assigns filenames
5231 to the @code{.debug_line} file name table. The syntax is:
5234 .file @var{fileno} @var{filename}
5237 The @var{fileno} operand should be a unique positive integer to use as the
5238 index of the entry in the table. The @var{filename} operand is a C string
5241 The detail of filename indices is exposed to the user because the filename
5242 table is shared with the @code{.debug_info} section of the DWARF2 debugging
5243 information, and thus the user must know the exact indices that table
5247 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
5249 @cindex @code{fill} directive
5250 @cindex writing patterns in memory
5251 @cindex patterns, writing in memory
5252 @var{repeat}, @var{size} and @var{value} are absolute expressions.
5253 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
5254 may be zero or more. @var{Size} may be zero or more, but if it is
5255 more than 8, then it is deemed to have the value 8, compatible with
5256 other people's assemblers. The contents of each @var{repeat} bytes
5257 is taken from an 8-byte number. The highest order 4 bytes are
5258 zero. The lowest order 4 bytes are @var{value} rendered in the
5259 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
5260 Each @var{size} bytes in a repetition is taken from the lowest order
5261 @var{size} bytes of this number. Again, this bizarre behavior is
5262 compatible with other people's assemblers.
5264 @var{size} and @var{value} are optional.
5265 If the second comma and @var{value} are absent, @var{value} is
5266 assumed zero. If the first comma and following tokens are absent,
5267 @var{size} is assumed to be 1.
5270 @section @code{.float @var{flonums}}
5272 @cindex floating point numbers (single)
5273 @cindex @code{float} directive
5274 This directive assembles zero or more flonums, separated by commas. It
5275 has the same effect as @code{.single}.
5277 The exact kind of floating point numbers emitted depends on how
5278 @command{@value{AS}} is configured.
5279 @xref{Machine Dependencies}.
5283 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5284 in @sc{ieee} format.
5289 @section @code{.func @var{name}[,@var{label}]}
5290 @cindex @code{func} directive
5291 @code{.func} emits debugging information to denote function @var{name}, and
5292 is ignored unless the file is assembled with debugging enabled.
5293 Only @samp{--gstabs[+]} is currently supported.
5294 @var{label} is the entry point of the function and if omitted @var{name}
5295 prepended with the @samp{leading char} is used.
5296 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5297 All functions are currently defined to have @code{void} return type.
5298 The function must be terminated with @code{.endfunc}.
5301 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5303 @cindex @code{global} directive
5304 @cindex symbol, making visible to linker
5305 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5306 @var{symbol} in your partial program, its value is made available to
5307 other partial programs that are linked with it. Otherwise,
5308 @var{symbol} takes its attributes from a symbol of the same name
5309 from another file linked into the same program.
5311 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5312 compatibility with other assemblers.
5315 On the HPPA, @code{.global} is not always enough to make it accessible to other
5316 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5317 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5322 @section @code{.gnu_attribute @var{tag},@var{value}}
5323 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5326 @section @code{.hidden @var{names}}
5328 @cindex @code{hidden} directive
5330 This is one of the ELF visibility directives. The other two are
5331 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5332 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5334 This directive overrides the named symbols default visibility (which is set by
5335 their binding: local, global or weak). The directive sets the visibility to
5336 @code{hidden} which means that the symbols are not visible to other components.
5337 Such symbols are always considered to be @code{protected} as well.
5341 @section @code{.hword @var{expressions}}
5343 @cindex @code{hword} directive
5344 @cindex integers, 16-bit
5345 @cindex numbers, 16-bit
5346 @cindex sixteen bit integers
5347 This expects zero or more @var{expressions}, and emits
5348 a 16 bit number for each.
5351 This directive is a synonym for @samp{.short}; depending on the target
5352 architecture, it may also be a synonym for @samp{.word}.
5356 This directive is a synonym for @samp{.short}.
5359 This directive is a synonym for both @samp{.short} and @samp{.word}.
5364 @section @code{.ident}
5366 @cindex @code{ident} directive
5368 This directive is used by some assemblers to place tags in object files. The
5369 behavior of this directive varies depending on the target. When using the
5370 a.out object file format, @command{@value{AS}} simply accepts the directive for
5371 source-file compatibility with existing assemblers, but does not emit anything
5372 for it. When using COFF, comments are emitted to the @code{.comment} or
5373 @code{.rdata} section, depending on the target. When using ELF, comments are
5374 emitted to the @code{.comment} section.
5377 @section @code{.if @var{absolute expression}}
5379 @cindex conditional assembly
5380 @cindex @code{if} directive
5381 @code{.if} marks the beginning of a section of code which is only
5382 considered part of the source program being assembled if the argument
5383 (which must be an @var{absolute expression}) is non-zero. The end of
5384 the conditional section of code must be marked by @code{.endif}
5385 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5386 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5387 If you have several conditions to check, @code{.elseif} may be used to avoid
5388 nesting blocks if/else within each subsequent @code{.else} block.
5390 The following variants of @code{.if} are also supported:
5392 @cindex @code{ifdef} directive
5393 @item .ifdef @var{symbol}
5394 Assembles the following section of code if the specified @var{symbol}
5395 has been defined. Note a symbol which has been referenced but not yet defined
5396 is considered to be undefined.
5398 @cindex @code{ifb} directive
5399 @item .ifb @var{text}
5400 Assembles the following section of code if the operand is blank (empty).
5402 @cindex @code{ifc} directive
5403 @item .ifc @var{string1},@var{string2}
5404 Assembles the following section of code if the two strings are the same. The
5405 strings may be optionally quoted with single quotes. If they are not quoted,
5406 the first string stops at the first comma, and the second string stops at the
5407 end of the line. Strings which contain whitespace should be quoted. The
5408 string comparison is case sensitive.
5410 @cindex @code{ifeq} directive
5411 @item .ifeq @var{absolute expression}
5412 Assembles the following section of code if the argument is zero.
5414 @cindex @code{ifeqs} directive
5415 @item .ifeqs @var{string1},@var{string2}
5416 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5418 @cindex @code{ifge} directive
5419 @item .ifge @var{absolute expression}
5420 Assembles the following section of code if the argument is greater than or
5423 @cindex @code{ifgt} directive
5424 @item .ifgt @var{absolute expression}
5425 Assembles the following section of code if the argument is greater than zero.
5427 @cindex @code{ifle} directive
5428 @item .ifle @var{absolute expression}
5429 Assembles the following section of code if the argument is less than or equal
5432 @cindex @code{iflt} directive
5433 @item .iflt @var{absolute expression}
5434 Assembles the following section of code if the argument is less than zero.
5436 @cindex @code{ifnb} directive
5437 @item .ifnb @var{text}
5438 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5439 following section of code if the operand is non-blank (non-empty).
5441 @cindex @code{ifnc} directive
5442 @item .ifnc @var{string1},@var{string2}.
5443 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5444 following section of code if the two strings are not the same.
5446 @cindex @code{ifndef} directive
5447 @cindex @code{ifnotdef} directive
5448 @item .ifndef @var{symbol}
5449 @itemx .ifnotdef @var{symbol}
5450 Assembles the following section of code if the specified @var{symbol}
5451 has not been defined. Both spelling variants are equivalent. Note a symbol
5452 which has been referenced but not yet defined is considered to be undefined.
5454 @cindex @code{ifne} directive
5455 @item .ifne @var{absolute expression}
5456 Assembles the following section of code if the argument is not equal to zero
5457 (in other words, this is equivalent to @code{.if}).
5459 @cindex @code{ifnes} directive
5460 @item .ifnes @var{string1},@var{string2}
5461 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5462 following section of code if the two strings are not the same.
5466 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5468 @cindex @code{incbin} directive
5469 @cindex binary files, including
5470 The @code{incbin} directive includes @var{file} verbatim at the current
5471 location. You can control the search paths used with the @samp{-I} command-line
5472 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5475 The @var{skip} argument skips a number of bytes from the start of the
5476 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5477 read. Note that the data is not aligned in any way, so it is the user's
5478 responsibility to make sure that proper alignment is provided both before and
5479 after the @code{incbin} directive.
5482 @section @code{.include "@var{file}"}
5484 @cindex @code{include} directive
5485 @cindex supporting files, including
5486 @cindex files, including
5487 This directive provides a way to include supporting files at specified
5488 points in your source program. The code from @var{file} is assembled as
5489 if it followed the point of the @code{.include}; when the end of the
5490 included file is reached, assembly of the original file continues. You
5491 can control the search paths used with the @samp{-I} command-line option
5492 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5496 @section @code{.int @var{expressions}}
5498 @cindex @code{int} directive
5499 @cindex integers, 32-bit
5500 Expect zero or more @var{expressions}, of any section, separated by commas.
5501 For each expression, emit a number that, at run time, is the value of that
5502 expression. The byte order and bit size of the number depends on what kind
5503 of target the assembly is for.
5507 On most forms of the H8/300, @code{.int} emits 16-bit
5508 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5515 @section @code{.internal @var{names}}
5517 @cindex @code{internal} directive
5519 This is one of the ELF visibility directives. The other two are
5520 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5521 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5523 This directive overrides the named symbols default visibility (which is set by
5524 their binding: local, global or weak). The directive sets the visibility to
5525 @code{internal} which means that the symbols are considered to be @code{hidden}
5526 (i.e., not visible to other components), and that some extra, processor specific
5527 processing must also be performed upon the symbols as well.
5531 @section @code{.irp @var{symbol},@var{values}}@dots{}
5533 @cindex @code{irp} directive
5534 Evaluate a sequence of statements assigning different values to @var{symbol}.
5535 The sequence of statements starts at the @code{.irp} directive, and is
5536 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5537 set to @var{value}, and the sequence of statements is assembled. If no
5538 @var{value} is listed, the sequence of statements is assembled once, with
5539 @var{symbol} set to the null string. To refer to @var{symbol} within the
5540 sequence of statements, use @var{\symbol}.
5542 For example, assembling
5550 is equivalent to assembling
5558 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5561 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5563 @cindex @code{irpc} directive
5564 Evaluate a sequence of statements assigning different values to @var{symbol}.
5565 The sequence of statements starts at the @code{.irpc} directive, and is
5566 terminated by an @code{.endr} directive. For each character in @var{value},
5567 @var{symbol} is set to the character, and the sequence of statements is
5568 assembled. If no @var{value} is listed, the sequence of statements is
5569 assembled once, with @var{symbol} set to the null string. To refer to
5570 @var{symbol} within the sequence of statements, use @var{\symbol}.
5572 For example, assembling
5580 is equivalent to assembling
5588 For some caveats with the spelling of @var{symbol}, see also the discussion
5592 @section @code{.lcomm @var{symbol} , @var{length}}
5594 @cindex @code{lcomm} directive
5595 @cindex local common symbols
5596 @cindex symbols, local common
5597 Reserve @var{length} (an absolute expression) bytes for a local common
5598 denoted by @var{symbol}. The section and value of @var{symbol} are
5599 those of the new local common. The addresses are allocated in the bss
5600 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5601 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5602 not visible to @code{@value{LD}}.
5605 Some targets permit a third argument to be used with @code{.lcomm}. This
5606 argument specifies the desired alignment of the symbol in the bss section.
5610 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5611 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5615 @section @code{.lflags}
5617 @cindex @code{lflags} directive (ignored)
5618 @command{@value{AS}} accepts this directive, for compatibility with other
5619 assemblers, but ignores it.
5621 @ifclear no-line-dir
5623 @section @code{.line @var{line-number}}
5625 @cindex @code{line} directive
5626 @cindex logical line number
5628 Change the logical line number. @var{line-number} must be an absolute
5629 expression. The next line has that logical line number. Therefore any other
5630 statements on the current line (after a statement separator character) are
5631 reported as on logical line number @var{line-number} @minus{} 1. One day
5632 @command{@value{AS}} will no longer support this directive: it is recognized only
5633 for compatibility with existing assembler programs.
5636 Even though this is a directive associated with the @code{a.out} or
5637 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5638 when producing COFF output, and treats @samp{.line} as though it
5639 were the COFF @samp{.ln} @emph{if} it is found outside a
5640 @code{.def}/@code{.endef} pair.
5642 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5643 used by compilers to generate auxiliary symbol information for
5648 @section @code{.linkonce [@var{type}]}
5650 @cindex @code{linkonce} directive
5651 @cindex common sections
5652 Mark the current section so that the linker only includes a single copy of it.
5653 This may be used to include the same section in several different object files,
5654 but ensure that the linker will only include it once in the final output file.
5655 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5656 Duplicate sections are detected based on the section name, so it should be
5659 This directive is only supported by a few object file formats; as of this
5660 writing, the only object file format which supports it is the Portable
5661 Executable format used on Windows NT.
5663 The @var{type} argument is optional. If specified, it must be one of the
5664 following strings. For example:
5668 Not all types may be supported on all object file formats.
5672 Silently discard duplicate sections. This is the default.
5675 Warn if there are duplicate sections, but still keep only one copy.
5678 Warn if any of the duplicates have different sizes.
5681 Warn if any of the duplicates do not have exactly the same contents.
5685 @section @code{.list}
5687 @cindex @code{list} directive
5688 @cindex listing control, turning on
5689 Control (in conjunction with the @code{.nolist} directive) whether or
5690 not assembly listings are generated. These two directives maintain an
5691 internal counter (which is zero initially). @code{.list} increments the
5692 counter, and @code{.nolist} decrements it. Assembly listings are
5693 generated whenever the counter is greater than zero.
5695 By default, listings are disabled. When you enable them (with the
5696 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5697 the initial value of the listing counter is one.
5700 @section @code{.ln @var{line-number}}
5702 @cindex @code{ln} directive
5703 @ifclear no-line-dir
5704 @samp{.ln} is a synonym for @samp{.line}.
5707 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5708 must be an absolute expression. The next line has that logical
5709 line number, so any other statements on the current line (after a
5710 statement separator character @code{;}) are reported as on logical
5711 line number @var{line-number} @minus{} 1.
5715 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5716 @cindex @code{loc} directive
5717 When emitting DWARF2 line number information,
5718 the @code{.loc} directive will add a row to the @code{.debug_line} line
5719 number matrix corresponding to the immediately following assembly
5720 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5721 arguments will be applied to the @code{.debug_line} state machine before
5724 The @var{options} are a sequence of the following tokens in any order:
5728 This option will set the @code{basic_block} register in the
5729 @code{.debug_line} state machine to @code{true}.
5732 This option will set the @code{prologue_end} register in the
5733 @code{.debug_line} state machine to @code{true}.
5735 @item epilogue_begin
5736 This option will set the @code{epilogue_begin} register in the
5737 @code{.debug_line} state machine to @code{true}.
5739 @item is_stmt @var{value}
5740 This option will set the @code{is_stmt} register in the
5741 @code{.debug_line} state machine to @code{value}, which must be
5744 @item isa @var{value}
5745 This directive will set the @code{isa} register in the @code{.debug_line}
5746 state machine to @var{value}, which must be an unsigned integer.
5748 @item discriminator @var{value}
5749 This directive will set the @code{discriminator} register in the @code{.debug_line}
5750 state machine to @var{value}, which must be an unsigned integer.
5752 @item view @var{value}
5753 This option causes a row to be added to @code{.debug_line} in reference to the
5754 current address (which might not be the same as that of the following assembly
5755 instruction), and to associate @var{value} with the @code{view} register in the
5756 @code{.debug_line} state machine. If @var{value} is a label, both the
5757 @code{view} register and the label are set to the number of prior @code{.loc}
5758 directives at the same program location. If @var{value} is the literal
5759 @code{0}, the @code{view} register is set to zero, and the assembler asserts
5760 that there aren't any prior @code{.loc} directives at the same program
5761 location. If @var{value} is the literal @code{-0}, the assembler arrange for
5762 the @code{view} register to be reset in this row, even if there are prior
5763 @code{.loc} directives at the same program location.
5767 @node Loc_mark_labels
5768 @section @code{.loc_mark_labels @var{enable}}
5769 @cindex @code{loc_mark_labels} directive
5770 When emitting DWARF2 line number information,
5771 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5772 to the @code{.debug_line} line number matrix with the @code{basic_block}
5773 register in the state machine set whenever a code label is seen.
5774 The @var{enable} argument should be either 1 or 0, to enable or disable
5775 this function respectively.
5779 @section @code{.local @var{names}}
5781 @cindex @code{local} directive
5782 This directive, which is available for ELF targets, marks each symbol in
5783 the comma-separated list of @code{names} as a local symbol so that it
5784 will not be externally visible. If the symbols do not already exist,
5785 they will be created.
5787 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5788 accept an alignment argument, which is the case for most ELF targets,
5789 the @code{.local} directive can be used in combination with @code{.comm}
5790 (@pxref{Comm}) to define aligned local common data.
5794 @section @code{.long @var{expressions}}
5796 @cindex @code{long} directive
5797 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5800 @c no one seems to know what this is for or whether this description is
5801 @c what it really ought to do
5803 @section @code{.lsym @var{symbol}, @var{expression}}
5805 @cindex @code{lsym} directive
5806 @cindex symbol, not referenced in assembly
5807 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5808 the hash table, ensuring it cannot be referenced by name during the
5809 rest of the assembly. This sets the attributes of the symbol to be
5810 the same as the expression value:
5812 @var{other} = @var{descriptor} = 0
5813 @var{type} = @r{(section of @var{expression})}
5814 @var{value} = @var{expression}
5817 The new symbol is not flagged as external.
5821 @section @code{.macro}
5824 The commands @code{.macro} and @code{.endm} allow you to define macros that
5825 generate assembly output. For example, this definition specifies a macro
5826 @code{sum} that puts a sequence of numbers into memory:
5829 .macro sum from=0, to=5
5838 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5850 @item .macro @var{macname}
5851 @itemx .macro @var{macname} @var{macargs} @dots{}
5852 @cindex @code{macro} directive
5853 Begin the definition of a macro called @var{macname}. If your macro
5854 definition requires arguments, specify their names after the macro name,
5855 separated by commas or spaces. You can qualify the macro argument to
5856 indicate whether all invocations must specify a non-blank value (through
5857 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5858 (through @samp{:@code{vararg}}). You can supply a default value for any
5859 macro argument by following the name with @samp{=@var{deflt}}. You
5860 cannot define two macros with the same @var{macname} unless it has been
5861 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5862 definitions. For example, these are all valid @code{.macro} statements:
5866 Begin the definition of a macro called @code{comm}, which takes no
5869 @item .macro plus1 p, p1
5870 @itemx .macro plus1 p p1
5871 Either statement begins the definition of a macro called @code{plus1},
5872 which takes two arguments; within the macro definition, write
5873 @samp{\p} or @samp{\p1} to evaluate the arguments.
5875 @item .macro reserve_str p1=0 p2
5876 Begin the definition of a macro called @code{reserve_str}, with two
5877 arguments. The first argument has a default value, but not the second.
5878 After the definition is complete, you can call the macro either as
5879 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5880 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5881 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5882 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5884 @item .macro m p1:req, p2=0, p3:vararg
5885 Begin the definition of a macro called @code{m}, with at least three
5886 arguments. The first argument must always have a value specified, but
5887 not the second, which instead has a default value. The third formal
5888 will get assigned all remaining arguments specified at invocation time.
5890 When you call a macro, you can specify the argument values either by
5891 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5892 @samp{sum to=17, from=9}.
5896 Note that since each of the @var{macargs} can be an identifier exactly
5897 as any other one permitted by the target architecture, there may be
5898 occasional problems if the target hand-crafts special meanings to certain
5899 characters when they occur in a special position. For example, if the colon
5900 (@code{:}) is generally permitted to be part of a symbol name, but the
5901 architecture specific code special-cases it when occurring as the final
5902 character of a symbol (to denote a label), then the macro parameter
5903 replacement code will have no way of knowing that and consider the whole
5904 construct (including the colon) an identifier, and check only this
5905 identifier for being the subject to parameter substitution. So for example
5906 this macro definition:
5914 might not work as expected. Invoking @samp{label foo} might not create a label
5915 called @samp{foo} but instead just insert the text @samp{\l:} into the
5916 assembler source, probably generating an error about an unrecognised
5919 Similarly problems might occur with the period character (@samp{.})
5920 which is often allowed inside opcode names (and hence identifier names). So
5921 for example constructing a macro to build an opcode from a base name and a
5922 length specifier like this:
5925 .macro opcode base length
5930 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5931 instruction but instead generate some kind of error as the assembler tries to
5932 interpret the text @samp{\base.\length}.
5934 There are several possible ways around this problem:
5937 @item Insert white space
5938 If it is possible to use white space characters then this is the simplest
5947 @item Use @samp{\()}
5948 The string @samp{\()} can be used to separate the end of a macro argument from
5949 the following text. eg:
5952 .macro opcode base length
5957 @item Use the alternate macro syntax mode
5958 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5959 used as a separator. eg:
5969 Note: this problem of correctly identifying string parameters to pseudo ops
5970 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5971 and @code{.irpc} (@pxref{Irpc}) as well.
5974 @cindex @code{endm} directive
5975 Mark the end of a macro definition.
5978 @cindex @code{exitm} directive
5979 Exit early from the current macro definition.
5981 @cindex number of macros executed
5982 @cindex macros, count executed
5984 @command{@value{AS}} maintains a counter of how many macros it has
5985 executed in this pseudo-variable; you can copy that number to your
5986 output with @samp{\@@}, but @emph{only within a macro definition}.
5988 @item LOCAL @var{name} [ , @dots{} ]
5989 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5990 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5991 @xref{Altmacro,,@code{.altmacro}}.
5995 @section @code{.mri @var{val}}
5997 @cindex @code{mri} directive
5998 @cindex MRI mode, temporarily
5999 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
6000 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
6001 affects code assembled until the next @code{.mri} directive, or until the end
6002 of the file. @xref{M, MRI mode, MRI mode}.
6005 @section @code{.noaltmacro}
6006 Disable alternate macro mode. @xref{Altmacro}.
6009 @section @code{.nolist}
6011 @cindex @code{nolist} directive
6012 @cindex listing control, turning off
6013 Control (in conjunction with the @code{.list} directive) whether or
6014 not assembly listings are generated. These two directives maintain an
6015 internal counter (which is zero initially). @code{.list} increments the
6016 counter, and @code{.nolist} decrements it. Assembly listings are
6017 generated whenever the counter is greater than zero.
6020 @section @code{.nops @var{size}[, @var{control}]}
6022 @cindex @code{nops} directive
6023 @cindex filling memory with no-op instructions
6024 This directive emits @var{size} bytes filled with no-op instructions.
6025 @var{size} is absolute expression, which must be a positve value.
6026 @var{control} controls how no-op instructions should be generated. If
6027 the comma and @var{control} are omitted, @var{control} is assumed to be
6030 Note: For Intel 80386 and AMD x86-64 targets, @var{control} specifies
6031 the size limit of a no-op instruction. The valid values of @var{control}
6032 are between 0 and 4 in 16-bit mode, between 0 and 7 when tuning for
6033 older processors in 32-bit mode, between 0 and 11 in 64-bit mode or when
6034 tuning for newer processors in 32-bit mode. When 0 is used, the no-op
6035 instruction size limit is set to the maximum supported size.
6038 @section @code{.octa @var{bignums}}
6040 @c FIXME: double size emitted for "octa" on some? Or warn?
6041 @cindex @code{octa} directive
6042 @cindex integer, 16-byte
6043 @cindex sixteen byte integer
6044 This directive expects zero or more bignums, separated by commas. For each
6045 bignum, it emits a 16-byte integer.
6047 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
6048 hence @emph{octa}-word for 16 bytes.
6051 @section @code{.offset @var{loc}}
6053 @cindex @code{offset} directive
6054 Set the location counter to @var{loc} in the absolute section. @var{loc} must
6055 be an absolute expression. This directive may be useful for defining
6056 symbols with absolute values. Do not confuse it with the @code{.org}
6060 @section @code{.org @var{new-lc} , @var{fill}}
6062 @cindex @code{org} directive
6063 @cindex location counter, advancing
6064 @cindex advancing location counter
6065 @cindex current address, advancing
6066 Advance the location counter of the current section to
6067 @var{new-lc}. @var{new-lc} is either an absolute expression or an
6068 expression with the same section as the current subsection. That is,
6069 you can't use @code{.org} to cross sections: if @var{new-lc} has the
6070 wrong section, the @code{.org} directive is ignored. To be compatible
6071 with former assemblers, if the section of @var{new-lc} is absolute,
6072 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
6073 is the same as the current subsection.
6075 @code{.org} may only increase the location counter, or leave it
6076 unchanged; you cannot use @code{.org} to move the location counter
6079 @c double negative used below "not undefined" because this is a specific
6080 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
6081 @c section. doc@cygnus.com 18feb91
6082 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
6083 may not be undefined. If you really detest this restriction we eagerly await
6084 a chance to share your improved assembler.
6086 Beware that the origin is relative to the start of the section, not
6087 to the start of the subsection. This is compatible with other
6088 people's assemblers.
6090 When the location counter (of the current subsection) is advanced, the
6091 intervening bytes are filled with @var{fill} which should be an
6092 absolute expression. If the comma and @var{fill} are omitted,
6093 @var{fill} defaults to zero.
6096 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
6098 @cindex padding the location counter given a power of two
6099 @cindex @code{p2align} directive
6100 Pad the location counter (in the current subsection) to a particular
6101 storage boundary. The first expression (which must be absolute) is the
6102 number of low-order zero bits the location counter must have after
6103 advancement. For example @samp{.p2align 3} advances the location
6104 counter until it a multiple of 8. If the location counter is already a
6105 multiple of 8, no change is needed.
6107 The second expression (also absolute) gives the fill value to be stored in the
6108 padding bytes. It (and the comma) may be omitted. If it is omitted, the
6109 padding bytes are normally zero. However, on most systems, if the section is
6110 marked as containing code and the fill value is omitted, the space is filled
6111 with no-op instructions.
6113 The third expression is also absolute, and is also optional. If it is present,
6114 it is the maximum number of bytes that should be skipped by this alignment
6115 directive. If doing the alignment would require skipping more bytes than the
6116 specified maximum, then the alignment is not done at all. You can omit the
6117 fill value (the second argument) entirely by simply using two commas after the
6118 required alignment; this can be useful if you want the alignment to be filled
6119 with no-op instructions when appropriate.
6121 @cindex @code{p2alignw} directive
6122 @cindex @code{p2alignl} directive
6123 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
6124 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
6125 pattern as a two byte word value. The @code{.p2alignl} directives treats the
6126 fill pattern as a four byte longword value. For example, @code{.p2alignw
6127 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
6128 filled in with the value 0x368d (the exact placement of the bytes depends upon
6129 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
6134 @section @code{.popsection}
6136 @cindex @code{popsection} directive
6137 @cindex Section Stack
6138 This is one of the ELF section stack manipulation directives. The others are
6139 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6140 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
6143 This directive replaces the current section (and subsection) with the top
6144 section (and subsection) on the section stack. This section is popped off the
6150 @section @code{.previous}
6152 @cindex @code{previous} directive
6153 @cindex Section Stack
6154 This is one of the ELF section stack manipulation directives. The others are
6155 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6156 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
6157 (@pxref{PopSection}).
6159 This directive swaps the current section (and subsection) with most recently
6160 referenced section/subsection pair prior to this one. Multiple
6161 @code{.previous} directives in a row will flip between two sections (and their
6162 subsections). For example:
6174 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
6180 # Now in section A subsection 1
6184 # Now in section B subsection 0
6187 # Now in section B subsection 1
6190 # Now in section B subsection 0
6194 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
6195 section B and 0x9abc into subsection 1 of section B.
6197 In terms of the section stack, this directive swaps the current section with
6198 the top section on the section stack.
6202 @section @code{.print @var{string}}
6204 @cindex @code{print} directive
6205 @command{@value{AS}} will print @var{string} on the standard output during
6206 assembly. You must put @var{string} in double quotes.
6210 @section @code{.protected @var{names}}
6212 @cindex @code{protected} directive
6214 This is one of the ELF visibility directives. The other two are
6215 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
6217 This directive overrides the named symbols default visibility (which is set by
6218 their binding: local, global or weak). The directive sets the visibility to
6219 @code{protected} which means that any references to the symbols from within the
6220 components that defines them must be resolved to the definition in that
6221 component, even if a definition in another component would normally preempt
6226 @section @code{.psize @var{lines} , @var{columns}}
6228 @cindex @code{psize} directive
6229 @cindex listing control: paper size
6230 @cindex paper size, for listings
6231 Use this directive to declare the number of lines---and, optionally, the
6232 number of columns---to use for each page, when generating listings.
6234 If you do not use @code{.psize}, listings use a default line-count
6235 of 60. You may omit the comma and @var{columns} specification; the
6236 default width is 200 columns.
6238 @command{@value{AS}} generates formfeeds whenever the specified number of
6239 lines is exceeded (or whenever you explicitly request one, using
6242 If you specify @var{lines} as @code{0}, no formfeeds are generated save
6243 those explicitly specified with @code{.eject}.
6246 @section @code{.purgem @var{name}}
6248 @cindex @code{purgem} directive
6249 Undefine the macro @var{name}, so that later uses of the string will not be
6250 expanded. @xref{Macro}.
6254 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
6256 @cindex @code{pushsection} directive
6257 @cindex Section Stack
6258 This is one of the ELF section stack manipulation directives. The others are
6259 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6260 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6263 This directive pushes the current section (and subsection) onto the
6264 top of the section stack, and then replaces the current section and
6265 subsection with @code{name} and @code{subsection}. The optional
6266 @code{flags}, @code{type} and @code{arguments} are treated the same
6267 as in the @code{.section} (@pxref{Section}) directive.
6271 @section @code{.quad @var{bignums}}
6273 @cindex @code{quad} directive
6274 @code{.quad} expects zero or more bignums, separated by commas. For
6275 each bignum, it emits
6277 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
6278 warning message; and just takes the lowest order 8 bytes of the bignum.
6279 @cindex eight-byte integer
6280 @cindex integer, 8-byte
6282 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
6283 hence @emph{quad}-word for 8 bytes.
6286 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
6287 warning message; and just takes the lowest order 16 bytes of the bignum.
6288 @cindex sixteen-byte integer
6289 @cindex integer, 16-byte
6293 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6295 @cindex @code{reloc} directive
6296 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6297 @var{expression}. If @var{offset} is a number, the relocation is generated in
6298 the current section. If @var{offset} is an expression that resolves to a
6299 symbol plus offset, the relocation is generated in the given symbol's section.
6300 @var{expression}, if present, must resolve to a symbol plus addend or to an
6301 absolute value, but note that not all targets support an addend. e.g. ELF REL
6302 targets such as i386 store an addend in the section contents rather than in the
6303 relocation. This low level interface does not support addends stored in the
6307 @section @code{.rept @var{count}}
6309 @cindex @code{rept} directive
6310 Repeat the sequence of lines between the @code{.rept} directive and the next
6311 @code{.endr} directive @var{count} times.
6313 For example, assembling
6321 is equivalent to assembling
6329 A count of zero is allowed, but nothing is generated. Negative counts are not
6330 allowed and if encountered will be treated as if they were zero.
6333 @section @code{.sbttl "@var{subheading}"}
6335 @cindex @code{sbttl} directive
6336 @cindex subtitles for listings
6337 @cindex listing control: subtitle
6338 Use @var{subheading} as the title (third line, immediately after the
6339 title line) when generating assembly listings.
6341 This directive affects subsequent pages, as well as the current page if
6342 it appears within ten lines of the top of a page.
6346 @section @code{.scl @var{class}}
6348 @cindex @code{scl} directive
6349 @cindex symbol storage class (COFF)
6350 @cindex COFF symbol storage class
6351 Set the storage-class value for a symbol. This directive may only be
6352 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6353 whether a symbol is static or external, or it may record further
6354 symbolic debugging information.
6359 @section @code{.section @var{name}}
6361 @cindex named section
6362 Use the @code{.section} directive to assemble the following code into a section
6365 This directive is only supported for targets that actually support arbitrarily
6366 named sections; on @code{a.out} targets, for example, it is not accepted, even
6367 with a standard @code{a.out} section name.
6371 @c only print the extra heading if both COFF and ELF are set
6372 @subheading COFF Version
6375 @cindex @code{section} directive (COFF version)
6376 For COFF targets, the @code{.section} directive is used in one of the following
6380 .section @var{name}[, "@var{flags}"]
6381 .section @var{name}[, @var{subsection}]
6384 If the optional argument is quoted, it is taken as flags to use for the
6385 section. Each flag is a single character. The following flags are recognized:
6389 bss section (uninitialized data)
6391 section is not loaded
6397 exclude section from linking
6403 shared section (meaningful for PE targets)
6405 ignored. (For compatibility with the ELF version)
6407 section is not readable (meaningful for PE targets)
6409 single-digit power-of-two section alignment (GNU extension)
6412 If no flags are specified, the default flags depend upon the section name. If
6413 the section name is not recognized, the default will be for the section to be
6414 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6415 from the section, rather than adding them, so if they are used on their own it
6416 will be as if no flags had been specified at all.
6418 If the optional argument to the @code{.section} directive is not quoted, it is
6419 taken as a subsection number (@pxref{Sub-Sections}).
6424 @c only print the extra heading if both COFF and ELF are set
6425 @subheading ELF Version
6428 @cindex Section Stack
6429 This is one of the ELF section stack manipulation directives. The others are
6430 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6431 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6432 @code{.previous} (@pxref{Previous}).
6434 @cindex @code{section} directive (ELF version)
6435 For ELF targets, the @code{.section} directive is used like this:
6438 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6441 @anchor{Section Name Substitutions}
6442 @kindex --sectname-subst
6443 @cindex section name substitution
6444 If the @samp{--sectname-subst} command-line option is provided, the @var{name}
6445 argument may contain a substitution sequence. Only @code{%S} is supported
6446 at the moment, and substitutes the current section name. For example:
6449 .macro exception_code
6450 .section %S.exception
6451 [exception code here]
6466 The two @code{exception_code} invocations above would create the
6467 @code{.text.exception} and @code{.init.exception} sections respectively.
6468 This is useful e.g. to discriminate between ancillary sections that are
6469 tied to setup code to be discarded after use from ancillary sections that
6470 need to stay resident without having to define multiple @code{exception_code}
6471 macros just for that purpose.
6473 The optional @var{flags} argument is a quoted string which may contain any
6474 combination of the following characters:
6478 section is allocatable
6480 section is a GNU_MBIND section
6482 section is excluded from executable and shared library.
6486 section is executable
6488 section is mergeable
6490 section contains zero terminated strings
6492 section is a member of a section group
6494 section is used for thread-local-storage
6496 section is a member of the previously-current section's group, if any
6497 @item @code{<number>}
6498 a numeric value indicating the bits to be set in the ELF section header's flags
6499 field. Note - if one or more of the alphabetic characters described above is
6500 also included in the flags field, their bit values will be ORed into the
6502 @item @code{<target specific>}
6503 some targets extend this list with their own flag characters
6506 Note - once a section's flags have been set they cannot be changed. There are
6507 a few exceptions to this rule however. Processor and application specific
6508 flags can be added to an already defined section. The @code{.interp},
6509 @code{.strtab} and @code{.symtab} sections can have the allocate flag
6510 (@code{a}) set after they are initially defined, and the @code{.note-GNU-stack}
6511 section may have the executable (@code{x}) flag added.
6513 The optional @var{type} argument may contain one of the following constants:
6517 section contains data
6519 section does not contain data (i.e., section only occupies space)
6521 section contains data which is used by things other than the program
6523 section contains an array of pointers to init functions
6525 section contains an array of pointers to finish functions
6526 @item @@preinit_array
6527 section contains an array of pointers to pre-init functions
6528 @item @@@code{<number>}
6529 a numeric value to be set as the ELF section header's type field.
6530 @item @@@code{<target specific>}
6531 some targets extend this list with their own types
6534 Many targets only support the first three section types. The type may be
6535 enclosed in double quotes if necessary.
6537 Note on targets where the @code{@@} character is the start of a comment (eg
6538 ARM) then another character is used instead. For example the ARM port uses the
6541 Note - some sections, eg @code{.text} and @code{.data} are considered to be
6542 special and have fixed types. Any attempt to declare them with a different
6543 type will generate an error from the assembler.
6545 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6546 be specified as well as an extra argument---@var{entsize}---like this:
6549 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6552 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6553 constants, each @var{entsize} octets long. Sections with both @code{M} and
6554 @code{S} must contain zero terminated strings where each character is
6555 @var{entsize} bytes long. The linker may remove duplicates within sections with
6556 the same name, same entity size and same flags. @var{entsize} must be an
6557 absolute expression. For sections with both @code{M} and @code{S}, a string
6558 which is a suffix of a larger string is considered a duplicate. Thus
6559 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6560 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6562 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6563 be present along with an additional field like this:
6566 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6569 The @var{GroupName} field specifies the name of the section group to which this
6570 particular section belongs. The optional linkage field can contain:
6574 indicates that only one copy of this section should be retained
6579 Note: if both the @var{M} and @var{G} flags are present then the fields for
6580 the Merge flag should come first, like this:
6583 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6586 If @var{flags} contains the @code{?} symbol then it may not also contain the
6587 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6588 present. Instead, @code{?} says to consider the section that's current before
6589 this directive. If that section used @code{G}, then the new section will use
6590 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6591 If not, then the @code{?} symbol has no effect.
6593 If no flags are specified, the default flags depend upon the section name. If
6594 the section name is not recognized, the default will be for the section to have
6595 none of the above flags: it will not be allocated in memory, nor writable, nor
6596 executable. The section will contain data.
6598 For ELF targets, the assembler supports another type of @code{.section}
6599 directive for compatibility with the Solaris assembler:
6602 .section "@var{name}"[, @var{flags}...]
6605 Note that the section name is quoted. There may be a sequence of comma
6610 section is allocatable
6614 section is executable
6616 section is excluded from executable and shared library.
6618 section is used for thread local storage
6621 This directive replaces the current section and subsection. See the
6622 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6623 some examples of how this directive and the other section stack directives
6629 @section @code{.set @var{symbol}, @var{expression}}
6631 @cindex @code{set} directive
6632 @cindex symbol value, setting
6633 Set the value of @var{symbol} to @var{expression}. This
6634 changes @var{symbol}'s value and type to conform to
6635 @var{expression}. If @var{symbol} was flagged as external, it remains
6636 flagged (@pxref{Symbol Attributes}).
6638 You may @code{.set} a symbol many times in the same assembly provided that the
6639 values given to the symbol are constants. Values that are based on expressions
6640 involving other symbols are allowed, but some targets may restrict this to only
6641 being done once per assembly. This is because those targets do not set the
6642 addresses of symbols at assembly time, but rather delay the assignment until a
6643 final link is performed. This allows the linker a chance to change the code in
6644 the files, changing the location of, and the relative distance between, various
6647 If you @code{.set} a global symbol, the value stored in the object
6648 file is the last value stored into it.
6651 On Z80 @code{set} is a real instruction, use
6652 @samp{@var{symbol} defl @var{expression}} instead.
6656 @section @code{.short @var{expressions}}
6658 @cindex @code{short} directive
6660 @code{.short} is normally the same as @samp{.word}.
6661 @xref{Word,,@code{.word}}.
6663 In some configurations, however, @code{.short} and @code{.word} generate
6664 numbers of different lengths. @xref{Machine Dependencies}.
6668 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6671 This expects zero or more @var{expressions}, and emits
6672 a 16 bit number for each.
6677 @section @code{.single @var{flonums}}
6679 @cindex @code{single} directive
6680 @cindex floating point numbers (single)
6681 This directive assembles zero or more flonums, separated by commas. It
6682 has the same effect as @code{.float}.
6684 The exact kind of floating point numbers emitted depends on how
6685 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6689 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6690 numbers in @sc{ieee} format.
6696 @section @code{.size}
6698 This directive is used to set the size associated with a symbol.
6702 @c only print the extra heading if both COFF and ELF are set
6703 @subheading COFF Version
6706 @cindex @code{size} directive (COFF version)
6707 For COFF targets, the @code{.size} directive is only permitted inside
6708 @code{.def}/@code{.endef} pairs. It is used like this:
6711 .size @var{expression}
6718 @c only print the extra heading if both COFF and ELF are set
6719 @subheading ELF Version
6722 @cindex @code{size} directive (ELF version)
6723 For ELF targets, the @code{.size} directive is used like this:
6726 .size @var{name} , @var{expression}
6729 This directive sets the size associated with a symbol @var{name}.
6730 The size in bytes is computed from @var{expression} which can make use of label
6731 arithmetic. This directive is typically used to set the size of function
6736 @ifclear no-space-dir
6738 @section @code{.skip @var{size} [,@var{fill}]}
6740 @cindex @code{skip} directive
6741 @cindex filling memory
6742 This directive emits @var{size} bytes, each of value @var{fill}. Both
6743 @var{size} and @var{fill} are absolute expressions. If the comma and
6744 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6749 @section @code{.sleb128 @var{expressions}}
6751 @cindex @code{sleb128} directive
6752 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6753 compact, variable length representation of numbers used by the DWARF
6754 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6756 @ifclear no-space-dir
6758 @section @code{.space @var{size} [,@var{fill}]}
6760 @cindex @code{space} directive
6761 @cindex filling memory
6762 This directive emits @var{size} bytes, each of value @var{fill}. Both
6763 @var{size} and @var{fill} are absolute expressions. If the comma
6764 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6769 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6770 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6771 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6772 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6780 @section @code{.stabd, .stabn, .stabs}
6782 @cindex symbolic debuggers, information for
6783 @cindex @code{stab@var{x}} directives
6784 There are three directives that begin @samp{.stab}.
6785 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6786 The symbols are not entered in the @command{@value{AS}} hash table: they
6787 cannot be referenced elsewhere in the source file.
6788 Up to five fields are required:
6792 This is the symbol's name. It may contain any character except
6793 @samp{\000}, so is more general than ordinary symbol names. Some
6794 debuggers used to code arbitrarily complex structures into symbol names
6798 An absolute expression. The symbol's type is set to the low 8 bits of
6799 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6800 and debuggers choke on silly bit patterns.
6803 An absolute expression. The symbol's ``other'' attribute is set to the
6804 low 8 bits of this expression.
6807 An absolute expression. The symbol's descriptor is set to the low 16
6808 bits of this expression.
6811 An absolute expression which becomes the symbol's value.
6814 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6815 or @code{.stabs} statement, the symbol has probably already been created;
6816 you get a half-formed symbol in your object file. This is
6817 compatible with earlier assemblers!
6820 @cindex @code{stabd} directive
6821 @item .stabd @var{type} , @var{other} , @var{desc}
6823 The ``name'' of the symbol generated is not even an empty string.
6824 It is a null pointer, for compatibility. Older assemblers used a
6825 null pointer so they didn't waste space in object files with empty
6828 The symbol's value is set to the location counter,
6829 relocatably. When your program is linked, the value of this symbol
6830 is the address of the location counter when the @code{.stabd} was
6833 @cindex @code{stabn} directive
6834 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6835 The name of the symbol is set to the empty string @code{""}.
6837 @cindex @code{stabs} directive
6838 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6839 All five fields are specified.
6845 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6846 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6848 @cindex string, copying to object file
6849 @cindex string8, copying to object file
6850 @cindex string16, copying to object file
6851 @cindex string32, copying to object file
6852 @cindex string64, copying to object file
6853 @cindex @code{string} directive
6854 @cindex @code{string8} directive
6855 @cindex @code{string16} directive
6856 @cindex @code{string32} directive
6857 @cindex @code{string64} directive
6859 Copy the characters in @var{str} to the object file. You may specify more than
6860 one string to copy, separated by commas. Unless otherwise specified for a
6861 particular machine, the assembler marks the end of each string with a 0 byte.
6862 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6864 The variants @code{string16}, @code{string32} and @code{string64} differ from
6865 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6866 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6867 are stored in target endianness byte order.
6873 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6874 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6879 @section @code{.struct @var{expression}}
6881 @cindex @code{struct} directive
6882 Switch to the absolute section, and set the section offset to @var{expression},
6883 which must be an absolute expression. You might use this as follows:
6892 This would define the symbol @code{field1} to have the value 0, the symbol
6893 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6894 value 8. Assembly would be left in the absolute section, and you would need to
6895 use a @code{.section} directive of some sort to change to some other section
6896 before further assembly.
6900 @section @code{.subsection @var{name}}
6902 @cindex @code{subsection} directive
6903 @cindex Section Stack
6904 This is one of the ELF section stack manipulation directives. The others are
6905 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6906 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6909 This directive replaces the current subsection with @code{name}. The current
6910 section is not changed. The replaced subsection is put onto the section stack
6911 in place of the then current top of stack subsection.
6916 @section @code{.symver}
6917 @cindex @code{symver} directive
6918 @cindex symbol versioning
6919 @cindex versions of symbols
6920 Use the @code{.symver} directive to bind symbols to specific version nodes
6921 within a source file. This is only supported on ELF platforms, and is
6922 typically used when assembling files to be linked into a shared library.
6923 There are cases where it may make sense to use this in objects to be bound
6924 into an application itself so as to override a versioned symbol from a
6927 For ELF targets, the @code{.symver} directive can be used like this:
6929 .symver @var{name}, @var{name2@@nodename}
6931 If the symbol @var{name} is defined within the file
6932 being assembled, the @code{.symver} directive effectively creates a symbol
6933 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6934 just don't try and create a regular alias is that the @var{@@} character isn't
6935 permitted in symbol names. The @var{name2} part of the name is the actual name
6936 of the symbol by which it will be externally referenced. The name @var{name}
6937 itself is merely a name of convenience that is used so that it is possible to
6938 have definitions for multiple versions of a function within a single source
6939 file, and so that the compiler can unambiguously know which version of a
6940 function is being mentioned. The @var{nodename} portion of the alias should be
6941 the name of a node specified in the version script supplied to the linker when
6942 building a shared library. If you are attempting to override a versioned
6943 symbol from a shared library, then @var{nodename} should correspond to the
6944 nodename of the symbol you are trying to override.
6946 If the symbol @var{name} is not defined within the file being assembled, all
6947 references to @var{name} will be changed to @var{name2@@nodename}. If no
6948 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6951 Another usage of the @code{.symver} directive is:
6953 .symver @var{name}, @var{name2@@@@nodename}
6955 In this case, the symbol @var{name} must exist and be defined within
6956 the file being assembled. It is similar to @var{name2@@nodename}. The
6957 difference is @var{name2@@@@nodename} will also be used to resolve
6958 references to @var{name2} by the linker.
6960 The third usage of the @code{.symver} directive is:
6962 .symver @var{name}, @var{name2@@@@@@nodename}
6964 When @var{name} is not defined within the
6965 file being assembled, it is treated as @var{name2@@nodename}. When
6966 @var{name} is defined within the file being assembled, the symbol
6967 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6972 @section @code{.tag @var{structname}}
6974 @cindex COFF structure debugging
6975 @cindex structure debugging, COFF
6976 @cindex @code{tag} directive
6977 This directive is generated by compilers to include auxiliary debugging
6978 information in the symbol table. It is only permitted inside
6979 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6980 definitions in the symbol table with instances of those structures.
6984 @section @code{.text @var{subsection}}
6986 @cindex @code{text} directive
6987 Tells @command{@value{AS}} to assemble the following statements onto the end of
6988 the text subsection numbered @var{subsection}, which is an absolute
6989 expression. If @var{subsection} is omitted, subsection number zero
6993 @section @code{.title "@var{heading}"}
6995 @cindex @code{title} directive
6996 @cindex listing control: title line
6997 Use @var{heading} as the title (second line, immediately after the
6998 source file name and pagenumber) when generating assembly listings.
7000 This directive affects subsequent pages, as well as the current page if
7001 it appears within ten lines of the top of a page.
7005 @section @code{.type}
7007 This directive is used to set the type of a symbol.
7011 @c only print the extra heading if both COFF and ELF are set
7012 @subheading COFF Version
7015 @cindex COFF symbol type
7016 @cindex symbol type, COFF
7017 @cindex @code{type} directive (COFF version)
7018 For COFF targets, this directive is permitted only within
7019 @code{.def}/@code{.endef} pairs. It is used like this:
7025 This records the integer @var{int} as the type attribute of a symbol table
7032 @c only print the extra heading if both COFF and ELF are set
7033 @subheading ELF Version
7036 @cindex ELF symbol type
7037 @cindex symbol type, ELF
7038 @cindex @code{type} directive (ELF version)
7039 For ELF targets, the @code{.type} directive is used like this:
7042 .type @var{name} , @var{type description}
7045 This sets the type of symbol @var{name} to be either a
7046 function symbol or an object symbol. There are five different syntaxes
7047 supported for the @var{type description} field, in order to provide
7048 compatibility with various other assemblers.
7050 Because some of the characters used in these syntaxes (such as @samp{@@} and
7051 @samp{#}) are comment characters for some architectures, some of the syntaxes
7052 below do not work on all architectures. The first variant will be accepted by
7053 the GNU assembler on all architectures so that variant should be used for
7054 maximum portability, if you do not need to assemble your code with other
7057 The syntaxes supported are:
7060 .type <name> STT_<TYPE_IN_UPPER_CASE>
7061 .type <name>,#<type>
7062 .type <name>,@@<type>
7063 .type <name>,%<type>
7064 .type <name>,"<type>"
7067 The types supported are:
7072 Mark the symbol as being a function name.
7075 @itemx gnu_indirect_function
7076 Mark the symbol as an indirect function when evaluated during reloc
7077 processing. (This is only supported on assemblers targeting GNU systems).
7081 Mark the symbol as being a data object.
7085 Mark the symbol as being a thread-local data object.
7089 Mark the symbol as being a common data object.
7093 Does not mark the symbol in any way. It is supported just for completeness.
7095 @item gnu_unique_object
7096 Marks the symbol as being a globally unique data object. The dynamic linker
7097 will make sure that in the entire process there is just one symbol with this
7098 name and type in use. (This is only supported on assemblers targeting GNU
7103 Note: Some targets support extra types in addition to those listed above.
7109 @section @code{.uleb128 @var{expressions}}
7111 @cindex @code{uleb128} directive
7112 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
7113 compact, variable length representation of numbers used by the DWARF
7114 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
7118 @section @code{.val @var{addr}}
7120 @cindex @code{val} directive
7121 @cindex COFF value attribute
7122 @cindex value attribute, COFF
7123 This directive, permitted only within @code{.def}/@code{.endef} pairs,
7124 records the address @var{addr} as the value attribute of a symbol table
7130 @section @code{.version "@var{string}"}
7132 @cindex @code{version} directive
7133 This directive creates a @code{.note} section and places into it an ELF
7134 formatted note of type NT_VERSION. The note's name is set to @code{string}.
7139 @section @code{.vtable_entry @var{table}, @var{offset}}
7141 @cindex @code{vtable_entry} directive
7142 This directive finds or creates a symbol @code{table} and creates a
7143 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
7146 @section @code{.vtable_inherit @var{child}, @var{parent}}
7148 @cindex @code{vtable_inherit} directive
7149 This directive finds the symbol @code{child} and finds or creates the symbol
7150 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
7151 parent whose addend is the value of the child symbol. As a special case the
7152 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
7156 @section @code{.warning "@var{string}"}
7157 @cindex warning directive
7158 Similar to the directive @code{.error}
7159 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
7162 @section @code{.weak @var{names}}
7164 @cindex @code{weak} directive
7165 This directive sets the weak attribute on the comma separated list of symbol
7166 @code{names}. If the symbols do not already exist, they will be created.
7168 On COFF targets other than PE, weak symbols are a GNU extension. This
7169 directive sets the weak attribute on the comma separated list of symbol
7170 @code{names}. If the symbols do not already exist, they will be created.
7172 On the PE target, weak symbols are supported natively as weak aliases.
7173 When a weak symbol is created that is not an alias, GAS creates an
7174 alternate symbol to hold the default value.
7177 @section @code{.weakref @var{alias}, @var{target}}
7179 @cindex @code{weakref} directive
7180 This directive creates an alias to the target symbol that enables the symbol to
7181 be referenced with weak-symbol semantics, but without actually making it weak.
7182 If direct references or definitions of the symbol are present, then the symbol
7183 will not be weak, but if all references to it are through weak references, the
7184 symbol will be marked as weak in the symbol table.
7186 The effect is equivalent to moving all references to the alias to a separate
7187 assembly source file, renaming the alias to the symbol in it, declaring the
7188 symbol as weak there, and running a reloadable link to merge the object files
7189 resulting from the assembly of the new source file and the old source file that
7190 had the references to the alias removed.
7192 The alias itself never makes to the symbol table, and is entirely handled
7193 within the assembler.
7196 @section @code{.word @var{expressions}}
7198 @cindex @code{word} directive
7199 This directive expects zero or more @var{expressions}, of any section,
7200 separated by commas.
7203 For each expression, @command{@value{AS}} emits a 32-bit number.
7206 For each expression, @command{@value{AS}} emits a 16-bit number.
7211 The size of the number emitted, and its byte order,
7212 depend on what target computer the assembly is for.
7215 @c on sparc the "special treatment to support compilers" doesn't
7216 @c happen---32-bit addressability, period; no long/short jumps.
7217 @ifset DIFF-TBL-KLUGE
7218 @cindex difference tables altered
7219 @cindex altered difference tables
7221 @emph{Warning: Special Treatment to support Compilers}
7225 Machines with a 32-bit address space, but that do less than 32-bit
7226 addressing, require the following special treatment. If the machine of
7227 interest to you does 32-bit addressing (or doesn't require it;
7228 @pxref{Machine Dependencies}), you can ignore this issue.
7231 In order to assemble compiler output into something that works,
7232 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
7233 Directives of the form @samp{.word sym1-sym2} are often emitted by
7234 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
7235 directive of the form @samp{.word sym1-sym2}, and the difference between
7236 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
7237 creates a @dfn{secondary jump table}, immediately before the next label.
7238 This secondary jump table is preceded by a short-jump to the
7239 first byte after the secondary table. This short-jump prevents the flow
7240 of control from accidentally falling into the new table. Inside the
7241 table is a long-jump to @code{sym2}. The original @samp{.word}
7242 contains @code{sym1} minus the address of the long-jump to
7245 If there were several occurrences of @samp{.word sym1-sym2} before the
7246 secondary jump table, all of them are adjusted. If there was a
7247 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
7248 long-jump to @code{sym4} is included in the secondary jump table,
7249 and the @code{.word} directives are adjusted to contain @code{sym3}
7250 minus the address of the long-jump to @code{sym4}; and so on, for as many
7251 entries in the original jump table as necessary.
7254 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
7255 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
7256 assembly language programmers.
7259 @c end DIFF-TBL-KLUGE
7261 @ifclear no-space-dir
7263 @section @code{.zero @var{size}}
7265 @cindex @code{zero} directive
7266 @cindex filling memory with zero bytes
7267 This directive emits @var{size} 0-valued bytes. @var{size} must be an absolute
7268 expression. This directive is actually an alias for the @samp{.skip} directive
7269 so in can take an optional second argument of the value to store in the bytes
7270 instead of zero. Using @samp{.zero} in this way would be confusing however.
7275 @section @code{.2byte @var{expression} [, @var{expression}]*}
7276 @cindex @code{2byte} directive
7277 @cindex two-byte integer
7278 @cindex integer, 2-byte
7280 This directive expects zero or more expressions, separated by commas. If there
7281 are no expressions then the directive does nothing. Otherwise each expression
7282 is evaluated in turn and placed in the next two bytes of the current output
7283 section, using the endian model of the target. If an expression will not fit
7284 in two bytes, a warning message is displayed and the least significant two
7285 bytes of the expression's value are used. If an expression cannot be evaluated
7286 at assembly time then relocations will be generated in order to compute the
7289 This directive does not apply any alignment before or after inserting the
7290 values. As a result of this, if relocations are generated, they may be
7291 different from those used for inserting values with a guaranteed alignment.
7293 This directive is only available for ELF targets,
7296 @section @code{.4byte @var{expression} [, @var{expression}]*}
7297 @cindex @code{4byte} directive
7298 @cindex four-byte integer
7299 @cindex integer, 4-byte
7301 Like the @option{.2byte} directive, except that it inserts unaligned, four byte
7302 long values into the output.
7305 @section @code{.8byte @var{expression} [, @var{expression}]*}
7306 @cindex @code{8byte} directive
7307 @cindex eight-byte integer
7308 @cindex integer, 8-byte
7310 Like the @option{.2byte} directive, except that it inserts unaligned, eight
7311 byte long bignum values into the output.
7316 @section Deprecated Directives
7318 @cindex deprecated directives
7319 @cindex obsolescent directives
7320 One day these directives won't work.
7321 They are included for compatibility with older assemblers.
7328 @node Object Attributes
7329 @chapter Object Attributes
7330 @cindex object attributes
7332 @command{@value{AS}} assembles source files written for a specific architecture
7333 into object files for that architecture. But not all object files are alike.
7334 Many architectures support incompatible variations. For instance, floating
7335 point arguments might be passed in floating point registers if the object file
7336 requires hardware floating point support---or floating point arguments might be
7337 passed in integer registers if the object file supports processors with no
7338 hardware floating point unit. Or, if two objects are built for different
7339 generations of the same architecture, the combination may require the
7340 newer generation at run-time.
7342 This information is useful during and after linking. At link time,
7343 @command{@value{LD}} can warn about incompatible object files. After link
7344 time, tools like @command{gdb} can use it to process the linked file
7347 Compatibility information is recorded as a series of object attributes. Each
7348 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
7349 string, and indicates who sets the meaning of the tag. The tag is an integer,
7350 and indicates what property the attribute describes. The value may be a string
7351 or an integer, and indicates how the property affects this object. Missing
7352 attributes are the same as attributes with a zero value or empty string value.
7354 Object attributes were developed as part of the ABI for the ARM Architecture.
7355 The file format is documented in @cite{ELF for the ARM Architecture}.
7358 * GNU Object Attributes:: @sc{gnu} Object Attributes
7359 * Defining New Object Attributes:: Defining New Object Attributes
7362 @node GNU Object Attributes
7363 @section @sc{gnu} Object Attributes
7365 The @code{.gnu_attribute} directive records an object attribute
7366 with vendor @samp{gnu}.
7368 Except for @samp{Tag_compatibility}, which has both an integer and a string for
7369 its value, @sc{gnu} attributes have a string value if the tag number is odd and
7370 an integer value if the tag number is even. The second bit (@code{@var{tag} &
7371 2} is set for architecture-independent attributes and clear for
7372 architecture-dependent ones.
7374 @subsection Common @sc{gnu} attributes
7376 These attributes are valid on all architectures.
7379 @item Tag_compatibility (32)
7380 The compatibility attribute takes an integer flag value and a vendor name. If
7381 the flag value is 0, the file is compatible with other toolchains. If it is 1,
7382 then the file is only compatible with the named toolchain. If it is greater
7383 than 1, the file can only be processed by other toolchains under some private
7384 arrangement indicated by the flag value and the vendor name.
7387 @subsection MIPS Attributes
7390 @item Tag_GNU_MIPS_ABI_FP (4)
7391 The floating-point ABI used by this object file. The value will be:
7395 0 for files not affected by the floating-point ABI.
7397 1 for files using the hardware floating-point ABI with a standard
7398 double-precision FPU.
7400 2 for files using the hardware floating-point ABI with a single-precision FPU.
7402 3 for files using the software floating-point ABI.
7404 4 for files using the deprecated hardware floating-point ABI which used 64-bit
7405 floating-point registers, 32-bit general-purpose registers and increased the
7406 number of callee-saved floating-point registers.
7408 5 for files using the hardware floating-point ABI with a double-precision FPU
7409 with either 32-bit or 64-bit floating-point registers and 32-bit
7410 general-purpose registers.
7412 6 for files using the hardware floating-point ABI with 64-bit floating-point
7413 registers and 32-bit general-purpose registers.
7415 7 for files using the hardware floating-point ABI with 64-bit floating-point
7416 registers, 32-bit general-purpose registers and a rule that forbids the
7417 direct use of odd-numbered single-precision floating-point registers.
7421 @subsection PowerPC Attributes
7424 @item Tag_GNU_Power_ABI_FP (4)
7425 The floating-point ABI used by this object file. The value will be:
7429 0 for files not affected by the floating-point ABI.
7431 1 for files using double-precision hardware floating-point ABI.
7433 2 for files using the software floating-point ABI.
7435 3 for files using single-precision hardware floating-point ABI.
7438 @item Tag_GNU_Power_ABI_Vector (8)
7439 The vector ABI used by this object file. The value will be:
7443 0 for files not affected by the vector ABI.
7445 1 for files using general purpose registers to pass vectors.
7447 2 for files using AltiVec registers to pass vectors.
7449 3 for files using SPE registers to pass vectors.
7453 @subsection IBM z Systems Attributes
7456 @item Tag_GNU_S390_ABI_Vector (8)
7457 The vector ABI used by this object file. The value will be:
7461 0 for files not affected by the vector ABI.
7463 1 for files using software vector ABI.
7465 2 for files using hardware vector ABI.
7469 @node Defining New Object Attributes
7470 @section Defining New Object Attributes
7472 If you want to define a new @sc{gnu} object attribute, here are the places you
7473 will need to modify. New attributes should be discussed on the @samp{binutils}
7478 This manual, which is the official register of attributes.
7480 The header for your architecture @file{include/elf}, to define the tag.
7482 The @file{bfd} support file for your architecture, to merge the attribute
7483 and issue any appropriate link warnings.
7485 Test cases in @file{ld/testsuite} for merging and link warnings.
7487 @file{binutils/readelf.c} to display your attribute.
7489 GCC, if you want the compiler to mark the attribute automatically.
7495 @node Machine Dependencies
7496 @chapter Machine Dependent Features
7498 @cindex machine dependencies
7499 The machine instruction sets are (almost by definition) different on
7500 each machine where @command{@value{AS}} runs. Floating point representations
7501 vary as well, and @command{@value{AS}} often supports a few additional
7502 directives or command-line options for compatibility with other
7503 assemblers on a particular platform. Finally, some versions of
7504 @command{@value{AS}} support special pseudo-instructions for branch
7507 This chapter discusses most of these differences, though it does not
7508 include details on any machine's instruction set. For details on that
7509 subject, see the hardware manufacturer's manual.
7513 * AArch64-Dependent:: AArch64 Dependent Features
7516 * Alpha-Dependent:: Alpha Dependent Features
7519 * ARC-Dependent:: ARC Dependent Features
7522 * ARM-Dependent:: ARM Dependent Features
7525 * AVR-Dependent:: AVR Dependent Features
7528 * Blackfin-Dependent:: Blackfin Dependent Features
7531 * CR16-Dependent:: CR16 Dependent Features
7534 * CRIS-Dependent:: CRIS Dependent Features
7537 * D10V-Dependent:: D10V Dependent Features
7540 * D30V-Dependent:: D30V Dependent Features
7543 * Epiphany-Dependent:: EPIPHANY Dependent Features
7546 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7549 * HPPA-Dependent:: HPPA Dependent Features
7552 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7555 * IA-64-Dependent:: Intel IA-64 Dependent Features
7558 * IP2K-Dependent:: IP2K Dependent Features
7561 * LM32-Dependent:: LM32 Dependent Features
7564 * M32C-Dependent:: M32C Dependent Features
7567 * M32R-Dependent:: M32R Dependent Features
7570 * M68K-Dependent:: M680x0 Dependent Features
7573 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7576 * S12Z-Dependent:: S12Z Dependent Features
7579 * Meta-Dependent :: Meta Dependent Features
7582 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7585 * MIPS-Dependent:: MIPS Dependent Features
7588 * MMIX-Dependent:: MMIX Dependent Features
7591 * MSP430-Dependent:: MSP430 Dependent Features
7594 * NDS32-Dependent:: Andes NDS32 Dependent Features
7597 * NiosII-Dependent:: Altera Nios II Dependent Features
7600 * NS32K-Dependent:: NS32K Dependent Features
7603 * PDP-11-Dependent:: PDP-11 Dependent Features
7606 * PJ-Dependent:: picoJava Dependent Features
7609 * PPC-Dependent:: PowerPC Dependent Features
7612 * PRU-Dependent:: PRU Dependent Features
7615 * RISC-V-Dependent:: RISC-V Dependent Features
7618 * RL78-Dependent:: RL78 Dependent Features
7621 * RX-Dependent:: RX Dependent Features
7624 * S/390-Dependent:: IBM S/390 Dependent Features
7627 * SCORE-Dependent:: SCORE Dependent Features
7630 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7633 * Sparc-Dependent:: SPARC Dependent Features
7636 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7639 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7642 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7645 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7648 * V850-Dependent:: V850 Dependent Features
7651 * Vax-Dependent:: VAX Dependent Features
7654 * Visium-Dependent:: Visium Dependent Features
7657 * WebAssembly-Dependent:: WebAssembly Dependent Features
7660 * XGATE-Dependent:: XGATE Dependent Features
7663 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7666 * Xtensa-Dependent:: Xtensa Dependent Features
7669 * Z80-Dependent:: Z80 Dependent Features
7672 * Z8000-Dependent:: Z8000 Dependent Features
7679 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7680 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7681 @c peculiarity: to preserve cross-references, there must be a node called
7682 @c "Machine Dependencies". Hence the conditional nodenames in each
7683 @c major node below. Node defaulting in makeinfo requires adjacency of
7684 @c node and sectioning commands; hence the repetition of @chapter BLAH
7685 @c in both conditional blocks.
7688 @include c-aarch64.texi
7692 @include c-alpha.texi
7708 @include c-bfin.texi
7712 @include c-cr16.texi
7716 @include c-cris.texi
7721 @node Machine Dependencies
7722 @chapter Machine Dependent Features
7724 The machine instruction sets are different on each Renesas chip family,
7725 and there are also some syntax differences among the families. This
7726 chapter describes the specific @command{@value{AS}} features for each
7730 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7731 * SH-Dependent:: Renesas SH Dependent Features
7738 @include c-d10v.texi
7742 @include c-d30v.texi
7746 @include c-epiphany.texi
7750 @include c-h8300.texi
7754 @include c-hppa.texi
7758 @include c-i386.texi
7762 @include c-ia64.texi
7766 @include c-ip2k.texi
7770 @include c-lm32.texi
7774 @include c-m32c.texi
7778 @include c-m32r.texi
7782 @include c-m68k.texi
7786 @include c-m68hc11.texi
7790 @include c-s12z.texi
7794 @include c-metag.texi
7798 @include c-microblaze.texi
7802 @include c-mips.texi
7806 @include c-mmix.texi
7810 @include c-msp430.texi
7814 @include c-nds32.texi
7818 @include c-nios2.texi
7822 @include c-ns32k.texi
7826 @include c-pdp11.texi
7842 @include c-riscv.texi
7846 @include c-rl78.texi
7854 @include c-s390.texi
7858 @include c-score.texi
7866 @include c-sparc.texi
7870 @include c-tic54x.texi
7874 @include c-tic6x.texi
7878 @include c-tilegx.texi
7882 @include c-tilepro.texi
7886 @include c-v850.texi
7894 @include c-visium.texi
7898 @include c-wasm32.texi
7902 @include c-xgate.texi
7906 @include c-xstormy16.texi
7910 @include c-xtensa.texi
7922 @c reverse effect of @down at top of generic Machine-Dep chapter
7926 @node Reporting Bugs
7927 @chapter Reporting Bugs
7928 @cindex bugs in assembler
7929 @cindex reporting bugs in assembler
7931 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7933 Reporting a bug may help you by bringing a solution to your problem, or it may
7934 not. But in any case the principal function of a bug report is to help the
7935 entire community by making the next version of @command{@value{AS}} work better.
7936 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7938 In order for a bug report to serve its purpose, you must include the
7939 information that enables us to fix the bug.
7942 * Bug Criteria:: Have you found a bug?
7943 * Bug Reporting:: How to report bugs
7947 @section Have You Found a Bug?
7948 @cindex bug criteria
7950 If you are not sure whether you have found a bug, here are some guidelines:
7953 @cindex fatal signal
7954 @cindex assembler crash
7955 @cindex crash of assembler
7957 If the assembler gets a fatal signal, for any input whatever, that is a
7958 @command{@value{AS}} bug. Reliable assemblers never crash.
7960 @cindex error on valid input
7962 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7964 @cindex invalid input
7966 If @command{@value{AS}} does not produce an error message for invalid input, that
7967 is a bug. However, you should note that your idea of ``invalid input'' might
7968 be our idea of ``an extension'' or ``support for traditional practice''.
7971 If you are an experienced user of assemblers, your suggestions for improvement
7972 of @command{@value{AS}} are welcome in any case.
7976 @section How to Report Bugs
7978 @cindex assembler bugs, reporting
7980 A number of companies and individuals offer support for @sc{gnu} products. If
7981 you obtained @command{@value{AS}} from a support organization, we recommend you
7982 contact that organization first.
7984 You can find contact information for many support companies and
7985 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7989 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7993 The fundamental principle of reporting bugs usefully is this:
7994 @strong{report all the facts}. If you are not sure whether to state a
7995 fact or leave it out, state it!
7997 Often people omit facts because they think they know what causes the problem
7998 and assume that some details do not matter. Thus, you might assume that the
7999 name of a symbol you use in an example does not matter. Well, probably it does
8000 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
8001 happens to fetch from the location where that name is stored in memory;
8002 perhaps, if the name were different, the contents of that location would fool
8003 the assembler into doing the right thing despite the bug. Play it safe and
8004 give a specific, complete example. That is the easiest thing for you to do,
8005 and the most helpful.
8007 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
8008 it is new to us. Therefore, always write your bug reports on the assumption
8009 that the bug has not been reported previously.
8011 Sometimes people give a few sketchy facts and ask, ``Does this ring a
8012 bell?'' This cannot help us fix a bug, so it is basically useless. We
8013 respond by asking for enough details to enable us to investigate.
8014 You might as well expedite matters by sending them to begin with.
8016 To enable us to fix the bug, you should include all these things:
8020 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
8021 it with the @samp{--version} argument.
8023 Without this, we will not know whether there is any point in looking for
8024 the bug in the current version of @command{@value{AS}}.
8027 Any patches you may have applied to the @command{@value{AS}} source.
8030 The type of machine you are using, and the operating system name and
8034 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
8038 The command arguments you gave the assembler to assemble your example and
8039 observe the bug. To guarantee you will not omit something important, list them
8040 all. A copy of the Makefile (or the output from make) is sufficient.
8042 If we were to try to guess the arguments, we would probably guess wrong
8043 and then we might not encounter the bug.
8046 A complete input file that will reproduce the bug. If the bug is observed when
8047 the assembler is invoked via a compiler, send the assembler source, not the
8048 high level language source. Most compilers will produce the assembler source
8049 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
8050 the options @samp{-v --save-temps}; this will save the assembler source in a
8051 file with an extension of @file{.s}, and also show you exactly how
8052 @command{@value{AS}} is being run.
8055 A description of what behavior you observe that you believe is
8056 incorrect. For example, ``It gets a fatal signal.''
8058 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
8059 will certainly notice it. But if the bug is incorrect output, we might not
8060 notice unless it is glaringly wrong. You might as well not give us a chance to
8063 Even if the problem you experience is a fatal signal, you should still say so
8064 explicitly. Suppose something strange is going on, such as, your copy of
8065 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
8066 library on your system. (This has happened!) Your copy might crash and ours
8067 would not. If you told us to expect a crash, then when ours fails to crash, we
8068 would know that the bug was not happening for us. If you had not told us to
8069 expect a crash, then we would not be able to draw any conclusion from our
8073 If you wish to suggest changes to the @command{@value{AS}} source, send us context
8074 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
8075 option. Always send diffs from the old file to the new file. If you even
8076 discuss something in the @command{@value{AS}} source, refer to it by context, not
8079 The line numbers in our development sources will not match those in your
8080 sources. Your line numbers would convey no useful information to us.
8083 Here are some things that are not necessary:
8087 A description of the envelope of the bug.
8089 Often people who encounter a bug spend a lot of time investigating
8090 which changes to the input file will make the bug go away and which
8091 changes will not affect it.
8093 This is often time consuming and not very useful, because the way we
8094 will find the bug is by running a single example under the debugger
8095 with breakpoints, not by pure deduction from a series of examples.
8096 We recommend that you save your time for something else.
8098 Of course, if you can find a simpler example to report @emph{instead}
8099 of the original one, that is a convenience for us. Errors in the
8100 output will be easier to spot, running under the debugger will take
8101 less time, and so on.
8103 However, simplification is not vital; if you do not want to do this,
8104 report the bug anyway and send us the entire test case you used.
8107 A patch for the bug.
8109 A patch for the bug does help us if it is a good one. But do not omit
8110 the necessary information, such as the test case, on the assumption that
8111 a patch is all we need. We might see problems with your patch and decide
8112 to fix the problem another way, or we might not understand it at all.
8114 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
8115 construct an example that will make the program follow a certain path through
8116 the code. If you do not send us the example, we will not be able to construct
8117 one, so we will not be able to verify that the bug is fixed.
8119 And if we cannot understand what bug you are trying to fix, or why your
8120 patch should be an improvement, we will not install it. A test case will
8121 help us to understand.
8124 A guess about what the bug is or what it depends on.
8126 Such guesses are usually wrong. Even we cannot guess right about such
8127 things without first using the debugger to find the facts.
8130 @node Acknowledgements
8131 @chapter Acknowledgements
8133 If you have contributed to GAS and your name isn't listed here,
8134 it is not meant as a slight. We just don't know about it. Send mail to the
8135 maintainer, and we'll correct the situation. Currently
8137 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
8139 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
8142 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
8143 information and the 68k series machines, most of the preprocessing pass, and
8144 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
8146 K. Richard Pixley maintained GAS for a while, adding various enhancements and
8147 many bug fixes, including merging support for several processors, breaking GAS
8148 up to handle multiple object file format back ends (including heavy rewrite,
8149 testing, an integration of the coff and b.out back ends), adding configuration
8150 including heavy testing and verification of cross assemblers and file splits
8151 and renaming, converted GAS to strictly ANSI C including full prototypes, added
8152 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
8153 port (including considerable amounts of reverse engineering), a SPARC opcode
8154 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
8155 assertions and made them work, much other reorganization, cleanup, and lint.
8157 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
8158 in format-specific I/O modules.
8160 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
8161 has done much work with it since.
8163 The Intel 80386 machine description was written by Eliot Dresselhaus.
8165 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
8167 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
8168 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
8170 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
8171 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
8172 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
8173 support a.out format.
8175 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
8176 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
8177 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
8178 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
8181 John Gilmore built the AMD 29000 support, added @code{.include} support, and
8182 simplified the configuration of which versions accept which directives. He
8183 updated the 68k machine description so that Motorola's opcodes always produced
8184 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
8185 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
8186 cross-compilation support, and one bug in relaxation that took a week and
8187 required the proverbial one-bit fix.
8189 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
8190 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
8191 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
8192 PowerPC assembler, and made a few other minor patches.
8194 Steve Chamberlain made GAS able to generate listings.
8196 Hewlett-Packard contributed support for the HP9000/300.
8198 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
8199 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
8200 formats). This work was supported by both the Center for Software Science at
8201 the University of Utah and Cygnus Support.
8203 Support for ELF format files has been worked on by Mark Eichin of Cygnus
8204 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
8205 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
8206 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
8207 and some initial 64-bit support).
8209 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
8211 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
8212 support for openVMS/Alpha.
8214 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
8217 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
8218 Inc.@: added support for Xtensa processors.
8220 Several engineers at Cygnus Support have also provided many small bug fixes and
8221 configuration enhancements.
8223 Jon Beniston added support for the Lattice Mico32 architecture.
8225 Many others have contributed large or small bugfixes and enhancements. If
8226 you have contributed significant work and are not mentioned on this list, and
8227 want to be, let us know. Some of the history has been lost; we are not
8228 intentionally leaving anyone out.
8230 @node GNU Free Documentation License
8231 @appendix GNU Free Documentation License
8235 @unnumbered AS Index
projects / deliverable / binutils-gdb.git / blob