1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright (C) 1991-2017 Free Software Foundation, Inc.
3 @c UPDATE!! On future updates--
4 @c (1) check for new machine-dep cmdline options in
5 @c md_parse_option definitions in config/tc-*.c
6 @c (2) for platform-specific directives, examine md_pseudo_op
8 @c (3) for object-format specific directives, examine obj_pseudo_op
10 @c (4) portable directives in potable[] in read.c
14 @macro gcctabopt{body}
17 @c defaults, config file may override:
22 @include asconfig.texi
27 @c common OR combinations of conditions
53 @set abnormal-separator
57 @settitle Using @value{AS}
60 @settitle Using @value{AS} (@value{TARGET})
62 @setchapternewpage odd
67 @c WARE! Some of the machine-dependent sections contain tables of machine
68 @c instructions. Except in multi-column format, these tables look silly.
69 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
70 @c the multi-col format is faked within @example sections.
72 @c Again unfortunately, the natural size that fits on a page, for these tables,
73 @c is different depending on whether or not smallbook is turned on.
74 @c This matters, because of order: text flow switches columns at each page
77 @c The format faked in this source works reasonably well for smallbook,
78 @c not well for the default large-page format. This manual expects that if you
79 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
80 @c tables in question. You can turn on one without the other at your
81 @c discretion, of course.
84 @c the insn tables look just as silly in info files regardless of smallbook,
85 @c might as well show 'em anyways.
89 @dircategory Software development
91 * As: (as). The GNU assembler.
92 * Gas: (as). The GNU assembler.
100 This file documents the GNU Assembler "@value{AS}".
102 @c man begin COPYRIGHT
103 Copyright @copyright{} 1991-2017 Free Software Foundation, Inc.
105 Permission is granted to copy, distribute and/or modify this document
106 under the terms of the GNU Free Documentation License, Version 1.3
107 or any later version published by the Free Software Foundation;
108 with no Invariant Sections, with no Front-Cover Texts, and with no
109 Back-Cover Texts. A copy of the license is included in the
110 section entitled ``GNU Free Documentation License''.
116 @title Using @value{AS}
117 @subtitle The @sc{gnu} Assembler
119 @subtitle for the @value{TARGET} family
121 @ifset VERSION_PACKAGE
123 @subtitle @value{VERSION_PACKAGE}
126 @subtitle Version @value{VERSION}
129 The Free Software Foundation Inc.@: thanks The Nice Computer
130 Company of Australia for loaning Dean Elsner to write the
131 first (Vax) version of @command{as} for Project @sc{gnu}.
132 The proprietors, management and staff of TNCCA thank FSF for
133 distracting the boss while they got some work
136 @author Dean Elsner, Jay Fenlason & friends
140 \hfill {\it Using {\tt @value{AS}}}\par
141 \hfill Edited by Cygnus Support\par
143 %"boxit" macro for figures:
144 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
145 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
146 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
147 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
148 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
151 @vskip 0pt plus 1filll
152 Copyright @copyright{} 1991-2017 Free Software Foundation, Inc.
154 Permission is granted to copy, distribute and/or modify this document
155 under the terms of the GNU Free Documentation License, Version 1.3
156 or any later version published by the Free Software Foundation;
157 with no Invariant Sections, with no Front-Cover Texts, and with no
158 Back-Cover Texts. A copy of the license is included in the
159 section entitled ``GNU Free Documentation License''.
166 @top Using @value{AS}
168 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}}
169 @ifset VERSION_PACKAGE
170 @value{VERSION_PACKAGE}
172 version @value{VERSION}.
174 This version of the file describes @command{@value{AS}} configured to generate
175 code for @value{TARGET} architectures.
178 This document is distributed under the terms of the GNU Free
179 Documentation License. A copy of the license is included in the
180 section entitled ``GNU Free Documentation License''.
183 * Overview:: Overview
184 * Invoking:: Command-Line Options
186 * Sections:: Sections and Relocation
188 * Expressions:: Expressions
189 * Pseudo Ops:: Assembler Directives
191 * Object Attributes:: Object Attributes
193 * Machine Dependencies:: Machine Dependent Features
194 * Reporting Bugs:: Reporting Bugs
195 * Acknowledgements:: Who Did What
196 * GNU Free Documentation License:: GNU Free Documentation License
197 * AS Index:: AS Index
204 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
206 This version of the manual describes @command{@value{AS}} configured to generate
207 code for @value{TARGET} architectures.
211 @cindex invocation summary
212 @cindex option summary
213 @cindex summary of options
214 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
215 see @ref{Invoking,,Command-Line Options}.
217 @c man title AS the portable GNU assembler.
221 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
225 @c We don't use deffn and friends for the following because they seem
226 @c to be limited to one line for the header.
228 @c man begin SYNOPSIS
229 @value{AS} [@b{-a}[@b{cdghlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
230 [@b{--compress-debug-sections}] [@b{--nocompress-debug-sections}]
231 [@b{--debug-prefix-map} @var{old}=@var{new}]
232 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
233 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--gdwarf-sections}]
234 [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
235 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
236 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
237 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}]
238 [@b{--no-pad-sections}]
239 [@b{-o} @var{objfile}] [@b{-R}]
240 [@b{--hash-size}=@var{NUM}] [@b{--reduce-memory-overheads}]
242 [@b{-v}] [@b{-version}] [@b{--version}]
243 [@b{-W}] [@b{--warn}] [@b{--fatal-warnings}] [@b{-w}] [@b{-x}]
244 [@b{-Z}] [@b{@@@var{FILE}}]
245 [@b{--sectname-subst}] [@b{--size-check=[error|warning]}]
246 [@b{--elf-stt-common=[no|yes]}]
247 [@b{--target-help}] [@var{target-options}]
248 [@b{--}|@var{files} @dots{}]
251 @c Target dependent options are listed below. Keep the list sorted.
252 @c Add an empty line for separation.
256 @emph{Target AArch64 options:}
258 [@b{-mabi}=@var{ABI}]
262 @emph{Target Alpha options:}
264 [@b{-mdebug} | @b{-no-mdebug}]
265 [@b{-replace} | @b{-noreplace}]
266 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
267 [@b{-F}] [@b{-32addr}]
271 @emph{Target ARC options:}
272 [@b{-mcpu=@var{cpu}}]
273 [@b{-mA6}|@b{-mARC600}|@b{-mARC601}|@b{-mA7}|@b{-mARC700}|@b{-mEM}|@b{-mHS}]
280 @emph{Target ARM options:}
281 @c Don't document the deprecated options
282 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
283 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
284 [@b{-mfpu}=@var{floating-point-format}]
285 [@b{-mfloat-abi}=@var{abi}]
286 [@b{-meabi}=@var{ver}]
289 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
290 @b{-mapcs-reentrant}]
291 [@b{-mthumb-interwork}] [@b{-k}]
295 @emph{Target Blackfin options:}
296 [@b{-mcpu}=@var{processor}[-@var{sirevision}]]
303 @emph{Target CRIS options:}
304 [@b{--underscore} | @b{--no-underscore}]
306 [@b{--emulation=criself} | @b{--emulation=crisaout}]
307 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
308 @c Deprecated -- deliberately not documented.
313 @emph{Target D10V options:}
318 @emph{Target D30V options:}
319 [@b{-O}|@b{-n}|@b{-N}]
323 @emph{Target EPIPHANY options:}
324 [@b{-mepiphany}|@b{-mepiphany16}]
328 @emph{Target H8/300 options:}
332 @c HPPA has no machine-dependent assembler options (yet).
336 @emph{Target i386 options:}
337 [@b{--32}|@b{--x32}|@b{--64}] [@b{-n}]
338 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
342 @emph{Target i960 options:}
343 @c see md_parse_option in tc-i960.c
344 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
346 [@b{-b}] [@b{-no-relax}]
350 @emph{Target IA-64 options:}
351 [@b{-mconstant-gp}|@b{-mauto-pic}]
352 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
354 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
355 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
356 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
357 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
361 @emph{Target IP2K options:}
362 [@b{-mip2022}|@b{-mip2022ext}]
366 @emph{Target M32C options:}
367 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
371 @emph{Target M32R options:}
372 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
377 @emph{Target M680X0 options:}
378 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
382 @emph{Target M68HC11 options:}
383 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}|@b{-mm9s12x}|@b{-mm9s12xg}]
384 [@b{-mshort}|@b{-mlong}]
385 [@b{-mshort-double}|@b{-mlong-double}]
386 [@b{--force-long-branches}] [@b{--short-branches}]
387 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
388 [@b{--print-opcodes}] [@b{--generate-example}]
392 @emph{Target MCORE options:}
393 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
394 [@b{-mcpu=[210|340]}]
398 @emph{Target Meta options:}
399 [@b{-mcpu=@var{cpu}}] [@b{-mfpu=@var{cpu}}] [@b{-mdsp=@var{cpu}}]
402 @emph{Target MICROBLAZE options:}
403 @c MicroBlaze has no machine-dependent assembler options.
407 @emph{Target MIPS options:}
408 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
409 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
410 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
411 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
412 [@b{-mfp64}] [@b{-mgp64}] [@b{-mfpxx}]
413 [@b{-modd-spreg}] [@b{-mno-odd-spreg}]
414 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
415 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
416 [@b{-mips32r3}] [@b{-mips32r5}] [@b{-mips32r6}] [@b{-mips64}] [@b{-mips64r2}]
417 [@b{-mips64r3}] [@b{-mips64r5}] [@b{-mips64r6}]
418 [@b{-construct-floats}] [@b{-no-construct-floats}]
419 [@b{-mignore-branch-isa}] [@b{-mno-ignore-branch-isa}]
420 [@b{-mnan=@var{encoding}}]
421 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
422 [@b{-mips16}] [@b{-no-mips16}]
423 [@b{-mmips16e2}] [@b{-mno-mips16e2}]
424 [@b{-mmicromips}] [@b{-mno-micromips}]
425 [@b{-msmartmips}] [@b{-mno-smartmips}]
426 [@b{-mips3d}] [@b{-no-mips3d}]
427 [@b{-mdmx}] [@b{-no-mdmx}]
428 [@b{-mdsp}] [@b{-mno-dsp}]
429 [@b{-mdspr2}] [@b{-mno-dspr2}]
430 [@b{-mdspr3}] [@b{-mno-dspr3}]
431 [@b{-mmsa}] [@b{-mno-msa}]
432 [@b{-mxpa}] [@b{-mno-xpa}]
433 [@b{-mmt}] [@b{-mno-mt}]
434 [@b{-mmcu}] [@b{-mno-mcu}]
435 [@b{-minsn32}] [@b{-mno-insn32}]
436 [@b{-mfix7000}] [@b{-mno-fix7000}]
437 [@b{-mfix-rm7000}] [@b{-mno-fix-rm7000}]
438 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
439 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
440 [@b{-mdebug}] [@b{-no-mdebug}]
441 [@b{-mpdr}] [@b{-mno-pdr}]
445 @emph{Target MMIX options:}
446 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
447 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
448 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
449 [@b{--linker-allocated-gregs}]
453 @emph{Target Nios II options:}
454 [@b{-relax-all}] [@b{-relax-section}] [@b{-no-relax}]
459 @emph{Target NDS32 options:}
460 [@b{-EL}] [@b{-EB}] [@b{-O}] [@b{-Os}] [@b{-mcpu=@var{cpu}}]
461 [@b{-misa=@var{isa}}] [@b{-mabi=@var{abi}}] [@b{-mall-ext}]
462 [@b{-m[no-]16-bit}] [@b{-m[no-]perf-ext}] [@b{-m[no-]perf2-ext}]
463 [@b{-m[no-]string-ext}] [@b{-m[no-]dsp-ext}] [@b{-m[no-]mac}] [@b{-m[no-]div}]
464 [@b{-m[no-]audio-isa-ext}] [@b{-m[no-]fpu-sp-ext}] [@b{-m[no-]fpu-dp-ext}]
465 [@b{-m[no-]fpu-fma}] [@b{-mfpu-freg=@var{FREG}}] [@b{-mreduced-regs}]
466 [@b{-mfull-regs}] [@b{-m[no-]dx-regs}] [@b{-mpic}] [@b{-mno-relax}]
471 @emph{Target PDP11 options:}
472 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
473 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
474 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
478 @emph{Target picoJava options:}
483 @emph{Target PowerPC options:}
485 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|@b{-m403}|@b{-m405}|
486 @b{-m440}|@b{-m464}|@b{-m476}|@b{-m7400}|@b{-m7410}|@b{-m7450}|@b{-m7455}|@b{-m750cl}|@b{-mppc64}|
487 @b{-m620}|@b{-me500}|@b{-e500x2}|@b{-me500mc}|@b{-me500mc64}|@b{-me5500}|@b{-me6500}|@b{-mppc64bridge}|
488 @b{-mbooke}|@b{-mpower4}|@b{-mpwr4}|@b{-mpower5}|@b{-mpwr5}|@b{-mpwr5x}|@b{-mpower6}|@b{-mpwr6}|
489 @b{-mpower7}|@b{-mpwr7}|@b{-mpower8}|@b{-mpwr8}|@b{-mpower9}|@b{-mpwr9}@b{-ma2}|
490 @b{-mcell}|@b{-mspe}|@b{-mtitan}|@b{-me300}|@b{-mcom}]
491 [@b{-many}] [@b{-maltivec}|@b{-mvsx}|@b{-mhtm}|@b{-mvle}]
492 [@b{-mregnames}|@b{-mno-regnames}]
493 [@b{-mrelocatable}|@b{-mrelocatable-lib}|@b{-K PIC}] [@b{-memb}]
494 [@b{-mlittle}|@b{-mlittle-endian}|@b{-le}|@b{-mbig}|@b{-mbig-endian}|@b{-be}]
495 [@b{-msolaris}|@b{-mno-solaris}]
496 [@b{-nops=@var{count}}]
500 @emph{Target PRU options:}
503 [@b{-mno-warn-regname-label}]
507 @emph{Target RL78 options:}
509 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
513 @emph{Target RX options:}
514 [@b{-mlittle-endian}|@b{-mbig-endian}]
515 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
516 [@b{-muse-conventional-section-names}]
517 [@b{-msmall-data-limit}]
520 [@b{-mint-register=@var{number}}]
521 [@b{-mgcc-abi}|@b{-mrx-abi}]
525 @emph{Target RISC-V options:}
526 [@b{-march}=@var{ISA}]
527 [@b{-mabi}=@var{ABI}]
531 @emph{Target s390 options:}
532 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
533 [@b{-mregnames}|@b{-mno-regnames}]
534 [@b{-mwarn-areg-zero}]
538 @emph{Target SCORE options:}
539 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
540 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
541 [@b{-march=score7}][@b{-march=score3}]
542 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
546 @emph{Target SPARC options:}
547 @c The order here is important. See c-sparc.texi.
548 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Aleon}|@b{-Asparclet}|@b{-Asparclite}
549 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av8plusb}|@b{-Av8plusc}|@b{-Av8plusd}
550 @b{-Av8plusv}|@b{-Av8plusm}|@b{-Av9}|@b{-Av9a}|@b{-Av9b}|@b{-Av9c}
551 @b{-Av9d}|@b{-Av9e}|@b{-Av9v}|@b{-Av9m}|@b{-Asparc}|@b{-Asparcvis}
552 @b{-Asparcvis2}|@b{-Asparcfmaf}|@b{-Asparcima}|@b{-Asparcvis3}
553 @b{-Asparcvisr}|@b{-Asparc5}]
554 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}]|@b{-xarch=v8plusb}|@b{-xarch=v8plusc}
555 @b{-xarch=v8plusd}|@b{-xarch=v8plusv}|@b{-xarch=v8plusm}|@b{-xarch=v9}
556 @b{-xarch=v9a}|@b{-xarch=v9b}|@b{-xarch=v9c}|@b{-xarch=v9d}|@b{-xarch=v9e}
557 @b{-xarch=v9v}|@b{-xarch=v9m}|@b{-xarch=sparc}|@b{-xarch=sparcvis}
558 @b{-xarch=sparcvis2}|@b{-xarch=sparcfmaf}|@b{-xarch=sparcima}
559 @b{-xarch=sparcvis3}|@b{-xarch=sparcvisr}|@b{-xarch=sparc5}
562 [@b{--enforce-aligned-data}][@b{--dcti-couples-detect}]
566 @emph{Target TIC54X options:}
567 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
568 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
572 @emph{Target TIC6X options:}
573 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
574 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
575 [@b{-mpic}|@b{-mno-pic}]
579 @emph{Target TILE-Gx options:}
580 [@b{-m32}|@b{-m64}][@b{-EB}][@b{-EL}]
583 @c TILEPro has no machine-dependent assembler options
587 @emph{Target Visium options:}
588 [@b{-mtune=@var{arch}}]
592 @emph{Target Xtensa options:}
593 [@b{--[no-]text-section-literals}] [@b{--[no-]auto-litpools}]
594 [@b{--[no-]absolute-literals}]
595 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
596 [@b{--[no-]transform}]
597 [@b{--rename-section} @var{oldname}=@var{newname}]
598 [@b{--[no-]trampolines}]
602 @emph{Target Z80 options:}
603 [@b{-z80}] [@b{-r800}]
604 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
605 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
606 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
607 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
608 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
609 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
613 @c Z8000 has no machine-dependent assembler options
622 @include at-file.texi
625 Turn on listings, in any of a variety of ways:
629 omit false conditionals
632 omit debugging directives
635 include general information, like @value{AS} version and options passed
638 include high-level source
644 include macro expansions
647 omit forms processing
653 set the name of the listing file
656 You may combine these options; for example, use @samp{-aln} for assembly
657 listing without forms processing. The @samp{=file} option, if used, must be
658 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
661 Begin in alternate macro mode.
663 @xref{Altmacro,,@code{.altmacro}}.
666 @item --compress-debug-sections
667 Compress DWARF debug sections using zlib with SHF_COMPRESSED from the
668 ELF ABI. The resulting object file may not be compatible with older
669 linkers and object file utilities. Note if compression would make a
670 given section @emph{larger} then it is not compressed.
673 @cindex @samp{--compress-debug-sections=} option
674 @item --compress-debug-sections=none
675 @itemx --compress-debug-sections=zlib
676 @itemx --compress-debug-sections=zlib-gnu
677 @itemx --compress-debug-sections=zlib-gabi
678 These options control how DWARF debug sections are compressed.
679 @option{--compress-debug-sections=none} is equivalent to
680 @option{--nocompress-debug-sections}.
681 @option{--compress-debug-sections=zlib} and
682 @option{--compress-debug-sections=zlib-gabi} are equivalent to
683 @option{--compress-debug-sections}.
684 @option{--compress-debug-sections=zlib-gnu} compresses DWARF debug
685 sections using zlib. The debug sections are renamed to begin with
686 @samp{.zdebug}. Note if compression would make a given section
687 @emph{larger} then it is not compressed nor renamed.
691 @item --nocompress-debug-sections
692 Do not compress DWARF debug sections. This is usually the default for all
693 targets except the x86/x86_64, but a configure time option can be used to
697 Ignored. This option is accepted for script compatibility with calls to
700 @item --debug-prefix-map @var{old}=@var{new}
701 When assembling files in directory @file{@var{old}}, record debugging
702 information describing them as in @file{@var{new}} instead.
704 @item --defsym @var{sym}=@var{value}
705 Define the symbol @var{sym} to be @var{value} before assembling the input file.
706 @var{value} must be an integer constant. As in C, a leading @samp{0x}
707 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
708 value. The value of the symbol can be overridden inside a source file via the
709 use of a @code{.set} pseudo-op.
712 ``fast''---skip whitespace and comment preprocessing (assume source is
717 Generate debugging information for each assembler source line using whichever
718 debug format is preferred by the target. This currently means either STABS,
722 Generate stabs debugging information for each assembler line. This
723 may help debugging assembler code, if the debugger can handle it.
726 Generate stabs debugging information for each assembler line, with GNU
727 extensions that probably only gdb can handle, and that could make other
728 debuggers crash or refuse to read your program. This
729 may help debugging assembler code. Currently the only GNU extension is
730 the location of the current working directory at assembling time.
733 Generate DWARF2 debugging information for each assembler line. This
734 may help debugging assembler code, if the debugger can handle it. Note---this
735 option is only supported by some targets, not all of them.
737 @item --gdwarf-sections
738 Instead of creating a .debug_line section, create a series of
739 .debug_line.@var{foo} sections where @var{foo} is the name of the
740 corresponding code section. For example a code section called @var{.text.func}
741 will have its dwarf line number information placed into a section called
742 @var{.debug_line.text.func}. If the code section is just called @var{.text}
743 then debug line section will still be called just @var{.debug_line} without any
747 @item --size-check=error
748 @itemx --size-check=warning
749 Issue an error or warning for invalid ELF .size directive.
751 @item --elf-stt-common=no
752 @itemx --elf-stt-common=yes
753 These options control whether the ELF assembler should generate common
754 symbols with the @code{STT_COMMON} type. The default can be controlled
755 by a configure option @option{--enable-elf-stt-common}.
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
1073 Intel 80960 processor.
1076 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
1077 Specify which variant of the 960 architecture is the target.
1080 Add code to collect statistics about branches taken.
1083 Do not alter compare-and-branch instructions for long displacements;
1090 The following options are available when @value{AS} is configured for the
1096 Specifies that the extended IP2022 instructions are allowed.
1099 Restores the default behaviour, which restricts the permitted instructions to
1100 just the basic IP2022 ones.
1106 The following options are available when @value{AS} is configured for the
1107 Renesas M32C and M16C processors.
1112 Assemble M32C instructions.
1115 Assemble M16C instructions (the default).
1118 Enable support for link-time relaxations.
1121 Support H'00 style hex constants in addition to 0x00 style.
1127 The following options are available when @value{AS} is configured for the
1128 Renesas M32R (formerly Mitsubishi M32R) series.
1133 Specify which processor in the M32R family is the target. The default
1134 is normally the M32R, but this option changes it to the M32RX.
1136 @item --warn-explicit-parallel-conflicts or --Wp
1137 Produce warning messages when questionable parallel constructs are
1140 @item --no-warn-explicit-parallel-conflicts or --Wnp
1141 Do not produce warning messages when questionable parallel constructs are
1148 The following options are available when @value{AS} is configured for the
1149 Motorola 68000 series.
1154 Shorten references to undefined symbols, to one word instead of two.
1156 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1157 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1158 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1159 Specify what processor in the 68000 family is the target. The default
1160 is normally the 68020, but this can be changed at configuration time.
1162 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1163 The target machine does (or does not) have a floating-point coprocessor.
1164 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1165 the basic 68000 is not compatible with the 68881, a combination of the
1166 two can be specified, since it's possible to do emulation of the
1167 coprocessor instructions with the main processor.
1169 @item -m68851 | -mno-68851
1170 The target machine does (or does not) have a memory-management
1171 unit coprocessor. The default is to assume an MMU for 68020 and up.
1179 @xref{Nios II Options}, for the options available when @value{AS} is configured
1180 for an Altera Nios II processor.
1184 @c man begin OPTIONS
1185 The following options are available when @value{AS} is configured for an
1186 Altera Nios II processor.
1188 @c man begin INCLUDE
1189 @include c-nios2.texi
1190 @c ended inside the included file
1196 For details about the PDP-11 machine dependent features options,
1197 see @ref{PDP-11-Options}.
1200 @item -mpic | -mno-pic
1201 Generate position-independent (or position-dependent) code. The
1202 default is @option{-mpic}.
1205 @itemx -mall-extensions
1206 Enable all instruction set extensions. This is the default.
1208 @item -mno-extensions
1209 Disable all instruction set extensions.
1211 @item -m@var{extension} | -mno-@var{extension}
1212 Enable (or disable) a particular instruction set extension.
1215 Enable the instruction set extensions supported by a particular CPU, and
1216 disable all other extensions.
1218 @item -m@var{machine}
1219 Enable the instruction set extensions supported by a particular machine
1220 model, and disable all other extensions.
1226 The following options are available when @value{AS} is configured for
1227 a picoJava processor.
1231 @cindex PJ endianness
1232 @cindex endianness, PJ
1233 @cindex big endian output, PJ
1235 Generate ``big endian'' format output.
1237 @cindex little endian output, PJ
1239 Generate ``little endian'' format output.
1247 @xref{PRU Options}, for the options available when @value{AS} is configured
1248 for a PRU processor.
1252 @c man begin OPTIONS
1253 The following options are available when @value{AS} is configured for a
1256 @c man begin INCLUDE
1258 @c ended inside the included file
1263 The following options are available when @value{AS} is configured for the
1264 Motorola 68HC11 or 68HC12 series.
1268 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1269 Specify what processor is the target. The default is
1270 defined by the configuration option when building the assembler.
1272 @item --xgate-ramoffset
1273 Instruct the linker to offset RAM addresses from S12X address space into
1274 XGATE address space.
1277 Specify to use the 16-bit integer ABI.
1280 Specify to use the 32-bit integer ABI.
1282 @item -mshort-double
1283 Specify to use the 32-bit double ABI.
1286 Specify to use the 64-bit double ABI.
1288 @item --force-long-branches
1289 Relative branches are turned into absolute ones. This concerns
1290 conditional branches, unconditional branches and branches to a
1293 @item -S | --short-branches
1294 Do not turn relative branches into absolute ones
1295 when the offset is out of range.
1297 @item --strict-direct-mode
1298 Do not turn the direct addressing mode into extended addressing mode
1299 when the instruction does not support direct addressing mode.
1301 @item --print-insn-syntax
1302 Print the syntax of instruction in case of error.
1304 @item --print-opcodes
1305 Print the list of instructions with syntax and then exit.
1307 @item --generate-example
1308 Print an example of instruction for each possible instruction and then exit.
1309 This option is only useful for testing @command{@value{AS}}.
1315 The following options are available when @command{@value{AS}} is configured
1316 for the SPARC architecture:
1319 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1320 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1321 Explicitly select a variant of the SPARC architecture.
1323 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1324 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1326 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1327 UltraSPARC extensions.
1329 @item -xarch=v8plus | -xarch=v8plusa
1330 For compatibility with the Solaris v9 assembler. These options are
1331 equivalent to -Av8plus and -Av8plusa, respectively.
1334 Warn when the assembler switches to another architecture.
1339 The following options are available when @value{AS} is configured for the 'c54x
1344 Enable extended addressing mode. All addresses and relocations will assume
1345 extended addressing (usually 23 bits).
1346 @item -mcpu=@var{CPU_VERSION}
1347 Sets the CPU version being compiled for.
1348 @item -merrors-to-file @var{FILENAME}
1349 Redirect error output to a file, for broken systems which don't support such
1350 behaviour in the shell.
1355 @c man begin OPTIONS
1356 The following options are available when @value{AS} is configured for
1361 This option sets the largest size of an object that can be referenced
1362 implicitly with the @code{gp} register. It is only accepted for targets that
1363 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1365 @cindex MIPS endianness
1366 @cindex endianness, MIPS
1367 @cindex big endian output, MIPS
1369 Generate ``big endian'' format output.
1371 @cindex little endian output, MIPS
1373 Generate ``little endian'' format output.
1391 Generate code for a particular MIPS Instruction Set Architecture level.
1392 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1393 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1394 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1395 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1396 @samp{-mips32r5}, @samp{-mips32r6}, @samp{-mips64}, @samp{-mips64r2},
1397 @samp{-mips64r3}, @samp{-mips64r5}, and @samp{-mips64r6} correspond to generic
1398 MIPS V, MIPS32, MIPS32 Release 2, MIPS32 Release 3, MIPS32 Release 5, MIPS32
1399 Release 6, MIPS64, MIPS64 Release 2, MIPS64 Release 3, MIPS64 Release 5, and
1400 MIPS64 Release 6 ISA processors, respectively.
1402 @item -march=@var{cpu}
1403 Generate code for a particular MIPS CPU.
1405 @item -mtune=@var{cpu}
1406 Schedule and tune for a particular MIPS CPU.
1410 Cause nops to be inserted if the read of the destination register
1411 of an mfhi or mflo instruction occurs in the following two instructions.
1414 @itemx -mno-fix-rm7000
1415 Cause nops to be inserted if a dmult or dmultu instruction is
1416 followed by a load instruction.
1420 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1421 section instead of the standard ELF .stabs sections.
1425 Control generation of @code{.pdr} sections.
1429 The register sizes are normally inferred from the ISA and ABI, but these
1430 flags force a certain group of registers to be treated as 32 bits wide at
1431 all times. @samp{-mgp32} controls the size of general-purpose registers
1432 and @samp{-mfp32} controls the size of floating-point registers.
1436 The register sizes are normally inferred from the ISA and ABI, but these
1437 flags force a certain group of registers to be treated as 64 bits wide at
1438 all times. @samp{-mgp64} controls the size of general-purpose registers
1439 and @samp{-mfp64} controls the size of floating-point registers.
1442 The register sizes are normally inferred from the ISA and ABI, but using
1443 this flag in combination with @samp{-mabi=32} enables an ABI variant
1444 which will operate correctly with floating-point registers which are
1448 @itemx -mno-odd-spreg
1449 Enable use of floating-point operations on odd-numbered single-precision
1450 registers when supported by the ISA. @samp{-mfpxx} implies
1451 @samp{-mno-odd-spreg}, otherwise the default is @samp{-modd-spreg}.
1455 Generate code for the MIPS 16 processor. This is equivalent to putting
1456 @code{.module mips16} at the start of the assembly file. @samp{-no-mips16}
1457 turns off this option.
1460 @itemx -mno-mips16e2
1461 Enable the use of MIPS16e2 instructions in MIPS16 mode. This is equivalent
1462 to putting @code{.module mips16e2} at the start of the assembly file.
1463 @samp{-mno-mips16e2} turns off this option.
1466 @itemx -mno-micromips
1467 Generate code for the microMIPS processor. This is equivalent to putting
1468 @code{.module micromips} at the start of the assembly file.
1469 @samp{-mno-micromips} turns off this option. This is equivalent to putting
1470 @code{.module nomicromips} at the start of the assembly file.
1473 @itemx -mno-smartmips
1474 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1475 equivalent to putting @code{.module smartmips} at the start of the assembly
1476 file. @samp{-mno-smartmips} turns off this option.
1480 Generate code for the MIPS-3D Application Specific Extension.
1481 This tells the assembler to accept MIPS-3D instructions.
1482 @samp{-no-mips3d} turns off this option.
1486 Generate code for the MDMX Application Specific Extension.
1487 This tells the assembler to accept MDMX instructions.
1488 @samp{-no-mdmx} turns off this option.
1492 Generate code for the DSP Release 1 Application Specific Extension.
1493 This tells the assembler to accept DSP Release 1 instructions.
1494 @samp{-mno-dsp} turns off this option.
1498 Generate code for the DSP Release 2 Application Specific Extension.
1499 This option implies @samp{-mdsp}.
1500 This tells the assembler to accept DSP Release 2 instructions.
1501 @samp{-mno-dspr2} turns off this option.
1505 Generate code for the DSP Release 3 Application Specific Extension.
1506 This option implies @samp{-mdsp} and @samp{-mdspr2}.
1507 This tells the assembler to accept DSP Release 3 instructions.
1508 @samp{-mno-dspr3} turns off this option.
1512 Generate code for the MIPS SIMD Architecture Extension.
1513 This tells the assembler to accept MSA instructions.
1514 @samp{-mno-msa} turns off this option.
1518 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1519 This tells the assembler to accept XPA instructions.
1520 @samp{-mno-xpa} turns off this option.
1524 Generate code for the MT Application Specific Extension.
1525 This tells the assembler to accept MT instructions.
1526 @samp{-mno-mt} turns off this option.
1530 Generate code for the MCU Application Specific Extension.
1531 This tells the assembler to accept MCU instructions.
1532 @samp{-mno-mcu} turns off this option.
1536 Only use 32-bit instruction encodings when generating code for the
1537 microMIPS processor. This option inhibits the use of any 16-bit
1538 instructions. This is equivalent to putting @code{.set insn32} at
1539 the start of the assembly file. @samp{-mno-insn32} turns off this
1540 option. This is equivalent to putting @code{.set noinsn32} at the
1541 start of the assembly file. By default @samp{-mno-insn32} is
1542 selected, allowing all instructions to be used.
1544 @item --construct-floats
1545 @itemx --no-construct-floats
1546 The @samp{--no-construct-floats} option disables the construction of
1547 double width floating point constants by loading the two halves of the
1548 value into the two single width floating point registers that make up
1549 the double width register. By default @samp{--construct-floats} is
1550 selected, allowing construction of these floating point constants.
1552 @item --relax-branch
1553 @itemx --no-relax-branch
1554 The @samp{--relax-branch} option enables the relaxation of out-of-range
1555 branches. By default @samp{--no-relax-branch} is selected, causing any
1556 out-of-range branches to produce an error.
1558 @item -mignore-branch-isa
1559 @itemx -mno-ignore-branch-isa
1560 Ignore branch checks for invalid transitions between ISA modes. The
1561 semantics of branches does not provide for an ISA mode switch, so in
1562 most cases the ISA mode a branch has been encoded for has to be the
1563 same as the ISA mode of the branch's target label. Therefore GAS has
1564 checks implemented that verify in branch assembly that the two ISA
1565 modes match. @samp{-mignore-branch-isa} disables these checks. By
1566 default @samp{-mno-ignore-branch-isa} is selected, causing any invalid
1567 branch requiring a transition between ISA modes to produce an error.
1569 @item -mnan=@var{encoding}
1570 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1571 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1574 @item --emulation=@var{name}
1575 This option was formerly used to switch between ELF and ECOFF output
1576 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1577 removed in GAS 2.24, so the option now serves little purpose.
1578 It is retained for backwards compatibility.
1580 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1581 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1582 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1583 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1584 preferred options instead.
1587 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1594 Control how to deal with multiplication overflow and division by zero.
1595 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1596 (and only work for Instruction Set Architecture level 2 and higher);
1597 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1601 When this option is used, @command{@value{AS}} will issue a warning every
1602 time it generates a nop instruction from a macro.
1608 The following options are available when @value{AS} is configured for
1614 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1615 The command line option @samp{-nojsri2bsr} can be used to disable it.
1619 Enable or disable the silicon filter behaviour. By default this is disabled.
1620 The default can be overridden by the @samp{-sifilter} command line option.
1623 Alter jump instructions for long displacements.
1625 @item -mcpu=[210|340]
1626 Select the cpu type on the target hardware. This controls which instructions
1630 Assemble for a big endian target.
1633 Assemble for a little endian target.
1642 @xref{Meta Options}, for the options available when @value{AS} is configured
1643 for a Meta processor.
1647 @c man begin OPTIONS
1648 The following options are available when @value{AS} is configured for a
1651 @c man begin INCLUDE
1652 @include c-metag.texi
1653 @c ended inside the included file
1658 @c man begin OPTIONS
1660 See the info pages for documentation of the MMIX-specific options.
1666 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1667 for a NDS32 processor.
1669 @c ended inside the included file
1673 @c man begin OPTIONS
1674 The following options are available when @value{AS} is configured for a
1677 @c man begin INCLUDE
1678 @include c-nds32.texi
1679 @c ended inside the included file
1686 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1687 for a PowerPC processor.
1691 @c man begin OPTIONS
1692 The following options are available when @value{AS} is configured for a
1695 @c man begin INCLUDE
1697 @c ended inside the included file
1705 @xref{RISC-V-Opts}, for the options available when @value{AS} is configured
1706 for a RISC-V processor.
1710 @c man begin OPTIONS
1711 The following options are available when @value{AS} is configured for a
1714 @c man begin INCLUDE
1715 @include c-riscv.texi
1716 @c ended inside the included file
1721 @c man begin OPTIONS
1723 See the info pages for documentation of the RX-specific options.
1727 The following options are available when @value{AS} is configured for the s390
1733 Select the word size, either 31/32 bits or 64 bits.
1736 Select the architecture mode, either the Enterprise System
1737 Architecture (esa) or the z/Architecture mode (zarch).
1738 @item -march=@var{processor}
1739 Specify which s390 processor variant is the target, @samp{g5} (or
1740 @samp{arch3}), @samp{g6}, @samp{z900} (or @samp{arch5}), @samp{z990} (or
1741 @samp{arch6}), @samp{z9-109}, @samp{z9-ec} (or @samp{arch7}), @samp{z10} (or
1742 @samp{arch8}), @samp{z196} (or @samp{arch9}), @samp{zEC12} (or @samp{arch10}),
1743 @samp{z13} (or @samp{arch11}), or @samp{arch12}.
1745 @itemx -mno-regnames
1746 Allow or disallow symbolic names for registers.
1747 @item -mwarn-areg-zero
1748 Warn whenever the operand for a base or index register has been specified
1749 but evaluates to zero.
1757 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1758 for a TMS320C6000 processor.
1762 @c man begin OPTIONS
1763 The following options are available when @value{AS} is configured for a
1764 TMS320C6000 processor.
1766 @c man begin INCLUDE
1767 @include c-tic6x.texi
1768 @c ended inside the included file
1776 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1777 for a TILE-Gx processor.
1781 @c man begin OPTIONS
1782 The following options are available when @value{AS} is configured for a TILE-Gx
1785 @c man begin INCLUDE
1786 @include c-tilegx.texi
1787 @c ended inside the included file
1795 @xref{Visium Options}, for the options available when @value{AS} is configured
1796 for a Visium processor.
1800 @c man begin OPTIONS
1801 The following option is available when @value{AS} is configured for a Visium
1804 @c man begin INCLUDE
1805 @include c-visium.texi
1806 @c ended inside the included file
1814 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1815 for an Xtensa processor.
1819 @c man begin OPTIONS
1820 The following options are available when @value{AS} is configured for an
1823 @c man begin INCLUDE
1824 @include c-xtensa.texi
1825 @c ended inside the included file
1830 @c man begin OPTIONS
1833 The following options are available when @value{AS} is configured for
1834 a Z80 family processor.
1837 Assemble for Z80 processor.
1839 Assemble for R800 processor.
1840 @item -ignore-undocumented-instructions
1842 Assemble undocumented Z80 instructions that also work on R800 without warning.
1843 @item -ignore-unportable-instructions
1845 Assemble all undocumented Z80 instructions without warning.
1846 @item -warn-undocumented-instructions
1848 Issue a warning for undocumented Z80 instructions that also work on R800.
1849 @item -warn-unportable-instructions
1851 Issue a warning for undocumented Z80 instructions that do not work on R800.
1852 @item -forbid-undocumented-instructions
1854 Treat all undocumented instructions as errors.
1855 @item -forbid-unportable-instructions
1857 Treat undocumented Z80 instructions that do not work on R800 as errors.
1864 * Manual:: Structure of this Manual
1865 * GNU Assembler:: The GNU Assembler
1866 * Object Formats:: Object File Formats
1867 * Command Line:: Command Line
1868 * Input Files:: Input Files
1869 * Object:: Output (Object) File
1870 * Errors:: Error and Warning Messages
1874 @section Structure of this Manual
1876 @cindex manual, structure and purpose
1877 This manual is intended to describe what you need to know to use
1878 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1879 notation for symbols, constants, and expressions; the directives that
1880 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1883 We also cover special features in the @value{TARGET}
1884 configuration of @command{@value{AS}}, including assembler directives.
1887 This manual also describes some of the machine-dependent features of
1888 various flavors of the assembler.
1891 @cindex machine instructions (not covered)
1892 On the other hand, this manual is @emph{not} intended as an introduction
1893 to programming in assembly language---let alone programming in general!
1894 In a similar vein, we make no attempt to introduce the machine
1895 architecture; we do @emph{not} describe the instruction set, standard
1896 mnemonics, registers or addressing modes that are standard to a
1897 particular architecture.
1899 You may want to consult the manufacturer's
1900 machine architecture manual for this information.
1904 For information on the H8/300 machine instruction set, see @cite{H8/300
1905 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1906 Programming Manual} (Renesas).
1909 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1910 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1911 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1912 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1915 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1919 @c I think this is premature---doc@cygnus.com, 17jan1991
1921 Throughout this manual, we assume that you are running @dfn{GNU},
1922 the portable operating system from the @dfn{Free Software
1923 Foundation, Inc.}. This restricts our attention to certain kinds of
1924 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1925 once this assumption is granted examples and definitions need less
1928 @command{@value{AS}} is part of a team of programs that turn a high-level
1929 human-readable series of instructions into a low-level
1930 computer-readable series of instructions. Different versions of
1931 @command{@value{AS}} are used for different kinds of computer.
1934 @c There used to be a section "Terminology" here, which defined
1935 @c "contents", "byte", "word", and "long". Defining "word" to any
1936 @c particular size is confusing when the .word directive may generate 16
1937 @c bits on one machine and 32 bits on another; in general, for the user
1938 @c version of this manual, none of these terms seem essential to define.
1939 @c They were used very little even in the former draft of the manual;
1940 @c this draft makes an effort to avoid them (except in names of
1944 @section The GNU Assembler
1946 @c man begin DESCRIPTION
1948 @sc{gnu} @command{as} is really a family of assemblers.
1950 This manual describes @command{@value{AS}}, a member of that family which is
1951 configured for the @value{TARGET} architectures.
1953 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1954 should find a fairly similar environment when you use it on another
1955 architecture. Each version has much in common with the others,
1956 including object file formats, most assembler directives (often called
1957 @dfn{pseudo-ops}) and assembler syntax.@refill
1959 @cindex purpose of @sc{gnu} assembler
1960 @command{@value{AS}} is primarily intended to assemble the output of the
1961 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1962 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1963 assemble correctly everything that other assemblers for the same
1964 machine would assemble.
1966 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1969 @c This remark should appear in generic version of manual; assumption
1970 @c here is that generic version sets M680x0.
1971 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1972 assembler for the same architecture; for example, we know of several
1973 incompatible versions of 680x0 assembly language syntax.
1978 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1979 program in one pass of the source file. This has a subtle impact on the
1980 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1982 @node Object Formats
1983 @section Object File Formats
1985 @cindex object file format
1986 The @sc{gnu} assembler can be configured to produce several alternative
1987 object file formats. For the most part, this does not affect how you
1988 write assembly language programs; but directives for debugging symbols
1989 are typically different in different file formats. @xref{Symbol
1990 Attributes,,Symbol Attributes}.
1993 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1994 @value{OBJ-NAME} format object files.
1996 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1998 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1999 @code{b.out} or COFF format object files.
2002 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
2003 SOM or ELF format object files.
2008 @section Command Line
2010 @cindex command line conventions
2012 After the program name @command{@value{AS}}, the command line may contain
2013 options and file names. Options may appear in any order, and may be
2014 before, after, or between file names. The order of file names is
2017 @cindex standard input, as input file
2019 @file{--} (two hyphens) by itself names the standard input file
2020 explicitly, as one of the files for @command{@value{AS}} to assemble.
2022 @cindex options, command line
2023 Except for @samp{--} any command line argument that begins with a
2024 hyphen (@samp{-}) is an option. Each option changes the behavior of
2025 @command{@value{AS}}. No option changes the way another option works. An
2026 option is a @samp{-} followed by one or more letters; the case of
2027 the letter is important. All options are optional.
2029 Some options expect exactly one file name to follow them. The file
2030 name may either immediately follow the option's letter (compatible
2031 with older assemblers) or it may be the next command argument (@sc{gnu}
2032 standard). These two command lines are equivalent:
2035 @value{AS} -o my-object-file.o mumble.s
2036 @value{AS} -omy-object-file.o mumble.s
2040 @section Input Files
2043 @cindex source program
2044 @cindex files, input
2045 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
2046 describe the program input to one run of @command{@value{AS}}. The program may
2047 be in one or more files; how the source is partitioned into files
2048 doesn't change the meaning of the source.
2050 @c I added "con" prefix to "catenation" just to prove I can overcome my
2051 @c APL training... doc@cygnus.com
2052 The source program is a concatenation of the text in all the files, in the
2055 @c man begin DESCRIPTION
2056 Each time you run @command{@value{AS}} it assembles exactly one source
2057 program. The source program is made up of one or more files.
2058 (The standard input is also a file.)
2060 You give @command{@value{AS}} a command line that has zero or more input file
2061 names. The input files are read (from left file name to right). A
2062 command line argument (in any position) that has no special meaning
2063 is taken to be an input file name.
2065 If you give @command{@value{AS}} no file names it attempts to read one input file
2066 from the @command{@value{AS}} standard input, which is normally your terminal. You
2067 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
2070 Use @samp{--} if you need to explicitly name the standard input file
2071 in your command line.
2073 If the source is empty, @command{@value{AS}} produces a small, empty object
2078 @subheading Filenames and Line-numbers
2080 @cindex input file linenumbers
2081 @cindex line numbers, in input files
2082 There are two ways of locating a line in the input file (or files) and
2083 either may be used in reporting error messages. One way refers to a line
2084 number in a physical file; the other refers to a line number in a
2085 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
2087 @dfn{Physical files} are those files named in the command line given
2088 to @command{@value{AS}}.
2090 @dfn{Logical files} are simply names declared explicitly by assembler
2091 directives; they bear no relation to physical files. Logical file names help
2092 error messages reflect the original source file, when @command{@value{AS}} source
2093 is itself synthesized from other files. @command{@value{AS}} understands the
2094 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
2095 @ref{File,,@code{.file}}.
2098 @section Output (Object) File
2104 Every time you run @command{@value{AS}} it produces an output file, which is
2105 your assembly language program translated into numbers. This file
2106 is the object file. Its default name is
2114 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
2116 You can give it another name by using the @option{-o} option. Conventionally,
2117 object file names end with @file{.o}. The default name is used for historical
2118 reasons: older assemblers were capable of assembling self-contained programs
2119 directly into a runnable program. (For some formats, this isn't currently
2120 possible, but it can be done for the @code{a.out} format.)
2124 The object file is meant for input to the linker @code{@value{LD}}. It contains
2125 assembled program code, information to help @code{@value{LD}} integrate
2126 the assembled program into a runnable file, and (optionally) symbolic
2127 information for the debugger.
2129 @c link above to some info file(s) like the description of a.out.
2130 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
2133 @section Error and Warning Messages
2135 @c man begin DESCRIPTION
2137 @cindex error messages
2138 @cindex warning messages
2139 @cindex messages from assembler
2140 @command{@value{AS}} may write warnings and error messages to the standard error
2141 file (usually your terminal). This should not happen when a compiler
2142 runs @command{@value{AS}} automatically. Warnings report an assumption made so
2143 that @command{@value{AS}} could keep assembling a flawed program; errors report a
2144 grave problem that stops the assembly.
2148 @cindex format of warning messages
2149 Warning messages have the format
2152 file_name:@b{NNN}:Warning Message Text
2156 @cindex file names and line numbers, in warnings/errors
2157 (where @b{NNN} is a line number). If both a logical file name
2158 (@pxref{File,,@code{.file}}) and a logical line number
2160 (@pxref{Line,,@code{.line}})
2162 have been given then they will be used, otherwise the file name and line number
2163 in the current assembler source file will be used. The message text is
2164 intended to be self explanatory (in the grand Unix tradition).
2166 Note the file name must be set via the logical version of the @code{.file}
2167 directive, not the DWARF2 version of the @code{.file} directive. For example:
2171 error_assembler_source
2177 produces this output:
2181 asm.s:2: Error: no such instruction: `error_assembler_source'
2182 foo.c:31: Error: no such instruction: `error_c_source'
2185 @cindex format of error messages
2186 Error messages have the format
2189 file_name:@b{NNN}:FATAL:Error Message Text
2192 The file name and line number are derived as for warning
2193 messages. The actual message text may be rather less explanatory
2194 because many of them aren't supposed to happen.
2197 @chapter Command-Line Options
2199 @cindex options, all versions of assembler
2200 This chapter describes command-line options available in @emph{all}
2201 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
2202 for options specific
2204 to the @value{TARGET} target.
2207 to particular machine architectures.
2210 @c man begin DESCRIPTION
2212 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2213 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2214 The assembler arguments must be separated from each other (and the @samp{-Wa})
2215 by commas. For example:
2218 gcc -c -g -O -Wa,-alh,-L file.c
2222 This passes two options to the assembler: @samp{-alh} (emit a listing to
2223 standard output with high-level and assembly source) and @samp{-L} (retain
2224 local symbols in the symbol table).
2226 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2227 command-line options are automatically passed to the assembler by the compiler.
2228 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2229 precisely what options it passes to each compilation pass, including the
2235 * a:: -a[cdghlns] enable listings
2236 * alternate:: --alternate enable alternate macro syntax
2237 * D:: -D for compatibility
2238 * f:: -f to work faster
2239 * I:: -I for .include search path
2240 @ifclear DIFF-TBL-KLUGE
2241 * K:: -K for compatibility
2243 @ifset DIFF-TBL-KLUGE
2244 * K:: -K for difference tables
2247 * L:: -L to retain local symbols
2248 * listing:: --listing-XXX to configure listing output
2249 * M:: -M or --mri to assemble in MRI compatibility mode
2250 * MD:: --MD for dependency tracking
2251 * no-pad-sections:: --no-pad-sections to stop section padding
2252 * o:: -o to name the object file
2253 * R:: -R to join data and text sections
2254 * statistics:: --statistics to see statistics about assembly
2255 * traditional-format:: --traditional-format for compatible output
2256 * v:: -v to announce version
2257 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2258 * Z:: -Z to make object file even after errors
2262 @section Enable Listings: @option{-a[cdghlns]}
2272 @cindex listings, enabling
2273 @cindex assembly listings, enabling
2275 These options enable listing output from the assembler. By itself,
2276 @samp{-a} requests high-level, assembly, and symbols listing.
2277 You can use other letters to select specific options for the list:
2278 @samp{-ah} requests a high-level language listing,
2279 @samp{-al} requests an output-program assembly listing, and
2280 @samp{-as} requests a symbol table listing.
2281 High-level listings require that a compiler debugging option like
2282 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2285 Use the @samp{-ag} option to print a first section with general assembly
2286 information, like @value{AS} version, switches passed, or time stamp.
2288 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2289 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2290 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2291 omitted from the listing.
2293 Use the @samp{-ad} option to omit debugging directives from the
2296 Once you have specified one of these options, you can further control
2297 listing output and its appearance using the directives @code{.list},
2298 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2300 The @samp{-an} option turns off all forms processing.
2301 If you do not request listing output with one of the @samp{-a} options, the
2302 listing-control directives have no effect.
2304 The letters after @samp{-a} may be combined into one option,
2305 @emph{e.g.}, @samp{-aln}.
2307 Note if the assembler source is coming from the standard input (e.g.,
2309 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
2310 is being used) then the listing will not contain any comments or preprocessor
2311 directives. This is because the listing code buffers input source lines from
2312 stdin only after they have been preprocessed by the assembler. This reduces
2313 memory usage and makes the code more efficient.
2316 @section @option{--alternate}
2319 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2322 @section @option{-D}
2325 This option has no effect whatsoever, but it is accepted to make it more
2326 likely that scripts written for other assemblers also work with
2327 @command{@value{AS}}.
2330 @section Work Faster: @option{-f}
2333 @cindex trusted compiler
2334 @cindex faster processing (@option{-f})
2335 @samp{-f} should only be used when assembling programs written by a
2336 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2337 and comment preprocessing on
2338 the input file(s) before assembling them. @xref{Preprocessing,
2342 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2343 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2348 @section @code{.include} Search Path: @option{-I} @var{path}
2350 @kindex -I @var{path}
2351 @cindex paths for @code{.include}
2352 @cindex search path for @code{.include}
2353 @cindex @code{include} directive search path
2354 Use this option to add a @var{path} to the list of directories
2355 @command{@value{AS}} searches for files specified in @code{.include}
2356 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2357 many times as necessary to include a variety of paths. The current
2358 working directory is always searched first; after that, @command{@value{AS}}
2359 searches any @samp{-I} directories in the same order as they were
2360 specified (left to right) on the command line.
2363 @section Difference Tables: @option{-K}
2366 @ifclear DIFF-TBL-KLUGE
2367 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2368 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2369 where it can be used to warn when the assembler alters the machine code
2370 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2371 family does not have the addressing limitations that sometimes lead to this
2372 alteration on other platforms.
2375 @ifset DIFF-TBL-KLUGE
2376 @cindex difference tables, warning
2377 @cindex warning for altered difference tables
2378 @command{@value{AS}} sometimes alters the code emitted for directives of the
2379 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2380 You can use the @samp{-K} option if you want a warning issued when this
2385 @section Include Local Symbols: @option{-L}
2388 @cindex local symbols, retaining in output
2389 Symbols beginning with system-specific local label prefixes, typically
2390 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2391 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2392 such symbols when debugging, because they are intended for the use of
2393 programs (like compilers) that compose assembler programs, not for your
2394 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2395 such symbols, so you do not normally debug with them.
2397 This option tells @command{@value{AS}} to retain those local symbols
2398 in the object file. Usually if you do this you also tell the linker
2399 @code{@value{LD}} to preserve those symbols.
2402 @section Configuring listing output: @option{--listing}
2404 The listing feature of the assembler can be enabled via the command line switch
2405 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2406 hex dump of the corresponding locations in the output object file, and displays
2407 them as a listing file. The format of this listing can be controlled by
2408 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2409 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2410 @code{.psize} (@pxref{Psize}), and
2411 @code{.eject} (@pxref{Eject}) and also by the following switches:
2414 @item --listing-lhs-width=@samp{number}
2415 @kindex --listing-lhs-width
2416 @cindex Width of first line disassembly output
2417 Sets the maximum width, in words, of the first line of the hex byte dump. This
2418 dump appears on the left hand side of the listing output.
2420 @item --listing-lhs-width2=@samp{number}
2421 @kindex --listing-lhs-width2
2422 @cindex Width of continuation lines of disassembly output
2423 Sets the maximum width, in words, of any further lines of the hex byte dump for
2424 a given input source line. If this value is not specified, it defaults to being
2425 the same as the value specified for @samp{--listing-lhs-width}. If neither
2426 switch is used the default is to one.
2428 @item --listing-rhs-width=@samp{number}
2429 @kindex --listing-rhs-width
2430 @cindex Width of source line output
2431 Sets the maximum width, in characters, of the source line that is displayed
2432 alongside the hex dump. The default value for this parameter is 100. The
2433 source line is displayed on the right hand side of the listing output.
2435 @item --listing-cont-lines=@samp{number}
2436 @kindex --listing-cont-lines
2437 @cindex Maximum number of continuation lines
2438 Sets the maximum number of continuation lines of hex dump that will be
2439 displayed for a given single line of source input. The default value is 4.
2443 @section Assemble in MRI Compatibility Mode: @option{-M}
2446 @cindex MRI compatibility mode
2447 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2448 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2449 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
2450 configured target) assembler from Microtec Research. The exact nature of the
2451 MRI syntax will not be documented here; see the MRI manuals for more
2452 information. Note in particular that the handling of macros and macro
2453 arguments is somewhat different. The purpose of this option is to permit
2454 assembling existing MRI assembler code using @command{@value{AS}}.
2456 The MRI compatibility is not complete. Certain operations of the MRI assembler
2457 depend upon its object file format, and can not be supported using other object
2458 file formats. Supporting these would require enhancing each object file format
2459 individually. These are:
2462 @item global symbols in common section
2464 The m68k MRI assembler supports common sections which are merged by the linker.
2465 Other object file formats do not support this. @command{@value{AS}} handles
2466 common sections by treating them as a single common symbol. It permits local
2467 symbols to be defined within a common section, but it can not support global
2468 symbols, since it has no way to describe them.
2470 @item complex relocations
2472 The MRI assemblers support relocations against a negated section address, and
2473 relocations which combine the start addresses of two or more sections. These
2474 are not support by other object file formats.
2476 @item @code{END} pseudo-op specifying start address
2478 The MRI @code{END} pseudo-op permits the specification of a start address.
2479 This is not supported by other object file formats. The start address may
2480 instead be specified using the @option{-e} option to the linker, or in a linker
2483 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2485 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2486 name to the output file. This is not supported by other object file formats.
2488 @item @code{ORG} pseudo-op
2490 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2491 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2492 which changes the location within the current section. Absolute sections are
2493 not supported by other object file formats. The address of a section may be
2494 assigned within a linker script.
2497 There are some other features of the MRI assembler which are not supported by
2498 @command{@value{AS}}, typically either because they are difficult or because they
2499 seem of little consequence. Some of these may be supported in future releases.
2503 @item EBCDIC strings
2505 EBCDIC strings are not supported.
2507 @item packed binary coded decimal
2509 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2510 and @code{DCB.P} pseudo-ops are not supported.
2512 @item @code{FEQU} pseudo-op
2514 The m68k @code{FEQU} pseudo-op is not supported.
2516 @item @code{NOOBJ} pseudo-op
2518 The m68k @code{NOOBJ} pseudo-op is not supported.
2520 @item @code{OPT} branch control options
2522 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2523 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2524 relaxes all branches, whether forward or backward, to an appropriate size, so
2525 these options serve no purpose.
2527 @item @code{OPT} list control options
2529 The following m68k @code{OPT} list control options are ignored: @code{C},
2530 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2531 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2533 @item other @code{OPT} options
2535 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2536 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2538 @item @code{OPT} @code{D} option is default
2540 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2541 @code{OPT NOD} may be used to turn it off.
2543 @item @code{XREF} pseudo-op.
2545 The m68k @code{XREF} pseudo-op is ignored.
2547 @item @code{.debug} pseudo-op
2549 The i960 @code{.debug} pseudo-op is not supported.
2551 @item @code{.extended} pseudo-op
2553 The i960 @code{.extended} pseudo-op is not supported.
2555 @item @code{.list} pseudo-op.
2557 The various options of the i960 @code{.list} pseudo-op are not supported.
2559 @item @code{.optimize} pseudo-op
2561 The i960 @code{.optimize} pseudo-op is not supported.
2563 @item @code{.output} pseudo-op
2565 The i960 @code{.output} pseudo-op is not supported.
2567 @item @code{.setreal} pseudo-op
2569 The i960 @code{.setreal} pseudo-op is not supported.
2574 @section Dependency Tracking: @option{--MD}
2577 @cindex dependency tracking
2580 @command{@value{AS}} can generate a dependency file for the file it creates. This
2581 file consists of a single rule suitable for @code{make} describing the
2582 dependencies of the main source file.
2584 The rule is written to the file named in its argument.
2586 This feature is used in the automatic updating of makefiles.
2588 @node no-pad-sections
2589 @section Output Section Padding
2590 @kindex --no-pad-sections
2591 @cindex output section padding
2592 Normally the assembler will pad the end of each output section up to its
2593 alignment boundary. But this can waste space, which can be significant on
2594 memory constrained targets. So the @option{--no-pad-sections} option will
2595 disable this behaviour.
2598 @section Name the Object File: @option{-o}
2601 @cindex naming object file
2602 @cindex object file name
2603 There is always one object file output when you run @command{@value{AS}}. By
2604 default it has the name
2607 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2621 You use this option (which takes exactly one filename) to give the
2622 object file a different name.
2624 Whatever the object file is called, @command{@value{AS}} overwrites any
2625 existing file of the same name.
2628 @section Join Data and Text Sections: @option{-R}
2631 @cindex data and text sections, joining
2632 @cindex text and data sections, joining
2633 @cindex joining text and data sections
2634 @cindex merging text and data sections
2635 @option{-R} tells @command{@value{AS}} to write the object file as if all
2636 data-section data lives in the text section. This is only done at
2637 the very last moment: your binary data are the same, but data
2638 section parts are relocated differently. The data section part of
2639 your object file is zero bytes long because all its bytes are
2640 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2642 When you specify @option{-R} it would be possible to generate shorter
2643 address displacements (because we do not have to cross between text and
2644 data section). We refrain from doing this simply for compatibility with
2645 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2648 When @command{@value{AS}} is configured for COFF or ELF output,
2649 this option is only useful if you use sections named @samp{.text} and
2654 @option{-R} is not supported for any of the HPPA targets. Using
2655 @option{-R} generates a warning from @command{@value{AS}}.
2659 @section Display Assembly Statistics: @option{--statistics}
2661 @kindex --statistics
2662 @cindex statistics, about assembly
2663 @cindex time, total for assembly
2664 @cindex space used, maximum for assembly
2665 Use @samp{--statistics} to display two statistics about the resources used by
2666 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2667 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2670 @node traditional-format
2671 @section Compatible Output: @option{--traditional-format}
2673 @kindex --traditional-format
2674 For some targets, the output of @command{@value{AS}} is different in some ways
2675 from the output of some existing assembler. This switch requests
2676 @command{@value{AS}} to use the traditional format instead.
2678 For example, it disables the exception frame optimizations which
2679 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2682 @section Announce Version: @option{-v}
2686 @cindex assembler version
2687 @cindex version of assembler
2688 You can find out what version of as is running by including the
2689 option @samp{-v} (which you can also spell as @samp{-version}) on the
2693 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2695 @command{@value{AS}} should never give a warning or error message when
2696 assembling compiler output. But programs written by people often
2697 cause @command{@value{AS}} to give a warning that a particular assumption was
2698 made. All such warnings are directed to the standard error file.
2702 @cindex suppressing warnings
2703 @cindex warnings, suppressing
2704 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2705 This only affects the warning messages: it does not change any particular of
2706 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2709 @kindex --fatal-warnings
2710 @cindex errors, caused by warnings
2711 @cindex warnings, causing error
2712 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2713 files that generate warnings to be in error.
2716 @cindex warnings, switching on
2717 You can switch these options off again by specifying @option{--warn}, which
2718 causes warnings to be output as usual.
2721 @section Generate Object File in Spite of Errors: @option{-Z}
2722 @cindex object file, after errors
2723 @cindex errors, continuing after
2724 After an error message, @command{@value{AS}} normally produces no output. If for
2725 some reason you are interested in object file output even after
2726 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2727 option. If there are any errors, @command{@value{AS}} continues anyways, and
2728 writes an object file after a final warning message of the form @samp{@var{n}
2729 errors, @var{m} warnings, generating bad object file.}
2734 @cindex machine-independent syntax
2735 @cindex syntax, machine-independent
2736 This chapter describes the machine-independent syntax allowed in a
2737 source file. @command{@value{AS}} syntax is similar to what many other
2738 assemblers use; it is inspired by the BSD 4.2
2743 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2747 * Preprocessing:: Preprocessing
2748 * Whitespace:: Whitespace
2749 * Comments:: Comments
2750 * Symbol Intro:: Symbols
2751 * Statements:: Statements
2752 * Constants:: Constants
2756 @section Preprocessing
2758 @cindex preprocessing
2759 The @command{@value{AS}} internal preprocessor:
2761 @cindex whitespace, removed by preprocessor
2763 adjusts and removes extra whitespace. It leaves one space or tab before
2764 the keywords on a line, and turns any other whitespace on the line into
2767 @cindex comments, removed by preprocessor
2769 removes all comments, replacing them with a single space, or an
2770 appropriate number of newlines.
2772 @cindex constants, converted by preprocessor
2774 converts character constants into the appropriate numeric values.
2777 It does not do macro processing, include file handling, or
2778 anything else you may get from your C compiler's preprocessor. You can
2779 do include file processing with the @code{.include} directive
2780 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2781 to get other ``CPP'' style preprocessing by giving the input file a
2782 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2783 Output, gcc info, Using GNU CC}.
2785 Excess whitespace, comments, and character constants
2786 cannot be used in the portions of the input text that are not
2789 @cindex turning preprocessing on and off
2790 @cindex preprocessing, turning on and off
2793 If the first line of an input file is @code{#NO_APP} or if you use the
2794 @samp{-f} option, whitespace and comments are not removed from the input file.
2795 Within an input file, you can ask for whitespace and comment removal in
2796 specific portions of the by putting a line that says @code{#APP} before the
2797 text that may contain whitespace or comments, and putting a line that says
2798 @code{#NO_APP} after this text. This feature is mainly intend to support
2799 @code{asm} statements in compilers whose output is otherwise free of comments
2806 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2807 Whitespace is used to separate symbols, and to make programs neater for
2808 people to read. Unless within character constants
2809 (@pxref{Characters,,Character Constants}), any whitespace means the same
2810 as exactly one space.
2816 There are two ways of rendering comments to @command{@value{AS}}. In both
2817 cases the comment is equivalent to one space.
2819 Anything from @samp{/*} through the next @samp{*/} is a comment.
2820 This means you may not nest these comments.
2824 The only way to include a newline ('\n') in a comment
2825 is to use this sort of comment.
2828 /* This sort of comment does not nest. */
2831 @cindex line comment character
2832 Anything from a @dfn{line comment} character up to the next newline is
2833 considered a comment and is ignored. The line comment character is target
2834 specific, and some targets multiple comment characters. Some targets also have
2835 line comment characters that only work if they are the first character on a
2836 line. Some targets use a sequence of two characters to introduce a line
2837 comment. Some targets can also change their line comment characters depending
2838 upon command line options that have been used. For more details see the
2839 @emph{Syntax} section in the documentation for individual targets.
2841 If the line comment character is the hash sign (@samp{#}) then it still has the
2842 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2843 to specify logical line numbers:
2846 @cindex lines starting with @code{#}
2847 @cindex logical line numbers
2848 To be compatible with past assemblers, lines that begin with @samp{#} have a
2849 special interpretation. Following the @samp{#} should be an absolute
2850 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2851 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2852 new logical file name. The rest of the line, if any, should be whitespace.
2854 If the first non-whitespace characters on the line are not numeric,
2855 the line is ignored. (Just like a comment.)
2858 # This is an ordinary comment.
2859 # 42-6 "new_file_name" # New logical file name
2860 # This is logical line # 36.
2862 This feature is deprecated, and may disappear from future versions
2863 of @command{@value{AS}}.
2868 @cindex characters used in symbols
2869 @ifclear SPECIAL-SYMS
2870 A @dfn{symbol} is one or more characters chosen from the set of all
2871 letters (both upper and lower case), digits and the three characters
2877 A @dfn{symbol} is one or more characters chosen from the set of all
2878 letters (both upper and lower case), digits and the three characters
2879 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2885 On most machines, you can also use @code{$} in symbol names; exceptions
2886 are noted in @ref{Machine Dependencies}.
2888 No symbol may begin with a digit. Case is significant.
2889 There is no length limit; all characters are significant. Multibyte characters
2890 are supported. Symbols are delimited by characters not in that set, or by the
2891 beginning of a file (since the source program must end with a newline, the end
2892 of a file is not a possible symbol delimiter). @xref{Symbols}.
2894 Symbol names may also be enclosed in double quote @code{"} characters. In such
2895 cases any characters are allowed, except for the NUL character. If a double
2896 quote character is to be included in the symbol name it must be preceeded by a
2897 backslash @code{\} character.
2898 @cindex length of symbols
2903 @cindex statements, structure of
2904 @cindex line separator character
2905 @cindex statement separator character
2907 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2908 @dfn{line separator character}. The line separator character is target
2909 specific and described in the @emph{Syntax} section of each
2910 target's documentation. Not all targets support a line separator character.
2911 The newline or line separator character is considered to be part of the
2912 preceding statement. Newlines and separators within character constants are an
2913 exception: they do not end statements.
2915 @cindex newline, required at file end
2916 @cindex EOF, newline must precede
2917 It is an error to end any statement with end-of-file: the last
2918 character of any input file should be a newline.@refill
2920 An empty statement is allowed, and may include whitespace. It is ignored.
2922 @cindex instructions and directives
2923 @cindex directives and instructions
2924 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2925 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2927 A statement begins with zero or more labels, optionally followed by a
2928 key symbol which determines what kind of statement it is. The key
2929 symbol determines the syntax of the rest of the statement. If the
2930 symbol begins with a dot @samp{.} then the statement is an assembler
2931 directive: typically valid for any computer. If the symbol begins with
2932 a letter the statement is an assembly language @dfn{instruction}: it
2933 assembles into a machine language instruction.
2935 Different versions of @command{@value{AS}} for different computers
2936 recognize different instructions. In fact, the same symbol may
2937 represent a different instruction in a different computer's assembly
2941 @cindex @code{:} (label)
2942 @cindex label (@code{:})
2943 A label is a symbol immediately followed by a colon (@code{:}).
2944 Whitespace before a label or after a colon is permitted, but you may not
2945 have whitespace between a label's symbol and its colon. @xref{Labels}.
2948 For HPPA targets, labels need not be immediately followed by a colon, but
2949 the definition of a label must begin in column zero. This also implies that
2950 only one label may be defined on each line.
2954 label: .directive followed by something
2955 another_label: # This is an empty statement.
2956 instruction operand_1, operand_2, @dots{}
2963 A constant is a number, written so that its value is known by
2964 inspection, without knowing any context. Like this:
2967 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2968 .ascii "Ring the bell\7" # A string constant.
2969 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2970 .float 0f-314159265358979323846264338327\
2971 95028841971.693993751E-40 # - pi, a flonum.
2976 * Characters:: Character Constants
2977 * Numbers:: Number Constants
2981 @subsection Character Constants
2983 @cindex character constants
2984 @cindex constants, character
2985 There are two kinds of character constants. A @dfn{character} stands
2986 for one character in one byte and its value may be used in
2987 numeric expressions. String constants (properly called string
2988 @emph{literals}) are potentially many bytes and their values may not be
2989 used in arithmetic expressions.
2993 * Chars:: Characters
2997 @subsubsection Strings
2999 @cindex string constants
3000 @cindex constants, string
3001 A @dfn{string} is written between double-quotes. It may contain
3002 double-quotes or null characters. The way to get special characters
3003 into a string is to @dfn{escape} these characters: precede them with
3004 a backslash @samp{\} character. For example @samp{\\} represents
3005 one backslash: the first @code{\} is an escape which tells
3006 @command{@value{AS}} to interpret the second character literally as a backslash
3007 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
3008 escape character). The complete list of escapes follows.
3010 @cindex escape codes, character
3011 @cindex character escape codes
3012 @c NOTE: Cindex entries must not start with a backlash character.
3013 @c NOTE: This confuses the pdf2texi script when it is creating the
3014 @c NOTE: index based upon the first character and so it generates:
3015 @c NOTE: \initial {\\}
3016 @c NOTE: which then results in the error message:
3017 @c NOTE: Argument of \\ has an extra }.
3018 @c NOTE: So in the index entries below a space character has been
3019 @c NOTE: prepended to avoid this problem.
3022 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
3024 @cindex @code{ \b} (backspace character)
3025 @cindex backspace (@code{\b})
3027 Mnemonic for backspace; for ASCII this is octal code 010.
3030 @c Mnemonic for EOText; for ASCII this is octal code 004.
3032 @cindex @code{ \f} (formfeed character)
3033 @cindex formfeed (@code{\f})
3035 Mnemonic for FormFeed; for ASCII this is octal code 014.
3037 @cindex @code{ \n} (newline character)
3038 @cindex newline (@code{\n})
3040 Mnemonic for newline; for ASCII this is octal code 012.
3043 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
3045 @cindex @code{ \r} (carriage return character)
3046 @cindex carriage return (@code{backslash-r})
3048 Mnemonic for carriage-Return; for ASCII this is octal code 015.
3051 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
3052 @c other assemblers.
3054 @cindex @code{ \t} (tab)
3055 @cindex tab (@code{\t})
3057 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
3060 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
3061 @c @item \x @var{digit} @var{digit} @var{digit}
3062 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
3064 @cindex @code{ \@var{ddd}} (octal character code)
3065 @cindex octal character code (@code{\@var{ddd}})
3066 @item \ @var{digit} @var{digit} @var{digit}
3067 An octal character code. The numeric code is 3 octal digits.
3068 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
3069 for example, @code{\008} has the value 010, and @code{\009} the value 011.
3071 @cindex @code{ \@var{xd...}} (hex character code)
3072 @cindex hex character code (@code{\@var{xd...}})
3073 @item \@code{x} @var{hex-digits...}
3074 A hex character code. All trailing hex digits are combined. Either upper or
3075 lower case @code{x} works.
3077 @cindex @code{ \\} (@samp{\} character)
3078 @cindex backslash (@code{\\})
3080 Represents one @samp{\} character.
3083 @c Represents one @samp{'} (accent acute) character.
3084 @c This is needed in single character literals
3085 @c (@xref{Characters,,Character Constants}.) to represent
3088 @cindex @code{ \"} (doublequote character)
3089 @cindex doublequote (@code{\"})
3091 Represents one @samp{"} character. Needed in strings to represent
3092 this character, because an unescaped @samp{"} would end the string.
3094 @item \ @var{anything-else}
3095 Any other character when escaped by @kbd{\} gives a warning, but
3096 assembles as if the @samp{\} was not present. The idea is that if
3097 you used an escape sequence you clearly didn't want the literal
3098 interpretation of the following character. However @command{@value{AS}} has no
3099 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
3100 code and warns you of the fact.
3103 Which characters are escapable, and what those escapes represent,
3104 varies widely among assemblers. The current set is what we think
3105 the BSD 4.2 assembler recognizes, and is a subset of what most C
3106 compilers recognize. If you are in doubt, do not use an escape
3110 @subsubsection Characters
3112 @cindex single character constant
3113 @cindex character, single
3114 @cindex constant, single character
3115 A single character may be written as a single quote immediately followed by
3116 that character. Some backslash escapes apply to characters, @code{\b},
3117 @code{\f}, @code{\n}, @code{\r}, @code{\t}, and @code{\"} with the same meaning
3118 as for strings, plus @code{\'} for a single quote. So if you want to write the
3119 character backslash, you must write @kbd{'\\} where the first @code{\} escapes
3120 the second @code{\}. As you can see, the quote is an acute accent, not a grave
3123 @ifclear abnormal-separator
3124 (or semicolon @samp{;})
3126 @ifset abnormal-separator
3128 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
3133 immediately following an acute accent is taken as a literal character
3134 and does not count as the end of a statement. The value of a character
3135 constant in a numeric expression is the machine's byte-wide code for
3136 that character. @command{@value{AS}} assumes your character code is ASCII:
3137 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
3140 @subsection Number Constants
3142 @cindex constants, number
3143 @cindex number constants
3144 @command{@value{AS}} distinguishes three kinds of numbers according to how they
3145 are stored in the target machine. @emph{Integers} are numbers that
3146 would fit into an @code{int} in the C language. @emph{Bignums} are
3147 integers, but they are stored in more than 32 bits. @emph{Flonums}
3148 are floating point numbers, described below.
3151 * Integers:: Integers
3156 * Bit Fields:: Bit Fields
3162 @subsubsection Integers
3164 @cindex constants, integer
3166 @cindex binary integers
3167 @cindex integers, binary
3168 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
3169 the binary digits @samp{01}.
3171 @cindex octal integers
3172 @cindex integers, octal
3173 An octal integer is @samp{0} followed by zero or more of the octal
3174 digits (@samp{01234567}).
3176 @cindex decimal integers
3177 @cindex integers, decimal
3178 A decimal integer starts with a non-zero digit followed by zero or
3179 more digits (@samp{0123456789}).
3181 @cindex hexadecimal integers
3182 @cindex integers, hexadecimal
3183 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
3184 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
3186 Integers have the usual values. To denote a negative integer, use
3187 the prefix operator @samp{-} discussed under expressions
3188 (@pxref{Prefix Ops,,Prefix Operators}).
3191 @subsubsection Bignums
3194 @cindex constants, bignum
3195 A @dfn{bignum} has the same syntax and semantics as an integer
3196 except that the number (or its negative) takes more than 32 bits to
3197 represent in binary. The distinction is made because in some places
3198 integers are permitted while bignums are not.
3201 @subsubsection Flonums
3203 @cindex floating point numbers
3204 @cindex constants, floating point
3206 @cindex precision, floating point
3207 A @dfn{flonum} represents a floating point number. The translation is
3208 indirect: a decimal floating point number from the text is converted by
3209 @command{@value{AS}} to a generic binary floating point number of more than
3210 sufficient precision. This generic floating point number is converted
3211 to a particular computer's floating point format (or formats) by a
3212 portion of @command{@value{AS}} specialized to that computer.
3214 A flonum is written by writing (in order)
3219 (@samp{0} is optional on the HPPA.)
3223 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
3225 @kbd{e} is recommended. Case is not important.
3227 @c FIXME: verify if flonum syntax really this vague for most cases
3228 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
3229 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
3232 On the H8/300, Renesas / SuperH SH,
3233 and AMD 29K architectures, the letter must be
3234 one of the letters @samp{DFPRSX} (in upper or lower case).
3236 On the ARC, the letter must be one of the letters @samp{DFRS}
3237 (in upper or lower case).
3239 On the Intel 960 architecture, the letter must be
3240 one of the letters @samp{DFT} (in upper or lower case).
3242 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3246 One of the letters @samp{DFRS} (in upper or lower case).
3249 One of the letters @samp{DFPRSX} (in upper or lower case).
3252 The letter @samp{E} (upper case only).
3255 One of the letters @samp{DFT} (in upper or lower case).
3260 An optional sign: either @samp{+} or @samp{-}.
3263 An optional @dfn{integer part}: zero or more decimal digits.
3266 An optional @dfn{fractional part}: @samp{.} followed by zero
3267 or more decimal digits.
3270 An optional exponent, consisting of:
3274 An @samp{E} or @samp{e}.
3275 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3276 @c principle this can perfectly well be different on different targets.
3278 Optional sign: either @samp{+} or @samp{-}.
3280 One or more decimal digits.
3285 At least one of the integer part or the fractional part must be
3286 present. The floating point number has the usual base-10 value.
3288 @command{@value{AS}} does all processing using integers. Flonums are computed
3289 independently of any floating point hardware in the computer running
3290 @command{@value{AS}}.
3294 @c Bit fields are written as a general facility but are also controlled
3295 @c by a conditional-compilation flag---which is as of now (21mar91)
3296 @c turned on only by the i960 config of GAS.
3298 @subsubsection Bit Fields
3301 @cindex constants, bit field
3302 You can also define numeric constants as @dfn{bit fields}.
3303 Specify two numbers separated by a colon---
3305 @var{mask}:@var{value}
3308 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
3311 The resulting number is then packed
3313 @c this conditional paren in case bit fields turned on elsewhere than 960
3314 (in host-dependent byte order)
3316 into a field whose width depends on which assembler directive has the
3317 bit-field as its argument. Overflow (a result from the bitwise and
3318 requiring more binary digits to represent) is not an error; instead,
3319 more constants are generated, of the specified width, beginning with the
3320 least significant digits.@refill
3322 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
3323 @code{.short}, and @code{.word} accept bit-field arguments.
3328 @chapter Sections and Relocation
3333 * Secs Background:: Background
3334 * Ld Sections:: Linker Sections
3335 * As Sections:: Assembler Internal Sections
3336 * Sub-Sections:: Sub-Sections
3340 @node Secs Background
3343 Roughly, a section is a range of addresses, with no gaps; all data
3344 ``in'' those addresses is treated the same for some particular purpose.
3345 For example there may be a ``read only'' section.
3347 @cindex linker, and assembler
3348 @cindex assembler, and linker
3349 The linker @code{@value{LD}} reads many object files (partial programs) and
3350 combines their contents to form a runnable program. When @command{@value{AS}}
3351 emits an object file, the partial program is assumed to start at address 0.
3352 @code{@value{LD}} assigns the final addresses for the partial program, so that
3353 different partial programs do not overlap. This is actually an
3354 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3357 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3358 addresses. These blocks slide to their run-time addresses as rigid
3359 units; their length does not change and neither does the order of bytes
3360 within them. Such a rigid unit is called a @emph{section}. Assigning
3361 run-time addresses to sections is called @dfn{relocation}. It includes
3362 the task of adjusting mentions of object-file addresses so they refer to
3363 the proper run-time addresses.
3365 For the H8/300, and for the Renesas / SuperH SH,
3366 @command{@value{AS}} pads sections if needed to
3367 ensure they end on a word (sixteen bit) boundary.
3370 @cindex standard assembler sections
3371 An object file written by @command{@value{AS}} has at least three sections, any
3372 of which may be empty. These are named @dfn{text}, @dfn{data} and
3377 When it generates COFF or ELF output,
3379 @command{@value{AS}} can also generate whatever other named sections you specify
3380 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3381 If you do not use any directives that place output in the @samp{.text}
3382 or @samp{.data} sections, these sections still exist, but are empty.
3387 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3389 @command{@value{AS}} can also generate whatever other named sections you
3390 specify using the @samp{.space} and @samp{.subspace} directives. See
3391 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3392 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3393 assembler directives.
3396 Additionally, @command{@value{AS}} uses different names for the standard
3397 text, data, and bss sections when generating SOM output. Program text
3398 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3399 BSS into @samp{$BSS$}.
3403 Within the object file, the text section starts at address @code{0}, the
3404 data section follows, and the bss section follows the data section.
3407 When generating either SOM or ELF output files on the HPPA, the text
3408 section starts at address @code{0}, the data section at address
3409 @code{0x4000000}, and the bss section follows the data section.
3412 To let @code{@value{LD}} know which data changes when the sections are
3413 relocated, and how to change that data, @command{@value{AS}} also writes to the
3414 object file details of the relocation needed. To perform relocation
3415 @code{@value{LD}} must know, each time an address in the object
3419 Where in the object file is the beginning of this reference to
3422 How long (in bytes) is this reference?
3424 Which section does the address refer to? What is the numeric value of
3426 (@var{address}) @minus{} (@var{start-address of section})?
3429 Is the reference to an address ``Program-Counter relative''?
3432 @cindex addresses, format of
3433 @cindex section-relative addressing
3434 In fact, every address @command{@value{AS}} ever uses is expressed as
3436 (@var{section}) + (@var{offset into section})
3439 Further, most expressions @command{@value{AS}} computes have this section-relative
3442 (For some object formats, such as SOM for the HPPA, some expressions are
3443 symbol-relative instead.)
3446 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3447 @var{N} into section @var{secname}.''
3449 Apart from text, data and bss sections you need to know about the
3450 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3451 addresses in the absolute section remain unchanged. For example, address
3452 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3453 @code{@value{LD}}. Although the linker never arranges two partial programs'
3454 data sections with overlapping addresses after linking, @emph{by definition}
3455 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3456 part of a program is always the same address when the program is running as
3457 address @code{@{absolute@ 239@}} in any other part of the program.
3459 The idea of sections is extended to the @dfn{undefined} section. Any
3460 address whose section is unknown at assembly time is by definition
3461 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3462 Since numbers are always defined, the only way to generate an undefined
3463 address is to mention an undefined symbol. A reference to a named
3464 common block would be such a symbol: its value is unknown at assembly
3465 time so it has section @emph{undefined}.
3467 By analogy the word @emph{section} is used to describe groups of sections in
3468 the linked program. @code{@value{LD}} puts all partial programs' text
3469 sections in contiguous addresses in the linked program. It is
3470 customary to refer to the @emph{text section} of a program, meaning all
3471 the addresses of all partial programs' text sections. Likewise for
3472 data and bss sections.
3474 Some sections are manipulated by @code{@value{LD}}; others are invented for
3475 use of @command{@value{AS}} and have no meaning except during assembly.
3478 @section Linker Sections
3479 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3484 @cindex named sections
3485 @cindex sections, named
3486 @item named sections
3489 @cindex text section
3490 @cindex data section
3494 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3495 separate but equal sections. Anything you can say of one section is
3498 When the program is running, however, it is
3499 customary for the text section to be unalterable. The
3500 text section is often shared among processes: it contains
3501 instructions, constants and the like. The data section of a running
3502 program is usually alterable: for example, C variables would be stored
3503 in the data section.
3508 This section contains zeroed bytes when your program begins running. It
3509 is used to hold uninitialized variables or common storage. The length of
3510 each partial program's bss section is important, but because it starts
3511 out containing zeroed bytes there is no need to store explicit zero
3512 bytes in the object file. The bss section was invented to eliminate
3513 those explicit zeros from object files.
3515 @cindex absolute section
3516 @item absolute section
3517 Address 0 of this section is always ``relocated'' to runtime address 0.
3518 This is useful if you want to refer to an address that @code{@value{LD}} must
3519 not change when relocating. In this sense we speak of absolute
3520 addresses being ``unrelocatable'': they do not change during relocation.
3522 @cindex undefined section
3523 @item undefined section
3524 This ``section'' is a catch-all for address references to objects not in
3525 the preceding sections.
3526 @c FIXME: ref to some other doc on obj-file formats could go here.
3529 @cindex relocation example
3530 An idealized example of three relocatable sections follows.
3532 The example uses the traditional section names @samp{.text} and @samp{.data}.
3534 Memory addresses are on the horizontal axis.
3538 @c END TEXI2ROFF-KILL
3541 partial program # 1: |ttttt|dddd|00|
3548 partial program # 2: |TTT|DDD|000|
3551 +--+---+-----+--+----+---+-----+~~
3552 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3553 +--+---+-----+--+----+---+-----+~~
3555 addresses: 0 @dots{}
3562 \line{\it Partial program \#1: \hfil}
3563 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3564 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3566 \line{\it Partial program \#2: \hfil}
3567 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3568 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3570 \line{\it linked program: \hfil}
3571 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3572 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3573 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3574 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3576 \line{\it addresses: \hfil}
3580 @c END TEXI2ROFF-KILL
3583 @section Assembler Internal Sections
3585 @cindex internal assembler sections
3586 @cindex sections in messages, internal
3587 These sections are meant only for the internal use of @command{@value{AS}}. They
3588 have no meaning at run-time. You do not really need to know about these
3589 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3590 warning messages, so it might be helpful to have an idea of their
3591 meanings to @command{@value{AS}}. These sections are used to permit the
3592 value of every expression in your assembly language program to be a
3593 section-relative address.
3596 @cindex assembler internal logic error
3597 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3598 An internal assembler logic error has been found. This means there is a
3599 bug in the assembler.
3601 @cindex expr (internal section)
3603 The assembler stores complex expression internally as combinations of
3604 symbols. When it needs to represent an expression as a symbol, it puts
3605 it in the expr section.
3607 @c FIXME item transfer[t] vector preload
3608 @c FIXME item transfer[t] vector postload
3609 @c FIXME item register
3613 @section Sub-Sections
3615 @cindex numbered subsections
3616 @cindex grouping data
3622 fall into two sections: text and data.
3624 You may have separate groups of
3626 data in named sections
3630 data in named sections
3636 that you want to end up near to each other in the object file, even though they
3637 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3638 use @dfn{subsections} for this purpose. Within each section, there can be
3639 numbered subsections with values from 0 to 8192. Objects assembled into the
3640 same subsection go into the object file together with other objects in the same
3641 subsection. For example, a compiler might want to store constants in the text
3642 section, but might not want to have them interspersed with the program being
3643 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3644 section of code being output, and a @samp{.text 1} before each group of
3645 constants being output.
3647 Subsections are optional. If you do not use subsections, everything
3648 goes in subsection number zero.
3651 Each subsection is zero-padded up to a multiple of four bytes.
3652 (Subsections may be padded a different amount on different flavors
3653 of @command{@value{AS}}.)
3657 On the H8/300 platform, each subsection is zero-padded to a word
3658 boundary (two bytes).
3659 The same is true on the Renesas SH.
3662 @c FIXME section padding (alignment)?
3663 @c Rich Pixley says padding here depends on target obj code format; that
3664 @c doesn't seem particularly useful to say without further elaboration,
3665 @c so for now I say nothing about it. If this is a generic BFD issue,
3666 @c these paragraphs might need to vanish from this manual, and be
3667 @c discussed in BFD chapter of binutils (or some such).
3671 Subsections appear in your object file in numeric order, lowest numbered
3672 to highest. (All this to be compatible with other people's assemblers.)
3673 The object file contains no representation of subsections; @code{@value{LD}} and
3674 other programs that manipulate object files see no trace of them.
3675 They just see all your text subsections as a text section, and all your
3676 data subsections as a data section.
3678 To specify which subsection you want subsequent statements assembled
3679 into, use a numeric argument to specify it, in a @samp{.text
3680 @var{expression}} or a @samp{.data @var{expression}} statement.
3683 When generating COFF output, you
3688 can also use an extra subsection
3689 argument with arbitrary named sections: @samp{.section @var{name},
3694 When generating ELF output, you
3699 can also use the @code{.subsection} directive (@pxref{SubSection})
3700 to specify a subsection: @samp{.subsection @var{expression}}.
3702 @var{Expression} should be an absolute expression
3703 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3704 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3705 begins in @code{text 0}. For instance:
3707 .text 0 # The default subsection is text 0 anyway.
3708 .ascii "This lives in the first text subsection. *"
3710 .ascii "But this lives in the second text subsection."
3712 .ascii "This lives in the data section,"
3713 .ascii "in the first data subsection."
3715 .ascii "This lives in the first text section,"
3716 .ascii "immediately following the asterisk (*)."
3719 Each section has a @dfn{location counter} incremented by one for every byte
3720 assembled into that section. Because subsections are merely a convenience
3721 restricted to @command{@value{AS}} there is no concept of a subsection location
3722 counter. There is no way to directly manipulate a location counter---but the
3723 @code{.align} directive changes it, and any label definition captures its
3724 current value. The location counter of the section where statements are being
3725 assembled is said to be the @dfn{active} location counter.
3728 @section bss Section
3731 @cindex common variable storage
3732 The bss section is used for local common variable storage.
3733 You may allocate address space in the bss section, but you may
3734 not dictate data to load into it before your program executes. When
3735 your program starts running, all the contents of the bss
3736 section are zeroed bytes.
3738 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3739 @ref{Lcomm,,@code{.lcomm}}.
3741 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3742 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3745 When assembling for a target which supports multiple sections, such as ELF or
3746 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3747 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3748 section. Typically the section will only contain symbol definitions and
3749 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3756 Symbols are a central concept: the programmer uses symbols to name
3757 things, the linker uses symbols to link, and the debugger uses symbols
3761 @cindex debuggers, and symbol order
3762 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3763 the same order they were declared. This may break some debuggers.
3768 * Setting Symbols:: Giving Symbols Other Values
3769 * Symbol Names:: Symbol Names
3770 * Dot:: The Special Dot Symbol
3771 * Symbol Attributes:: Symbol Attributes
3778 A @dfn{label} is written as a symbol immediately followed by a colon
3779 @samp{:}. The symbol then represents the current value of the
3780 active location counter, and is, for example, a suitable instruction
3781 operand. You are warned if you use the same symbol to represent two
3782 different locations: the first definition overrides any other
3786 On the HPPA, the usual form for a label need not be immediately followed by a
3787 colon, but instead must start in column zero. Only one label may be defined on
3788 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3789 provides a special directive @code{.label} for defining labels more flexibly.
3792 @node Setting Symbols
3793 @section Giving Symbols Other Values
3795 @cindex assigning values to symbols
3796 @cindex symbol values, assigning
3797 A symbol can be given an arbitrary value by writing a symbol, followed
3798 by an equals sign @samp{=}, followed by an expression
3799 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3800 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3801 equals sign @samp{=}@samp{=} here represents an equivalent of the
3802 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3805 Blackfin does not support symbol assignment with @samp{=}.
3809 @section Symbol Names
3811 @cindex symbol names
3812 @cindex names, symbol
3813 @ifclear SPECIAL-SYMS
3814 Symbol names begin with a letter or with one of @samp{._}. On most
3815 machines, you can also use @code{$} in symbol names; exceptions are
3816 noted in @ref{Machine Dependencies}. That character may be followed by any
3817 string of digits, letters, dollar signs (unless otherwise noted for a
3818 particular target machine), and underscores.
3822 Symbol names begin with a letter or with one of @samp{._}. On the
3823 Renesas SH you can also use @code{$} in symbol names. That
3824 character may be followed by any string of digits, letters, dollar signs (save
3825 on the H8/300), and underscores.
3829 Case of letters is significant: @code{foo} is a different symbol name
3832 Symbol names do not start with a digit. An exception to this rule is made for
3833 Local Labels. See below.
3835 Multibyte characters are supported. To generate a symbol name containing
3836 multibyte characters enclose it within double quotes and use escape codes. cf
3837 @xref{Strings}. Generating a multibyte symbol name from a label is not
3838 currently supported.
3840 Each symbol has exactly one name. Each name in an assembly language program
3841 refers to exactly one symbol. You may use that symbol name any number of times
3844 @subheading Local Symbol Names
3846 @cindex local symbol names
3847 @cindex symbol names, local
3848 A local symbol is any symbol beginning with certain local label prefixes.
3849 By default, the local label prefix is @samp{.L} for ELF systems or
3850 @samp{L} for traditional a.out systems, but each target may have its own
3851 set of local label prefixes.
3853 On the HPPA local symbols begin with @samp{L$}.
3856 Local symbols are defined and used within the assembler, but they are
3857 normally not saved in object files. Thus, they are not visible when debugging.
3858 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols})
3859 to retain the local symbols in the object files.
3861 @subheading Local Labels
3863 @cindex local labels
3864 @cindex temporary symbol names
3865 @cindex symbol names, temporary
3866 Local labels are different from local symbols. Local labels help compilers and
3867 programmers use names temporarily. They create symbols which are guaranteed to
3868 be unique over the entire scope of the input source code and which can be
3869 referred to by a simple notation. To define a local label, write a label of
3870 the form @samp{@b{N}:} (where @b{N} represents any non-negative integer).
3871 To refer to the most recent previous definition of that label write
3872 @samp{@b{N}b}, using the same number as when you defined the label. To refer
3873 to the next definition of a local label, write @samp{@b{N}f}. The @samp{b}
3874 stands for ``backwards'' and the @samp{f} stands for ``forwards''.
3876 There is no restriction on how you can use these labels, and you can reuse them
3877 too. So that it is possible to repeatedly define the same local label (using
3878 the same number @samp{@b{N}}), although you can only refer to the most recently
3879 defined local label of that number (for a backwards reference) or the next
3880 definition of a specific local label for a forward reference. It is also worth
3881 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3882 implemented in a slightly more efficient manner than the others.
3893 Which is the equivalent of:
3896 label_1: branch label_3
3897 label_2: branch label_1
3898 label_3: branch label_4
3899 label_4: branch label_3
3902 Local label names are only a notational device. They are immediately
3903 transformed into more conventional symbol names before the assembler uses them.
3904 The symbol names are stored in the symbol table, appear in error messages, and
3905 are optionally emitted to the object file. The names are constructed using
3909 @item @emph{local label prefix}
3910 All local symbols begin with the system-specific local label prefix.
3911 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3912 that start with the local label prefix. These labels are
3913 used for symbols you are never intended to see. If you use the
3914 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3915 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3916 you may use them in debugging.
3919 This is the number that was used in the local label definition. So if the
3920 label is written @samp{55:} then the number is @samp{55}.
3923 This unusual character is included so you do not accidentally invent a symbol
3924 of the same name. The character has ASCII value of @samp{\002} (control-B).
3926 @item @emph{ordinal number}
3927 This is a serial number to keep the labels distinct. The first definition of
3928 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3929 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3930 the number @samp{1} and its 15th definition gets @samp{15} as well.
3933 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3934 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3936 @subheading Dollar Local Labels
3937 @cindex dollar local symbols
3939 On some targets @code{@value{AS}} also supports an even more local form of
3940 local labels called dollar labels. These labels go out of scope (i.e., they
3941 become undefined) as soon as a non-local label is defined. Thus they remain
3942 valid for only a small region of the input source code. Normal local labels,
3943 by contrast, remain in scope for the entire file, or until they are redefined
3944 by another occurrence of the same local label.
3946 Dollar labels are defined in exactly the same way as ordinary local labels,
3947 except that they have a dollar sign suffix to their numeric value, e.g.,
3950 They can also be distinguished from ordinary local labels by their transformed
3951 names which use ASCII character @samp{\001} (control-A) as the magic character
3952 to distinguish them from ordinary labels. For example, the fifth definition of
3953 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3956 @section The Special Dot Symbol
3958 @cindex dot (symbol)
3959 @cindex @code{.} (symbol)
3960 @cindex current address
3961 @cindex location counter
3962 The special symbol @samp{.} refers to the current address that
3963 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3964 .long .} defines @code{melvin} to contain its own address.
3965 Assigning a value to @code{.} is treated the same as a @code{.org}
3967 @ifclear no-space-dir
3968 Thus, the expression @samp{.=.+4} is the same as saying
3972 @node Symbol Attributes
3973 @section Symbol Attributes
3975 @cindex symbol attributes
3976 @cindex attributes, symbol
3977 Every symbol has, as well as its name, the attributes ``Value'' and
3978 ``Type''. Depending on output format, symbols can also have auxiliary
3981 The detailed definitions are in @file{a.out.h}.
3984 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3985 all these attributes, and probably won't warn you. This makes the
3986 symbol an externally defined symbol, which is generally what you
3990 * Symbol Value:: Value
3991 * Symbol Type:: Type
3994 * a.out Symbols:: Symbol Attributes: @code{a.out}
3998 * a.out Symbols:: Symbol Attributes: @code{a.out}
4001 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
4006 * COFF Symbols:: Symbol Attributes for COFF
4009 * SOM Symbols:: Symbol Attributes for SOM
4016 @cindex value of a symbol
4017 @cindex symbol value
4018 The value of a symbol is (usually) 32 bits. For a symbol which labels a
4019 location in the text, data, bss or absolute sections the value is the
4020 number of addresses from the start of that section to the label.
4021 Naturally for text, data and bss sections the value of a symbol changes
4022 as @code{@value{LD}} changes section base addresses during linking. Absolute
4023 symbols' values do not change during linking: that is why they are
4026 The value of an undefined symbol is treated in a special way. If it is
4027 0 then the symbol is not defined in this assembler source file, and
4028 @code{@value{LD}} tries to determine its value from other files linked into the
4029 same program. You make this kind of symbol simply by mentioning a symbol
4030 name without defining it. A non-zero value represents a @code{.comm}
4031 common declaration. The value is how much common storage to reserve, in
4032 bytes (addresses). The symbol refers to the first address of the
4038 @cindex type of a symbol
4040 The type attribute of a symbol contains relocation (section)
4041 information, any flag settings indicating that a symbol is external, and
4042 (optionally), other information for linkers and debuggers. The exact
4043 format depends on the object-code output format in use.
4048 @c The following avoids a "widow" subsection title. @group would be
4049 @c better if it were available outside examples.
4052 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
4054 @cindex @code{b.out} symbol attributes
4055 @cindex symbol attributes, @code{b.out}
4056 These symbol attributes appear only when @command{@value{AS}} is configured for
4057 one of the Berkeley-descended object output formats---@code{a.out} or
4063 @subsection Symbol Attributes: @code{a.out}
4065 @cindex @code{a.out} symbol attributes
4066 @cindex symbol attributes, @code{a.out}
4072 @subsection Symbol Attributes: @code{a.out}
4074 @cindex @code{a.out} symbol attributes
4075 @cindex symbol attributes, @code{a.out}
4079 * Symbol Desc:: Descriptor
4080 * Symbol Other:: Other
4084 @subsubsection Descriptor
4086 @cindex descriptor, of @code{a.out} symbol
4087 This is an arbitrary 16-bit value. You may establish a symbol's
4088 descriptor value by using a @code{.desc} statement
4089 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
4090 @command{@value{AS}}.
4093 @subsubsection Other
4095 @cindex other attribute, of @code{a.out} symbol
4096 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
4101 @subsection Symbol Attributes for COFF
4103 @cindex COFF symbol attributes
4104 @cindex symbol attributes, COFF
4106 The COFF format supports a multitude of auxiliary symbol attributes;
4107 like the primary symbol attributes, they are set between @code{.def} and
4108 @code{.endef} directives.
4110 @subsubsection Primary Attributes
4112 @cindex primary attributes, COFF symbols
4113 The symbol name is set with @code{.def}; the value and type,
4114 respectively, with @code{.val} and @code{.type}.
4116 @subsubsection Auxiliary Attributes
4118 @cindex auxiliary attributes, COFF symbols
4119 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
4120 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
4121 table information for COFF.
4126 @subsection Symbol Attributes for SOM
4128 @cindex SOM symbol attributes
4129 @cindex symbol attributes, SOM
4131 The SOM format for the HPPA supports a multitude of symbol attributes set with
4132 the @code{.EXPORT} and @code{.IMPORT} directives.
4134 The attributes are described in @cite{HP9000 Series 800 Assembly
4135 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
4136 @code{EXPORT} assembler directive documentation.
4140 @chapter Expressions
4144 @cindex numeric values
4145 An @dfn{expression} specifies an address or numeric value.
4146 Whitespace may precede and/or follow an expression.
4148 The result of an expression must be an absolute number, or else an offset into
4149 a particular section. If an expression is not absolute, and there is not
4150 enough information when @command{@value{AS}} sees the expression to know its
4151 section, a second pass over the source program might be necessary to interpret
4152 the expression---but the second pass is currently not implemented.
4153 @command{@value{AS}} aborts with an error message in this situation.
4156 * Empty Exprs:: Empty Expressions
4157 * Integer Exprs:: Integer Expressions
4161 @section Empty Expressions
4163 @cindex empty expressions
4164 @cindex expressions, empty
4165 An empty expression has no value: it is just whitespace or null.
4166 Wherever an absolute expression is required, you may omit the
4167 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
4168 is compatible with other assemblers.
4171 @section Integer Expressions
4173 @cindex integer expressions
4174 @cindex expressions, integer
4175 An @dfn{integer expression} is one or more @emph{arguments} delimited
4176 by @emph{operators}.
4179 * Arguments:: Arguments
4180 * Operators:: Operators
4181 * Prefix Ops:: Prefix Operators
4182 * Infix Ops:: Infix Operators
4186 @subsection Arguments
4188 @cindex expression arguments
4189 @cindex arguments in expressions
4190 @cindex operands in expressions
4191 @cindex arithmetic operands
4192 @dfn{Arguments} are symbols, numbers or subexpressions. In other
4193 contexts arguments are sometimes called ``arithmetic operands''. In
4194 this manual, to avoid confusing them with the ``instruction operands'' of
4195 the machine language, we use the term ``argument'' to refer to parts of
4196 expressions only, reserving the word ``operand'' to refer only to machine
4197 instruction operands.
4199 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
4200 @var{section} is one of text, data, bss, absolute,
4201 or undefined. @var{NNN} is a signed, 2's complement 32 bit
4204 Numbers are usually integers.
4206 A number can be a flonum or bignum. In this case, you are warned
4207 that only the low order 32 bits are used, and @command{@value{AS}} pretends
4208 these 32 bits are an integer. You may write integer-manipulating
4209 instructions that act on exotic constants, compatible with other
4212 @cindex subexpressions
4213 Subexpressions are a left parenthesis @samp{(} followed by an integer
4214 expression, followed by a right parenthesis @samp{)}; or a prefix
4215 operator followed by an argument.
4218 @subsection Operators
4220 @cindex operators, in expressions
4221 @cindex arithmetic functions
4222 @cindex functions, in expressions
4223 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
4224 operators are followed by an argument. Infix operators appear
4225 between their arguments. Operators may be preceded and/or followed by
4229 @subsection Prefix Operator
4231 @cindex prefix operators
4232 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
4233 one argument, which must be absolute.
4235 @c the tex/end tex stuff surrounding this small table is meant to make
4236 @c it align, on the printed page, with the similar table in the next
4237 @c section (which is inside an enumerate).
4239 \global\advance\leftskip by \itemindent
4244 @dfn{Negation}. Two's complement negation.
4246 @dfn{Complementation}. Bitwise not.
4250 \global\advance\leftskip by -\itemindent
4254 @subsection Infix Operators
4256 @cindex infix operators
4257 @cindex operators, permitted arguments
4258 @dfn{Infix operators} take two arguments, one on either side. Operators
4259 have precedence, but operations with equal precedence are performed left
4260 to right. Apart from @code{+} or @option{-}, both arguments must be
4261 absolute, and the result is absolute.
4264 @cindex operator precedence
4265 @cindex precedence of operators
4272 @dfn{Multiplication}.
4275 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4281 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4284 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4288 Intermediate precedence
4293 @dfn{Bitwise Inclusive Or}.
4299 @dfn{Bitwise Exclusive Or}.
4302 @dfn{Bitwise Or Not}.
4309 @cindex addition, permitted arguments
4310 @cindex plus, permitted arguments
4311 @cindex arguments for addition
4313 @dfn{Addition}. If either argument is absolute, the result has the section of
4314 the other argument. You may not add together arguments from different
4317 @cindex subtraction, permitted arguments
4318 @cindex minus, permitted arguments
4319 @cindex arguments for subtraction
4321 @dfn{Subtraction}. If the right argument is absolute, the
4322 result has the section of the left argument.
4323 If both arguments are in the same section, the result is absolute.
4324 You may not subtract arguments from different sections.
4325 @c FIXME is there still something useful to say about undefined - undefined ?
4327 @cindex comparison expressions
4328 @cindex expressions, comparison
4333 @dfn{Is Not Equal To}
4337 @dfn{Is Greater Than}
4339 @dfn{Is Greater Than Or Equal To}
4341 @dfn{Is Less Than Or Equal To}
4343 The comparison operators can be used as infix operators. A true results has a
4344 value of -1 whereas a false result has a value of 0. Note, these operators
4345 perform signed comparisons.
4348 @item Lowest Precedence
4357 These two logical operations can be used to combine the results of sub
4358 expressions. Note, unlike the comparison operators a true result returns a
4359 value of 1 but a false results does still return 0. Also note that the logical
4360 or operator has a slightly lower precedence than logical and.
4365 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4366 address; you can only have a defined section in one of the two arguments.
4369 @chapter Assembler Directives
4371 @cindex directives, machine independent
4372 @cindex pseudo-ops, machine independent
4373 @cindex machine independent directives
4374 All assembler directives have names that begin with a period (@samp{.}).
4375 The names are case insensitive for most targets, and usually written
4378 This chapter discusses directives that are available regardless of the
4379 target machine configuration for the @sc{gnu} assembler.
4381 Some machine configurations provide additional directives.
4382 @xref{Machine Dependencies}.
4385 @ifset machine-directives
4386 @xref{Machine Dependencies}, for additional directives.
4391 * Abort:: @code{.abort}
4393 * ABORT (COFF):: @code{.ABORT}
4396 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4397 * Altmacro:: @code{.altmacro}
4398 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4399 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4400 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4401 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, etc
4402 * Byte:: @code{.byte @var{expressions}}
4403 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4404 * Comm:: @code{.comm @var{symbol} , @var{length} }
4405 * Data:: @code{.data @var{subsection}}
4407 * Def:: @code{.def @var{name}}
4410 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4416 * Double:: @code{.double @var{flonums}}
4417 * Eject:: @code{.eject}
4418 * Else:: @code{.else}
4419 * Elseif:: @code{.elseif}
4422 * Endef:: @code{.endef}
4425 * Endfunc:: @code{.endfunc}
4426 * Endif:: @code{.endif}
4427 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4428 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4429 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4431 * Error:: @code{.error @var{string}}
4432 * Exitm:: @code{.exitm}
4433 * Extern:: @code{.extern}
4434 * Fail:: @code{.fail}
4435 * File:: @code{.file}
4436 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4437 * Float:: @code{.float @var{flonums}}
4438 * Func:: @code{.func}
4439 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4441 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4442 * Hidden:: @code{.hidden @var{names}}
4445 * hword:: @code{.hword @var{expressions}}
4446 * Ident:: @code{.ident}
4447 * If:: @code{.if @var{absolute expression}}
4448 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4449 * Include:: @code{.include "@var{file}"}
4450 * Int:: @code{.int @var{expressions}}
4452 * Internal:: @code{.internal @var{names}}
4455 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4456 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4457 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4458 * Lflags:: @code{.lflags}
4459 @ifclear no-line-dir
4460 * Line:: @code{.line @var{line-number}}
4463 * Linkonce:: @code{.linkonce [@var{type}]}
4464 * List:: @code{.list}
4465 * Ln:: @code{.ln @var{line-number}}
4466 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4467 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4469 * Local:: @code{.local @var{names}}
4472 * Long:: @code{.long @var{expressions}}
4474 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4477 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4478 * MRI:: @code{.mri @var{val}}
4479 * Noaltmacro:: @code{.noaltmacro}
4480 * Nolist:: @code{.nolist}
4481 * Octa:: @code{.octa @var{bignums}}
4482 * Offset:: @code{.offset @var{loc}}
4483 * Org:: @code{.org @var{new-lc}, @var{fill}}
4484 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4486 * PopSection:: @code{.popsection}
4487 * Previous:: @code{.previous}
4490 * Print:: @code{.print @var{string}}
4492 * Protected:: @code{.protected @var{names}}
4495 * Psize:: @code{.psize @var{lines}, @var{columns}}
4496 * Purgem:: @code{.purgem @var{name}}
4498 * PushSection:: @code{.pushsection @var{name}}
4501 * Quad:: @code{.quad @var{bignums}}
4502 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4503 * Rept:: @code{.rept @var{count}}
4504 * Sbttl:: @code{.sbttl "@var{subheading}"}
4506 * Scl:: @code{.scl @var{class}}
4509 * Section:: @code{.section @var{name}[, @var{flags}]}
4512 * Set:: @code{.set @var{symbol}, @var{expression}}
4513 * Short:: @code{.short @var{expressions}}
4514 * Single:: @code{.single @var{flonums}}
4516 * Size:: @code{.size [@var{name} , @var{expression}]}
4518 @ifclear no-space-dir
4519 * Skip:: @code{.skip @var{size} , @var{fill}}
4522 * Sleb128:: @code{.sleb128 @var{expressions}}
4523 @ifclear no-space-dir
4524 * Space:: @code{.space @var{size} , @var{fill}}
4527 * Stab:: @code{.stabd, .stabn, .stabs}
4530 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4531 * Struct:: @code{.struct @var{expression}}
4533 * SubSection:: @code{.subsection}
4534 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4538 * Tag:: @code{.tag @var{structname}}
4541 * Text:: @code{.text @var{subsection}}
4542 * Title:: @code{.title "@var{heading}"}
4544 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4547 * Uleb128:: @code{.uleb128 @var{expressions}}
4549 * Val:: @code{.val @var{addr}}
4553 * Version:: @code{.version "@var{string}"}
4554 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4555 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4558 * Warning:: @code{.warning @var{string}}
4559 * Weak:: @code{.weak @var{names}}
4560 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4561 * Word:: @code{.word @var{expressions}}
4562 @ifclear no-space-dir
4563 * Zero:: @code{.zero @var{size}}
4566 * 2byte:: @code{.2byte @var{expressions}}
4567 * 4byte:: @code{.4byte @var{expressions}}
4568 * 8byte:: @code{.8byte @var{bignums}}
4570 * Deprecated:: Deprecated Directives
4574 @section @code{.abort}
4576 @cindex @code{abort} directive
4577 @cindex stopping the assembly
4578 This directive stops the assembly immediately. It is for
4579 compatibility with other assemblers. The original idea was that the
4580 assembly language source would be piped into the assembler. If the sender
4581 of the source quit, it could use this directive tells @command{@value{AS}} to
4582 quit also. One day @code{.abort} will not be supported.
4586 @section @code{.ABORT} (COFF)
4588 @cindex @code{ABORT} directive
4589 When producing COFF output, @command{@value{AS}} accepts this directive as a
4590 synonym for @samp{.abort}.
4593 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4599 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4601 @cindex padding the location counter
4602 @cindex @code{align} directive
4603 Pad the location counter (in the current subsection) to a particular storage
4604 boundary. The first expression (which must be absolute) is the alignment
4605 required, as described below.
4607 The second expression (also absolute) gives the fill value to be stored in the
4608 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4609 padding bytes are normally zero. However, on some systems, if the section is
4610 marked as containing code and the fill value is omitted, the space is filled
4611 with no-op instructions.
4613 The third expression is also absolute, and is also optional. If it is present,
4614 it is the maximum number of bytes that should be skipped by this alignment
4615 directive. If doing the alignment would require skipping more bytes than the
4616 specified maximum, then the alignment is not done at all. You can omit the
4617 fill value (the second argument) entirely by simply using two commas after the
4618 required alignment; this can be useful if you want the alignment to be filled
4619 with no-op instructions when appropriate.
4621 The way the required alignment is specified varies from system to system.
4622 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or1k,
4623 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4624 alignment request in bytes. For example @samp{.align 8} advances
4625 the location counter until it is a multiple of 8. If the location counter
4626 is already a multiple of 8, no change is needed. For the tic54x, the
4627 first expression is the alignment request in words.
4629 For other systems, including ppc, i386 using a.out format, arm and
4630 strongarm, it is the
4631 number of low-order zero bits the location counter must have after
4632 advancement. For example @samp{.align 3} advances the location
4633 counter until it a multiple of 8. If the location counter is already a
4634 multiple of 8, no change is needed.
4636 This inconsistency is due to the different behaviors of the various
4637 native assemblers for these systems which GAS must emulate.
4638 GAS also provides @code{.balign} and @code{.p2align} directives,
4639 described later, which have a consistent behavior across all
4640 architectures (but are specific to GAS).
4643 @section @code{.altmacro}
4644 Enable alternate macro mode, enabling:
4647 @item LOCAL @var{name} [ , @dots{} ]
4648 One additional directive, @code{LOCAL}, is available. It is used to
4649 generate a string replacement for each of the @var{name} arguments, and
4650 replace any instances of @var{name} in each macro expansion. The
4651 replacement string is unique in the assembly, and different for each
4652 separate macro expansion. @code{LOCAL} allows you to write macros that
4653 define symbols, without fear of conflict between separate macro expansions.
4655 @item String delimiters
4656 You can write strings delimited in these other ways besides
4657 @code{"@var{string}"}:
4660 @item '@var{string}'
4661 You can delimit strings with single-quote characters.
4663 @item <@var{string}>
4664 You can delimit strings with matching angle brackets.
4667 @item single-character string escape
4668 To include any single character literally in a string (even if the
4669 character would otherwise have some special meaning), you can prefix the
4670 character with @samp{!} (an exclamation mark). For example, you can
4671 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4673 @item Expression results as strings
4674 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4675 and use the result as a string.
4679 @section @code{.ascii "@var{string}"}@dots{}
4681 @cindex @code{ascii} directive
4682 @cindex string literals
4683 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4684 separated by commas. It assembles each string (with no automatic
4685 trailing zero byte) into consecutive addresses.
4688 @section @code{.asciz "@var{string}"}@dots{}
4690 @cindex @code{asciz} directive
4691 @cindex zero-terminated strings
4692 @cindex null-terminated strings
4693 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4694 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4697 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4699 @cindex padding the location counter given number of bytes
4700 @cindex @code{balign} directive
4701 Pad the location counter (in the current subsection) to a particular
4702 storage boundary. The first expression (which must be absolute) is the
4703 alignment request in bytes. For example @samp{.balign 8} advances
4704 the location counter until it is a multiple of 8. If the location counter
4705 is already a multiple of 8, no change is needed.
4707 The second expression (also absolute) gives the fill value to be stored in the
4708 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4709 padding bytes are normally zero. However, on some systems, if the section is
4710 marked as containing code and the fill value is omitted, the space is filled
4711 with no-op instructions.
4713 The third expression is also absolute, and is also optional. If it is present,
4714 it is the maximum number of bytes that should be skipped by this alignment
4715 directive. If doing the alignment would require skipping more bytes than the
4716 specified maximum, then the alignment is not done at all. You can omit the
4717 fill value (the second argument) entirely by simply using two commas after the
4718 required alignment; this can be useful if you want the alignment to be filled
4719 with no-op instructions when appropriate.
4721 @cindex @code{balignw} directive
4722 @cindex @code{balignl} directive
4723 The @code{.balignw} and @code{.balignl} directives are variants of the
4724 @code{.balign} directive. The @code{.balignw} directive treats the fill
4725 pattern as a two byte word value. The @code{.balignl} directives treats the
4726 fill pattern as a four byte longword value. For example, @code{.balignw
4727 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4728 filled in with the value 0x368d (the exact placement of the bytes depends upon
4729 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4732 @node Bundle directives
4733 @section Bundle directives
4734 @subsection @code{.bundle_align_mode @var{abs-expr}}
4735 @cindex @code{bundle_align_mode} directive
4737 @cindex instruction bundle
4738 @cindex aligned instruction bundle
4739 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4740 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4741 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4742 disabled (which is the default state). If the argument it not zero, it
4743 gives the size of an instruction bundle as a power of two (as for the
4744 @code{.p2align} directive, @pxref{P2align}).
4746 For some targets, it's an ABI requirement that no instruction may span a
4747 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4748 instructions that starts on an aligned boundary. For example, if
4749 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4750 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4751 effect, no single instruction may span a boundary between bundles. If an
4752 instruction would start too close to the end of a bundle for the length of
4753 that particular instruction to fit within the bundle, then the space at the
4754 end of that bundle is filled with no-op instructions so the instruction
4755 starts in the next bundle. As a corollary, it's an error if any single
4756 instruction's encoding is longer than the bundle size.
4758 @subsection @code{.bundle_lock} and @code{.bundle_unlock}
4759 @cindex @code{bundle_lock} directive
4760 @cindex @code{bundle_unlock} directive
4761 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4762 allow explicit control over instruction bundle padding. These directives
4763 are only valid when @code{.bundle_align_mode} has been used to enable
4764 aligned instruction bundle mode. It's an error if they appear when
4765 @code{.bundle_align_mode} has not been used at all, or when the last
4766 directive was @w{@code{.bundle_align_mode 0}}.
4768 @cindex bundle-locked
4769 For some targets, it's an ABI requirement that certain instructions may
4770 appear only as part of specified permissible sequences of multiple
4771 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4772 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4773 instruction sequence. For purposes of aligned instruction bundle mode, a
4774 sequence starting with @code{.bundle_lock} and ending with
4775 @code{.bundle_unlock} is treated as a single instruction. That is, the
4776 entire sequence must fit into a single bundle and may not span a bundle
4777 boundary. If necessary, no-op instructions will be inserted before the
4778 first instruction of the sequence so that the whole sequence starts on an
4779 aligned bundle boundary. It's an error if the sequence is longer than the
4782 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4783 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4784 nested. That is, a second @code{.bundle_lock} directive before the next
4785 @code{.bundle_unlock} directive has no effect except that it must be
4786 matched by another closing @code{.bundle_unlock} so that there is the
4787 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4790 @section @code{.byte @var{expressions}}
4792 @cindex @code{byte} directive
4793 @cindex integers, one byte
4794 @code{.byte} expects zero or more expressions, separated by commas.
4795 Each expression is assembled into the next byte.
4797 @node CFI directives
4798 @section CFI directives
4799 @subsection @code{.cfi_sections @var{section_list}}
4800 @cindex @code{cfi_sections} directive
4801 @code{.cfi_sections} may be used to specify whether CFI directives
4802 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4803 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4804 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4805 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4806 directive is not used is @code{.cfi_sections .eh_frame}.
4808 On targets that support compact unwinding tables these can be generated
4809 by specifying @code{.eh_frame_entry} instead of @code{.eh_frame}.
4811 Some targets may support an additional name, such as @code{.c6xabi.exidx}
4812 which is used by the @value{TIC6X} target.
4814 The @code{.cfi_sections} directive can be repeated, with the same or different
4815 arguments, provided that CFI generation has not yet started. Once CFI
4816 generation has started however the section list is fixed and any attempts to
4817 redefine it will result in an error.
4819 @subsection @code{.cfi_startproc [simple]}
4820 @cindex @code{cfi_startproc} directive
4821 @code{.cfi_startproc} is used at the beginning of each function that
4822 should have an entry in @code{.eh_frame}. It initializes some internal
4823 data structures. Don't forget to close the function by
4824 @code{.cfi_endproc}.
4826 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4827 it also emits some architecture dependent initial CFI instructions.
4829 @subsection @code{.cfi_endproc}
4830 @cindex @code{cfi_endproc} directive
4831 @code{.cfi_endproc} is used at the end of a function where it closes its
4832 unwind entry previously opened by
4833 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4835 @subsection @code{.cfi_personality @var{encoding} [, @var{exp}]}
4836 @cindex @code{cfi_personality} directive
4837 @code{.cfi_personality} defines personality routine and its encoding.
4838 @var{encoding} must be a constant determining how the personality
4839 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4840 argument is not present, otherwise second argument should be
4841 a constant or a symbol name. When using indirect encodings,
4842 the symbol provided should be the location where personality
4843 can be loaded from, not the personality routine itself.
4844 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4845 no personality routine.
4847 @subsection @code{.cfi_personality_id @var{id}}
4848 @cindex @code{cfi_personality_id} directive
4849 @code{cfi_personality_id} defines a personality routine by its index as
4850 defined in a compact unwinding format.
4851 Only valid when generating compact EH frames (i.e.
4852 with @code{.cfi_sections eh_frame_entry}.
4854 @subsection @code{.cfi_fde_data [@var{opcode1} [, @dots{}]]}
4855 @cindex @code{cfi_fde_data} directive
4856 @code{cfi_fde_data} is used to describe the compact unwind opcodes to be
4857 used for the current function. These are emitted inline in the
4858 @code{.eh_frame_entry} section if small enough and there is no LSDA, or
4859 in the @code{.gnu.extab} section otherwise.
4860 Only valid when generating compact EH frames (i.e.
4861 with @code{.cfi_sections eh_frame_entry}.
4863 @subsection @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4864 @code{.cfi_lsda} defines LSDA and its encoding.
4865 @var{encoding} must be a constant determining how the LSDA
4866 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), the second
4867 argument is not present, otherwise the second argument should be a constant
4868 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4869 meaning that no LSDA is present.
4871 @subsection @code{.cfi_inline_lsda} [@var{align}]
4872 @code{.cfi_inline_lsda} marks the start of a LSDA data section and
4873 switches to the corresponding @code{.gnu.extab} section.
4874 Must be preceded by a CFI block containing a @code{.cfi_lsda} directive.
4875 Only valid when generating compact EH frames (i.e.
4876 with @code{.cfi_sections eh_frame_entry}.
4878 The table header and unwinding opcodes will be generated at this point,
4879 so that they are immediately followed by the LSDA data. The symbol
4880 referenced by the @code{.cfi_lsda} directive should still be defined
4881 in case a fallback FDE based encoding is used. The LSDA data is terminated
4882 by a section directive.
4884 The optional @var{align} argument specifies the alignment required.
4885 The alignment is specified as a power of two, as with the
4886 @code{.p2align} directive.
4888 @subsection @code{.cfi_def_cfa @var{register}, @var{offset}}
4889 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4890 address from @var{register} and add @var{offset} to it}.
4892 @subsection @code{.cfi_def_cfa_register @var{register}}
4893 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4894 now on @var{register} will be used instead of the old one. Offset
4897 @subsection @code{.cfi_def_cfa_offset @var{offset}}
4898 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4899 remains the same, but @var{offset} is new. Note that it is the
4900 absolute offset that will be added to a defined register to compute
4903 @subsection @code{.cfi_adjust_cfa_offset @var{offset}}
4904 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4905 value that is added/subtracted from the previous offset.
4907 @subsection @code{.cfi_offset @var{register}, @var{offset}}
4908 Previous value of @var{register} is saved at offset @var{offset} from
4911 @subsection @code{.cfi_val_offset @var{register}, @var{offset}}
4912 Previous value of @var{register} is CFA + @var{offset}.
4914 @subsection @code{.cfi_rel_offset @var{register}, @var{offset}}
4915 Previous value of @var{register} is saved at offset @var{offset} from
4916 the current CFA register. This is transformed to @code{.cfi_offset}
4917 using the known displacement of the CFA register from the CFA.
4918 This is often easier to use, because the number will match the
4919 code it's annotating.
4921 @subsection @code{.cfi_register @var{register1}, @var{register2}}
4922 Previous value of @var{register1} is saved in register @var{register2}.
4924 @subsection @code{.cfi_restore @var{register}}
4925 @code{.cfi_restore} says that the rule for @var{register} is now the
4926 same as it was at the beginning of the function, after all initial
4927 instruction added by @code{.cfi_startproc} were executed.
4929 @subsection @code{.cfi_undefined @var{register}}
4930 From now on the previous value of @var{register} can't be restored anymore.
4932 @subsection @code{.cfi_same_value @var{register}}
4933 Current value of @var{register} is the same like in the previous frame,
4934 i.e. no restoration needed.
4936 @subsection @code{.cfi_remember_state} and @code{.cfi_restore_state}
4937 @code{.cfi_remember_state} pushes the set of rules for every register onto an
4938 implicit stack, while @code{.cfi_restore_state} pops them off the stack and
4939 places them in the current row. This is useful for situations where you have
4940 multiple @code{.cfi_*} directives that need to be undone due to the control
4941 flow of the program. For example, we could have something like this (assuming
4942 the CFA is the value of @code{rbp}):
4952 .cfi_def_cfa %rsp, 8
4955 /* Do something else */
4958 Here, we want the @code{.cfi} directives to affect only the rows corresponding
4959 to the instructions before @code{label}. This means we'd have to add multiple
4960 @code{.cfi} directives after @code{label} to recreate the original save
4961 locations of the registers, as well as setting the CFA back to the value of
4962 @code{rbp}. This would be clumsy, and result in a larger binary size. Instead,
4974 .cfi_def_cfa %rsp, 8
4978 /* Do something else */
4981 That way, the rules for the instructions after @code{label} will be the same
4982 as before the first @code{.cfi_restore} without having to use multiple
4983 @code{.cfi} directives.
4985 @subsection @code{.cfi_return_column @var{register}}
4986 Change return column @var{register}, i.e. the return address is either
4987 directly in @var{register} or can be accessed by rules for @var{register}.
4989 @subsection @code{.cfi_signal_frame}
4990 Mark current function as signal trampoline.
4992 @subsection @code{.cfi_window_save}
4993 SPARC register window has been saved.
4995 @subsection @code{.cfi_escape} @var{expression}[, @dots{}]
4996 Allows the user to add arbitrary bytes to the unwind info. One
4997 might use this to add OS-specific CFI opcodes, or generic CFI
4998 opcodes that GAS does not yet support.
5000 @subsection @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
5001 The current value of @var{register} is @var{label}. The value of @var{label}
5002 will be encoded in the output file according to @var{encoding}; see the
5003 description of @code{.cfi_personality} for details on this encoding.
5005 The usefulness of equating a register to a fixed label is probably
5006 limited to the return address register. Here, it can be useful to
5007 mark a code segment that has only one return address which is reached
5008 by a direct branch and no copy of the return address exists in memory
5009 or another register.
5012 @section @code{.comm @var{symbol} , @var{length} }
5014 @cindex @code{comm} directive
5015 @cindex symbol, common
5016 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
5017 common symbol in one object file may be merged with a defined or common symbol
5018 of the same name in another object file. If @code{@value{LD}} does not see a
5019 definition for the symbol--just one or more common symbols--then it will
5020 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
5021 absolute expression. If @code{@value{LD}} sees multiple common symbols with
5022 the same name, and they do not all have the same size, it will allocate space
5023 using the largest size.
5026 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
5027 an optional third argument. This is the desired alignment of the symbol,
5028 specified for ELF as a byte boundary (for example, an alignment of 16 means
5029 that the least significant 4 bits of the address should be zero), and for PE
5030 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
5031 boundary). The alignment must be an absolute expression, and it must be a
5032 power of two. If @code{@value{LD}} allocates uninitialized memory for the
5033 common symbol, it will use the alignment when placing the symbol. If no
5034 alignment is specified, @command{@value{AS}} will set the alignment to the
5035 largest power of two less than or equal to the size of the symbol, up to a
5036 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
5037 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
5038 @samp{--section-alignment} option; image file sections in PE are aligned to
5039 multiples of 4096, which is far too large an alignment for ordinary variables.
5040 It is rather the default alignment for (non-debug) sections within object
5041 (@samp{*.o}) files, which are less strictly aligned.}.
5045 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
5046 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
5050 @section @code{.data @var{subsection}}
5052 @cindex @code{data} directive
5053 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
5054 end of the data subsection numbered @var{subsection} (which is an
5055 absolute expression). If @var{subsection} is omitted, it defaults
5060 @section @code{.def @var{name}}
5062 @cindex @code{def} directive
5063 @cindex COFF symbols, debugging
5064 @cindex debugging COFF symbols
5065 Begin defining debugging information for a symbol @var{name}; the
5066 definition extends until the @code{.endef} directive is encountered.
5069 This directive is only observed when @command{@value{AS}} is configured for COFF
5070 format output; when producing @code{b.out}, @samp{.def} is recognized,
5077 @section @code{.desc @var{symbol}, @var{abs-expression}}
5079 @cindex @code{desc} directive
5080 @cindex COFF symbol descriptor
5081 @cindex symbol descriptor, COFF
5082 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
5083 to the low 16 bits of an absolute expression.
5086 The @samp{.desc} directive is not available when @command{@value{AS}} is
5087 configured for COFF output; it is only for @code{a.out} or @code{b.out}
5088 object format. For the sake of compatibility, @command{@value{AS}} accepts
5089 it, but produces no output, when configured for COFF.
5095 @section @code{.dim}
5097 @cindex @code{dim} directive
5098 @cindex COFF auxiliary symbol information
5099 @cindex auxiliary symbol information, COFF
5100 This directive is generated by compilers to include auxiliary debugging
5101 information in the symbol table. It is only permitted inside
5102 @code{.def}/@code{.endef} pairs.
5105 @samp{.dim} is only meaningful when generating COFF format output; when
5106 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5112 @section @code{.double @var{flonums}}
5114 @cindex @code{double} directive
5115 @cindex floating point numbers (double)
5116 @code{.double} expects zero or more flonums, separated by commas. It
5117 assembles floating point numbers.
5119 The exact kind of floating point numbers emitted depends on how
5120 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5124 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
5125 in @sc{ieee} format.
5130 @section @code{.eject}
5132 @cindex @code{eject} directive
5133 @cindex new page, in listings
5134 @cindex page, in listings
5135 @cindex listing control: new page
5136 Force a page break at this point, when generating assembly listings.
5139 @section @code{.else}
5141 @cindex @code{else} directive
5142 @code{.else} is part of the @command{@value{AS}} support for conditional
5143 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
5144 of code to be assembled if the condition for the preceding @code{.if}
5148 @section @code{.elseif}
5150 @cindex @code{elseif} directive
5151 @code{.elseif} is part of the @command{@value{AS}} support for conditional
5152 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
5153 @code{.if} block that would otherwise fill the entire @code{.else} section.
5156 @section @code{.end}
5158 @cindex @code{end} directive
5159 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
5160 process anything in the file past the @code{.end} directive.
5164 @section @code{.endef}
5166 @cindex @code{endef} directive
5167 This directive flags the end of a symbol definition begun with
5171 @samp{.endef} is only meaningful when generating COFF format output; if
5172 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
5173 directive but ignores it.
5178 @section @code{.endfunc}
5179 @cindex @code{endfunc} directive
5180 @code{.endfunc} marks the end of a function specified with @code{.func}.
5183 @section @code{.endif}
5185 @cindex @code{endif} directive
5186 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
5187 it marks the end of a block of code that is only assembled
5188 conditionally. @xref{If,,@code{.if}}.
5191 @section @code{.equ @var{symbol}, @var{expression}}
5193 @cindex @code{equ} directive
5194 @cindex assigning values to symbols
5195 @cindex symbols, assigning values to
5196 This directive sets the value of @var{symbol} to @var{expression}.
5197 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
5200 The syntax for @code{equ} on the HPPA is
5201 @samp{@var{symbol} .equ @var{expression}}.
5205 The syntax for @code{equ} on the Z80 is
5206 @samp{@var{symbol} equ @var{expression}}.
5207 On the Z80 it is an error if @var{symbol} is already defined,
5208 but the symbol is not protected from later redefinition.
5209 Compare @ref{Equiv}.
5213 @section @code{.equiv @var{symbol}, @var{expression}}
5214 @cindex @code{equiv} directive
5215 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
5216 the assembler will signal an error if @var{symbol} is already defined. Note a
5217 symbol which has been referenced but not actually defined is considered to be
5220 Except for the contents of the error message, this is roughly equivalent to
5227 plus it protects the symbol from later redefinition.
5230 @section @code{.eqv @var{symbol}, @var{expression}}
5231 @cindex @code{eqv} directive
5232 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
5233 evaluate the expression or any part of it immediately. Instead each time
5234 the resulting symbol is used in an expression, a snapshot of its current
5238 @section @code{.err}
5239 @cindex @code{err} directive
5240 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
5241 message and, unless the @option{-Z} option was used, it will not generate an
5242 object file. This can be used to signal an error in conditionally compiled code.
5245 @section @code{.error "@var{string}"}
5246 @cindex error directive
5248 Similarly to @code{.err}, this directive emits an error, but you can specify a
5249 string that will be emitted as the error message. If you don't specify the
5250 message, it defaults to @code{".error directive invoked in source file"}.
5251 @xref{Errors, ,Error and Warning Messages}.
5254 .error "This code has not been assembled and tested."
5258 @section @code{.exitm}
5259 Exit early from the current macro definition. @xref{Macro}.
5262 @section @code{.extern}
5264 @cindex @code{extern} directive
5265 @code{.extern} is accepted in the source program---for compatibility
5266 with other assemblers---but it is ignored. @command{@value{AS}} treats
5267 all undefined symbols as external.
5270 @section @code{.fail @var{expression}}
5272 @cindex @code{fail} directive
5273 Generates an error or a warning. If the value of the @var{expression} is 500
5274 or more, @command{@value{AS}} will print a warning message. If the value is less
5275 than 500, @command{@value{AS}} will print an error message. The message will
5276 include the value of @var{expression}. This can occasionally be useful inside
5277 complex nested macros or conditional assembly.
5280 @section @code{.file}
5281 @cindex @code{file} directive
5283 @ifclear no-file-dir
5284 There are two different versions of the @code{.file} directive. Targets
5285 that support DWARF2 line number information use the DWARF2 version of
5286 @code{.file}. Other targets use the default version.
5288 @subheading Default Version
5290 @cindex logical file name
5291 @cindex file name, logical
5292 This version of the @code{.file} directive tells @command{@value{AS}} that we
5293 are about to start a new logical file. The syntax is:
5299 @var{string} is the new file name. In general, the filename is
5300 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
5301 to specify an empty file name, you must give the quotes--@code{""}. This
5302 statement may go away in future: it is only recognized to be compatible with
5303 old @command{@value{AS}} programs.
5305 @subheading DWARF2 Version
5308 When emitting DWARF2 line number information, @code{.file} assigns filenames
5309 to the @code{.debug_line} file name table. The syntax is:
5312 .file @var{fileno} @var{filename}
5315 The @var{fileno} operand should be a unique positive integer to use as the
5316 index of the entry in the table. The @var{filename} operand is a C string
5319 The detail of filename indices is exposed to the user because the filename
5320 table is shared with the @code{.debug_info} section of the DWARF2 debugging
5321 information, and thus the user must know the exact indices that table
5325 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
5327 @cindex @code{fill} directive
5328 @cindex writing patterns in memory
5329 @cindex patterns, writing in memory
5330 @var{repeat}, @var{size} and @var{value} are absolute expressions.
5331 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
5332 may be zero or more. @var{Size} may be zero or more, but if it is
5333 more than 8, then it is deemed to have the value 8, compatible with
5334 other people's assemblers. The contents of each @var{repeat} bytes
5335 is taken from an 8-byte number. The highest order 4 bytes are
5336 zero. The lowest order 4 bytes are @var{value} rendered in the
5337 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
5338 Each @var{size} bytes in a repetition is taken from the lowest order
5339 @var{size} bytes of this number. Again, this bizarre behavior is
5340 compatible with other people's assemblers.
5342 @var{size} and @var{value} are optional.
5343 If the second comma and @var{value} are absent, @var{value} is
5344 assumed zero. If the first comma and following tokens are absent,
5345 @var{size} is assumed to be 1.
5348 @section @code{.float @var{flonums}}
5350 @cindex floating point numbers (single)
5351 @cindex @code{float} directive
5352 This directive assembles zero or more flonums, separated by commas. It
5353 has the same effect as @code{.single}.
5355 The exact kind of floating point numbers emitted depends on how
5356 @command{@value{AS}} is configured.
5357 @xref{Machine Dependencies}.
5361 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5362 in @sc{ieee} format.
5367 @section @code{.func @var{name}[,@var{label}]}
5368 @cindex @code{func} directive
5369 @code{.func} emits debugging information to denote function @var{name}, and
5370 is ignored unless the file is assembled with debugging enabled.
5371 Only @samp{--gstabs[+]} is currently supported.
5372 @var{label} is the entry point of the function and if omitted @var{name}
5373 prepended with the @samp{leading char} is used.
5374 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5375 All functions are currently defined to have @code{void} return type.
5376 The function must be terminated with @code{.endfunc}.
5379 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5381 @cindex @code{global} directive
5382 @cindex symbol, making visible to linker
5383 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5384 @var{symbol} in your partial program, its value is made available to
5385 other partial programs that are linked with it. Otherwise,
5386 @var{symbol} takes its attributes from a symbol of the same name
5387 from another file linked into the same program.
5389 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5390 compatibility with other assemblers.
5393 On the HPPA, @code{.global} is not always enough to make it accessible to other
5394 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5395 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5400 @section @code{.gnu_attribute @var{tag},@var{value}}
5401 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5404 @section @code{.hidden @var{names}}
5406 @cindex @code{hidden} directive
5408 This is one of the ELF visibility directives. The other two are
5409 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5410 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5412 This directive overrides the named symbols default visibility (which is set by
5413 their binding: local, global or weak). The directive sets the visibility to
5414 @code{hidden} which means that the symbols are not visible to other components.
5415 Such symbols are always considered to be @code{protected} as well.
5419 @section @code{.hword @var{expressions}}
5421 @cindex @code{hword} directive
5422 @cindex integers, 16-bit
5423 @cindex numbers, 16-bit
5424 @cindex sixteen bit integers
5425 This expects zero or more @var{expressions}, and emits
5426 a 16 bit number for each.
5429 This directive is a synonym for @samp{.short}; depending on the target
5430 architecture, it may also be a synonym for @samp{.word}.
5434 This directive is a synonym for @samp{.short}.
5437 This directive is a synonym for both @samp{.short} and @samp{.word}.
5442 @section @code{.ident}
5444 @cindex @code{ident} directive
5446 This directive is used by some assemblers to place tags in object files. The
5447 behavior of this directive varies depending on the target. When using the
5448 a.out object file format, @command{@value{AS}} simply accepts the directive for
5449 source-file compatibility with existing assemblers, but does not emit anything
5450 for it. When using COFF, comments are emitted to the @code{.comment} or
5451 @code{.rdata} section, depending on the target. When using ELF, comments are
5452 emitted to the @code{.comment} section.
5455 @section @code{.if @var{absolute expression}}
5457 @cindex conditional assembly
5458 @cindex @code{if} directive
5459 @code{.if} marks the beginning of a section of code which is only
5460 considered part of the source program being assembled if the argument
5461 (which must be an @var{absolute expression}) is non-zero. The end of
5462 the conditional section of code must be marked by @code{.endif}
5463 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5464 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5465 If you have several conditions to check, @code{.elseif} may be used to avoid
5466 nesting blocks if/else within each subsequent @code{.else} block.
5468 The following variants of @code{.if} are also supported:
5470 @cindex @code{ifdef} directive
5471 @item .ifdef @var{symbol}
5472 Assembles the following section of code if the specified @var{symbol}
5473 has been defined. Note a symbol which has been referenced but not yet defined
5474 is considered to be undefined.
5476 @cindex @code{ifb} directive
5477 @item .ifb @var{text}
5478 Assembles the following section of code if the operand is blank (empty).
5480 @cindex @code{ifc} directive
5481 @item .ifc @var{string1},@var{string2}
5482 Assembles the following section of code if the two strings are the same. The
5483 strings may be optionally quoted with single quotes. If they are not quoted,
5484 the first string stops at the first comma, and the second string stops at the
5485 end of the line. Strings which contain whitespace should be quoted. The
5486 string comparison is case sensitive.
5488 @cindex @code{ifeq} directive
5489 @item .ifeq @var{absolute expression}
5490 Assembles the following section of code if the argument is zero.
5492 @cindex @code{ifeqs} directive
5493 @item .ifeqs @var{string1},@var{string2}
5494 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5496 @cindex @code{ifge} directive
5497 @item .ifge @var{absolute expression}
5498 Assembles the following section of code if the argument is greater than or
5501 @cindex @code{ifgt} directive
5502 @item .ifgt @var{absolute expression}
5503 Assembles the following section of code if the argument is greater than zero.
5505 @cindex @code{ifle} directive
5506 @item .ifle @var{absolute expression}
5507 Assembles the following section of code if the argument is less than or equal
5510 @cindex @code{iflt} directive
5511 @item .iflt @var{absolute expression}
5512 Assembles the following section of code if the argument is less than zero.
5514 @cindex @code{ifnb} directive
5515 @item .ifnb @var{text}
5516 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5517 following section of code if the operand is non-blank (non-empty).
5519 @cindex @code{ifnc} directive
5520 @item .ifnc @var{string1},@var{string2}.
5521 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5522 following section of code if the two strings are not the same.
5524 @cindex @code{ifndef} directive
5525 @cindex @code{ifnotdef} directive
5526 @item .ifndef @var{symbol}
5527 @itemx .ifnotdef @var{symbol}
5528 Assembles the following section of code if the specified @var{symbol}
5529 has not been defined. Both spelling variants are equivalent. Note a symbol
5530 which has been referenced but not yet defined is considered to be undefined.
5532 @cindex @code{ifne} directive
5533 @item .ifne @var{absolute expression}
5534 Assembles the following section of code if the argument is not equal to zero
5535 (in other words, this is equivalent to @code{.if}).
5537 @cindex @code{ifnes} directive
5538 @item .ifnes @var{string1},@var{string2}
5539 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5540 following section of code if the two strings are not the same.
5544 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5546 @cindex @code{incbin} directive
5547 @cindex binary files, including
5548 The @code{incbin} directive includes @var{file} verbatim at the current
5549 location. You can control the search paths used with the @samp{-I} command-line
5550 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5553 The @var{skip} argument skips a number of bytes from the start of the
5554 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5555 read. Note that the data is not aligned in any way, so it is the user's
5556 responsibility to make sure that proper alignment is provided both before and
5557 after the @code{incbin} directive.
5560 @section @code{.include "@var{file}"}
5562 @cindex @code{include} directive
5563 @cindex supporting files, including
5564 @cindex files, including
5565 This directive provides a way to include supporting files at specified
5566 points in your source program. The code from @var{file} is assembled as
5567 if it followed the point of the @code{.include}; when the end of the
5568 included file is reached, assembly of the original file continues. You
5569 can control the search paths used with the @samp{-I} command-line option
5570 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5574 @section @code{.int @var{expressions}}
5576 @cindex @code{int} directive
5577 @cindex integers, 32-bit
5578 Expect zero or more @var{expressions}, of any section, separated by commas.
5579 For each expression, emit a number that, at run time, is the value of that
5580 expression. The byte order and bit size of the number depends on what kind
5581 of target the assembly is for.
5585 On most forms of the H8/300, @code{.int} emits 16-bit
5586 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5593 @section @code{.internal @var{names}}
5595 @cindex @code{internal} directive
5597 This is one of the ELF visibility directives. The other two are
5598 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5599 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5601 This directive overrides the named symbols default visibility (which is set by
5602 their binding: local, global or weak). The directive sets the visibility to
5603 @code{internal} which means that the symbols are considered to be @code{hidden}
5604 (i.e., not visible to other components), and that some extra, processor specific
5605 processing must also be performed upon the symbols as well.
5609 @section @code{.irp @var{symbol},@var{values}}@dots{}
5611 @cindex @code{irp} directive
5612 Evaluate a sequence of statements assigning different values to @var{symbol}.
5613 The sequence of statements starts at the @code{.irp} directive, and is
5614 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5615 set to @var{value}, and the sequence of statements is assembled. If no
5616 @var{value} is listed, the sequence of statements is assembled once, with
5617 @var{symbol} set to the null string. To refer to @var{symbol} within the
5618 sequence of statements, use @var{\symbol}.
5620 For example, assembling
5628 is equivalent to assembling
5636 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5639 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5641 @cindex @code{irpc} directive
5642 Evaluate a sequence of statements assigning different values to @var{symbol}.
5643 The sequence of statements starts at the @code{.irpc} directive, and is
5644 terminated by an @code{.endr} directive. For each character in @var{value},
5645 @var{symbol} is set to the character, and the sequence of statements is
5646 assembled. If no @var{value} is listed, the sequence of statements is
5647 assembled once, with @var{symbol} set to the null string. To refer to
5648 @var{symbol} within the sequence of statements, use @var{\symbol}.
5650 For example, assembling
5658 is equivalent to assembling
5666 For some caveats with the spelling of @var{symbol}, see also the discussion
5670 @section @code{.lcomm @var{symbol} , @var{length}}
5672 @cindex @code{lcomm} directive
5673 @cindex local common symbols
5674 @cindex symbols, local common
5675 Reserve @var{length} (an absolute expression) bytes for a local common
5676 denoted by @var{symbol}. The section and value of @var{symbol} are
5677 those of the new local common. The addresses are allocated in the bss
5678 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5679 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5680 not visible to @code{@value{LD}}.
5683 Some targets permit a third argument to be used with @code{.lcomm}. This
5684 argument specifies the desired alignment of the symbol in the bss section.
5688 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5689 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5693 @section @code{.lflags}
5695 @cindex @code{lflags} directive (ignored)
5696 @command{@value{AS}} accepts this directive, for compatibility with other
5697 assemblers, but ignores it.
5699 @ifclear no-line-dir
5701 @section @code{.line @var{line-number}}
5703 @cindex @code{line} directive
5704 @cindex logical line number
5706 Change the logical line number. @var{line-number} must be an absolute
5707 expression. The next line has that logical line number. Therefore any other
5708 statements on the current line (after a statement separator character) are
5709 reported as on logical line number @var{line-number} @minus{} 1. One day
5710 @command{@value{AS}} will no longer support this directive: it is recognized only
5711 for compatibility with existing assembler programs.
5714 Even though this is a directive associated with the @code{a.out} or
5715 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5716 when producing COFF output, and treats @samp{.line} as though it
5717 were the COFF @samp{.ln} @emph{if} it is found outside a
5718 @code{.def}/@code{.endef} pair.
5720 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5721 used by compilers to generate auxiliary symbol information for
5726 @section @code{.linkonce [@var{type}]}
5728 @cindex @code{linkonce} directive
5729 @cindex common sections
5730 Mark the current section so that the linker only includes a single copy of it.
5731 This may be used to include the same section in several different object files,
5732 but ensure that the linker will only include it once in the final output file.
5733 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5734 Duplicate sections are detected based on the section name, so it should be
5737 This directive is only supported by a few object file formats; as of this
5738 writing, the only object file format which supports it is the Portable
5739 Executable format used on Windows NT.
5741 The @var{type} argument is optional. If specified, it must be one of the
5742 following strings. For example:
5746 Not all types may be supported on all object file formats.
5750 Silently discard duplicate sections. This is the default.
5753 Warn if there are duplicate sections, but still keep only one copy.
5756 Warn if any of the duplicates have different sizes.
5759 Warn if any of the duplicates do not have exactly the same contents.
5763 @section @code{.list}
5765 @cindex @code{list} directive
5766 @cindex listing control, turning on
5767 Control (in conjunction with the @code{.nolist} directive) whether or
5768 not assembly listings are generated. These two directives maintain an
5769 internal counter (which is zero initially). @code{.list} increments the
5770 counter, and @code{.nolist} decrements it. Assembly listings are
5771 generated whenever the counter is greater than zero.
5773 By default, listings are disabled. When you enable them (with the
5774 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5775 the initial value of the listing counter is one.
5778 @section @code{.ln @var{line-number}}
5780 @cindex @code{ln} directive
5781 @ifclear no-line-dir
5782 @samp{.ln} is a synonym for @samp{.line}.
5785 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5786 must be an absolute expression. The next line has that logical
5787 line number, so any other statements on the current line (after a
5788 statement separator character @code{;}) are reported as on logical
5789 line number @var{line-number} @minus{} 1.
5792 This directive is accepted, but ignored, when @command{@value{AS}} is
5793 configured for @code{b.out}; its effect is only associated with COFF
5799 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5800 @cindex @code{loc} directive
5801 When emitting DWARF2 line number information,
5802 the @code{.loc} directive will add a row to the @code{.debug_line} line
5803 number matrix corresponding to the immediately following assembly
5804 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5805 arguments will be applied to the @code{.debug_line} state machine before
5808 The @var{options} are a sequence of the following tokens in any order:
5812 This option will set the @code{basic_block} register in the
5813 @code{.debug_line} state machine to @code{true}.
5816 This option will set the @code{prologue_end} register in the
5817 @code{.debug_line} state machine to @code{true}.
5819 @item epilogue_begin
5820 This option will set the @code{epilogue_begin} register in the
5821 @code{.debug_line} state machine to @code{true}.
5823 @item is_stmt @var{value}
5824 This option will set the @code{is_stmt} register in the
5825 @code{.debug_line} state machine to @code{value}, which must be
5828 @item isa @var{value}
5829 This directive will set the @code{isa} register in the @code{.debug_line}
5830 state machine to @var{value}, which must be an unsigned integer.
5832 @item discriminator @var{value}
5833 This directive will set the @code{discriminator} register in the @code{.debug_line}
5834 state machine to @var{value}, which must be an unsigned integer.
5838 @node Loc_mark_labels
5839 @section @code{.loc_mark_labels @var{enable}}
5840 @cindex @code{loc_mark_labels} directive
5841 When emitting DWARF2 line number information,
5842 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5843 to the @code{.debug_line} line number matrix with the @code{basic_block}
5844 register in the state machine set whenever a code label is seen.
5845 The @var{enable} argument should be either 1 or 0, to enable or disable
5846 this function respectively.
5850 @section @code{.local @var{names}}
5852 @cindex @code{local} directive
5853 This directive, which is available for ELF targets, marks each symbol in
5854 the comma-separated list of @code{names} as a local symbol so that it
5855 will not be externally visible. If the symbols do not already exist,
5856 they will be created.
5858 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5859 accept an alignment argument, which is the case for most ELF targets,
5860 the @code{.local} directive can be used in combination with @code{.comm}
5861 (@pxref{Comm}) to define aligned local common data.
5865 @section @code{.long @var{expressions}}
5867 @cindex @code{long} directive
5868 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5871 @c no one seems to know what this is for or whether this description is
5872 @c what it really ought to do
5874 @section @code{.lsym @var{symbol}, @var{expression}}
5876 @cindex @code{lsym} directive
5877 @cindex symbol, not referenced in assembly
5878 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5879 the hash table, ensuring it cannot be referenced by name during the
5880 rest of the assembly. This sets the attributes of the symbol to be
5881 the same as the expression value:
5883 @var{other} = @var{descriptor} = 0
5884 @var{type} = @r{(section of @var{expression})}
5885 @var{value} = @var{expression}
5888 The new symbol is not flagged as external.
5892 @section @code{.macro}
5895 The commands @code{.macro} and @code{.endm} allow you to define macros that
5896 generate assembly output. For example, this definition specifies a macro
5897 @code{sum} that puts a sequence of numbers into memory:
5900 .macro sum from=0, to=5
5909 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5921 @item .macro @var{macname}
5922 @itemx .macro @var{macname} @var{macargs} @dots{}
5923 @cindex @code{macro} directive
5924 Begin the definition of a macro called @var{macname}. If your macro
5925 definition requires arguments, specify their names after the macro name,
5926 separated by commas or spaces. You can qualify the macro argument to
5927 indicate whether all invocations must specify a non-blank value (through
5928 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5929 (through @samp{:@code{vararg}}). You can supply a default value for any
5930 macro argument by following the name with @samp{=@var{deflt}}. You
5931 cannot define two macros with the same @var{macname} unless it has been
5932 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5933 definitions. For example, these are all valid @code{.macro} statements:
5937 Begin the definition of a macro called @code{comm}, which takes no
5940 @item .macro plus1 p, p1
5941 @itemx .macro plus1 p p1
5942 Either statement begins the definition of a macro called @code{plus1},
5943 which takes two arguments; within the macro definition, write
5944 @samp{\p} or @samp{\p1} to evaluate the arguments.
5946 @item .macro reserve_str p1=0 p2
5947 Begin the definition of a macro called @code{reserve_str}, with two
5948 arguments. The first argument has a default value, but not the second.
5949 After the definition is complete, you can call the macro either as
5950 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5951 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5952 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5953 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5955 @item .macro m p1:req, p2=0, p3:vararg
5956 Begin the definition of a macro called @code{m}, with at least three
5957 arguments. The first argument must always have a value specified, but
5958 not the second, which instead has a default value. The third formal
5959 will get assigned all remaining arguments specified at invocation time.
5961 When you call a macro, you can specify the argument values either by
5962 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5963 @samp{sum to=17, from=9}.
5967 Note that since each of the @var{macargs} can be an identifier exactly
5968 as any other one permitted by the target architecture, there may be
5969 occasional problems if the target hand-crafts special meanings to certain
5970 characters when they occur in a special position. For example, if the colon
5971 (@code{:}) is generally permitted to be part of a symbol name, but the
5972 architecture specific code special-cases it when occurring as the final
5973 character of a symbol (to denote a label), then the macro parameter
5974 replacement code will have no way of knowing that and consider the whole
5975 construct (including the colon) an identifier, and check only this
5976 identifier for being the subject to parameter substitution. So for example
5977 this macro definition:
5985 might not work as expected. Invoking @samp{label foo} might not create a label
5986 called @samp{foo} but instead just insert the text @samp{\l:} into the
5987 assembler source, probably generating an error about an unrecognised
5990 Similarly problems might occur with the period character (@samp{.})
5991 which is often allowed inside opcode names (and hence identifier names). So
5992 for example constructing a macro to build an opcode from a base name and a
5993 length specifier like this:
5996 .macro opcode base length
6001 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
6002 instruction but instead generate some kind of error as the assembler tries to
6003 interpret the text @samp{\base.\length}.
6005 There are several possible ways around this problem:
6008 @item Insert white space
6009 If it is possible to use white space characters then this is the simplest
6018 @item Use @samp{\()}
6019 The string @samp{\()} can be used to separate the end of a macro argument from
6020 the following text. eg:
6023 .macro opcode base length
6028 @item Use the alternate macro syntax mode
6029 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
6030 used as a separator. eg:
6040 Note: this problem of correctly identifying string parameters to pseudo ops
6041 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
6042 and @code{.irpc} (@pxref{Irpc}) as well.
6045 @cindex @code{endm} directive
6046 Mark the end of a macro definition.
6049 @cindex @code{exitm} directive
6050 Exit early from the current macro definition.
6052 @cindex number of macros executed
6053 @cindex macros, count executed
6055 @command{@value{AS}} maintains a counter of how many macros it has
6056 executed in this pseudo-variable; you can copy that number to your
6057 output with @samp{\@@}, but @emph{only within a macro definition}.
6059 @item LOCAL @var{name} [ , @dots{} ]
6060 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
6061 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
6062 @xref{Altmacro,,@code{.altmacro}}.
6066 @section @code{.mri @var{val}}
6068 @cindex @code{mri} directive
6069 @cindex MRI mode, temporarily
6070 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
6071 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
6072 affects code assembled until the next @code{.mri} directive, or until the end
6073 of the file. @xref{M, MRI mode, MRI mode}.
6076 @section @code{.noaltmacro}
6077 Disable alternate macro mode. @xref{Altmacro}.
6080 @section @code{.nolist}
6082 @cindex @code{nolist} directive
6083 @cindex listing control, turning off
6084 Control (in conjunction with the @code{.list} directive) whether or
6085 not assembly listings are generated. These two directives maintain an
6086 internal counter (which is zero initially). @code{.list} increments the
6087 counter, and @code{.nolist} decrements it. Assembly listings are
6088 generated whenever the counter is greater than zero.
6091 @section @code{.octa @var{bignums}}
6093 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
6094 @cindex @code{octa} directive
6095 @cindex integer, 16-byte
6096 @cindex sixteen byte integer
6097 This directive expects zero or more bignums, separated by commas. For each
6098 bignum, it emits a 16-byte integer.
6100 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
6101 hence @emph{octa}-word for 16 bytes.
6104 @section @code{.offset @var{loc}}
6106 @cindex @code{offset} directive
6107 Set the location counter to @var{loc} in the absolute section. @var{loc} must
6108 be an absolute expression. This directive may be useful for defining
6109 symbols with absolute values. Do not confuse it with the @code{.org}
6113 @section @code{.org @var{new-lc} , @var{fill}}
6115 @cindex @code{org} directive
6116 @cindex location counter, advancing
6117 @cindex advancing location counter
6118 @cindex current address, advancing
6119 Advance the location counter of the current section to
6120 @var{new-lc}. @var{new-lc} is either an absolute expression or an
6121 expression with the same section as the current subsection. That is,
6122 you can't use @code{.org} to cross sections: if @var{new-lc} has the
6123 wrong section, the @code{.org} directive is ignored. To be compatible
6124 with former assemblers, if the section of @var{new-lc} is absolute,
6125 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
6126 is the same as the current subsection.
6128 @code{.org} may only increase the location counter, or leave it
6129 unchanged; you cannot use @code{.org} to move the location counter
6132 @c double negative used below "not undefined" because this is a specific
6133 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
6134 @c section. doc@cygnus.com 18feb91
6135 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
6136 may not be undefined. If you really detest this restriction we eagerly await
6137 a chance to share your improved assembler.
6139 Beware that the origin is relative to the start of the section, not
6140 to the start of the subsection. This is compatible with other
6141 people's assemblers.
6143 When the location counter (of the current subsection) is advanced, the
6144 intervening bytes are filled with @var{fill} which should be an
6145 absolute expression. If the comma and @var{fill} are omitted,
6146 @var{fill} defaults to zero.
6149 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
6151 @cindex padding the location counter given a power of two
6152 @cindex @code{p2align} directive
6153 Pad the location counter (in the current subsection) to a particular
6154 storage boundary. The first expression (which must be absolute) is the
6155 number of low-order zero bits the location counter must have after
6156 advancement. For example @samp{.p2align 3} advances the location
6157 counter until it a multiple of 8. If the location counter is already a
6158 multiple of 8, no change is needed.
6160 The second expression (also absolute) gives the fill value to be stored in the
6161 padding bytes. It (and the comma) may be omitted. If it is omitted, the
6162 padding bytes are normally zero. However, on some systems, if the section is
6163 marked as containing code and the fill value is omitted, the space is filled
6164 with no-op instructions.
6166 The third expression is also absolute, and is also optional. If it is present,
6167 it is the maximum number of bytes that should be skipped by this alignment
6168 directive. If doing the alignment would require skipping more bytes than the
6169 specified maximum, then the alignment is not done at all. You can omit the
6170 fill value (the second argument) entirely by simply using two commas after the
6171 required alignment; this can be useful if you want the alignment to be filled
6172 with no-op instructions when appropriate.
6174 @cindex @code{p2alignw} directive
6175 @cindex @code{p2alignl} directive
6176 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
6177 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
6178 pattern as a two byte word value. The @code{.p2alignl} directives treats the
6179 fill pattern as a four byte longword value. For example, @code{.p2alignw
6180 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
6181 filled in with the value 0x368d (the exact placement of the bytes depends upon
6182 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
6187 @section @code{.popsection}
6189 @cindex @code{popsection} directive
6190 @cindex Section Stack
6191 This is one of the ELF section stack manipulation directives. The others are
6192 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6193 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
6196 This directive replaces the current section (and subsection) with the top
6197 section (and subsection) on the section stack. This section is popped off the
6203 @section @code{.previous}
6205 @cindex @code{previous} directive
6206 @cindex Section Stack
6207 This is one of the ELF section stack manipulation directives. The others are
6208 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6209 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
6210 (@pxref{PopSection}).
6212 This directive swaps the current section (and subsection) with most recently
6213 referenced section/subsection pair prior to this one. Multiple
6214 @code{.previous} directives in a row will flip between two sections (and their
6215 subsections). For example:
6227 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
6233 # Now in section A subsection 1
6237 # Now in section B subsection 0
6240 # Now in section B subsection 1
6243 # Now in section B subsection 0
6247 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
6248 section B and 0x9abc into subsection 1 of section B.
6250 In terms of the section stack, this directive swaps the current section with
6251 the top section on the section stack.
6255 @section @code{.print @var{string}}
6257 @cindex @code{print} directive
6258 @command{@value{AS}} will print @var{string} on the standard output during
6259 assembly. You must put @var{string} in double quotes.
6263 @section @code{.protected @var{names}}
6265 @cindex @code{protected} directive
6267 This is one of the ELF visibility directives. The other two are
6268 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
6270 This directive overrides the named symbols default visibility (which is set by
6271 their binding: local, global or weak). The directive sets the visibility to
6272 @code{protected} which means that any references to the symbols from within the
6273 components that defines them must be resolved to the definition in that
6274 component, even if a definition in another component would normally preempt
6279 @section @code{.psize @var{lines} , @var{columns}}
6281 @cindex @code{psize} directive
6282 @cindex listing control: paper size
6283 @cindex paper size, for listings
6284 Use this directive to declare the number of lines---and, optionally, the
6285 number of columns---to use for each page, when generating listings.
6287 If you do not use @code{.psize}, listings use a default line-count
6288 of 60. You may omit the comma and @var{columns} specification; the
6289 default width is 200 columns.
6291 @command{@value{AS}} generates formfeeds whenever the specified number of
6292 lines is exceeded (or whenever you explicitly request one, using
6295 If you specify @var{lines} as @code{0}, no formfeeds are generated save
6296 those explicitly specified with @code{.eject}.
6299 @section @code{.purgem @var{name}}
6301 @cindex @code{purgem} directive
6302 Undefine the macro @var{name}, so that later uses of the string will not be
6303 expanded. @xref{Macro}.
6307 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
6309 @cindex @code{pushsection} directive
6310 @cindex Section Stack
6311 This is one of the ELF section stack manipulation directives. The others are
6312 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6313 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6316 This directive pushes the current section (and subsection) onto the
6317 top of the section stack, and then replaces the current section and
6318 subsection with @code{name} and @code{subsection}. The optional
6319 @code{flags}, @code{type} and @code{arguments} are treated the same
6320 as in the @code{.section} (@pxref{Section}) directive.
6324 @section @code{.quad @var{bignums}}
6326 @cindex @code{quad} directive
6327 @code{.quad} expects zero or more bignums, separated by commas. For
6328 each bignum, it emits
6330 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
6331 warning message; and just takes the lowest order 8 bytes of the bignum.
6332 @cindex eight-byte integer
6333 @cindex integer, 8-byte
6335 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
6336 hence @emph{quad}-word for 8 bytes.
6339 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
6340 warning message; and just takes the lowest order 16 bytes of the bignum.
6341 @cindex sixteen-byte integer
6342 @cindex integer, 16-byte
6346 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6348 @cindex @code{reloc} directive
6349 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6350 @var{expression}. If @var{offset} is a number, the relocation is generated in
6351 the current section. If @var{offset} is an expression that resolves to a
6352 symbol plus offset, the relocation is generated in the given symbol's section.
6353 @var{expression}, if present, must resolve to a symbol plus addend or to an
6354 absolute value, but note that not all targets support an addend. e.g. ELF REL
6355 targets such as i386 store an addend in the section contents rather than in the
6356 relocation. This low level interface does not support addends stored in the
6360 @section @code{.rept @var{count}}
6362 @cindex @code{rept} directive
6363 Repeat the sequence of lines between the @code{.rept} directive and the next
6364 @code{.endr} directive @var{count} times.
6366 For example, assembling
6374 is equivalent to assembling
6383 @section @code{.sbttl "@var{subheading}"}
6385 @cindex @code{sbttl} directive
6386 @cindex subtitles for listings
6387 @cindex listing control: subtitle
6388 Use @var{subheading} as the title (third line, immediately after the
6389 title line) when generating assembly listings.
6391 This directive affects subsequent pages, as well as the current page if
6392 it appears within ten lines of the top of a page.
6396 @section @code{.scl @var{class}}
6398 @cindex @code{scl} directive
6399 @cindex symbol storage class (COFF)
6400 @cindex COFF symbol storage class
6401 Set the storage-class value for a symbol. This directive may only be
6402 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6403 whether a symbol is static or external, or it may record further
6404 symbolic debugging information.
6407 The @samp{.scl} directive is primarily associated with COFF output; when
6408 configured to generate @code{b.out} output format, @command{@value{AS}}
6409 accepts this directive but ignores it.
6415 @section @code{.section @var{name}}
6417 @cindex named section
6418 Use the @code{.section} directive to assemble the following code into a section
6421 This directive is only supported for targets that actually support arbitrarily
6422 named sections; on @code{a.out} targets, for example, it is not accepted, even
6423 with a standard @code{a.out} section name.
6427 @c only print the extra heading if both COFF and ELF are set
6428 @subheading COFF Version
6431 @cindex @code{section} directive (COFF version)
6432 For COFF targets, the @code{.section} directive is used in one of the following
6436 .section @var{name}[, "@var{flags}"]
6437 .section @var{name}[, @var{subsection}]
6440 If the optional argument is quoted, it is taken as flags to use for the
6441 section. Each flag is a single character. The following flags are recognized:
6445 bss section (uninitialized data)
6447 section is not loaded
6453 exclude section from linking
6459 shared section (meaningful for PE targets)
6461 ignored. (For compatibility with the ELF version)
6463 section is not readable (meaningful for PE targets)
6465 single-digit power-of-two section alignment (GNU extension)
6468 If no flags are specified, the default flags depend upon the section name. If
6469 the section name is not recognized, the default will be for the section to be
6470 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6471 from the section, rather than adding them, so if they are used on their own it
6472 will be as if no flags had been specified at all.
6474 If the optional argument to the @code{.section} directive is not quoted, it is
6475 taken as a subsection number (@pxref{Sub-Sections}).
6480 @c only print the extra heading if both COFF and ELF are set
6481 @subheading ELF Version
6484 @cindex Section Stack
6485 This is one of the ELF section stack manipulation directives. The others are
6486 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6487 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6488 @code{.previous} (@pxref{Previous}).
6490 @cindex @code{section} directive (ELF version)
6491 For ELF targets, the @code{.section} directive is used like this:
6494 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6497 @anchor{Section Name Substitutions}
6498 @kindex --sectname-subst
6499 @cindex section name substitution
6500 If the @samp{--sectname-subst} command-line option is provided, the @var{name}
6501 argument may contain a substitution sequence. Only @code{%S} is supported
6502 at the moment, and substitutes the current section name. For example:
6505 .macro exception_code
6506 .section %S.exception
6507 [exception code here]
6522 The two @code{exception_code} invocations above would create the
6523 @code{.text.exception} and @code{.init.exception} sections respectively.
6524 This is useful e.g. to discriminate between ancillary sections that are
6525 tied to setup code to be discarded after use from ancillary sections that
6526 need to stay resident without having to define multiple @code{exception_code}
6527 macros just for that purpose.
6529 The optional @var{flags} argument is a quoted string which may contain any
6530 combination of the following characters:
6534 section is allocatable
6536 section is a GNU_MBIND section
6538 section is excluded from executable and shared library.
6542 section is executable
6544 section is mergeable
6546 section contains zero terminated strings
6548 section is a member of a section group
6550 section is used for thread-local-storage
6552 section is a member of the previously-current section's group, if any
6553 @item @code{<number>}
6554 a numeric value indicating the bits to be set in the ELF section header's flags
6555 field. Note - if one or more of the alphabetic characters described above is
6556 also included in the flags field, their bit values will be ORed into the
6558 @item @code{<target specific>}
6559 some targets extend this list with their own flag characters
6562 Note - once a section's flags have been set they cannot be changed. There are
6563 a few exceptions to this rule however. Processor and application specific
6564 flags can be added to an already defined section. The @code{.interp},
6565 @code{.strtab} and @code{.symtab} sections can have the allocate flag
6566 (@code{a}) set after they are initially defined, and the @code{.note-GNU-stack}
6567 section may have the executable (@code{x}) flag added.
6569 The optional @var{type} argument may contain one of the following constants:
6573 section contains data
6575 section does not contain data (i.e., section only occupies space)
6577 section contains data which is used by things other than the program
6579 section contains an array of pointers to init functions
6581 section contains an array of pointers to finish functions
6582 @item @@preinit_array
6583 section contains an array of pointers to pre-init functions
6584 @item @@@code{<number>}
6585 a numeric value to be set as the ELF section header's type field.
6586 @item @@@code{<target specific>}
6587 some targets extend this list with their own types
6590 Many targets only support the first three section types. The type may be
6591 enclosed in double quotes if necessary.
6593 Note on targets where the @code{@@} character is the start of a comment (eg
6594 ARM) then another character is used instead. For example the ARM port uses the
6597 Note - some sections, eg @code{.text} and @code{.data} are considered to be
6598 special and have fixed types. Any attempt to declare them with a different
6599 type will generate an error from the assembler.
6601 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6602 be specified as well as an extra argument---@var{entsize}---like this:
6605 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6608 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6609 constants, each @var{entsize} octets long. Sections with both @code{M} and
6610 @code{S} must contain zero terminated strings where each character is
6611 @var{entsize} bytes long. The linker may remove duplicates within sections with
6612 the same name, same entity size and same flags. @var{entsize} must be an
6613 absolute expression. For sections with both @code{M} and @code{S}, a string
6614 which is a suffix of a larger string is considered a duplicate. Thus
6615 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6616 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6618 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6619 be present along with an additional field like this:
6622 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6625 The @var{GroupName} field specifies the name of the section group to which this
6626 particular section belongs. The optional linkage field can contain:
6630 indicates that only one copy of this section should be retained
6635 Note: if both the @var{M} and @var{G} flags are present then the fields for
6636 the Merge flag should come first, like this:
6639 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6642 If @var{flags} contains the @code{?} symbol then it may not also contain the
6643 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6644 present. Instead, @code{?} says to consider the section that's current before
6645 this directive. If that section used @code{G}, then the new section will use
6646 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6647 If not, then the @code{?} symbol has no effect.
6649 If no flags are specified, the default flags depend upon the section name. If
6650 the section name is not recognized, the default will be for the section to have
6651 none of the above flags: it will not be allocated in memory, nor writable, nor
6652 executable. The section will contain data.
6654 For ELF targets, the assembler supports another type of @code{.section}
6655 directive for compatibility with the Solaris assembler:
6658 .section "@var{name}"[, @var{flags}...]
6661 Note that the section name is quoted. There may be a sequence of comma
6666 section is allocatable
6670 section is executable
6672 section is excluded from executable and shared library.
6674 section is used for thread local storage
6677 This directive replaces the current section and subsection. See the
6678 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6679 some examples of how this directive and the other section stack directives
6685 @section @code{.set @var{symbol}, @var{expression}}
6687 @cindex @code{set} directive
6688 @cindex symbol value, setting
6689 Set the value of @var{symbol} to @var{expression}. This
6690 changes @var{symbol}'s value and type to conform to
6691 @var{expression}. If @var{symbol} was flagged as external, it remains
6692 flagged (@pxref{Symbol Attributes}).
6694 You may @code{.set} a symbol many times in the same assembly provided that the
6695 values given to the symbol are constants. Values that are based on expressions
6696 involving other symbols are allowed, but some targets may restrict this to only
6697 being done once per assembly. This is because those targets do not set the
6698 addresses of symbols at assembly time, but rather delay the assignment until a
6699 final link is performed. This allows the linker a chance to change the code in
6700 the files, changing the location of, and the relative distance between, various
6703 If you @code{.set} a global symbol, the value stored in the object
6704 file is the last value stored into it.
6707 On Z80 @code{set} is a real instruction, use
6708 @samp{@var{symbol} defl @var{expression}} instead.
6712 @section @code{.short @var{expressions}}
6714 @cindex @code{short} directive
6716 @code{.short} is normally the same as @samp{.word}.
6717 @xref{Word,,@code{.word}}.
6719 In some configurations, however, @code{.short} and @code{.word} generate
6720 numbers of different lengths. @xref{Machine Dependencies}.
6724 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6727 This expects zero or more @var{expressions}, and emits
6728 a 16 bit number for each.
6733 @section @code{.single @var{flonums}}
6735 @cindex @code{single} directive
6736 @cindex floating point numbers (single)
6737 This directive assembles zero or more flonums, separated by commas. It
6738 has the same effect as @code{.float}.
6740 The exact kind of floating point numbers emitted depends on how
6741 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6745 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6746 numbers in @sc{ieee} format.
6752 @section @code{.size}
6754 This directive is used to set the size associated with a symbol.
6758 @c only print the extra heading if both COFF and ELF are set
6759 @subheading COFF Version
6762 @cindex @code{size} directive (COFF version)
6763 For COFF targets, the @code{.size} directive is only permitted inside
6764 @code{.def}/@code{.endef} pairs. It is used like this:
6767 .size @var{expression}
6771 @samp{.size} is only meaningful when generating COFF format output; when
6772 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6779 @c only print the extra heading if both COFF and ELF are set
6780 @subheading ELF Version
6783 @cindex @code{size} directive (ELF version)
6784 For ELF targets, the @code{.size} directive is used like this:
6787 .size @var{name} , @var{expression}
6790 This directive sets the size associated with a symbol @var{name}.
6791 The size in bytes is computed from @var{expression} which can make use of label
6792 arithmetic. This directive is typically used to set the size of function
6797 @ifclear no-space-dir
6799 @section @code{.skip @var{size} , @var{fill}}
6801 @cindex @code{skip} directive
6802 @cindex filling memory
6803 This directive emits @var{size} bytes, each of value @var{fill}. Both
6804 @var{size} and @var{fill} are absolute expressions. If the comma and
6805 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6810 @section @code{.sleb128 @var{expressions}}
6812 @cindex @code{sleb128} directive
6813 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6814 compact, variable length representation of numbers used by the DWARF
6815 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6817 @ifclear no-space-dir
6819 @section @code{.space @var{size} , @var{fill}}
6821 @cindex @code{space} directive
6822 @cindex filling memory
6823 This directive emits @var{size} bytes, each of value @var{fill}. Both
6824 @var{size} and @var{fill} are absolute expressions. If the comma
6825 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6830 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6831 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6832 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6833 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6841 @section @code{.stabd, .stabn, .stabs}
6843 @cindex symbolic debuggers, information for
6844 @cindex @code{stab@var{x}} directives
6845 There are three directives that begin @samp{.stab}.
6846 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6847 The symbols are not entered in the @command{@value{AS}} hash table: they
6848 cannot be referenced elsewhere in the source file.
6849 Up to five fields are required:
6853 This is the symbol's name. It may contain any character except
6854 @samp{\000}, so is more general than ordinary symbol names. Some
6855 debuggers used to code arbitrarily complex structures into symbol names
6859 An absolute expression. The symbol's type is set to the low 8 bits of
6860 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6861 and debuggers choke on silly bit patterns.
6864 An absolute expression. The symbol's ``other'' attribute is set to the
6865 low 8 bits of this expression.
6868 An absolute expression. The symbol's descriptor is set to the low 16
6869 bits of this expression.
6872 An absolute expression which becomes the symbol's value.
6875 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6876 or @code{.stabs} statement, the symbol has probably already been created;
6877 you get a half-formed symbol in your object file. This is
6878 compatible with earlier assemblers!
6881 @cindex @code{stabd} directive
6882 @item .stabd @var{type} , @var{other} , @var{desc}
6884 The ``name'' of the symbol generated is not even an empty string.
6885 It is a null pointer, for compatibility. Older assemblers used a
6886 null pointer so they didn't waste space in object files with empty
6889 The symbol's value is set to the location counter,
6890 relocatably. When your program is linked, the value of this symbol
6891 is the address of the location counter when the @code{.stabd} was
6894 @cindex @code{stabn} directive
6895 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6896 The name of the symbol is set to the empty string @code{""}.
6898 @cindex @code{stabs} directive
6899 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6900 All five fields are specified.
6906 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6907 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6909 @cindex string, copying to object file
6910 @cindex string8, copying to object file
6911 @cindex string16, copying to object file
6912 @cindex string32, copying to object file
6913 @cindex string64, copying to object file
6914 @cindex @code{string} directive
6915 @cindex @code{string8} directive
6916 @cindex @code{string16} directive
6917 @cindex @code{string32} directive
6918 @cindex @code{string64} directive
6920 Copy the characters in @var{str} to the object file. You may specify more than
6921 one string to copy, separated by commas. Unless otherwise specified for a
6922 particular machine, the assembler marks the end of each string with a 0 byte.
6923 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6925 The variants @code{string16}, @code{string32} and @code{string64} differ from
6926 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6927 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6928 are stored in target endianness byte order.
6934 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6935 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6940 @section @code{.struct @var{expression}}
6942 @cindex @code{struct} directive
6943 Switch to the absolute section, and set the section offset to @var{expression},
6944 which must be an absolute expression. You might use this as follows:
6953 This would define the symbol @code{field1} to have the value 0, the symbol
6954 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6955 value 8. Assembly would be left in the absolute section, and you would need to
6956 use a @code{.section} directive of some sort to change to some other section
6957 before further assembly.
6961 @section @code{.subsection @var{name}}
6963 @cindex @code{subsection} directive
6964 @cindex Section Stack
6965 This is one of the ELF section stack manipulation directives. The others are
6966 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6967 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6970 This directive replaces the current subsection with @code{name}. The current
6971 section is not changed. The replaced subsection is put onto the section stack
6972 in place of the then current top of stack subsection.
6977 @section @code{.symver}
6978 @cindex @code{symver} directive
6979 @cindex symbol versioning
6980 @cindex versions of symbols
6981 Use the @code{.symver} directive to bind symbols to specific version nodes
6982 within a source file. This is only supported on ELF platforms, and is
6983 typically used when assembling files to be linked into a shared library.
6984 There are cases where it may make sense to use this in objects to be bound
6985 into an application itself so as to override a versioned symbol from a
6988 For ELF targets, the @code{.symver} directive can be used like this:
6990 .symver @var{name}, @var{name2@@nodename}
6992 If the symbol @var{name} is defined within the file
6993 being assembled, the @code{.symver} directive effectively creates a symbol
6994 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6995 just don't try and create a regular alias is that the @var{@@} character isn't
6996 permitted in symbol names. The @var{name2} part of the name is the actual name
6997 of the symbol by which it will be externally referenced. The name @var{name}
6998 itself is merely a name of convenience that is used so that it is possible to
6999 have definitions for multiple versions of a function within a single source
7000 file, and so that the compiler can unambiguously know which version of a
7001 function is being mentioned. The @var{nodename} portion of the alias should be
7002 the name of a node specified in the version script supplied to the linker when
7003 building a shared library. If you are attempting to override a versioned
7004 symbol from a shared library, then @var{nodename} should correspond to the
7005 nodename of the symbol you are trying to override.
7007 If the symbol @var{name} is not defined within the file being assembled, all
7008 references to @var{name} will be changed to @var{name2@@nodename}. If no
7009 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
7012 Another usage of the @code{.symver} directive is:
7014 .symver @var{name}, @var{name2@@@@nodename}
7016 In this case, the symbol @var{name} must exist and be defined within
7017 the file being assembled. It is similar to @var{name2@@nodename}. The
7018 difference is @var{name2@@@@nodename} will also be used to resolve
7019 references to @var{name2} by the linker.
7021 The third usage of the @code{.symver} directive is:
7023 .symver @var{name}, @var{name2@@@@@@nodename}
7025 When @var{name} is not defined within the
7026 file being assembled, it is treated as @var{name2@@nodename}. When
7027 @var{name} is defined within the file being assembled, the symbol
7028 name, @var{name}, will be changed to @var{name2@@@@nodename}.
7033 @section @code{.tag @var{structname}}
7035 @cindex COFF structure debugging
7036 @cindex structure debugging, COFF
7037 @cindex @code{tag} directive
7038 This directive is generated by compilers to include auxiliary debugging
7039 information in the symbol table. It is only permitted inside
7040 @code{.def}/@code{.endef} pairs. Tags are used to link structure
7041 definitions in the symbol table with instances of those structures.
7044 @samp{.tag} is only used when generating COFF format output; when
7045 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
7051 @section @code{.text @var{subsection}}
7053 @cindex @code{text} directive
7054 Tells @command{@value{AS}} to assemble the following statements onto the end of
7055 the text subsection numbered @var{subsection}, which is an absolute
7056 expression. If @var{subsection} is omitted, subsection number zero
7060 @section @code{.title "@var{heading}"}
7062 @cindex @code{title} directive
7063 @cindex listing control: title line
7064 Use @var{heading} as the title (second line, immediately after the
7065 source file name and pagenumber) when generating assembly listings.
7067 This directive affects subsequent pages, as well as the current page if
7068 it appears within ten lines of the top of a page.
7072 @section @code{.type}
7074 This directive is used to set the type of a symbol.
7078 @c only print the extra heading if both COFF and ELF are set
7079 @subheading COFF Version
7082 @cindex COFF symbol type
7083 @cindex symbol type, COFF
7084 @cindex @code{type} directive (COFF version)
7085 For COFF targets, this directive is permitted only within
7086 @code{.def}/@code{.endef} pairs. It is used like this:
7092 This records the integer @var{int} as the type attribute of a symbol table
7096 @samp{.type} is associated only with COFF format output; when
7097 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
7098 directive but ignores it.
7104 @c only print the extra heading if both COFF and ELF are set
7105 @subheading ELF Version
7108 @cindex ELF symbol type
7109 @cindex symbol type, ELF
7110 @cindex @code{type} directive (ELF version)
7111 For ELF targets, the @code{.type} directive is used like this:
7114 .type @var{name} , @var{type description}
7117 This sets the type of symbol @var{name} to be either a
7118 function symbol or an object symbol. There are five different syntaxes
7119 supported for the @var{type description} field, in order to provide
7120 compatibility with various other assemblers.
7122 Because some of the characters used in these syntaxes (such as @samp{@@} and
7123 @samp{#}) are comment characters for some architectures, some of the syntaxes
7124 below do not work on all architectures. The first variant will be accepted by
7125 the GNU assembler on all architectures so that variant should be used for
7126 maximum portability, if you do not need to assemble your code with other
7129 The syntaxes supported are:
7132 .type <name> STT_<TYPE_IN_UPPER_CASE>
7133 .type <name>,#<type>
7134 .type <name>,@@<type>
7135 .type <name>,%<type>
7136 .type <name>,"<type>"
7139 The types supported are:
7144 Mark the symbol as being a function name.
7147 @itemx gnu_indirect_function
7148 Mark the symbol as an indirect function when evaluated during reloc
7149 processing. (This is only supported on assemblers targeting GNU systems).
7153 Mark the symbol as being a data object.
7157 Mark the symbol as being a thread-local data object.
7161 Mark the symbol as being a common data object.
7165 Does not mark the symbol in any way. It is supported just for completeness.
7167 @item gnu_unique_object
7168 Marks the symbol as being a globally unique data object. The dynamic linker
7169 will make sure that in the entire process there is just one symbol with this
7170 name and type in use. (This is only supported on assemblers targeting GNU
7175 Note: Some targets support extra types in addition to those listed above.
7181 @section @code{.uleb128 @var{expressions}}
7183 @cindex @code{uleb128} directive
7184 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
7185 compact, variable length representation of numbers used by the DWARF
7186 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
7190 @section @code{.val @var{addr}}
7192 @cindex @code{val} directive
7193 @cindex COFF value attribute
7194 @cindex value attribute, COFF
7195 This directive, permitted only within @code{.def}/@code{.endef} pairs,
7196 records the address @var{addr} as the value attribute of a symbol table
7200 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
7201 configured for @code{b.out}, it accepts this directive but ignores it.
7207 @section @code{.version "@var{string}"}
7209 @cindex @code{version} directive
7210 This directive creates a @code{.note} section and places into it an ELF
7211 formatted note of type NT_VERSION. The note's name is set to @code{string}.
7216 @section @code{.vtable_entry @var{table}, @var{offset}}
7218 @cindex @code{vtable_entry} directive
7219 This directive finds or creates a symbol @code{table} and creates a
7220 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
7223 @section @code{.vtable_inherit @var{child}, @var{parent}}
7225 @cindex @code{vtable_inherit} directive
7226 This directive finds the symbol @code{child} and finds or creates the symbol
7227 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
7228 parent whose addend is the value of the child symbol. As a special case the
7229 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
7233 @section @code{.warning "@var{string}"}
7234 @cindex warning directive
7235 Similar to the directive @code{.error}
7236 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
7239 @section @code{.weak @var{names}}
7241 @cindex @code{weak} directive
7242 This directive sets the weak attribute on the comma separated list of symbol
7243 @code{names}. If the symbols do not already exist, they will be created.
7245 On COFF targets other than PE, weak symbols are a GNU extension. This
7246 directive sets the weak attribute on the comma separated list of symbol
7247 @code{names}. If the symbols do not already exist, they will be created.
7249 On the PE target, weak symbols are supported natively as weak aliases.
7250 When a weak symbol is created that is not an alias, GAS creates an
7251 alternate symbol to hold the default value.
7254 @section @code{.weakref @var{alias}, @var{target}}
7256 @cindex @code{weakref} directive
7257 This directive creates an alias to the target symbol that enables the symbol to
7258 be referenced with weak-symbol semantics, but without actually making it weak.
7259 If direct references or definitions of the symbol are present, then the symbol
7260 will not be weak, but if all references to it are through weak references, the
7261 symbol will be marked as weak in the symbol table.
7263 The effect is equivalent to moving all references to the alias to a separate
7264 assembly source file, renaming the alias to the symbol in it, declaring the
7265 symbol as weak there, and running a reloadable link to merge the object files
7266 resulting from the assembly of the new source file and the old source file that
7267 had the references to the alias removed.
7269 The alias itself never makes to the symbol table, and is entirely handled
7270 within the assembler.
7273 @section @code{.word @var{expressions}}
7275 @cindex @code{word} directive
7276 This directive expects zero or more @var{expressions}, of any section,
7277 separated by commas.
7280 For each expression, @command{@value{AS}} emits a 32-bit number.
7283 For each expression, @command{@value{AS}} emits a 16-bit number.
7288 The size of the number emitted, and its byte order,
7289 depend on what target computer the assembly is for.
7292 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
7293 @c happen---32-bit addressability, period; no long/short jumps.
7294 @ifset DIFF-TBL-KLUGE
7295 @cindex difference tables altered
7296 @cindex altered difference tables
7298 @emph{Warning: Special Treatment to support Compilers}
7302 Machines with a 32-bit address space, but that do less than 32-bit
7303 addressing, require the following special treatment. If the machine of
7304 interest to you does 32-bit addressing (or doesn't require it;
7305 @pxref{Machine Dependencies}), you can ignore this issue.
7308 In order to assemble compiler output into something that works,
7309 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
7310 Directives of the form @samp{.word sym1-sym2} are often emitted by
7311 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
7312 directive of the form @samp{.word sym1-sym2}, and the difference between
7313 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
7314 creates a @dfn{secondary jump table}, immediately before the next label.
7315 This secondary jump table is preceded by a short-jump to the
7316 first byte after the secondary table. This short-jump prevents the flow
7317 of control from accidentally falling into the new table. Inside the
7318 table is a long-jump to @code{sym2}. The original @samp{.word}
7319 contains @code{sym1} minus the address of the long-jump to
7322 If there were several occurrences of @samp{.word sym1-sym2} before the
7323 secondary jump table, all of them are adjusted. If there was a
7324 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
7325 long-jump to @code{sym4} is included in the secondary jump table,
7326 and the @code{.word} directives are adjusted to contain @code{sym3}
7327 minus the address of the long-jump to @code{sym4}; and so on, for as many
7328 entries in the original jump table as necessary.
7331 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
7332 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
7333 assembly language programmers.
7336 @c end DIFF-TBL-KLUGE
7338 @ifclear no-space-dir
7340 @section @code{.zero @var{size}}
7342 @cindex @code{zero} directive
7343 @cindex filling memory with zero bytes
7344 This directive emits @var{size} 0-valued bytes. @var{size} must be an absolute
7345 expression. This directive is actually an alias for the @samp{.skip} directive
7346 so in can take an optional second argument of the value to store in the bytes
7347 instead of zero. Using @samp{.zero} in this way would be confusing however.
7352 @section @code{.2byte @var{expression} [, @var{expression}]*}
7353 @cindex @code{2byte} directive
7354 @cindex two-byte integer
7355 @cindex integer, 2-byte
7357 This directive expects zero or more expressions, separated by commas. If there
7358 are no expressions then the directive does nothing. Otherwise each expression
7359 is evaluated in turn and placed in the next two bytes of the current output
7360 section, using the endian model of the target. If an expression will not fit
7361 in two bytes, a warning message is displayed and the least significant two
7362 bytes of the expression's value are used. If an expression cannot be evaluated
7363 at assembly time then relocations will be generated in order to compute the
7366 This directive does not apply any alignment before or after inserting the
7367 values. As a result of this, if relocations are generated, they may be
7368 different from those used for inserting values with a guaranteed alignment.
7370 This directive is only available for ELF targets,
7373 @section @code{.4byte @var{expression} [, @var{expression}]*}
7374 @cindex @code{4byte} directive
7375 @cindex four-byte integer
7376 @cindex integer, 4-byte
7378 Like the @option{.2byte} directive, except that it inserts unaligned, four byte
7379 long values into the output.
7382 @section @code{.8byte @var{expression} [, @var{expression}]*}
7383 @cindex @code{8byte} directive
7384 @cindex eight-byte integer
7385 @cindex integer, 8-byte
7387 Like the @option{.8byte} directive, except that it inserts unaligned, eight
7388 byte long bignum values into the output.
7393 @section Deprecated Directives
7395 @cindex deprecated directives
7396 @cindex obsolescent directives
7397 One day these directives won't work.
7398 They are included for compatibility with older assemblers.
7405 @node Object Attributes
7406 @chapter Object Attributes
7407 @cindex object attributes
7409 @command{@value{AS}} assembles source files written for a specific architecture
7410 into object files for that architecture. But not all object files are alike.
7411 Many architectures support incompatible variations. For instance, floating
7412 point arguments might be passed in floating point registers if the object file
7413 requires hardware floating point support---or floating point arguments might be
7414 passed in integer registers if the object file supports processors with no
7415 hardware floating point unit. Or, if two objects are built for different
7416 generations of the same architecture, the combination may require the
7417 newer generation at run-time.
7419 This information is useful during and after linking. At link time,
7420 @command{@value{LD}} can warn about incompatible object files. After link
7421 time, tools like @command{gdb} can use it to process the linked file
7424 Compatibility information is recorded as a series of object attributes. Each
7425 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
7426 string, and indicates who sets the meaning of the tag. The tag is an integer,
7427 and indicates what property the attribute describes. The value may be a string
7428 or an integer, and indicates how the property affects this object. Missing
7429 attributes are the same as attributes with a zero value or empty string value.
7431 Object attributes were developed as part of the ABI for the ARM Architecture.
7432 The file format is documented in @cite{ELF for the ARM Architecture}.
7435 * GNU Object Attributes:: @sc{gnu} Object Attributes
7436 * Defining New Object Attributes:: Defining New Object Attributes
7439 @node GNU Object Attributes
7440 @section @sc{gnu} Object Attributes
7442 The @code{.gnu_attribute} directive records an object attribute
7443 with vendor @samp{gnu}.
7445 Except for @samp{Tag_compatibility}, which has both an integer and a string for
7446 its value, @sc{gnu} attributes have a string value if the tag number is odd and
7447 an integer value if the tag number is even. The second bit (@code{@var{tag} &
7448 2} is set for architecture-independent attributes and clear for
7449 architecture-dependent ones.
7451 @subsection Common @sc{gnu} attributes
7453 These attributes are valid on all architectures.
7456 @item Tag_compatibility (32)
7457 The compatibility attribute takes an integer flag value and a vendor name. If
7458 the flag value is 0, the file is compatible with other toolchains. If it is 1,
7459 then the file is only compatible with the named toolchain. If it is greater
7460 than 1, the file can only be processed by other toolchains under some private
7461 arrangement indicated by the flag value and the vendor name.
7464 @subsection MIPS Attributes
7467 @item Tag_GNU_MIPS_ABI_FP (4)
7468 The floating-point ABI used by this object file. The value will be:
7472 0 for files not affected by the floating-point ABI.
7474 1 for files using the hardware floating-point ABI with a standard
7475 double-precision FPU.
7477 2 for files using the hardware floating-point ABI with a single-precision FPU.
7479 3 for files using the software floating-point ABI.
7481 4 for files using the deprecated hardware floating-point ABI which used 64-bit
7482 floating-point registers, 32-bit general-purpose registers and increased the
7483 number of callee-saved floating-point registers.
7485 5 for files using the hardware floating-point ABI with a double-precision FPU
7486 with either 32-bit or 64-bit floating-point registers and 32-bit
7487 general-purpose registers.
7489 6 for files using the hardware floating-point ABI with 64-bit floating-point
7490 registers and 32-bit general-purpose registers.
7492 7 for files using the hardware floating-point ABI with 64-bit floating-point
7493 registers, 32-bit general-purpose registers and a rule that forbids the
7494 direct use of odd-numbered single-precision floating-point registers.
7498 @subsection PowerPC Attributes
7501 @item Tag_GNU_Power_ABI_FP (4)
7502 The floating-point ABI used by this object file. The value will be:
7506 0 for files not affected by the floating-point ABI.
7508 1 for files using double-precision hardware floating-point ABI.
7510 2 for files using the software floating-point ABI.
7512 3 for files using single-precision hardware floating-point ABI.
7515 @item Tag_GNU_Power_ABI_Vector (8)
7516 The vector ABI used by this object file. The value will be:
7520 0 for files not affected by the vector ABI.
7522 1 for files using general purpose registers to pass vectors.
7524 2 for files using AltiVec registers to pass vectors.
7526 3 for files using SPE registers to pass vectors.
7530 @subsection IBM z Systems Attributes
7533 @item Tag_GNU_S390_ABI_Vector (8)
7534 The vector ABI used by this object file. The value will be:
7538 0 for files not affected by the vector ABI.
7540 1 for files using software vector ABI.
7542 2 for files using hardware vector ABI.
7546 @node Defining New Object Attributes
7547 @section Defining New Object Attributes
7549 If you want to define a new @sc{gnu} object attribute, here are the places you
7550 will need to modify. New attributes should be discussed on the @samp{binutils}
7555 This manual, which is the official register of attributes.
7557 The header for your architecture @file{include/elf}, to define the tag.
7559 The @file{bfd} support file for your architecture, to merge the attribute
7560 and issue any appropriate link warnings.
7562 Test cases in @file{ld/testsuite} for merging and link warnings.
7564 @file{binutils/readelf.c} to display your attribute.
7566 GCC, if you want the compiler to mark the attribute automatically.
7572 @node Machine Dependencies
7573 @chapter Machine Dependent Features
7575 @cindex machine dependencies
7576 The machine instruction sets are (almost by definition) different on
7577 each machine where @command{@value{AS}} runs. Floating point representations
7578 vary as well, and @command{@value{AS}} often supports a few additional
7579 directives or command-line options for compatibility with other
7580 assemblers on a particular platform. Finally, some versions of
7581 @command{@value{AS}} support special pseudo-instructions for branch
7584 This chapter discusses most of these differences, though it does not
7585 include details on any machine's instruction set. For details on that
7586 subject, see the hardware manufacturer's manual.
7590 * AArch64-Dependent:: AArch64 Dependent Features
7593 * Alpha-Dependent:: Alpha Dependent Features
7596 * ARC-Dependent:: ARC Dependent Features
7599 * ARM-Dependent:: ARM Dependent Features
7602 * AVR-Dependent:: AVR Dependent Features
7605 * Blackfin-Dependent:: Blackfin Dependent Features
7608 * CR16-Dependent:: CR16 Dependent Features
7611 * CRIS-Dependent:: CRIS Dependent Features
7614 * D10V-Dependent:: D10V Dependent Features
7617 * D30V-Dependent:: D30V Dependent Features
7620 * Epiphany-Dependent:: EPIPHANY Dependent Features
7623 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7626 * HPPA-Dependent:: HPPA Dependent Features
7629 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
7632 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7635 * i860-Dependent:: Intel 80860 Dependent Features
7638 * i960-Dependent:: Intel 80960 Dependent Features
7641 * IA-64-Dependent:: Intel IA-64 Dependent Features
7644 * IP2K-Dependent:: IP2K Dependent Features
7647 * LM32-Dependent:: LM32 Dependent Features
7650 * M32C-Dependent:: M32C Dependent Features
7653 * M32R-Dependent:: M32R Dependent Features
7656 * M68K-Dependent:: M680x0 Dependent Features
7659 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7662 * Meta-Dependent :: Meta Dependent Features
7665 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7668 * MIPS-Dependent:: MIPS Dependent Features
7671 * MMIX-Dependent:: MMIX Dependent Features
7674 * MSP430-Dependent:: MSP430 Dependent Features
7677 * NDS32-Dependent:: Andes NDS32 Dependent Features
7680 * NiosII-Dependent:: Altera Nios II Dependent Features
7683 * NS32K-Dependent:: NS32K Dependent Features
7686 * PDP-11-Dependent:: PDP-11 Dependent Features
7689 * PJ-Dependent:: picoJava Dependent Features
7692 * PPC-Dependent:: PowerPC Dependent Features
7695 * PRU-Dependent:: PRU Dependent Features
7698 * RL78-Dependent:: RL78 Dependent Features
7701 * RISC-V-Dependent:: RISC-V Dependent Features
7704 * RX-Dependent:: RX Dependent Features
7707 * S/390-Dependent:: IBM S/390 Dependent Features
7710 * SCORE-Dependent:: SCORE Dependent Features
7713 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7714 * SH64-Dependent:: SuperH SH64 Dependent Features
7717 * Sparc-Dependent:: SPARC Dependent Features
7720 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7723 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7726 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7729 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7732 * V850-Dependent:: V850 Dependent Features
7735 * Vax-Dependent:: VAX Dependent Features
7738 * Visium-Dependent:: Visium Dependent Features
7741 * WebAssembly-Dependent:: WebAssembly Dependent Features
7744 * XGATE-Dependent:: XGATE Dependent Features
7747 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7750 * Xtensa-Dependent:: Xtensa Dependent Features
7753 * Z80-Dependent:: Z80 Dependent Features
7756 * Z8000-Dependent:: Z8000 Dependent Features
7763 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7764 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7765 @c peculiarity: to preserve cross-references, there must be a node called
7766 @c "Machine Dependencies". Hence the conditional nodenames in each
7767 @c major node below. Node defaulting in makeinfo requires adjacency of
7768 @c node and sectioning commands; hence the repetition of @chapter BLAH
7769 @c in both conditional blocks.
7772 @include c-aarch64.texi
7776 @include c-alpha.texi
7792 @include c-bfin.texi
7796 @include c-cr16.texi
7800 @include c-cris.texi
7805 @node Machine Dependencies
7806 @chapter Machine Dependent Features
7808 The machine instruction sets are different on each Renesas chip family,
7809 and there are also some syntax differences among the families. This
7810 chapter describes the specific @command{@value{AS}} features for each
7814 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7815 * SH-Dependent:: Renesas SH Dependent Features
7822 @include c-d10v.texi
7826 @include c-d30v.texi
7830 @include c-epiphany.texi
7834 @include c-h8300.texi
7838 @include c-hppa.texi
7842 @include c-i370.texi
7846 @include c-i386.texi
7850 @include c-i860.texi
7854 @include c-i960.texi
7858 @include c-ia64.texi
7862 @include c-ip2k.texi
7866 @include c-lm32.texi
7870 @include c-m32c.texi
7874 @include c-m32r.texi
7878 @include c-m68k.texi
7882 @include c-m68hc11.texi
7886 @include c-metag.texi
7890 @include c-microblaze.texi
7894 @include c-mips.texi
7898 @include c-mmix.texi
7902 @include c-msp430.texi
7906 @include c-nds32.texi
7910 @include c-nios2.texi
7914 @include c-ns32k.texi
7918 @include c-pdp11.texi
7934 @include c-rl78.texi
7938 @include c-riscv.texi
7946 @include c-s390.texi
7950 @include c-score.texi
7955 @include c-sh64.texi
7959 @include c-sparc.texi
7963 @include c-tic54x.texi
7967 @include c-tic6x.texi
7971 @include c-tilegx.texi
7975 @include c-tilepro.texi
7979 @include c-v850.texi
7987 @include c-visium.texi
7991 @include c-wasm32.texi
7995 @include c-xgate.texi
7999 @include c-xstormy16.texi
8003 @include c-xtensa.texi
8015 @c reverse effect of @down at top of generic Machine-Dep chapter
8019 @node Reporting Bugs
8020 @chapter Reporting Bugs
8021 @cindex bugs in assembler
8022 @cindex reporting bugs in assembler
8024 Your bug reports play an essential role in making @command{@value{AS}} reliable.
8026 Reporting a bug may help you by bringing a solution to your problem, or it may
8027 not. But in any case the principal function of a bug report is to help the
8028 entire community by making the next version of @command{@value{AS}} work better.
8029 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
8031 In order for a bug report to serve its purpose, you must include the
8032 information that enables us to fix the bug.
8035 * Bug Criteria:: Have you found a bug?
8036 * Bug Reporting:: How to report bugs
8040 @section Have You Found a Bug?
8041 @cindex bug criteria
8043 If you are not sure whether you have found a bug, here are some guidelines:
8046 @cindex fatal signal
8047 @cindex assembler crash
8048 @cindex crash of assembler
8050 If the assembler gets a fatal signal, for any input whatever, that is a
8051 @command{@value{AS}} bug. Reliable assemblers never crash.
8053 @cindex error on valid input
8055 If @command{@value{AS}} produces an error message for valid input, that is a bug.
8057 @cindex invalid input
8059 If @command{@value{AS}} does not produce an error message for invalid input, that
8060 is a bug. However, you should note that your idea of ``invalid input'' might
8061 be our idea of ``an extension'' or ``support for traditional practice''.
8064 If you are an experienced user of assemblers, your suggestions for improvement
8065 of @command{@value{AS}} are welcome in any case.
8069 @section How to Report Bugs
8071 @cindex assembler bugs, reporting
8073 A number of companies and individuals offer support for @sc{gnu} products. If
8074 you obtained @command{@value{AS}} from a support organization, we recommend you
8075 contact that organization first.
8077 You can find contact information for many support companies and
8078 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
8082 In any event, we also recommend that you send bug reports for @command{@value{AS}}
8086 The fundamental principle of reporting bugs usefully is this:
8087 @strong{report all the facts}. If you are not sure whether to state a
8088 fact or leave it out, state it!
8090 Often people omit facts because they think they know what causes the problem
8091 and assume that some details do not matter. Thus, you might assume that the
8092 name of a symbol you use in an example does not matter. Well, probably it does
8093 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
8094 happens to fetch from the location where that name is stored in memory;
8095 perhaps, if the name were different, the contents of that location would fool
8096 the assembler into doing the right thing despite the bug. Play it safe and
8097 give a specific, complete example. That is the easiest thing for you to do,
8098 and the most helpful.
8100 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
8101 it is new to us. Therefore, always write your bug reports on the assumption
8102 that the bug has not been reported previously.
8104 Sometimes people give a few sketchy facts and ask, ``Does this ring a
8105 bell?'' This cannot help us fix a bug, so it is basically useless. We
8106 respond by asking for enough details to enable us to investigate.
8107 You might as well expedite matters by sending them to begin with.
8109 To enable us to fix the bug, you should include all these things:
8113 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
8114 it with the @samp{--version} argument.
8116 Without this, we will not know whether there is any point in looking for
8117 the bug in the current version of @command{@value{AS}}.
8120 Any patches you may have applied to the @command{@value{AS}} source.
8123 The type of machine you are using, and the operating system name and
8127 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
8131 The command arguments you gave the assembler to assemble your example and
8132 observe the bug. To guarantee you will not omit something important, list them
8133 all. A copy of the Makefile (or the output from make) is sufficient.
8135 If we were to try to guess the arguments, we would probably guess wrong
8136 and then we might not encounter the bug.
8139 A complete input file that will reproduce the bug. If the bug is observed when
8140 the assembler is invoked via a compiler, send the assembler source, not the
8141 high level language source. Most compilers will produce the assembler source
8142 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
8143 the options @samp{-v --save-temps}; this will save the assembler source in a
8144 file with an extension of @file{.s}, and also show you exactly how
8145 @command{@value{AS}} is being run.
8148 A description of what behavior you observe that you believe is
8149 incorrect. For example, ``It gets a fatal signal.''
8151 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
8152 will certainly notice it. But if the bug is incorrect output, we might not
8153 notice unless it is glaringly wrong. You might as well not give us a chance to
8156 Even if the problem you experience is a fatal signal, you should still say so
8157 explicitly. Suppose something strange is going on, such as, your copy of
8158 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
8159 library on your system. (This has happened!) Your copy might crash and ours
8160 would not. If you told us to expect a crash, then when ours fails to crash, we
8161 would know that the bug was not happening for us. If you had not told us to
8162 expect a crash, then we would not be able to draw any conclusion from our
8166 If you wish to suggest changes to the @command{@value{AS}} source, send us context
8167 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
8168 option. Always send diffs from the old file to the new file. If you even
8169 discuss something in the @command{@value{AS}} source, refer to it by context, not
8172 The line numbers in our development sources will not match those in your
8173 sources. Your line numbers would convey no useful information to us.
8176 Here are some things that are not necessary:
8180 A description of the envelope of the bug.
8182 Often people who encounter a bug spend a lot of time investigating
8183 which changes to the input file will make the bug go away and which
8184 changes will not affect it.
8186 This is often time consuming and not very useful, because the way we
8187 will find the bug is by running a single example under the debugger
8188 with breakpoints, not by pure deduction from a series of examples.
8189 We recommend that you save your time for something else.
8191 Of course, if you can find a simpler example to report @emph{instead}
8192 of the original one, that is a convenience for us. Errors in the
8193 output will be easier to spot, running under the debugger will take
8194 less time, and so on.
8196 However, simplification is not vital; if you do not want to do this,
8197 report the bug anyway and send us the entire test case you used.
8200 A patch for the bug.
8202 A patch for the bug does help us if it is a good one. But do not omit
8203 the necessary information, such as the test case, on the assumption that
8204 a patch is all we need. We might see problems with your patch and decide
8205 to fix the problem another way, or we might not understand it at all.
8207 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
8208 construct an example that will make the program follow a certain path through
8209 the code. If you do not send us the example, we will not be able to construct
8210 one, so we will not be able to verify that the bug is fixed.
8212 And if we cannot understand what bug you are trying to fix, or why your
8213 patch should be an improvement, we will not install it. A test case will
8214 help us to understand.
8217 A guess about what the bug is or what it depends on.
8219 Such guesses are usually wrong. Even we cannot guess right about such
8220 things without first using the debugger to find the facts.
8223 @node Acknowledgements
8224 @chapter Acknowledgements
8226 If you have contributed to GAS and your name isn't listed here,
8227 it is not meant as a slight. We just don't know about it. Send mail to the
8228 maintainer, and we'll correct the situation. Currently
8230 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
8232 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
8235 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
8236 information and the 68k series machines, most of the preprocessing pass, and
8237 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
8239 K. Richard Pixley maintained GAS for a while, adding various enhancements and
8240 many bug fixes, including merging support for several processors, breaking GAS
8241 up to handle multiple object file format back ends (including heavy rewrite,
8242 testing, an integration of the coff and b.out back ends), adding configuration
8243 including heavy testing and verification of cross assemblers and file splits
8244 and renaming, converted GAS to strictly ANSI C including full prototypes, added
8245 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
8246 port (including considerable amounts of reverse engineering), a SPARC opcode
8247 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
8248 assertions and made them work, much other reorganization, cleanup, and lint.
8250 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
8251 in format-specific I/O modules.
8253 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
8254 has done much work with it since.
8256 The Intel 80386 machine description was written by Eliot Dresselhaus.
8258 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
8260 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
8261 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
8263 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
8264 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
8265 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
8266 support a.out format.
8268 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
8269 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
8270 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
8271 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
8274 John Gilmore built the AMD 29000 support, added @code{.include} support, and
8275 simplified the configuration of which versions accept which directives. He
8276 updated the 68k machine description so that Motorola's opcodes always produced
8277 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
8278 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
8279 cross-compilation support, and one bug in relaxation that took a week and
8280 required the proverbial one-bit fix.
8282 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
8283 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
8284 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
8285 PowerPC assembler, and made a few other minor patches.
8287 Steve Chamberlain made GAS able to generate listings.
8289 Hewlett-Packard contributed support for the HP9000/300.
8291 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
8292 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
8293 formats). This work was supported by both the Center for Software Science at
8294 the University of Utah and Cygnus Support.
8296 Support for ELF format files has been worked on by Mark Eichin of Cygnus
8297 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
8298 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
8299 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
8300 and some initial 64-bit support).
8302 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
8304 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
8305 support for openVMS/Alpha.
8307 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
8310 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
8311 Inc.@: added support for Xtensa processors.
8313 Several engineers at Cygnus Support have also provided many small bug fixes and
8314 configuration enhancements.
8316 Jon Beniston added support for the Lattice Mico32 architecture.
8318 Many others have contributed large or small bugfixes and enhancements. If
8319 you have contributed significant work and are not mentioned on this list, and
8320 want to be, let us know. Some of the history has been lost; we are not
8321 intentionally leaving anyone out.
8323 @node GNU Free Documentation License
8324 @appendix GNU Free Documentation License
8328 @unnumbered AS Index