1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright (C) 1991-2015 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{} 1991-2015 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{} 1991-2015 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}] [@b{-o}
238 @var{objfile}] [@b{-R}] [@b{--reduce-memory-overheads}] [@b{--statistics}]
239 [@b{-v}] [@b{-version}] [@b{--version}] [@b{-W}] [@b{--warn}]
240 [@b{--fatal-warnings}] [@b{-w}] [@b{-x}] [@b{-Z}] [@b{@@@var{FILE}}]
241 [@b{--sectname-subst}] [@b{--size-check=[error|warning]}]
242 [@b{--target-help}] [@var{target-options}]
243 [@b{--}|@var{files} @dots{}]
245 @c Target dependent options are listed below. Keep the list sorted.
246 @c Add an empty line for separation.
249 @emph{Target AArch64 options:}
251 [@b{-mabi}=@var{ABI}]
255 @emph{Target Alpha options:}
257 [@b{-mdebug} | @b{-no-mdebug}]
258 [@b{-replace} | @b{-noreplace}]
259 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
260 [@b{-F}] [@b{-32addr}]
264 @emph{Target ARC options:}
270 @emph{Target ARM options:}
271 @c Don't document the deprecated options
272 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
273 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
274 [@b{-mfpu}=@var{floating-point-format}]
275 [@b{-mfloat-abi}=@var{abi}]
276 [@b{-meabi}=@var{ver}]
279 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
280 @b{-mapcs-reentrant}]
281 [@b{-mthumb-interwork}] [@b{-k}]
285 @emph{Target Blackfin options:}
286 [@b{-mcpu}=@var{processor}[-@var{sirevision}]]
293 @emph{Target CRIS options:}
294 [@b{--underscore} | @b{--no-underscore}]
296 [@b{--emulation=criself} | @b{--emulation=crisaout}]
297 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
298 @c Deprecated -- deliberately not documented.
303 @emph{Target D10V options:}
308 @emph{Target D30V options:}
309 [@b{-O}|@b{-n}|@b{-N}]
313 @emph{Target EPIPHANY options:}
314 [@b{-mepiphany}|@b{-mepiphany16}]
318 @emph{Target H8/300 options:}
322 @c HPPA has no machine-dependent assembler options (yet).
326 @emph{Target i386 options:}
327 [@b{--32}|@b{--x32}|@b{--64}] [@b{-n}]
328 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
332 @emph{Target i960 options:}
333 @c see md_parse_option in tc-i960.c
334 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
336 [@b{-b}] [@b{-no-relax}]
340 @emph{Target IA-64 options:}
341 [@b{-mconstant-gp}|@b{-mauto-pic}]
342 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
344 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
345 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
346 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
347 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
351 @emph{Target IP2K options:}
352 [@b{-mip2022}|@b{-mip2022ext}]
356 @emph{Target M32C options:}
357 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
361 @emph{Target M32R options:}
362 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
367 @emph{Target M680X0 options:}
368 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
372 @emph{Target M68HC11 options:}
373 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}|@b{-mm9s12x}|@b{-mm9s12xg}]
374 [@b{-mshort}|@b{-mlong}]
375 [@b{-mshort-double}|@b{-mlong-double}]
376 [@b{--force-long-branches}] [@b{--short-branches}]
377 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
378 [@b{--print-opcodes}] [@b{--generate-example}]
382 @emph{Target MCORE options:}
383 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
384 [@b{-mcpu=[210|340]}]
388 @emph{Target Meta options:}
389 [@b{-mcpu=@var{cpu}}] [@b{-mfpu=@var{cpu}}] [@b{-mdsp=@var{cpu}}]
392 @emph{Target MICROBLAZE options:}
393 @c MicroBlaze has no machine-dependent assembler options.
397 @emph{Target MIPS options:}
398 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
399 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
400 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
401 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
402 [@b{-mfp64}] [@b{-mgp64}] [@b{-mfpxx}]
403 [@b{-modd-spreg}] [@b{-mno-odd-spreg}]
404 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
405 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
406 [@b{-mips32r3}] [@b{-mips32r5}] [@b{-mips32r6}] [@b{-mips64}] [@b{-mips64r2}]
407 [@b{-mips64r3}] [@b{-mips64r5}] [@b{-mips64r6}]
408 [@b{-construct-floats}] [@b{-no-construct-floats}]
409 [@b{-mnan=@var{encoding}}]
410 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
411 [@b{-mips16}] [@b{-no-mips16}]
412 [@b{-mmicromips}] [@b{-mno-micromips}]
413 [@b{-msmartmips}] [@b{-mno-smartmips}]
414 [@b{-mips3d}] [@b{-no-mips3d}]
415 [@b{-mdmx}] [@b{-no-mdmx}]
416 [@b{-mdsp}] [@b{-mno-dsp}]
417 [@b{-mdspr2}] [@b{-mno-dspr2}]
418 [@b{-mmsa}] [@b{-mno-msa}]
419 [@b{-mxpa}] [@b{-mno-xpa}]
420 [@b{-mmt}] [@b{-mno-mt}]
421 [@b{-mmcu}] [@b{-mno-mcu}]
422 [@b{-minsn32}] [@b{-mno-insn32}]
423 [@b{-mfix7000}] [@b{-mno-fix7000}]
424 [@b{-mfix-rm7000}] [@b{-mno-fix-rm7000}]
425 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
426 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
427 [@b{-mdebug}] [@b{-no-mdebug}]
428 [@b{-mpdr}] [@b{-mno-pdr}]
432 @emph{Target MMIX options:}
433 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
434 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
435 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
436 [@b{--linker-allocated-gregs}]
440 @emph{Target Nios II options:}
441 [@b{-relax-all}] [@b{-relax-section}] [@b{-no-relax}]
446 @emph{Target NDS32 options:}
447 [@b{-EL}] [@b{-EB}] [@b{-O}] [@b{-Os}] [@b{-mcpu=@var{cpu}}]
448 [@b{-misa=@var{isa}}] [@b{-mabi=@var{abi}}] [@b{-mall-ext}]
449 [@b{-m[no-]16-bit}] [@b{-m[no-]perf-ext}] [@b{-m[no-]perf2-ext}]
450 [@b{-m[no-]string-ext}] [@b{-m[no-]dsp-ext}] [@b{-m[no-]mac}] [@b{-m[no-]div}]
451 [@b{-m[no-]audio-isa-ext}] [@b{-m[no-]fpu-sp-ext}] [@b{-m[no-]fpu-dp-ext}]
452 [@b{-m[no-]fpu-fma}] [@b{-mfpu-freg=@var{FREG}}] [@b{-mreduced-regs}]
453 [@b{-mfull-regs}] [@b{-m[no-]dx-regs}] [@b{-mpic}] [@b{-mno-relax}]
458 @emph{Target PDP11 options:}
459 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
460 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
461 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
465 @emph{Target picoJava options:}
470 @emph{Target PowerPC options:}
472 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|@b{-m403}|@b{-m405}|
473 @b{-m440}|@b{-m464}|@b{-m476}|@b{-m7400}|@b{-m7410}|@b{-m7450}|@b{-m7455}|@b{-m750cl}|@b{-mppc64}|
474 @b{-m620}|@b{-me500}|@b{-e500x2}|@b{-me500mc}|@b{-me500mc64}|@b{-me5500}|@b{-me6500}|@b{-mppc64bridge}|
475 @b{-mbooke}|@b{-mpower4}|@b{-mpwr4}|@b{-mpower5}|@b{-mpwr5}|@b{-mpwr5x}|@b{-mpower6}|@b{-mpwr6}|
476 @b{-mpower7}|@b{-mpwr7}|@b{-mpower8}|@b{-mpwr8}|@b{-ma2}|@b{-mcell}|@b{-mspe}|@b{-mtitan}|@b{-me300}|@b{-mcom}]
477 [@b{-many}] [@b{-maltivec}|@b{-mvsx}|@b{-mhtm}|@b{-mvle}]
478 [@b{-mregnames}|@b{-mno-regnames}]
479 [@b{-mrelocatable}|@b{-mrelocatable-lib}|@b{-K PIC}] [@b{-memb}]
480 [@b{-mlittle}|@b{-mlittle-endian}|@b{-le}|@b{-mbig}|@b{-mbig-endian}|@b{-be}]
481 [@b{-msolaris}|@b{-mno-solaris}]
482 [@b{-nops=@var{count}}]
486 @emph{Target RL78 options:}
488 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
492 @emph{Target RX options:}
493 [@b{-mlittle-endian}|@b{-mbig-endian}]
494 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
495 [@b{-muse-conventional-section-names}]
496 [@b{-msmall-data-limit}]
499 [@b{-mint-register=@var{number}}]
500 [@b{-mgcc-abi}|@b{-mrx-abi}]
504 @emph{Target s390 options:}
505 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
506 [@b{-mregnames}|@b{-mno-regnames}]
507 [@b{-mwarn-areg-zero}]
511 @emph{Target SCORE options:}
512 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
513 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
514 [@b{-march=score7}][@b{-march=score3}]
515 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
519 @emph{Target SPARC options:}
520 @c The order here is important. See c-sparc.texi.
521 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
522 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
523 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
528 @emph{Target TIC54X options:}
529 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
530 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
534 @emph{Target TIC6X options:}
535 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
536 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
537 [@b{-mpic}|@b{-mno-pic}]
541 @emph{Target TILE-Gx options:}
542 [@b{-m32}|@b{-m64}][@b{-EB}][@b{-EL}]
545 @c TILEPro has no machine-dependent assembler options
549 @emph{Target Visium options:}
550 [@b{-mtune=@var{arch}}]
554 @emph{Target Xtensa options:}
555 [@b{--[no-]text-section-literals}] [@b{--[no-]absolute-literals}]
556 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
557 [@b{--[no-]transform}]
558 [@b{--rename-section} @var{oldname}=@var{newname}]
559 [@b{--[no-]trampolines}]
563 @emph{Target Z80 options:}
564 [@b{-z80}] [@b{-r800}]
565 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
566 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
567 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
568 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
569 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
570 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
574 @c Z8000 has no machine-dependent assembler options
583 @include at-file.texi
586 Turn on listings, in any of a variety of ways:
590 omit false conditionals
593 omit debugging directives
596 include general information, like @value{AS} version and options passed
599 include high-level source
605 include macro expansions
608 omit forms processing
614 set the name of the listing file
617 You may combine these options; for example, use @samp{-aln} for assembly
618 listing without forms processing. The @samp{=file} option, if used, must be
619 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
622 Begin in alternate macro mode.
624 @xref{Altmacro,,@code{.altmacro}}.
627 @item --compress-debug-sections
628 Compress DWARF debug sections using zlib. The debug sections are renamed
629 to begin with @samp{.zdebug}, and the resulting object file may not be
630 compatible with older linkers and object file utilities. Note if compression
631 would make a given section @emph{larger} then it is not compressed or renamed.
634 @cindex @samp{--compress-debug-sections=} option
635 @item --compress-debug-sections=none
636 @itemx --compress-debug-sections=zlib
637 @itemx --compress-debug-sections=zlib-gnu
638 @itemx --compress-debug-sections=zlib-gabi
639 These options control how DWARF debug sections are compressed.
640 @option{--compress-debug-sections=none} is equivalent to
641 @option{--nocompress-debug-sections}.
642 @option{--compress-debug-sections=zlib} and
643 @option{--compress-debug-sections=zlib-gnu} are equivalent to
644 @option{--compress-debug-sections}.
645 @option{--compress-debug-sections=zlib-gabi} compresses
646 DWARF debug sections with SHF_COMPRESSED from the ELF ABI.
649 @item --nocompress-debug-sections
650 Do not compress DWARF debug sections. This is the default.
653 Ignored. This option is accepted for script compatibility with calls to
656 @item --debug-prefix-map @var{old}=@var{new}
657 When assembling files in directory @file{@var{old}}, record debugging
658 information describing them as in @file{@var{new}} instead.
660 @item --defsym @var{sym}=@var{value}
661 Define the symbol @var{sym} to be @var{value} before assembling the input file.
662 @var{value} must be an integer constant. As in C, a leading @samp{0x}
663 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
664 value. The value of the symbol can be overridden inside a source file via the
665 use of a @code{.set} pseudo-op.
668 ``fast''---skip whitespace and comment preprocessing (assume source is
673 Generate debugging information for each assembler source line using whichever
674 debug format is preferred by the target. This currently means either STABS,
678 Generate stabs debugging information for each assembler line. This
679 may help debugging assembler code, if the debugger can handle it.
682 Generate stabs debugging information for each assembler line, with GNU
683 extensions that probably only gdb can handle, and that could make other
684 debuggers crash or refuse to read your program. This
685 may help debugging assembler code. Currently the only GNU extension is
686 the location of the current working directory at assembling time.
689 Generate DWARF2 debugging information for each assembler line. This
690 may help debugging assembler code, if the debugger can handle it. Note---this
691 option is only supported by some targets, not all of them.
693 @item --gdwarf-sections
694 Instead of creating a .debug_line section, create a series of
695 .debug_line.@var{foo} sections where @var{foo} is the name of the
696 corresponding code section. For example a code section called @var{.text.func}
697 will have its dwarf line number information placed into a section called
698 @var{.debug_line.text.func}. If the code section is just called @var{.text}
699 then debug line section will still be called just @var{.debug_line} without any
702 @item --size-check=error
703 @itemx --size-check=warning
704 Issue an error or warning for invalid ELF .size directive.
707 Print a summary of the command line options and exit.
710 Print a summary of all target specific options and exit.
713 Add directory @var{dir} to the search list for @code{.include} directives.
716 Don't warn about signed overflow.
719 @ifclear DIFF-TBL-KLUGE
720 This option is accepted but has no effect on the @value{TARGET} family.
722 @ifset DIFF-TBL-KLUGE
723 Issue warnings when difference tables altered for long displacements.
728 Keep (in the symbol table) local symbols. These symbols start with
729 system-specific local label prefixes, typically @samp{.L} for ELF systems
730 or @samp{L} for traditional a.out systems.
735 @item --listing-lhs-width=@var{number}
736 Set the maximum width, in words, of the output data column for an assembler
737 listing to @var{number}.
739 @item --listing-lhs-width2=@var{number}
740 Set the maximum width, in words, of the output data column for continuation
741 lines in an assembler listing to @var{number}.
743 @item --listing-rhs-width=@var{number}
744 Set the maximum width of an input source line, as displayed in a listing, to
747 @item --listing-cont-lines=@var{number}
748 Set the maximum number of lines printed in a listing for a single line of input
751 @item -o @var{objfile}
752 Name the object-file output from @command{@value{AS}} @var{objfile}.
755 Fold the data section into the text section.
757 @kindex --hash-size=@var{number}
758 Set the default size of GAS's hash tables to a prime number close to
759 @var{number}. Increasing this value can reduce the length of time it takes the
760 assembler to perform its tasks, at the expense of increasing the assembler's
761 memory requirements. Similarly reducing this value can reduce the memory
762 requirements at the expense of speed.
764 @item --reduce-memory-overheads
765 This option reduces GAS's memory requirements, at the expense of making the
766 assembly processes slower. Currently this switch is a synonym for
767 @samp{--hash-size=4051}, but in the future it may have other effects as well.
770 @item --sectname-subst
771 Honor substitution sequences in section names.
773 @xref{Section Name Substitutions,,@code{.section @var{name}}}.
778 Print the maximum space (in bytes) and total time (in seconds) used by
781 @item --strip-local-absolute
782 Remove local absolute symbols from the outgoing symbol table.
786 Print the @command{as} version.
789 Print the @command{as} version and exit.
793 Suppress warning messages.
795 @item --fatal-warnings
796 Treat warnings as errors.
799 Don't suppress warning messages or treat them as errors.
808 Generate an object file even after errors.
810 @item -- | @var{files} @dots{}
811 Standard input, or source files to assemble.
819 @xref{AArch64 Options}, for the options available when @value{AS} is configured
820 for the 64-bit mode of the ARM Architecture (AArch64).
825 The following options are available when @value{AS} is configured for the
826 64-bit mode of the ARM Architecture (AArch64).
829 @include c-aarch64.texi
830 @c ended inside the included file
838 @xref{Alpha Options}, for the options available when @value{AS} is configured
839 for an Alpha processor.
844 The following options are available when @value{AS} is configured for an Alpha
848 @include c-alpha.texi
849 @c ended inside the included file
856 The following options are available when @value{AS} is configured for
861 This option selects the core processor variant.
863 Select either big-endian (-EB) or little-endian (-EL) output.
868 The following options are available when @value{AS} is configured for the ARM
872 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
873 Specify which ARM processor variant is the target.
874 @item -march=@var{architecture}[+@var{extension}@dots{}]
875 Specify which ARM architecture variant is used by the target.
876 @item -mfpu=@var{floating-point-format}
877 Select which Floating Point architecture is the target.
878 @item -mfloat-abi=@var{abi}
879 Select which floating point ABI is in use.
881 Enable Thumb only instruction decoding.
882 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
883 Select which procedure calling convention is in use.
885 Select either big-endian (-EB) or little-endian (-EL) output.
886 @item -mthumb-interwork
887 Specify that the code has been generated with interworking between Thumb and
890 Turns on CodeComposer Studio assembly syntax compatibility mode.
892 Specify that PIC code has been generated.
900 @xref{Blackfin Options}, for the options available when @value{AS} is
901 configured for the Blackfin processor family.
906 The following options are available when @value{AS} is configured for
907 the Blackfin processor family.
911 @c ended inside the included file
918 See the info pages for documentation of the CRIS-specific options.
922 The following options are available when @value{AS} is configured for
925 @cindex D10V optimization
926 @cindex optimization, D10V
928 Optimize output by parallelizing instructions.
933 The following options are available when @value{AS} is configured for a D30V
936 @cindex D30V optimization
937 @cindex optimization, D30V
939 Optimize output by parallelizing instructions.
943 Warn when nops are generated.
945 @cindex D30V nops after 32-bit multiply
947 Warn when a nop after a 32-bit multiply instruction is generated.
953 The following options are available when @value{AS} is configured for the
954 Adapteva EPIPHANY series.
957 @xref{Epiphany Options}, for the options available when @value{AS} is
958 configured for an Epiphany processor.
963 The following options are available when @value{AS} is configured for
964 an Epiphany processor.
967 @include c-epiphany.texi
968 @c ended inside the included file
976 @xref{H8/300 Options}, for the options available when @value{AS} is configured
977 for an H8/300 processor.
982 The following options are available when @value{AS} is configured for an H8/300
986 @include c-h8300.texi
987 @c ended inside the included file
995 @xref{i386-Options}, for the options available when @value{AS} is
996 configured for an i386 processor.
1000 @c man begin OPTIONS
1001 The following options are available when @value{AS} is configured for
1004 @c man begin INCLUDE
1005 @include c-i386.texi
1006 @c ended inside the included file
1011 @c man begin OPTIONS
1013 The following options are available when @value{AS} is configured for the
1014 Intel 80960 processor.
1017 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
1018 Specify which variant of the 960 architecture is the target.
1021 Add code to collect statistics about branches taken.
1024 Do not alter compare-and-branch instructions for long displacements;
1031 The following options are available when @value{AS} is configured for the
1037 Specifies that the extended IP2022 instructions are allowed.
1040 Restores the default behaviour, which restricts the permitted instructions to
1041 just the basic IP2022 ones.
1047 The following options are available when @value{AS} is configured for the
1048 Renesas M32C and M16C processors.
1053 Assemble M32C instructions.
1056 Assemble M16C instructions (the default).
1059 Enable support for link-time relaxations.
1062 Support H'00 style hex constants in addition to 0x00 style.
1068 The following options are available when @value{AS} is configured for the
1069 Renesas M32R (formerly Mitsubishi M32R) series.
1074 Specify which processor in the M32R family is the target. The default
1075 is normally the M32R, but this option changes it to the M32RX.
1077 @item --warn-explicit-parallel-conflicts or --Wp
1078 Produce warning messages when questionable parallel constructs are
1081 @item --no-warn-explicit-parallel-conflicts or --Wnp
1082 Do not produce warning messages when questionable parallel constructs are
1089 The following options are available when @value{AS} is configured for the
1090 Motorola 68000 series.
1095 Shorten references to undefined symbols, to one word instead of two.
1097 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1098 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1099 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1100 Specify what processor in the 68000 family is the target. The default
1101 is normally the 68020, but this can be changed at configuration time.
1103 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1104 The target machine does (or does not) have a floating-point coprocessor.
1105 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1106 the basic 68000 is not compatible with the 68881, a combination of the
1107 two can be specified, since it's possible to do emulation of the
1108 coprocessor instructions with the main processor.
1110 @item -m68851 | -mno-68851
1111 The target machine does (or does not) have a memory-management
1112 unit coprocessor. The default is to assume an MMU for 68020 and up.
1120 @xref{Nios II Options}, for the options available when @value{AS} is configured
1121 for an Altera Nios II processor.
1125 @c man begin OPTIONS
1126 The following options are available when @value{AS} is configured for an
1127 Altera Nios II processor.
1129 @c man begin INCLUDE
1130 @include c-nios2.texi
1131 @c ended inside the included file
1137 For details about the PDP-11 machine dependent features options,
1138 see @ref{PDP-11-Options}.
1141 @item -mpic | -mno-pic
1142 Generate position-independent (or position-dependent) code. The
1143 default is @option{-mpic}.
1146 @itemx -mall-extensions
1147 Enable all instruction set extensions. This is the default.
1149 @item -mno-extensions
1150 Disable all instruction set extensions.
1152 @item -m@var{extension} | -mno-@var{extension}
1153 Enable (or disable) a particular instruction set extension.
1156 Enable the instruction set extensions supported by a particular CPU, and
1157 disable all other extensions.
1159 @item -m@var{machine}
1160 Enable the instruction set extensions supported by a particular machine
1161 model, and disable all other extensions.
1167 The following options are available when @value{AS} is configured for
1168 a picoJava processor.
1172 @cindex PJ endianness
1173 @cindex endianness, PJ
1174 @cindex big endian output, PJ
1176 Generate ``big endian'' format output.
1178 @cindex little endian output, PJ
1180 Generate ``little endian'' format output.
1186 The following options are available when @value{AS} is configured for the
1187 Motorola 68HC11 or 68HC12 series.
1191 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1192 Specify what processor is the target. The default is
1193 defined by the configuration option when building the assembler.
1195 @item --xgate-ramoffset
1196 Instruct the linker to offset RAM addresses from S12X address space into
1197 XGATE address space.
1200 Specify to use the 16-bit integer ABI.
1203 Specify to use the 32-bit integer ABI.
1205 @item -mshort-double
1206 Specify to use the 32-bit double ABI.
1209 Specify to use the 64-bit double ABI.
1211 @item --force-long-branches
1212 Relative branches are turned into absolute ones. This concerns
1213 conditional branches, unconditional branches and branches to a
1216 @item -S | --short-branches
1217 Do not turn relative branches into absolute ones
1218 when the offset is out of range.
1220 @item --strict-direct-mode
1221 Do not turn the direct addressing mode into extended addressing mode
1222 when the instruction does not support direct addressing mode.
1224 @item --print-insn-syntax
1225 Print the syntax of instruction in case of error.
1227 @item --print-opcodes
1228 Print the list of instructions with syntax and then exit.
1230 @item --generate-example
1231 Print an example of instruction for each possible instruction and then exit.
1232 This option is only useful for testing @command{@value{AS}}.
1238 The following options are available when @command{@value{AS}} is configured
1239 for the SPARC architecture:
1242 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1243 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1244 Explicitly select a variant of the SPARC architecture.
1246 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1247 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1249 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1250 UltraSPARC extensions.
1252 @item -xarch=v8plus | -xarch=v8plusa
1253 For compatibility with the Solaris v9 assembler. These options are
1254 equivalent to -Av8plus and -Av8plusa, respectively.
1257 Warn when the assembler switches to another architecture.
1262 The following options are available when @value{AS} is configured for the 'c54x
1267 Enable extended addressing mode. All addresses and relocations will assume
1268 extended addressing (usually 23 bits).
1269 @item -mcpu=@var{CPU_VERSION}
1270 Sets the CPU version being compiled for.
1271 @item -merrors-to-file @var{FILENAME}
1272 Redirect error output to a file, for broken systems which don't support such
1273 behaviour in the shell.
1278 The following options are available when @value{AS} is configured for
1283 This option sets the largest size of an object that can be referenced
1284 implicitly with the @code{gp} register. It is only accepted for targets that
1285 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1287 @cindex MIPS endianness
1288 @cindex endianness, MIPS
1289 @cindex big endian output, MIPS
1291 Generate ``big endian'' format output.
1293 @cindex little endian output, MIPS
1295 Generate ``little endian'' format output.
1313 Generate code for a particular MIPS Instruction Set Architecture level.
1314 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1315 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1316 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1317 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1318 @samp{-mips32r5}, @samp{-mips32r6}, @samp{-mips64}, @samp{-mips64r2},
1319 @samp{-mips64r3}, @samp{-mips64r5}, and @samp{-mips64r6} correspond to generic
1320 MIPS V, MIPS32, MIPS32 Release 2, MIPS32 Release 3, MIPS32 Release 5, MIPS32
1321 Release 6, MIPS64, MIPS64 Release 2, MIPS64 Release 3, MIPS64 Release 5, and
1322 MIPS64 Release 6 ISA processors, respectively.
1324 @item -march=@var{cpu}
1325 Generate code for a particular MIPS CPU.
1327 @item -mtune=@var{cpu}
1328 Schedule and tune for a particular MIPS CPU.
1332 Cause nops to be inserted if the read of the destination register
1333 of an mfhi or mflo instruction occurs in the following two instructions.
1336 @itemx -mno-fix-rm7000
1337 Cause nops to be inserted if a dmult or dmultu instruction is
1338 followed by a load instruction.
1342 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1343 section instead of the standard ELF .stabs sections.
1347 Control generation of @code{.pdr} sections.
1351 The register sizes are normally inferred from the ISA and ABI, but these
1352 flags force a certain group of registers to be treated as 32 bits wide at
1353 all times. @samp{-mgp32} controls the size of general-purpose registers
1354 and @samp{-mfp32} controls the size of floating-point registers.
1358 The register sizes are normally inferred from the ISA and ABI, but these
1359 flags force a certain group of registers to be treated as 64 bits wide at
1360 all times. @samp{-mgp64} controls the size of general-purpose registers
1361 and @samp{-mfp64} controls the size of floating-point registers.
1364 The register sizes are normally inferred from the ISA and ABI, but using
1365 this flag in combination with @samp{-mabi=32} enables an ABI variant
1366 which will operate correctly with floating-point registers which are
1370 @itemx -mno-odd-spreg
1371 Enable use of floating-point operations on odd-numbered single-precision
1372 registers when supported by the ISA. @samp{-mfpxx} implies
1373 @samp{-mno-odd-spreg}, otherwise the default is @samp{-modd-spreg}.
1377 Generate code for the MIPS 16 processor. This is equivalent to putting
1378 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1379 turns off this option.
1382 @itemx -mno-micromips
1383 Generate code for the microMIPS processor. This is equivalent to putting
1384 @code{.set micromips} at the start of the assembly file. @samp{-mno-micromips}
1385 turns off this option. This is equivalent to putting @code{.set nomicromips}
1386 at the start of the assembly file.
1389 @itemx -mno-smartmips
1390 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1391 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1392 @samp{-mno-smartmips} turns off this option.
1396 Generate code for the MIPS-3D Application Specific Extension.
1397 This tells the assembler to accept MIPS-3D instructions.
1398 @samp{-no-mips3d} turns off this option.
1402 Generate code for the MDMX Application Specific Extension.
1403 This tells the assembler to accept MDMX instructions.
1404 @samp{-no-mdmx} turns off this option.
1408 Generate code for the DSP Release 1 Application Specific Extension.
1409 This tells the assembler to accept DSP Release 1 instructions.
1410 @samp{-mno-dsp} turns off this option.
1414 Generate code for the DSP Release 2 Application Specific Extension.
1415 This option implies -mdsp.
1416 This tells the assembler to accept DSP Release 2 instructions.
1417 @samp{-mno-dspr2} turns off this option.
1421 Generate code for the MIPS SIMD Architecture Extension.
1422 This tells the assembler to accept MSA instructions.
1423 @samp{-mno-msa} turns off this option.
1427 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1428 This tells the assembler to accept XPA instructions.
1429 @samp{-mno-xpa} turns off this option.
1433 Generate code for the MT Application Specific Extension.
1434 This tells the assembler to accept MT instructions.
1435 @samp{-mno-mt} turns off this option.
1439 Generate code for the MCU Application Specific Extension.
1440 This tells the assembler to accept MCU instructions.
1441 @samp{-mno-mcu} turns off this option.
1445 Only use 32-bit instruction encodings when generating code for the
1446 microMIPS processor. This option inhibits the use of any 16-bit
1447 instructions. This is equivalent to putting @code{.set insn32} at
1448 the start of the assembly file. @samp{-mno-insn32} turns off this
1449 option. This is equivalent to putting @code{.set noinsn32} at the
1450 start of the assembly file. By default @samp{-mno-insn32} is
1451 selected, allowing all instructions to be used.
1453 @item --construct-floats
1454 @itemx --no-construct-floats
1455 The @samp{--no-construct-floats} option disables the construction of
1456 double width floating point constants by loading the two halves of the
1457 value into the two single width floating point registers that make up
1458 the double width register. By default @samp{--construct-floats} is
1459 selected, allowing construction of these floating point constants.
1461 @item --relax-branch
1462 @itemx --no-relax-branch
1463 The @samp{--relax-branch} option enables the relaxation of out-of-range
1464 branches. By default @samp{--no-relax-branch} is selected, causing any
1465 out-of-range branches to produce an error.
1467 @item -mnan=@var{encoding}
1468 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1469 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1472 @item --emulation=@var{name}
1473 This option was formerly used to switch between ELF and ECOFF output
1474 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1475 removed in GAS 2.24, so the option now serves little purpose.
1476 It is retained for backwards compatibility.
1478 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1479 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1480 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1481 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1482 preferred options instead.
1485 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1492 Control how to deal with multiplication overflow and division by zero.
1493 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1494 (and only work for Instruction Set Architecture level 2 and higher);
1495 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1499 When this option is used, @command{@value{AS}} will issue a warning every
1500 time it generates a nop instruction from a macro.
1505 The following options are available when @value{AS} is configured for
1511 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1512 The command line option @samp{-nojsri2bsr} can be used to disable it.
1516 Enable or disable the silicon filter behaviour. By default this is disabled.
1517 The default can be overridden by the @samp{-sifilter} command line option.
1520 Alter jump instructions for long displacements.
1522 @item -mcpu=[210|340]
1523 Select the cpu type on the target hardware. This controls which instructions
1527 Assemble for a big endian target.
1530 Assemble for a little endian target.
1539 @xref{Meta Options}, for the options available when @value{AS} is configured
1540 for a Meta processor.
1544 @c man begin OPTIONS
1545 The following options are available when @value{AS} is configured for a
1548 @c man begin INCLUDE
1549 @include c-metag.texi
1550 @c ended inside the included file
1555 @c man begin OPTIONS
1557 See the info pages for documentation of the MMIX-specific options.
1563 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1564 for a NDS32 processor.
1566 @c ended inside the included file
1570 @c man begin OPTIONS
1571 The following options are available when @value{AS} is configured for a
1574 @c man begin INCLUDE
1575 @include c-nds32.texi
1576 @c ended inside the included file
1583 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1584 for a PowerPC processor.
1588 @c man begin OPTIONS
1589 The following options are available when @value{AS} is configured for a
1592 @c man begin INCLUDE
1594 @c ended inside the included file
1599 @c man begin OPTIONS
1601 See the info pages for documentation of the RX-specific options.
1605 The following options are available when @value{AS} is configured for the s390
1611 Select the word size, either 31/32 bits or 64 bits.
1614 Select the architecture mode, either the Enterprise System
1615 Architecture (esa) or the z/Architecture mode (zarch).
1616 @item -march=@var{processor}
1617 Specify which s390 processor variant is the target, @samp{g6}, @samp{g6},
1618 @samp{z900}, @samp{z990}, @samp{z9-109}, @samp{z9-ec}, @samp{z10},
1619 @samp{z196}, @samp{zEC12}, or @samp{z13}.
1621 @itemx -mno-regnames
1622 Allow or disallow symbolic names for registers.
1623 @item -mwarn-areg-zero
1624 Warn whenever the operand for a base or index register has been specified
1625 but evaluates to zero.
1633 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1634 for a TMS320C6000 processor.
1638 @c man begin OPTIONS
1639 The following options are available when @value{AS} is configured for a
1640 TMS320C6000 processor.
1642 @c man begin INCLUDE
1643 @include c-tic6x.texi
1644 @c ended inside the included file
1652 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1653 for a TILE-Gx processor.
1657 @c man begin OPTIONS
1658 The following options are available when @value{AS} is configured for a TILE-Gx
1661 @c man begin INCLUDE
1662 @include c-tilegx.texi
1663 @c ended inside the included file
1671 @xref{Visium Options}, for the options available when @value{AS} is configured
1672 for a Visium processor.
1676 @c man begin OPTIONS
1677 The following option is available when @value{AS} is configured for a Visium
1680 @c man begin INCLUDE
1681 @include c-visium.texi
1682 @c ended inside the included file
1690 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1691 for an Xtensa processor.
1695 @c man begin OPTIONS
1696 The following options are available when @value{AS} is configured for an
1699 @c man begin INCLUDE
1700 @include c-xtensa.texi
1701 @c ended inside the included file
1706 @c man begin OPTIONS
1709 The following options are available when @value{AS} is configured for
1710 a Z80 family processor.
1713 Assemble for Z80 processor.
1715 Assemble for R800 processor.
1716 @item -ignore-undocumented-instructions
1718 Assemble undocumented Z80 instructions that also work on R800 without warning.
1719 @item -ignore-unportable-instructions
1721 Assemble all undocumented Z80 instructions without warning.
1722 @item -warn-undocumented-instructions
1724 Issue a warning for undocumented Z80 instructions that also work on R800.
1725 @item -warn-unportable-instructions
1727 Issue a warning for undocumented Z80 instructions that do not work on R800.
1728 @item -forbid-undocumented-instructions
1730 Treat all undocumented instructions as errors.
1731 @item -forbid-unportable-instructions
1733 Treat undocumented Z80 instructions that do not work on R800 as errors.
1740 * Manual:: Structure of this Manual
1741 * GNU Assembler:: The GNU Assembler
1742 * Object Formats:: Object File Formats
1743 * Command Line:: Command Line
1744 * Input Files:: Input Files
1745 * Object:: Output (Object) File
1746 * Errors:: Error and Warning Messages
1750 @section Structure of this Manual
1752 @cindex manual, structure and purpose
1753 This manual is intended to describe what you need to know to use
1754 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1755 notation for symbols, constants, and expressions; the directives that
1756 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1759 We also cover special features in the @value{TARGET}
1760 configuration of @command{@value{AS}}, including assembler directives.
1763 This manual also describes some of the machine-dependent features of
1764 various flavors of the assembler.
1767 @cindex machine instructions (not covered)
1768 On the other hand, this manual is @emph{not} intended as an introduction
1769 to programming in assembly language---let alone programming in general!
1770 In a similar vein, we make no attempt to introduce the machine
1771 architecture; we do @emph{not} describe the instruction set, standard
1772 mnemonics, registers or addressing modes that are standard to a
1773 particular architecture.
1775 You may want to consult the manufacturer's
1776 machine architecture manual for this information.
1780 For information on the H8/300 machine instruction set, see @cite{H8/300
1781 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1782 Programming Manual} (Renesas).
1785 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1786 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1787 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1788 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1791 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1795 @c I think this is premature---doc@cygnus.com, 17jan1991
1797 Throughout this manual, we assume that you are running @dfn{GNU},
1798 the portable operating system from the @dfn{Free Software
1799 Foundation, Inc.}. This restricts our attention to certain kinds of
1800 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1801 once this assumption is granted examples and definitions need less
1804 @command{@value{AS}} is part of a team of programs that turn a high-level
1805 human-readable series of instructions into a low-level
1806 computer-readable series of instructions. Different versions of
1807 @command{@value{AS}} are used for different kinds of computer.
1810 @c There used to be a section "Terminology" here, which defined
1811 @c "contents", "byte", "word", and "long". Defining "word" to any
1812 @c particular size is confusing when the .word directive may generate 16
1813 @c bits on one machine and 32 bits on another; in general, for the user
1814 @c version of this manual, none of these terms seem essential to define.
1815 @c They were used very little even in the former draft of the manual;
1816 @c this draft makes an effort to avoid them (except in names of
1820 @section The GNU Assembler
1822 @c man begin DESCRIPTION
1824 @sc{gnu} @command{as} is really a family of assemblers.
1826 This manual describes @command{@value{AS}}, a member of that family which is
1827 configured for the @value{TARGET} architectures.
1829 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1830 should find a fairly similar environment when you use it on another
1831 architecture. Each version has much in common with the others,
1832 including object file formats, most assembler directives (often called
1833 @dfn{pseudo-ops}) and assembler syntax.@refill
1835 @cindex purpose of @sc{gnu} assembler
1836 @command{@value{AS}} is primarily intended to assemble the output of the
1837 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1838 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1839 assemble correctly everything that other assemblers for the same
1840 machine would assemble.
1842 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1845 @c This remark should appear in generic version of manual; assumption
1846 @c here is that generic version sets M680x0.
1847 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1848 assembler for the same architecture; for example, we know of several
1849 incompatible versions of 680x0 assembly language syntax.
1854 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1855 program in one pass of the source file. This has a subtle impact on the
1856 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1858 @node Object Formats
1859 @section Object File Formats
1861 @cindex object file format
1862 The @sc{gnu} assembler can be configured to produce several alternative
1863 object file formats. For the most part, this does not affect how you
1864 write assembly language programs; but directives for debugging symbols
1865 are typically different in different file formats. @xref{Symbol
1866 Attributes,,Symbol Attributes}.
1869 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1870 @value{OBJ-NAME} format object files.
1872 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1874 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1875 @code{b.out} or COFF format object files.
1878 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1879 SOM or ELF format object files.
1884 @section Command Line
1886 @cindex command line conventions
1888 After the program name @command{@value{AS}}, the command line may contain
1889 options and file names. Options may appear in any order, and may be
1890 before, after, or between file names. The order of file names is
1893 @cindex standard input, as input file
1895 @file{--} (two hyphens) by itself names the standard input file
1896 explicitly, as one of the files for @command{@value{AS}} to assemble.
1898 @cindex options, command line
1899 Except for @samp{--} any command line argument that begins with a
1900 hyphen (@samp{-}) is an option. Each option changes the behavior of
1901 @command{@value{AS}}. No option changes the way another option works. An
1902 option is a @samp{-} followed by one or more letters; the case of
1903 the letter is important. All options are optional.
1905 Some options expect exactly one file name to follow them. The file
1906 name may either immediately follow the option's letter (compatible
1907 with older assemblers) or it may be the next command argument (@sc{gnu}
1908 standard). These two command lines are equivalent:
1911 @value{AS} -o my-object-file.o mumble.s
1912 @value{AS} -omy-object-file.o mumble.s
1916 @section Input Files
1919 @cindex source program
1920 @cindex files, input
1921 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1922 describe the program input to one run of @command{@value{AS}}. The program may
1923 be in one or more files; how the source is partitioned into files
1924 doesn't change the meaning of the source.
1926 @c I added "con" prefix to "catenation" just to prove I can overcome my
1927 @c APL training... doc@cygnus.com
1928 The source program is a concatenation of the text in all the files, in the
1931 @c man begin DESCRIPTION
1932 Each time you run @command{@value{AS}} it assembles exactly one source
1933 program. The source program is made up of one or more files.
1934 (The standard input is also a file.)
1936 You give @command{@value{AS}} a command line that has zero or more input file
1937 names. The input files are read (from left file name to right). A
1938 command line argument (in any position) that has no special meaning
1939 is taken to be an input file name.
1941 If you give @command{@value{AS}} no file names it attempts to read one input file
1942 from the @command{@value{AS}} standard input, which is normally your terminal. You
1943 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1946 Use @samp{--} if you need to explicitly name the standard input file
1947 in your command line.
1949 If the source is empty, @command{@value{AS}} produces a small, empty object
1954 @subheading Filenames and Line-numbers
1956 @cindex input file linenumbers
1957 @cindex line numbers, in input files
1958 There are two ways of locating a line in the input file (or files) and
1959 either may be used in reporting error messages. One way refers to a line
1960 number in a physical file; the other refers to a line number in a
1961 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1963 @dfn{Physical files} are those files named in the command line given
1964 to @command{@value{AS}}.
1966 @dfn{Logical files} are simply names declared explicitly by assembler
1967 directives; they bear no relation to physical files. Logical file names help
1968 error messages reflect the original source file, when @command{@value{AS}} source
1969 is itself synthesized from other files. @command{@value{AS}} understands the
1970 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1971 @ref{File,,@code{.file}}.
1974 @section Output (Object) File
1980 Every time you run @command{@value{AS}} it produces an output file, which is
1981 your assembly language program translated into numbers. This file
1982 is the object file. Its default name is
1990 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1992 You can give it another name by using the @option{-o} option. Conventionally,
1993 object file names end with @file{.o}. The default name is used for historical
1994 reasons: older assemblers were capable of assembling self-contained programs
1995 directly into a runnable program. (For some formats, this isn't currently
1996 possible, but it can be done for the @code{a.out} format.)
2000 The object file is meant for input to the linker @code{@value{LD}}. It contains
2001 assembled program code, information to help @code{@value{LD}} integrate
2002 the assembled program into a runnable file, and (optionally) symbolic
2003 information for the debugger.
2005 @c link above to some info file(s) like the description of a.out.
2006 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
2009 @section Error and Warning Messages
2011 @c man begin DESCRIPTION
2013 @cindex error messages
2014 @cindex warning messages
2015 @cindex messages from assembler
2016 @command{@value{AS}} may write warnings and error messages to the standard error
2017 file (usually your terminal). This should not happen when a compiler
2018 runs @command{@value{AS}} automatically. Warnings report an assumption made so
2019 that @command{@value{AS}} could keep assembling a flawed program; errors report a
2020 grave problem that stops the assembly.
2024 @cindex format of warning messages
2025 Warning messages have the format
2028 file_name:@b{NNN}:Warning Message Text
2032 @cindex line numbers, in warnings/errors
2033 (where @b{NNN} is a line number). If a logical file name has been given
2034 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
2035 the current input file is used. If a logical line number was given
2037 (@pxref{Line,,@code{.line}})
2039 then it is used to calculate the number printed,
2040 otherwise the actual line in the current source file is printed. The
2041 message text is intended to be self explanatory (in the grand Unix
2044 @cindex format of error messages
2045 Error messages have the format
2047 file_name:@b{NNN}:FATAL:Error Message Text
2049 The file name and line number are derived as for warning
2050 messages. The actual message text may be rather less explanatory
2051 because many of them aren't supposed to happen.
2054 @chapter Command-Line Options
2056 @cindex options, all versions of assembler
2057 This chapter describes command-line options available in @emph{all}
2058 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
2059 for options specific
2061 to the @value{TARGET} target.
2064 to particular machine architectures.
2067 @c man begin DESCRIPTION
2069 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2070 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2071 The assembler arguments must be separated from each other (and the @samp{-Wa})
2072 by commas. For example:
2075 gcc -c -g -O -Wa,-alh,-L file.c
2079 This passes two options to the assembler: @samp{-alh} (emit a listing to
2080 standard output with high-level and assembly source) and @samp{-L} (retain
2081 local symbols in the symbol table).
2083 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2084 command-line options are automatically passed to the assembler by the compiler.
2085 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2086 precisely what options it passes to each compilation pass, including the
2092 * a:: -a[cdghlns] enable listings
2093 * alternate:: --alternate enable alternate macro syntax
2094 * D:: -D for compatibility
2095 * f:: -f to work faster
2096 * I:: -I for .include search path
2097 @ifclear DIFF-TBL-KLUGE
2098 * K:: -K for compatibility
2100 @ifset DIFF-TBL-KLUGE
2101 * K:: -K for difference tables
2104 * L:: -L to retain local symbols
2105 * listing:: --listing-XXX to configure listing output
2106 * M:: -M or --mri to assemble in MRI compatibility mode
2107 * MD:: --MD for dependency tracking
2108 * o:: -o to name the object file
2109 * R:: -R to join data and text sections
2110 * statistics:: --statistics to see statistics about assembly
2111 * traditional-format:: --traditional-format for compatible output
2112 * v:: -v to announce version
2113 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2114 * Z:: -Z to make object file even after errors
2118 @section Enable Listings: @option{-a[cdghlns]}
2128 @cindex listings, enabling
2129 @cindex assembly listings, enabling
2131 These options enable listing output from the assembler. By itself,
2132 @samp{-a} requests high-level, assembly, and symbols listing.
2133 You can use other letters to select specific options for the list:
2134 @samp{-ah} requests a high-level language listing,
2135 @samp{-al} requests an output-program assembly listing, and
2136 @samp{-as} requests a symbol table listing.
2137 High-level listings require that a compiler debugging option like
2138 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2141 Use the @samp{-ag} option to print a first section with general assembly
2142 information, like @value{AS} version, switches passed, or time stamp.
2144 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2145 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2146 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2147 omitted from the listing.
2149 Use the @samp{-ad} option to omit debugging directives from the
2152 Once you have specified one of these options, you can further control
2153 listing output and its appearance using the directives @code{.list},
2154 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2156 The @samp{-an} option turns off all forms processing.
2157 If you do not request listing output with one of the @samp{-a} options, the
2158 listing-control directives have no effect.
2160 The letters after @samp{-a} may be combined into one option,
2161 @emph{e.g.}, @samp{-aln}.
2163 Note if the assembler source is coming from the standard input (e.g.,
2165 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
2166 is being used) then the listing will not contain any comments or preprocessor
2167 directives. This is because the listing code buffers input source lines from
2168 stdin only after they have been preprocessed by the assembler. This reduces
2169 memory usage and makes the code more efficient.
2172 @section @option{--alternate}
2175 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2178 @section @option{-D}
2181 This option has no effect whatsoever, but it is accepted to make it more
2182 likely that scripts written for other assemblers also work with
2183 @command{@value{AS}}.
2186 @section Work Faster: @option{-f}
2189 @cindex trusted compiler
2190 @cindex faster processing (@option{-f})
2191 @samp{-f} should only be used when assembling programs written by a
2192 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2193 and comment preprocessing on
2194 the input file(s) before assembling them. @xref{Preprocessing,
2198 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2199 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2204 @section @code{.include} Search Path: @option{-I} @var{path}
2206 @kindex -I @var{path}
2207 @cindex paths for @code{.include}
2208 @cindex search path for @code{.include}
2209 @cindex @code{include} directive search path
2210 Use this option to add a @var{path} to the list of directories
2211 @command{@value{AS}} searches for files specified in @code{.include}
2212 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2213 many times as necessary to include a variety of paths. The current
2214 working directory is always searched first; after that, @command{@value{AS}}
2215 searches any @samp{-I} directories in the same order as they were
2216 specified (left to right) on the command line.
2219 @section Difference Tables: @option{-K}
2222 @ifclear DIFF-TBL-KLUGE
2223 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2224 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2225 where it can be used to warn when the assembler alters the machine code
2226 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2227 family does not have the addressing limitations that sometimes lead to this
2228 alteration on other platforms.
2231 @ifset DIFF-TBL-KLUGE
2232 @cindex difference tables, warning
2233 @cindex warning for altered difference tables
2234 @command{@value{AS}} sometimes alters the code emitted for directives of the
2235 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2236 You can use the @samp{-K} option if you want a warning issued when this
2241 @section Include Local Symbols: @option{-L}
2244 @cindex local symbols, retaining in output
2245 Symbols beginning with system-specific local label prefixes, typically
2246 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2247 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2248 such symbols when debugging, because they are intended for the use of
2249 programs (like compilers) that compose assembler programs, not for your
2250 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2251 such symbols, so you do not normally debug with them.
2253 This option tells @command{@value{AS}} to retain those local symbols
2254 in the object file. Usually if you do this you also tell the linker
2255 @code{@value{LD}} to preserve those symbols.
2258 @section Configuring listing output: @option{--listing}
2260 The listing feature of the assembler can be enabled via the command line switch
2261 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2262 hex dump of the corresponding locations in the output object file, and displays
2263 them as a listing file. The format of this listing can be controlled by
2264 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2265 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2266 @code{.psize} (@pxref{Psize}), and
2267 @code{.eject} (@pxref{Eject}) and also by the following switches:
2270 @item --listing-lhs-width=@samp{number}
2271 @kindex --listing-lhs-width
2272 @cindex Width of first line disassembly output
2273 Sets the maximum width, in words, of the first line of the hex byte dump. This
2274 dump appears on the left hand side of the listing output.
2276 @item --listing-lhs-width2=@samp{number}
2277 @kindex --listing-lhs-width2
2278 @cindex Width of continuation lines of disassembly output
2279 Sets the maximum width, in words, of any further lines of the hex byte dump for
2280 a given input source line. If this value is not specified, it defaults to being
2281 the same as the value specified for @samp{--listing-lhs-width}. If neither
2282 switch is used the default is to one.
2284 @item --listing-rhs-width=@samp{number}
2285 @kindex --listing-rhs-width
2286 @cindex Width of source line output
2287 Sets the maximum width, in characters, of the source line that is displayed
2288 alongside the hex dump. The default value for this parameter is 100. The
2289 source line is displayed on the right hand side of the listing output.
2291 @item --listing-cont-lines=@samp{number}
2292 @kindex --listing-cont-lines
2293 @cindex Maximum number of continuation lines
2294 Sets the maximum number of continuation lines of hex dump that will be
2295 displayed for a given single line of source input. The default value is 4.
2299 @section Assemble in MRI Compatibility Mode: @option{-M}
2302 @cindex MRI compatibility mode
2303 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2304 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2305 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
2306 configured target) assembler from Microtec Research. The exact nature of the
2307 MRI syntax will not be documented here; see the MRI manuals for more
2308 information. Note in particular that the handling of macros and macro
2309 arguments is somewhat different. The purpose of this option is to permit
2310 assembling existing MRI assembler code using @command{@value{AS}}.
2312 The MRI compatibility is not complete. Certain operations of the MRI assembler
2313 depend upon its object file format, and can not be supported using other object
2314 file formats. Supporting these would require enhancing each object file format
2315 individually. These are:
2318 @item global symbols in common section
2320 The m68k MRI assembler supports common sections which are merged by the linker.
2321 Other object file formats do not support this. @command{@value{AS}} handles
2322 common sections by treating them as a single common symbol. It permits local
2323 symbols to be defined within a common section, but it can not support global
2324 symbols, since it has no way to describe them.
2326 @item complex relocations
2328 The MRI assemblers support relocations against a negated section address, and
2329 relocations which combine the start addresses of two or more sections. These
2330 are not support by other object file formats.
2332 @item @code{END} pseudo-op specifying start address
2334 The MRI @code{END} pseudo-op permits the specification of a start address.
2335 This is not supported by other object file formats. The start address may
2336 instead be specified using the @option{-e} option to the linker, or in a linker
2339 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2341 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2342 name to the output file. This is not supported by other object file formats.
2344 @item @code{ORG} pseudo-op
2346 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2347 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2348 which changes the location within the current section. Absolute sections are
2349 not supported by other object file formats. The address of a section may be
2350 assigned within a linker script.
2353 There are some other features of the MRI assembler which are not supported by
2354 @command{@value{AS}}, typically either because they are difficult or because they
2355 seem of little consequence. Some of these may be supported in future releases.
2359 @item EBCDIC strings
2361 EBCDIC strings are not supported.
2363 @item packed binary coded decimal
2365 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2366 and @code{DCB.P} pseudo-ops are not supported.
2368 @item @code{FEQU} pseudo-op
2370 The m68k @code{FEQU} pseudo-op is not supported.
2372 @item @code{NOOBJ} pseudo-op
2374 The m68k @code{NOOBJ} pseudo-op is not supported.
2376 @item @code{OPT} branch control options
2378 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2379 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2380 relaxes all branches, whether forward or backward, to an appropriate size, so
2381 these options serve no purpose.
2383 @item @code{OPT} list control options
2385 The following m68k @code{OPT} list control options are ignored: @code{C},
2386 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2387 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2389 @item other @code{OPT} options
2391 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2392 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2394 @item @code{OPT} @code{D} option is default
2396 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2397 @code{OPT NOD} may be used to turn it off.
2399 @item @code{XREF} pseudo-op.
2401 The m68k @code{XREF} pseudo-op is ignored.
2403 @item @code{.debug} pseudo-op
2405 The i960 @code{.debug} pseudo-op is not supported.
2407 @item @code{.extended} pseudo-op
2409 The i960 @code{.extended} pseudo-op is not supported.
2411 @item @code{.list} pseudo-op.
2413 The various options of the i960 @code{.list} pseudo-op are not supported.
2415 @item @code{.optimize} pseudo-op
2417 The i960 @code{.optimize} pseudo-op is not supported.
2419 @item @code{.output} pseudo-op
2421 The i960 @code{.output} pseudo-op is not supported.
2423 @item @code{.setreal} pseudo-op
2425 The i960 @code{.setreal} pseudo-op is not supported.
2430 @section Dependency Tracking: @option{--MD}
2433 @cindex dependency tracking
2436 @command{@value{AS}} can generate a dependency file for the file it creates. This
2437 file consists of a single rule suitable for @code{make} describing the
2438 dependencies of the main source file.
2440 The rule is written to the file named in its argument.
2442 This feature is used in the automatic updating of makefiles.
2445 @section Name the Object File: @option{-o}
2448 @cindex naming object file
2449 @cindex object file name
2450 There is always one object file output when you run @command{@value{AS}}. By
2451 default it has the name
2454 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2468 You use this option (which takes exactly one filename) to give the
2469 object file a different name.
2471 Whatever the object file is called, @command{@value{AS}} overwrites any
2472 existing file of the same name.
2475 @section Join Data and Text Sections: @option{-R}
2478 @cindex data and text sections, joining
2479 @cindex text and data sections, joining
2480 @cindex joining text and data sections
2481 @cindex merging text and data sections
2482 @option{-R} tells @command{@value{AS}} to write the object file as if all
2483 data-section data lives in the text section. This is only done at
2484 the very last moment: your binary data are the same, but data
2485 section parts are relocated differently. The data section part of
2486 your object file is zero bytes long because all its bytes are
2487 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2489 When you specify @option{-R} it would be possible to generate shorter
2490 address displacements (because we do not have to cross between text and
2491 data section). We refrain from doing this simply for compatibility with
2492 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2495 When @command{@value{AS}} is configured for COFF or ELF output,
2496 this option is only useful if you use sections named @samp{.text} and
2501 @option{-R} is not supported for any of the HPPA targets. Using
2502 @option{-R} generates a warning from @command{@value{AS}}.
2506 @section Display Assembly Statistics: @option{--statistics}
2508 @kindex --statistics
2509 @cindex statistics, about assembly
2510 @cindex time, total for assembly
2511 @cindex space used, maximum for assembly
2512 Use @samp{--statistics} to display two statistics about the resources used by
2513 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2514 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2517 @node traditional-format
2518 @section Compatible Output: @option{--traditional-format}
2520 @kindex --traditional-format
2521 For some targets, the output of @command{@value{AS}} is different in some ways
2522 from the output of some existing assembler. This switch requests
2523 @command{@value{AS}} to use the traditional format instead.
2525 For example, it disables the exception frame optimizations which
2526 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2529 @section Announce Version: @option{-v}
2533 @cindex assembler version
2534 @cindex version of assembler
2535 You can find out what version of as is running by including the
2536 option @samp{-v} (which you can also spell as @samp{-version}) on the
2540 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2542 @command{@value{AS}} should never give a warning or error message when
2543 assembling compiler output. But programs written by people often
2544 cause @command{@value{AS}} to give a warning that a particular assumption was
2545 made. All such warnings are directed to the standard error file.
2549 @cindex suppressing warnings
2550 @cindex warnings, suppressing
2551 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2552 This only affects the warning messages: it does not change any particular of
2553 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2556 @kindex --fatal-warnings
2557 @cindex errors, caused by warnings
2558 @cindex warnings, causing error
2559 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2560 files that generate warnings to be in error.
2563 @cindex warnings, switching on
2564 You can switch these options off again by specifying @option{--warn}, which
2565 causes warnings to be output as usual.
2568 @section Generate Object File in Spite of Errors: @option{-Z}
2569 @cindex object file, after errors
2570 @cindex errors, continuing after
2571 After an error message, @command{@value{AS}} normally produces no output. If for
2572 some reason you are interested in object file output even after
2573 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2574 option. If there are any errors, @command{@value{AS}} continues anyways, and
2575 writes an object file after a final warning message of the form @samp{@var{n}
2576 errors, @var{m} warnings, generating bad object file.}
2581 @cindex machine-independent syntax
2582 @cindex syntax, machine-independent
2583 This chapter describes the machine-independent syntax allowed in a
2584 source file. @command{@value{AS}} syntax is similar to what many other
2585 assemblers use; it is inspired by the BSD 4.2
2590 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2594 * Preprocessing:: Preprocessing
2595 * Whitespace:: Whitespace
2596 * Comments:: Comments
2597 * Symbol Intro:: Symbols
2598 * Statements:: Statements
2599 * Constants:: Constants
2603 @section Preprocessing
2605 @cindex preprocessing
2606 The @command{@value{AS}} internal preprocessor:
2608 @cindex whitespace, removed by preprocessor
2610 adjusts and removes extra whitespace. It leaves one space or tab before
2611 the keywords on a line, and turns any other whitespace on the line into
2614 @cindex comments, removed by preprocessor
2616 removes all comments, replacing them with a single space, or an
2617 appropriate number of newlines.
2619 @cindex constants, converted by preprocessor
2621 converts character constants into the appropriate numeric values.
2624 It does not do macro processing, include file handling, or
2625 anything else you may get from your C compiler's preprocessor. You can
2626 do include file processing with the @code{.include} directive
2627 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2628 to get other ``CPP'' style preprocessing by giving the input file a
2629 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2630 Output, gcc.info, Using GNU CC}.
2632 Excess whitespace, comments, and character constants
2633 cannot be used in the portions of the input text that are not
2636 @cindex turning preprocessing on and off
2637 @cindex preprocessing, turning on and off
2640 If the first line of an input file is @code{#NO_APP} or if you use the
2641 @samp{-f} option, whitespace and comments are not removed from the input file.
2642 Within an input file, you can ask for whitespace and comment removal in
2643 specific portions of the by putting a line that says @code{#APP} before the
2644 text that may contain whitespace or comments, and putting a line that says
2645 @code{#NO_APP} after this text. This feature is mainly intend to support
2646 @code{asm} statements in compilers whose output is otherwise free of comments
2653 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2654 Whitespace is used to separate symbols, and to make programs neater for
2655 people to read. Unless within character constants
2656 (@pxref{Characters,,Character Constants}), any whitespace means the same
2657 as exactly one space.
2663 There are two ways of rendering comments to @command{@value{AS}}. In both
2664 cases the comment is equivalent to one space.
2666 Anything from @samp{/*} through the next @samp{*/} is a comment.
2667 This means you may not nest these comments.
2671 The only way to include a newline ('\n') in a comment
2672 is to use this sort of comment.
2675 /* This sort of comment does not nest. */
2678 @cindex line comment character
2679 Anything from a @dfn{line comment} character up to the next newline is
2680 considered a comment and is ignored. The line comment character is target
2681 specific, and some targets multiple comment characters. Some targets also have
2682 line comment characters that only work if they are the first character on a
2683 line. Some targets use a sequence of two characters to introduce a line
2684 comment. Some targets can also change their line comment characters depending
2685 upon command line options that have been used. For more details see the
2686 @emph{Syntax} section in the documentation for individual targets.
2688 If the line comment character is the hash sign (@samp{#}) then it still has the
2689 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2690 to specify logical line numbers:
2693 @cindex lines starting with @code{#}
2694 @cindex logical line numbers
2695 To be compatible with past assemblers, lines that begin with @samp{#} have a
2696 special interpretation. Following the @samp{#} should be an absolute
2697 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2698 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2699 new logical file name. The rest of the line, if any, should be whitespace.
2701 If the first non-whitespace characters on the line are not numeric,
2702 the line is ignored. (Just like a comment.)
2705 # This is an ordinary comment.
2706 # 42-6 "new_file_name" # New logical file name
2707 # This is logical line # 36.
2709 This feature is deprecated, and may disappear from future versions
2710 of @command{@value{AS}}.
2715 @cindex characters used in symbols
2716 @ifclear SPECIAL-SYMS
2717 A @dfn{symbol} is one or more characters chosen from the set of all
2718 letters (both upper and lower case), digits and the three characters
2724 A @dfn{symbol} is one or more characters chosen from the set of all
2725 letters (both upper and lower case), digits and the three characters
2726 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2732 On most machines, you can also use @code{$} in symbol names; exceptions
2733 are noted in @ref{Machine Dependencies}.
2735 No symbol may begin with a digit. Case is significant.
2736 There is no length limit: all characters are significant. Multibyte characters
2737 are supported. Symbols are delimited by characters not in that set, or by the
2738 beginning of a file (since the source program must end with a newline, the end
2739 of a file is not a possible symbol delimiter). @xref{Symbols}.
2740 @cindex length of symbols
2745 @cindex statements, structure of
2746 @cindex line separator character
2747 @cindex statement separator character
2749 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2750 @dfn{line separator character}. The line separator character is target
2751 specific and described in the @emph{Syntax} section of each
2752 target's documentation. Not all targets support a line separator character.
2753 The newline or line separator character is considered to be part of the
2754 preceding statement. Newlines and separators within character constants are an
2755 exception: they do not end statements.
2757 @cindex newline, required at file end
2758 @cindex EOF, newline must precede
2759 It is an error to end any statement with end-of-file: the last
2760 character of any input file should be a newline.@refill
2762 An empty statement is allowed, and may include whitespace. It is ignored.
2764 @cindex instructions and directives
2765 @cindex directives and instructions
2766 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2767 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2769 A statement begins with zero or more labels, optionally followed by a
2770 key symbol which determines what kind of statement it is. The key
2771 symbol determines the syntax of the rest of the statement. If the
2772 symbol begins with a dot @samp{.} then the statement is an assembler
2773 directive: typically valid for any computer. If the symbol begins with
2774 a letter the statement is an assembly language @dfn{instruction}: it
2775 assembles into a machine language instruction.
2777 Different versions of @command{@value{AS}} for different computers
2778 recognize different instructions. In fact, the same symbol may
2779 represent a different instruction in a different computer's assembly
2783 @cindex @code{:} (label)
2784 @cindex label (@code{:})
2785 A label is a symbol immediately followed by a colon (@code{:}).
2786 Whitespace before a label or after a colon is permitted, but you may not
2787 have whitespace between a label's symbol and its colon. @xref{Labels}.
2790 For HPPA targets, labels need not be immediately followed by a colon, but
2791 the definition of a label must begin in column zero. This also implies that
2792 only one label may be defined on each line.
2796 label: .directive followed by something
2797 another_label: # This is an empty statement.
2798 instruction operand_1, operand_2, @dots{}
2805 A constant is a number, written so that its value is known by
2806 inspection, without knowing any context. Like this:
2809 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2810 .ascii "Ring the bell\7" # A string constant.
2811 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2812 .float 0f-314159265358979323846264338327\
2813 95028841971.693993751E-40 # - pi, a flonum.
2818 * Characters:: Character Constants
2819 * Numbers:: Number Constants
2823 @subsection Character Constants
2825 @cindex character constants
2826 @cindex constants, character
2827 There are two kinds of character constants. A @dfn{character} stands
2828 for one character in one byte and its value may be used in
2829 numeric expressions. String constants (properly called string
2830 @emph{literals}) are potentially many bytes and their values may not be
2831 used in arithmetic expressions.
2835 * Chars:: Characters
2839 @subsubsection Strings
2841 @cindex string constants
2842 @cindex constants, string
2843 A @dfn{string} is written between double-quotes. It may contain
2844 double-quotes or null characters. The way to get special characters
2845 into a string is to @dfn{escape} these characters: precede them with
2846 a backslash @samp{\} character. For example @samp{\\} represents
2847 one backslash: the first @code{\} is an escape which tells
2848 @command{@value{AS}} to interpret the second character literally as a backslash
2849 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2850 escape character). The complete list of escapes follows.
2852 @cindex escape codes, character
2853 @cindex character escape codes
2856 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2858 @cindex @code{\b} (backspace character)
2859 @cindex backspace (@code{\b})
2861 Mnemonic for backspace; for ASCII this is octal code 010.
2864 @c Mnemonic for EOText; for ASCII this is octal code 004.
2866 @cindex @code{\f} (formfeed character)
2867 @cindex formfeed (@code{\f})
2869 Mnemonic for FormFeed; for ASCII this is octal code 014.
2871 @cindex @code{\n} (newline character)
2872 @cindex newline (@code{\n})
2874 Mnemonic for newline; for ASCII this is octal code 012.
2877 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2879 @cindex @code{\r} (carriage return character)
2880 @cindex carriage return (@code{\r})
2882 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2885 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2886 @c other assemblers.
2888 @cindex @code{\t} (tab)
2889 @cindex tab (@code{\t})
2891 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2894 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2895 @c @item \x @var{digit} @var{digit} @var{digit}
2896 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2898 @cindex @code{\@var{ddd}} (octal character code)
2899 @cindex octal character code (@code{\@var{ddd}})
2900 @item \ @var{digit} @var{digit} @var{digit}
2901 An octal character code. The numeric code is 3 octal digits.
2902 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2903 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2905 @cindex @code{\@var{xd...}} (hex character code)
2906 @cindex hex character code (@code{\@var{xd...}})
2907 @item \@code{x} @var{hex-digits...}
2908 A hex character code. All trailing hex digits are combined. Either upper or
2909 lower case @code{x} works.
2911 @cindex @code{\\} (@samp{\} character)
2912 @cindex backslash (@code{\\})
2914 Represents one @samp{\} character.
2917 @c Represents one @samp{'} (accent acute) character.
2918 @c This is needed in single character literals
2919 @c (@xref{Characters,,Character Constants}.) to represent
2922 @cindex @code{\"} (doublequote character)
2923 @cindex doublequote (@code{\"})
2925 Represents one @samp{"} character. Needed in strings to represent
2926 this character, because an unescaped @samp{"} would end the string.
2928 @item \ @var{anything-else}
2929 Any other character when escaped by @kbd{\} gives a warning, but
2930 assembles as if the @samp{\} was not present. The idea is that if
2931 you used an escape sequence you clearly didn't want the literal
2932 interpretation of the following character. However @command{@value{AS}} has no
2933 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2934 code and warns you of the fact.
2937 Which characters are escapable, and what those escapes represent,
2938 varies widely among assemblers. The current set is what we think
2939 the BSD 4.2 assembler recognizes, and is a subset of what most C
2940 compilers recognize. If you are in doubt, do not use an escape
2944 @subsubsection Characters
2946 @cindex single character constant
2947 @cindex character, single
2948 @cindex constant, single character
2949 A single character may be written as a single quote immediately
2950 followed by that character. The same escapes apply to characters as
2951 to strings. So if you want to write the character backslash, you
2952 must write @kbd{'\\} where the first @code{\} escapes the second
2953 @code{\}. As you can see, the quote is an acute accent, not a
2954 grave accent. A newline
2956 @ifclear abnormal-separator
2957 (or semicolon @samp{;})
2959 @ifset abnormal-separator
2961 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2966 immediately following an acute accent is taken as a literal character
2967 and does not count as the end of a statement. The value of a character
2968 constant in a numeric expression is the machine's byte-wide code for
2969 that character. @command{@value{AS}} assumes your character code is ASCII:
2970 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2973 @subsection Number Constants
2975 @cindex constants, number
2976 @cindex number constants
2977 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2978 are stored in the target machine. @emph{Integers} are numbers that
2979 would fit into an @code{int} in the C language. @emph{Bignums} are
2980 integers, but they are stored in more than 32 bits. @emph{Flonums}
2981 are floating point numbers, described below.
2984 * Integers:: Integers
2989 * Bit Fields:: Bit Fields
2995 @subsubsection Integers
2997 @cindex constants, integer
2999 @cindex binary integers
3000 @cindex integers, binary
3001 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
3002 the binary digits @samp{01}.
3004 @cindex octal integers
3005 @cindex integers, octal
3006 An octal integer is @samp{0} followed by zero or more of the octal
3007 digits (@samp{01234567}).
3009 @cindex decimal integers
3010 @cindex integers, decimal
3011 A decimal integer starts with a non-zero digit followed by zero or
3012 more digits (@samp{0123456789}).
3014 @cindex hexadecimal integers
3015 @cindex integers, hexadecimal
3016 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
3017 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
3019 Integers have the usual values. To denote a negative integer, use
3020 the prefix operator @samp{-} discussed under expressions
3021 (@pxref{Prefix Ops,,Prefix Operators}).
3024 @subsubsection Bignums
3027 @cindex constants, bignum
3028 A @dfn{bignum} has the same syntax and semantics as an integer
3029 except that the number (or its negative) takes more than 32 bits to
3030 represent in binary. The distinction is made because in some places
3031 integers are permitted while bignums are not.
3034 @subsubsection Flonums
3036 @cindex floating point numbers
3037 @cindex constants, floating point
3039 @cindex precision, floating point
3040 A @dfn{flonum} represents a floating point number. The translation is
3041 indirect: a decimal floating point number from the text is converted by
3042 @command{@value{AS}} to a generic binary floating point number of more than
3043 sufficient precision. This generic floating point number is converted
3044 to a particular computer's floating point format (or formats) by a
3045 portion of @command{@value{AS}} specialized to that computer.
3047 A flonum is written by writing (in order)
3052 (@samp{0} is optional on the HPPA.)
3056 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
3058 @kbd{e} is recommended. Case is not important.
3060 @c FIXME: verify if flonum syntax really this vague for most cases
3061 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
3062 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
3065 On the H8/300, Renesas / SuperH SH,
3066 and AMD 29K architectures, the letter must be
3067 one of the letters @samp{DFPRSX} (in upper or lower case).
3069 On the ARC, the letter must be one of the letters @samp{DFRS}
3070 (in upper or lower case).
3072 On the Intel 960 architecture, the letter must be
3073 one of the letters @samp{DFT} (in upper or lower case).
3075 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3079 One of the letters @samp{DFRS} (in upper or lower case).
3082 One of the letters @samp{DFPRSX} (in upper or lower case).
3085 The letter @samp{E} (upper case only).
3088 One of the letters @samp{DFT} (in upper or lower case).
3093 An optional sign: either @samp{+} or @samp{-}.
3096 An optional @dfn{integer part}: zero or more decimal digits.
3099 An optional @dfn{fractional part}: @samp{.} followed by zero
3100 or more decimal digits.
3103 An optional exponent, consisting of:
3107 An @samp{E} or @samp{e}.
3108 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3109 @c principle this can perfectly well be different on different targets.
3111 Optional sign: either @samp{+} or @samp{-}.
3113 One or more decimal digits.
3118 At least one of the integer part or the fractional part must be
3119 present. The floating point number has the usual base-10 value.
3121 @command{@value{AS}} does all processing using integers. Flonums are computed
3122 independently of any floating point hardware in the computer running
3123 @command{@value{AS}}.
3127 @c Bit fields are written as a general facility but are also controlled
3128 @c by a conditional-compilation flag---which is as of now (21mar91)
3129 @c turned on only by the i960 config of GAS.
3131 @subsubsection Bit Fields
3134 @cindex constants, bit field
3135 You can also define numeric constants as @dfn{bit fields}.
3136 Specify two numbers separated by a colon---
3138 @var{mask}:@var{value}
3141 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
3144 The resulting number is then packed
3146 @c this conditional paren in case bit fields turned on elsewhere than 960
3147 (in host-dependent byte order)
3149 into a field whose width depends on which assembler directive has the
3150 bit-field as its argument. Overflow (a result from the bitwise and
3151 requiring more binary digits to represent) is not an error; instead,
3152 more constants are generated, of the specified width, beginning with the
3153 least significant digits.@refill
3155 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
3156 @code{.short}, and @code{.word} accept bit-field arguments.
3161 @chapter Sections and Relocation
3166 * Secs Background:: Background
3167 * Ld Sections:: Linker Sections
3168 * As Sections:: Assembler Internal Sections
3169 * Sub-Sections:: Sub-Sections
3173 @node Secs Background
3176 Roughly, a section is a range of addresses, with no gaps; all data
3177 ``in'' those addresses is treated the same for some particular purpose.
3178 For example there may be a ``read only'' section.
3180 @cindex linker, and assembler
3181 @cindex assembler, and linker
3182 The linker @code{@value{LD}} reads many object files (partial programs) and
3183 combines their contents to form a runnable program. When @command{@value{AS}}
3184 emits an object file, the partial program is assumed to start at address 0.
3185 @code{@value{LD}} assigns the final addresses for the partial program, so that
3186 different partial programs do not overlap. This is actually an
3187 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3190 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3191 addresses. These blocks slide to their run-time addresses as rigid
3192 units; their length does not change and neither does the order of bytes
3193 within them. Such a rigid unit is called a @emph{section}. Assigning
3194 run-time addresses to sections is called @dfn{relocation}. It includes
3195 the task of adjusting mentions of object-file addresses so they refer to
3196 the proper run-time addresses.
3198 For the H8/300, and for the Renesas / SuperH SH,
3199 @command{@value{AS}} pads sections if needed to
3200 ensure they end on a word (sixteen bit) boundary.
3203 @cindex standard assembler sections
3204 An object file written by @command{@value{AS}} has at least three sections, any
3205 of which may be empty. These are named @dfn{text}, @dfn{data} and
3210 When it generates COFF or ELF output,
3212 @command{@value{AS}} can also generate whatever other named sections you specify
3213 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3214 If you do not use any directives that place output in the @samp{.text}
3215 or @samp{.data} sections, these sections still exist, but are empty.
3220 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3222 @command{@value{AS}} can also generate whatever other named sections you
3223 specify using the @samp{.space} and @samp{.subspace} directives. See
3224 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3225 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3226 assembler directives.
3229 Additionally, @command{@value{AS}} uses different names for the standard
3230 text, data, and bss sections when generating SOM output. Program text
3231 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3232 BSS into @samp{$BSS$}.
3236 Within the object file, the text section starts at address @code{0}, the
3237 data section follows, and the bss section follows the data section.
3240 When generating either SOM or ELF output files on the HPPA, the text
3241 section starts at address @code{0}, the data section at address
3242 @code{0x4000000}, and the bss section follows the data section.
3245 To let @code{@value{LD}} know which data changes when the sections are
3246 relocated, and how to change that data, @command{@value{AS}} also writes to the
3247 object file details of the relocation needed. To perform relocation
3248 @code{@value{LD}} must know, each time an address in the object
3252 Where in the object file is the beginning of this reference to
3255 How long (in bytes) is this reference?
3257 Which section does the address refer to? What is the numeric value of
3259 (@var{address}) @minus{} (@var{start-address of section})?
3262 Is the reference to an address ``Program-Counter relative''?
3265 @cindex addresses, format of
3266 @cindex section-relative addressing
3267 In fact, every address @command{@value{AS}} ever uses is expressed as
3269 (@var{section}) + (@var{offset into section})
3272 Further, most expressions @command{@value{AS}} computes have this section-relative
3275 (For some object formats, such as SOM for the HPPA, some expressions are
3276 symbol-relative instead.)
3279 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3280 @var{N} into section @var{secname}.''
3282 Apart from text, data and bss sections you need to know about the
3283 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3284 addresses in the absolute section remain unchanged. For example, address
3285 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3286 @code{@value{LD}}. Although the linker never arranges two partial programs'
3287 data sections with overlapping addresses after linking, @emph{by definition}
3288 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3289 part of a program is always the same address when the program is running as
3290 address @code{@{absolute@ 239@}} in any other part of the program.
3292 The idea of sections is extended to the @dfn{undefined} section. Any
3293 address whose section is unknown at assembly time is by definition
3294 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3295 Since numbers are always defined, the only way to generate an undefined
3296 address is to mention an undefined symbol. A reference to a named
3297 common block would be such a symbol: its value is unknown at assembly
3298 time so it has section @emph{undefined}.
3300 By analogy the word @emph{section} is used to describe groups of sections in
3301 the linked program. @code{@value{LD}} puts all partial programs' text
3302 sections in contiguous addresses in the linked program. It is
3303 customary to refer to the @emph{text section} of a program, meaning all
3304 the addresses of all partial programs' text sections. Likewise for
3305 data and bss sections.
3307 Some sections are manipulated by @code{@value{LD}}; others are invented for
3308 use of @command{@value{AS}} and have no meaning except during assembly.
3311 @section Linker Sections
3312 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3317 @cindex named sections
3318 @cindex sections, named
3319 @item named sections
3322 @cindex text section
3323 @cindex data section
3327 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3328 separate but equal sections. Anything you can say of one section is
3331 When the program is running, however, it is
3332 customary for the text section to be unalterable. The
3333 text section is often shared among processes: it contains
3334 instructions, constants and the like. The data section of a running
3335 program is usually alterable: for example, C variables would be stored
3336 in the data section.
3341 This section contains zeroed bytes when your program begins running. It
3342 is used to hold uninitialized variables or common storage. The length of
3343 each partial program's bss section is important, but because it starts
3344 out containing zeroed bytes there is no need to store explicit zero
3345 bytes in the object file. The bss section was invented to eliminate
3346 those explicit zeros from object files.
3348 @cindex absolute section
3349 @item absolute section
3350 Address 0 of this section is always ``relocated'' to runtime address 0.
3351 This is useful if you want to refer to an address that @code{@value{LD}} must
3352 not change when relocating. In this sense we speak of absolute
3353 addresses being ``unrelocatable'': they do not change during relocation.
3355 @cindex undefined section
3356 @item undefined section
3357 This ``section'' is a catch-all for address references to objects not in
3358 the preceding sections.
3359 @c FIXME: ref to some other doc on obj-file formats could go here.
3362 @cindex relocation example
3363 An idealized example of three relocatable sections follows.
3365 The example uses the traditional section names @samp{.text} and @samp{.data}.
3367 Memory addresses are on the horizontal axis.
3371 @c END TEXI2ROFF-KILL
3374 partial program # 1: |ttttt|dddd|00|
3381 partial program # 2: |TTT|DDD|000|
3384 +--+---+-----+--+----+---+-----+~~
3385 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3386 +--+---+-----+--+----+---+-----+~~
3388 addresses: 0 @dots{}
3395 \line{\it Partial program \#1: \hfil}
3396 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3397 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3399 \line{\it Partial program \#2: \hfil}
3400 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3401 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3403 \line{\it linked program: \hfil}
3404 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3405 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3406 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3407 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3409 \line{\it addresses: \hfil}
3413 @c END TEXI2ROFF-KILL
3416 @section Assembler Internal Sections
3418 @cindex internal assembler sections
3419 @cindex sections in messages, internal
3420 These sections are meant only for the internal use of @command{@value{AS}}. They
3421 have no meaning at run-time. You do not really need to know about these
3422 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3423 warning messages, so it might be helpful to have an idea of their
3424 meanings to @command{@value{AS}}. These sections are used to permit the
3425 value of every expression in your assembly language program to be a
3426 section-relative address.
3429 @cindex assembler internal logic error
3430 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3431 An internal assembler logic error has been found. This means there is a
3432 bug in the assembler.
3434 @cindex expr (internal section)
3436 The assembler stores complex expression internally as combinations of
3437 symbols. When it needs to represent an expression as a symbol, it puts
3438 it in the expr section.
3440 @c FIXME item transfer[t] vector preload
3441 @c FIXME item transfer[t] vector postload
3442 @c FIXME item register
3446 @section Sub-Sections
3448 @cindex numbered subsections
3449 @cindex grouping data
3455 fall into two sections: text and data.
3457 You may have separate groups of
3459 data in named sections
3463 data in named sections
3469 that you want to end up near to each other in the object file, even though they
3470 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3471 use @dfn{subsections} for this purpose. Within each section, there can be
3472 numbered subsections with values from 0 to 8192. Objects assembled into the
3473 same subsection go into the object file together with other objects in the same
3474 subsection. For example, a compiler might want to store constants in the text
3475 section, but might not want to have them interspersed with the program being
3476 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3477 section of code being output, and a @samp{.text 1} before each group of
3478 constants being output.
3480 Subsections are optional. If you do not use subsections, everything
3481 goes in subsection number zero.
3484 Each subsection is zero-padded up to a multiple of four bytes.
3485 (Subsections may be padded a different amount on different flavors
3486 of @command{@value{AS}}.)
3490 On the H8/300 platform, each subsection is zero-padded to a word
3491 boundary (two bytes).
3492 The same is true on the Renesas SH.
3495 @c FIXME section padding (alignment)?
3496 @c Rich Pixley says padding here depends on target obj code format; that
3497 @c doesn't seem particularly useful to say without further elaboration,
3498 @c so for now I say nothing about it. If this is a generic BFD issue,
3499 @c these paragraphs might need to vanish from this manual, and be
3500 @c discussed in BFD chapter of binutils (or some such).
3504 Subsections appear in your object file in numeric order, lowest numbered
3505 to highest. (All this to be compatible with other people's assemblers.)
3506 The object file contains no representation of subsections; @code{@value{LD}} and
3507 other programs that manipulate object files see no trace of them.
3508 They just see all your text subsections as a text section, and all your
3509 data subsections as a data section.
3511 To specify which subsection you want subsequent statements assembled
3512 into, use a numeric argument to specify it, in a @samp{.text
3513 @var{expression}} or a @samp{.data @var{expression}} statement.
3516 When generating COFF output, you
3521 can also use an extra subsection
3522 argument with arbitrary named sections: @samp{.section @var{name},
3527 When generating ELF output, you
3532 can also use the @code{.subsection} directive (@pxref{SubSection})
3533 to specify a subsection: @samp{.subsection @var{expression}}.
3535 @var{Expression} should be an absolute expression
3536 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3537 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3538 begins in @code{text 0}. For instance:
3540 .text 0 # The default subsection is text 0 anyway.
3541 .ascii "This lives in the first text subsection. *"
3543 .ascii "But this lives in the second text subsection."
3545 .ascii "This lives in the data section,"
3546 .ascii "in the first data subsection."
3548 .ascii "This lives in the first text section,"
3549 .ascii "immediately following the asterisk (*)."
3552 Each section has a @dfn{location counter} incremented by one for every byte
3553 assembled into that section. Because subsections are merely a convenience
3554 restricted to @command{@value{AS}} there is no concept of a subsection location
3555 counter. There is no way to directly manipulate a location counter---but the
3556 @code{.align} directive changes it, and any label definition captures its
3557 current value. The location counter of the section where statements are being
3558 assembled is said to be the @dfn{active} location counter.
3561 @section bss Section
3564 @cindex common variable storage
3565 The bss section is used for local common variable storage.
3566 You may allocate address space in the bss section, but you may
3567 not dictate data to load into it before your program executes. When
3568 your program starts running, all the contents of the bss
3569 section are zeroed bytes.
3571 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3572 @ref{Lcomm,,@code{.lcomm}}.
3574 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3575 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3578 When assembling for a target which supports multiple sections, such as ELF or
3579 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3580 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3581 section. Typically the section will only contain symbol definitions and
3582 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3589 Symbols are a central concept: the programmer uses symbols to name
3590 things, the linker uses symbols to link, and the debugger uses symbols
3594 @cindex debuggers, and symbol order
3595 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3596 the same order they were declared. This may break some debuggers.
3601 * Setting Symbols:: Giving Symbols Other Values
3602 * Symbol Names:: Symbol Names
3603 * Dot:: The Special Dot Symbol
3604 * Symbol Attributes:: Symbol Attributes
3611 A @dfn{label} is written as a symbol immediately followed by a colon
3612 @samp{:}. The symbol then represents the current value of the
3613 active location counter, and is, for example, a suitable instruction
3614 operand. You are warned if you use the same symbol to represent two
3615 different locations: the first definition overrides any other
3619 On the HPPA, the usual form for a label need not be immediately followed by a
3620 colon, but instead must start in column zero. Only one label may be defined on
3621 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3622 provides a special directive @code{.label} for defining labels more flexibly.
3625 @node Setting Symbols
3626 @section Giving Symbols Other Values
3628 @cindex assigning values to symbols
3629 @cindex symbol values, assigning
3630 A symbol can be given an arbitrary value by writing a symbol, followed
3631 by an equals sign @samp{=}, followed by an expression
3632 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3633 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3634 equals sign @samp{=}@samp{=} here represents an equivalent of the
3635 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3638 Blackfin does not support symbol assignment with @samp{=}.
3642 @section Symbol Names
3644 @cindex symbol names
3645 @cindex names, symbol
3646 @ifclear SPECIAL-SYMS
3647 Symbol names begin with a letter or with one of @samp{._}. On most
3648 machines, you can also use @code{$} in symbol names; exceptions are
3649 noted in @ref{Machine Dependencies}. That character may be followed by any
3650 string of digits, letters, dollar signs (unless otherwise noted for a
3651 particular target machine), and underscores.
3655 Symbol names begin with a letter or with one of @samp{._}. On the
3656 Renesas SH you can also use @code{$} in symbol names. That
3657 character may be followed by any string of digits, letters, dollar signs (save
3658 on the H8/300), and underscores.
3662 Case of letters is significant: @code{foo} is a different symbol name
3665 Symbol names do not start with a digit. An exception to this rule is made for
3666 Local Labels. See below.
3668 Multibyte characters are supported. To generate a symbol name containing
3669 multibyte characters enclose it within double quotes and use escape codes. cf
3670 @xref{Strings}. Generating a multibyte symbol name from a label is not
3671 currently supported.
3673 Each symbol has exactly one name. Each name in an assembly language program
3674 refers to exactly one symbol. You may use that symbol name any number of times
3677 @subheading Local Symbol Names
3679 @cindex local symbol names
3680 @cindex symbol names, local
3681 A local symbol is any symbol beginning with certain local label prefixes.
3682 By default, the local label prefix is @samp{.L} for ELF systems or
3683 @samp{L} for traditional a.out systems, but each target may have its own
3684 set of local label prefixes.
3686 On the HPPA local symbols begin with @samp{L$}.
3689 Local symbols are defined and used within the assembler, but they are
3690 normally not saved in object files. Thus, they are not visible when debugging.
3691 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols})
3692 to retain the local symbols in the object files.
3694 @subheading Local Labels
3696 @cindex local labels
3697 @cindex temporary symbol names
3698 @cindex symbol names, temporary
3699 Local labels are different from local symbols. Local labels help compilers and
3700 programmers use names temporarily. They create symbols which are guaranteed to
3701 be unique over the entire scope of the input source code and which can be
3702 referred to by a simple notation. To define a local label, write a label of
3703 the form @samp{@b{N}:} (where @b{N} represents any positive integer). To refer
3704 to the most recent previous definition of that label write @samp{@b{N}b}, using
3705 the same number as when you defined the label. To refer to the next definition
3706 of a local label, write @samp{@b{N}f}---the @samp{b} stands for ``backwards''
3707 and the @samp{f} stands for ``forwards''.
3709 There is no restriction on how you can use these labels, and you can reuse them
3710 too. So that it is possible to repeatedly define the same local label (using
3711 the same number @samp{@b{N}}), although you can only refer to the most recently
3712 defined local label of that number (for a backwards reference) or the next
3713 definition of a specific local label for a forward reference. It is also worth
3714 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3715 implemented in a slightly more efficient manner than the others.
3726 Which is the equivalent of:
3729 label_1: branch label_3
3730 label_2: branch label_1
3731 label_3: branch label_4
3732 label_4: branch label_3
3735 Local label names are only a notational device. They are immediately
3736 transformed into more conventional symbol names before the assembler uses them.
3737 The symbol names are stored in the symbol table, appear in error messages, and
3738 are optionally emitted to the object file. The names are constructed using
3742 @item @emph{local label prefix}
3743 All local symbols begin with the system-specific local label prefix.
3744 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3745 that start with the local label prefix. These labels are
3746 used for symbols you are never intended to see. If you use the
3747 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3748 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3749 you may use them in debugging.
3752 This is the number that was used in the local label definition. So if the
3753 label is written @samp{55:} then the number is @samp{55}.
3756 This unusual character is included so you do not accidentally invent a symbol
3757 of the same name. The character has ASCII value of @samp{\002} (control-B).
3759 @item @emph{ordinal number}
3760 This is a serial number to keep the labels distinct. The first definition of
3761 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3762 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3763 the number @samp{1} and its 15th definition gets @samp{15} as well.
3766 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3767 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3769 @subheading Dollar Local Labels
3770 @cindex dollar local symbols
3772 On some targets @code{@value{AS}} also supports an even more local form of
3773 local labels called dollar labels. These labels go out of scope (i.e., they
3774 become undefined) as soon as a non-local label is defined. Thus they remain
3775 valid for only a small region of the input source code. Normal local labels,
3776 by contrast, remain in scope for the entire file, or until they are redefined
3777 by another occurrence of the same local label.
3779 Dollar labels are defined in exactly the same way as ordinary local labels,
3780 except that they have a dollar sign suffix to their numeric value, e.g.,
3783 They can also be distinguished from ordinary local labels by their transformed
3784 names which use ASCII character @samp{\001} (control-A) as the magic character
3785 to distinguish them from ordinary labels. For example, the fifth definition of
3786 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3789 @section The Special Dot Symbol
3791 @cindex dot (symbol)
3792 @cindex @code{.} (symbol)
3793 @cindex current address
3794 @cindex location counter
3795 The special symbol @samp{.} refers to the current address that
3796 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3797 .long .} defines @code{melvin} to contain its own address.
3798 Assigning a value to @code{.} is treated the same as a @code{.org}
3800 @ifclear no-space-dir
3801 Thus, the expression @samp{.=.+4} is the same as saying
3805 @node Symbol Attributes
3806 @section Symbol Attributes
3808 @cindex symbol attributes
3809 @cindex attributes, symbol
3810 Every symbol has, as well as its name, the attributes ``Value'' and
3811 ``Type''. Depending on output format, symbols can also have auxiliary
3814 The detailed definitions are in @file{a.out.h}.
3817 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3818 all these attributes, and probably won't warn you. This makes the
3819 symbol an externally defined symbol, which is generally what you
3823 * Symbol Value:: Value
3824 * Symbol Type:: Type
3827 * a.out Symbols:: Symbol Attributes: @code{a.out}
3831 * a.out Symbols:: Symbol Attributes: @code{a.out}
3834 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3839 * COFF Symbols:: Symbol Attributes for COFF
3842 * SOM Symbols:: Symbol Attributes for SOM
3849 @cindex value of a symbol
3850 @cindex symbol value
3851 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3852 location in the text, data, bss or absolute sections the value is the
3853 number of addresses from the start of that section to the label.
3854 Naturally for text, data and bss sections the value of a symbol changes
3855 as @code{@value{LD}} changes section base addresses during linking. Absolute
3856 symbols' values do not change during linking: that is why they are
3859 The value of an undefined symbol is treated in a special way. If it is
3860 0 then the symbol is not defined in this assembler source file, and
3861 @code{@value{LD}} tries to determine its value from other files linked into the
3862 same program. You make this kind of symbol simply by mentioning a symbol
3863 name without defining it. A non-zero value represents a @code{.comm}
3864 common declaration. The value is how much common storage to reserve, in
3865 bytes (addresses). The symbol refers to the first address of the
3871 @cindex type of a symbol
3873 The type attribute of a symbol contains relocation (section)
3874 information, any flag settings indicating that a symbol is external, and
3875 (optionally), other information for linkers and debuggers. The exact
3876 format depends on the object-code output format in use.
3881 @c The following avoids a "widow" subsection title. @group would be
3882 @c better if it were available outside examples.
3885 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3887 @cindex @code{b.out} symbol attributes
3888 @cindex symbol attributes, @code{b.out}
3889 These symbol attributes appear only when @command{@value{AS}} is configured for
3890 one of the Berkeley-descended object output formats---@code{a.out} or
3896 @subsection Symbol Attributes: @code{a.out}
3898 @cindex @code{a.out} symbol attributes
3899 @cindex symbol attributes, @code{a.out}
3905 @subsection Symbol Attributes: @code{a.out}
3907 @cindex @code{a.out} symbol attributes
3908 @cindex symbol attributes, @code{a.out}
3912 * Symbol Desc:: Descriptor
3913 * Symbol Other:: Other
3917 @subsubsection Descriptor
3919 @cindex descriptor, of @code{a.out} symbol
3920 This is an arbitrary 16-bit value. You may establish a symbol's
3921 descriptor value by using a @code{.desc} statement
3922 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3923 @command{@value{AS}}.
3926 @subsubsection Other
3928 @cindex other attribute, of @code{a.out} symbol
3929 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3934 @subsection Symbol Attributes for COFF
3936 @cindex COFF symbol attributes
3937 @cindex symbol attributes, COFF
3939 The COFF format supports a multitude of auxiliary symbol attributes;
3940 like the primary symbol attributes, they are set between @code{.def} and
3941 @code{.endef} directives.
3943 @subsubsection Primary Attributes
3945 @cindex primary attributes, COFF symbols
3946 The symbol name is set with @code{.def}; the value and type,
3947 respectively, with @code{.val} and @code{.type}.
3949 @subsubsection Auxiliary Attributes
3951 @cindex auxiliary attributes, COFF symbols
3952 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3953 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3954 table information for COFF.
3959 @subsection Symbol Attributes for SOM
3961 @cindex SOM symbol attributes
3962 @cindex symbol attributes, SOM
3964 The SOM format for the HPPA supports a multitude of symbol attributes set with
3965 the @code{.EXPORT} and @code{.IMPORT} directives.
3967 The attributes are described in @cite{HP9000 Series 800 Assembly
3968 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3969 @code{EXPORT} assembler directive documentation.
3973 @chapter Expressions
3977 @cindex numeric values
3978 An @dfn{expression} specifies an address or numeric value.
3979 Whitespace may precede and/or follow an expression.
3981 The result of an expression must be an absolute number, or else an offset into
3982 a particular section. If an expression is not absolute, and there is not
3983 enough information when @command{@value{AS}} sees the expression to know its
3984 section, a second pass over the source program might be necessary to interpret
3985 the expression---but the second pass is currently not implemented.
3986 @command{@value{AS}} aborts with an error message in this situation.
3989 * Empty Exprs:: Empty Expressions
3990 * Integer Exprs:: Integer Expressions
3994 @section Empty Expressions
3996 @cindex empty expressions
3997 @cindex expressions, empty
3998 An empty expression has no value: it is just whitespace or null.
3999 Wherever an absolute expression is required, you may omit the
4000 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
4001 is compatible with other assemblers.
4004 @section Integer Expressions
4006 @cindex integer expressions
4007 @cindex expressions, integer
4008 An @dfn{integer expression} is one or more @emph{arguments} delimited
4009 by @emph{operators}.
4012 * Arguments:: Arguments
4013 * Operators:: Operators
4014 * Prefix Ops:: Prefix Operators
4015 * Infix Ops:: Infix Operators
4019 @subsection Arguments
4021 @cindex expression arguments
4022 @cindex arguments in expressions
4023 @cindex operands in expressions
4024 @cindex arithmetic operands
4025 @dfn{Arguments} are symbols, numbers or subexpressions. In other
4026 contexts arguments are sometimes called ``arithmetic operands''. In
4027 this manual, to avoid confusing them with the ``instruction operands'' of
4028 the machine language, we use the term ``argument'' to refer to parts of
4029 expressions only, reserving the word ``operand'' to refer only to machine
4030 instruction operands.
4032 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
4033 @var{section} is one of text, data, bss, absolute,
4034 or undefined. @var{NNN} is a signed, 2's complement 32 bit
4037 Numbers are usually integers.
4039 A number can be a flonum or bignum. In this case, you are warned
4040 that only the low order 32 bits are used, and @command{@value{AS}} pretends
4041 these 32 bits are an integer. You may write integer-manipulating
4042 instructions that act on exotic constants, compatible with other
4045 @cindex subexpressions
4046 Subexpressions are a left parenthesis @samp{(} followed by an integer
4047 expression, followed by a right parenthesis @samp{)}; or a prefix
4048 operator followed by an argument.
4051 @subsection Operators
4053 @cindex operators, in expressions
4054 @cindex arithmetic functions
4055 @cindex functions, in expressions
4056 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
4057 operators are followed by an argument. Infix operators appear
4058 between their arguments. Operators may be preceded and/or followed by
4062 @subsection Prefix Operator
4064 @cindex prefix operators
4065 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
4066 one argument, which must be absolute.
4068 @c the tex/end tex stuff surrounding this small table is meant to make
4069 @c it align, on the printed page, with the similar table in the next
4070 @c section (which is inside an enumerate).
4072 \global\advance\leftskip by \itemindent
4077 @dfn{Negation}. Two's complement negation.
4079 @dfn{Complementation}. Bitwise not.
4083 \global\advance\leftskip by -\itemindent
4087 @subsection Infix Operators
4089 @cindex infix operators
4090 @cindex operators, permitted arguments
4091 @dfn{Infix operators} take two arguments, one on either side. Operators
4092 have precedence, but operations with equal precedence are performed left
4093 to right. Apart from @code{+} or @option{-}, both arguments must be
4094 absolute, and the result is absolute.
4097 @cindex operator precedence
4098 @cindex precedence of operators
4105 @dfn{Multiplication}.
4108 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4114 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4117 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4121 Intermediate precedence
4126 @dfn{Bitwise Inclusive Or}.
4132 @dfn{Bitwise Exclusive Or}.
4135 @dfn{Bitwise Or Not}.
4142 @cindex addition, permitted arguments
4143 @cindex plus, permitted arguments
4144 @cindex arguments for addition
4146 @dfn{Addition}. If either argument is absolute, the result has the section of
4147 the other argument. You may not add together arguments from different
4150 @cindex subtraction, permitted arguments
4151 @cindex minus, permitted arguments
4152 @cindex arguments for subtraction
4154 @dfn{Subtraction}. If the right argument is absolute, the
4155 result has the section of the left argument.
4156 If both arguments are in the same section, the result is absolute.
4157 You may not subtract arguments from different sections.
4158 @c FIXME is there still something useful to say about undefined - undefined ?
4160 @cindex comparison expressions
4161 @cindex expressions, comparison
4166 @dfn{Is Not Equal To}
4170 @dfn{Is Greater Than}
4172 @dfn{Is Greater Than Or Equal To}
4174 @dfn{Is Less Than Or Equal To}
4176 The comparison operators can be used as infix operators. A true results has a
4177 value of -1 whereas a false result has a value of 0. Note, these operators
4178 perform signed comparisons.
4181 @item Lowest Precedence
4190 These two logical operations can be used to combine the results of sub
4191 expressions. Note, unlike the comparison operators a true result returns a
4192 value of 1 but a false results does still return 0. Also note that the logical
4193 or operator has a slightly lower precedence than logical and.
4198 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4199 address; you can only have a defined section in one of the two arguments.
4202 @chapter Assembler Directives
4204 @cindex directives, machine independent
4205 @cindex pseudo-ops, machine independent
4206 @cindex machine independent directives
4207 All assembler directives have names that begin with a period (@samp{.}).
4208 The names are case insensitive for most targets, and usually written
4211 This chapter discusses directives that are available regardless of the
4212 target machine configuration for the @sc{gnu} assembler.
4214 Some machine configurations provide additional directives.
4215 @xref{Machine Dependencies}.
4218 @ifset machine-directives
4219 @xref{Machine Dependencies}, for additional directives.
4224 * Abort:: @code{.abort}
4226 * ABORT (COFF):: @code{.ABORT}
4229 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4230 * Altmacro:: @code{.altmacro}
4231 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4232 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4233 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4234 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, etc
4235 * Byte:: @code{.byte @var{expressions}}
4236 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4237 * Comm:: @code{.comm @var{symbol} , @var{length} }
4238 * Data:: @code{.data @var{subsection}}
4240 * Def:: @code{.def @var{name}}
4243 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4249 * Double:: @code{.double @var{flonums}}
4250 * Eject:: @code{.eject}
4251 * Else:: @code{.else}
4252 * Elseif:: @code{.elseif}
4255 * Endef:: @code{.endef}
4258 * Endfunc:: @code{.endfunc}
4259 * Endif:: @code{.endif}
4260 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4261 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4262 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4264 * Error:: @code{.error @var{string}}
4265 * Exitm:: @code{.exitm}
4266 * Extern:: @code{.extern}
4267 * Fail:: @code{.fail}
4268 * File:: @code{.file}
4269 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4270 * Float:: @code{.float @var{flonums}}
4271 * Func:: @code{.func}
4272 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4274 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4275 * Hidden:: @code{.hidden @var{names}}
4278 * hword:: @code{.hword @var{expressions}}
4279 * Ident:: @code{.ident}
4280 * If:: @code{.if @var{absolute expression}}
4281 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4282 * Include:: @code{.include "@var{file}"}
4283 * Int:: @code{.int @var{expressions}}
4285 * Internal:: @code{.internal @var{names}}
4288 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4289 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4290 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4291 * Lflags:: @code{.lflags}
4292 @ifclear no-line-dir
4293 * Line:: @code{.line @var{line-number}}
4296 * Linkonce:: @code{.linkonce [@var{type}]}
4297 * List:: @code{.list}
4298 * Ln:: @code{.ln @var{line-number}}
4299 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4300 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4302 * Local:: @code{.local @var{names}}
4305 * Long:: @code{.long @var{expressions}}
4307 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4310 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4311 * MRI:: @code{.mri @var{val}}
4312 * Noaltmacro:: @code{.noaltmacro}
4313 * Nolist:: @code{.nolist}
4314 * Octa:: @code{.octa @var{bignums}}
4315 * Offset:: @code{.offset @var{loc}}
4316 * Org:: @code{.org @var{new-lc}, @var{fill}}
4317 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4319 * PopSection:: @code{.popsection}
4320 * Previous:: @code{.previous}
4323 * Print:: @code{.print @var{string}}
4325 * Protected:: @code{.protected @var{names}}
4328 * Psize:: @code{.psize @var{lines}, @var{columns}}
4329 * Purgem:: @code{.purgem @var{name}}
4331 * PushSection:: @code{.pushsection @var{name}}
4334 * Quad:: @code{.quad @var{bignums}}
4335 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4336 * Rept:: @code{.rept @var{count}}
4337 * Sbttl:: @code{.sbttl "@var{subheading}"}
4339 * Scl:: @code{.scl @var{class}}
4342 * Section:: @code{.section @var{name}[, @var{flags}]}
4345 * Set:: @code{.set @var{symbol}, @var{expression}}
4346 * Short:: @code{.short @var{expressions}}
4347 * Single:: @code{.single @var{flonums}}
4349 * Size:: @code{.size [@var{name} , @var{expression}]}
4351 @ifclear no-space-dir
4352 * Skip:: @code{.skip @var{size} , @var{fill}}
4355 * Sleb128:: @code{.sleb128 @var{expressions}}
4356 @ifclear no-space-dir
4357 * Space:: @code{.space @var{size} , @var{fill}}
4360 * Stab:: @code{.stabd, .stabn, .stabs}
4363 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4364 * Struct:: @code{.struct @var{expression}}
4366 * SubSection:: @code{.subsection}
4367 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4371 * Tag:: @code{.tag @var{structname}}
4374 * Text:: @code{.text @var{subsection}}
4375 * Title:: @code{.title "@var{heading}"}
4377 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4380 * Uleb128:: @code{.uleb128 @var{expressions}}
4382 * Val:: @code{.val @var{addr}}
4386 * Version:: @code{.version "@var{string}"}
4387 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4388 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4391 * Warning:: @code{.warning @var{string}}
4392 * Weak:: @code{.weak @var{names}}
4393 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4394 * Word:: @code{.word @var{expressions}}
4395 @ifclear no-space-dir
4396 * Zero:: @code{.zero @var{size}}
4398 * Deprecated:: Deprecated Directives
4402 @section @code{.abort}
4404 @cindex @code{abort} directive
4405 @cindex stopping the assembly
4406 This directive stops the assembly immediately. It is for
4407 compatibility with other assemblers. The original idea was that the
4408 assembly language source would be piped into the assembler. If the sender
4409 of the source quit, it could use this directive tells @command{@value{AS}} to
4410 quit also. One day @code{.abort} will not be supported.
4414 @section @code{.ABORT} (COFF)
4416 @cindex @code{ABORT} directive
4417 When producing COFF output, @command{@value{AS}} accepts this directive as a
4418 synonym for @samp{.abort}.
4421 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4427 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4429 @cindex padding the location counter
4430 @cindex @code{align} directive
4431 Pad the location counter (in the current subsection) to a particular storage
4432 boundary. The first expression (which must be absolute) is the alignment
4433 required, as described below.
4435 The second expression (also absolute) gives the fill value to be stored in the
4436 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4437 padding bytes are normally zero. However, on some systems, if the section is
4438 marked as containing code and the fill value is omitted, the space is filled
4439 with no-op instructions.
4441 The third expression is also absolute, and is also optional. If it is present,
4442 it is the maximum number of bytes that should be skipped by this alignment
4443 directive. If doing the alignment would require skipping more bytes than the
4444 specified maximum, then the alignment is not done at all. You can omit the
4445 fill value (the second argument) entirely by simply using two commas after the
4446 required alignment; this can be useful if you want the alignment to be filled
4447 with no-op instructions when appropriate.
4449 The way the required alignment is specified varies from system to system.
4450 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or1k,
4451 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4452 alignment request in bytes. For example @samp{.align 8} advances
4453 the location counter until it is a multiple of 8. If the location counter
4454 is already a multiple of 8, no change is needed. For the tic54x, the
4455 first expression is the alignment request in words.
4457 For other systems, including ppc, i386 using a.out format, arm and
4458 strongarm, it is the
4459 number of low-order zero bits the location counter must have after
4460 advancement. For example @samp{.align 3} advances the location
4461 counter until it a multiple of 8. If the location counter is already a
4462 multiple of 8, no change is needed.
4464 This inconsistency is due to the different behaviors of the various
4465 native assemblers for these systems which GAS must emulate.
4466 GAS also provides @code{.balign} and @code{.p2align} directives,
4467 described later, which have a consistent behavior across all
4468 architectures (but are specific to GAS).
4471 @section @code{.altmacro}
4472 Enable alternate macro mode, enabling:
4475 @item LOCAL @var{name} [ , @dots{} ]
4476 One additional directive, @code{LOCAL}, is available. It is used to
4477 generate a string replacement for each of the @var{name} arguments, and
4478 replace any instances of @var{name} in each macro expansion. The
4479 replacement string is unique in the assembly, and different for each
4480 separate macro expansion. @code{LOCAL} allows you to write macros that
4481 define symbols, without fear of conflict between separate macro expansions.
4483 @item String delimiters
4484 You can write strings delimited in these other ways besides
4485 @code{"@var{string}"}:
4488 @item '@var{string}'
4489 You can delimit strings with single-quote characters.
4491 @item <@var{string}>
4492 You can delimit strings with matching angle brackets.
4495 @item single-character string escape
4496 To include any single character literally in a string (even if the
4497 character would otherwise have some special meaning), you can prefix the
4498 character with @samp{!} (an exclamation mark). For example, you can
4499 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4501 @item Expression results as strings
4502 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4503 and use the result as a string.
4507 @section @code{.ascii "@var{string}"}@dots{}
4509 @cindex @code{ascii} directive
4510 @cindex string literals
4511 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4512 separated by commas. It assembles each string (with no automatic
4513 trailing zero byte) into consecutive addresses.
4516 @section @code{.asciz "@var{string}"}@dots{}
4518 @cindex @code{asciz} directive
4519 @cindex zero-terminated strings
4520 @cindex null-terminated strings
4521 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4522 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4525 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4527 @cindex padding the location counter given number of bytes
4528 @cindex @code{balign} directive
4529 Pad the location counter (in the current subsection) to a particular
4530 storage boundary. The first expression (which must be absolute) is the
4531 alignment request in bytes. For example @samp{.balign 8} advances
4532 the location counter until it is a multiple of 8. If the location counter
4533 is already a multiple of 8, no change is needed.
4535 The second expression (also absolute) gives the fill value to be stored in the
4536 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4537 padding bytes are normally zero. However, on some systems, if the section is
4538 marked as containing code and the fill value is omitted, the space is filled
4539 with no-op instructions.
4541 The third expression is also absolute, and is also optional. If it is present,
4542 it is the maximum number of bytes that should be skipped by this alignment
4543 directive. If doing the alignment would require skipping more bytes than the
4544 specified maximum, then the alignment is not done at all. You can omit the
4545 fill value (the second argument) entirely by simply using two commas after the
4546 required alignment; this can be useful if you want the alignment to be filled
4547 with no-op instructions when appropriate.
4549 @cindex @code{balignw} directive
4550 @cindex @code{balignl} directive
4551 The @code{.balignw} and @code{.balignl} directives are variants of the
4552 @code{.balign} directive. The @code{.balignw} directive treats the fill
4553 pattern as a two byte word value. The @code{.balignl} directives treats the
4554 fill pattern as a four byte longword value. For example, @code{.balignw
4555 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4556 filled in with the value 0x368d (the exact placement of the bytes depends upon
4557 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4560 @node Bundle directives
4561 @section Bundle directives
4562 @subsection @code{.bundle_align_mode @var{abs-expr}}
4563 @cindex @code{bundle_align_mode} directive
4565 @cindex instruction bundle
4566 @cindex aligned instruction bundle
4567 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4568 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4569 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4570 disabled (which is the default state). If the argument it not zero, it
4571 gives the size of an instruction bundle as a power of two (as for the
4572 @code{.p2align} directive, @pxref{P2align}).
4574 For some targets, it's an ABI requirement that no instruction may span a
4575 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4576 instructions that starts on an aligned boundary. For example, if
4577 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4578 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4579 effect, no single instruction may span a boundary between bundles. If an
4580 instruction would start too close to the end of a bundle for the length of
4581 that particular instruction to fit within the bundle, then the space at the
4582 end of that bundle is filled with no-op instructions so the instruction
4583 starts in the next bundle. As a corollary, it's an error if any single
4584 instruction's encoding is longer than the bundle size.
4586 @subsection @code{.bundle_lock} and @code{.bundle_unlock}
4587 @cindex @code{bundle_lock} directive
4588 @cindex @code{bundle_unlock} directive
4589 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4590 allow explicit control over instruction bundle padding. These directives
4591 are only valid when @code{.bundle_align_mode} has been used to enable
4592 aligned instruction bundle mode. It's an error if they appear when
4593 @code{.bundle_align_mode} has not been used at all, or when the last
4594 directive was @w{@code{.bundle_align_mode 0}}.
4596 @cindex bundle-locked
4597 For some targets, it's an ABI requirement that certain instructions may
4598 appear only as part of specified permissible sequences of multiple
4599 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4600 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4601 instruction sequence. For purposes of aligned instruction bundle mode, a
4602 sequence starting with @code{.bundle_lock} and ending with
4603 @code{.bundle_unlock} is treated as a single instruction. That is, the
4604 entire sequence must fit into a single bundle and may not span a bundle
4605 boundary. If necessary, no-op instructions will be inserted before the
4606 first instruction of the sequence so that the whole sequence starts on an
4607 aligned bundle boundary. It's an error if the sequence is longer than the
4610 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4611 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4612 nested. That is, a second @code{.bundle_lock} directive before the next
4613 @code{.bundle_unlock} directive has no effect except that it must be
4614 matched by another closing @code{.bundle_unlock} so that there is the
4615 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4618 @section @code{.byte @var{expressions}}
4620 @cindex @code{byte} directive
4621 @cindex integers, one byte
4622 @code{.byte} expects zero or more expressions, separated by commas.
4623 Each expression is assembled into the next byte.
4625 @node CFI directives
4626 @section CFI directives
4627 @subsection @code{.cfi_sections @var{section_list}}
4628 @cindex @code{cfi_sections} directive
4629 @code{.cfi_sections} may be used to specify whether CFI directives
4630 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4631 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4632 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4633 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4634 directive is not used is @code{.cfi_sections .eh_frame}.
4636 On targets that support compact unwinding tables these can be generated
4637 by specifying @code{.eh_frame_entry} instead of @code{.eh_frame}.
4639 @subsection @code{.cfi_startproc [simple]}
4640 @cindex @code{cfi_startproc} directive
4641 @code{.cfi_startproc} is used at the beginning of each function that
4642 should have an entry in @code{.eh_frame}. It initializes some internal
4643 data structures. Don't forget to close the function by
4644 @code{.cfi_endproc}.
4646 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4647 it also emits some architecture dependent initial CFI instructions.
4649 @subsection @code{.cfi_endproc}
4650 @cindex @code{cfi_endproc} directive
4651 @code{.cfi_endproc} is used at the end of a function where it closes its
4652 unwind entry previously opened by
4653 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4655 @subsection @code{.cfi_personality @var{encoding} [, @var{exp}]}
4656 @cindex @code{cfi_personality} directive
4657 @code{.cfi_personality} defines personality routine and its encoding.
4658 @var{encoding} must be a constant determining how the personality
4659 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4660 argument is not present, otherwise second argument should be
4661 a constant or a symbol name. When using indirect encodings,
4662 the symbol provided should be the location where personality
4663 can be loaded from, not the personality routine itself.
4664 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4665 no personality routine.
4667 @subsection @code{.cfi_personality_id @var{id}}
4668 @cindex @code{cfi_personality_id} directive
4669 @code{cfi_personality_id} defines a personality routine by its index as
4670 defined in a compact unwinding format.
4671 Only valid when generating compact EH frames (i.e.
4672 with @code{.cfi_sections eh_frame_entry}.
4674 @subsection @code{.cfi_fde_data [@var{opcode1} [, @dots{}]]}
4675 @cindex @code{cfi_fde_data} directive
4676 @code{cfi_fde_data} is used to describe the compact unwind opcodes to be
4677 used for the current function. These are emitted inline in the
4678 @code{.eh_frame_entry} section if small enough and there is no LSDA, or
4679 in the @code{.gnu.extab} section otherwise.
4680 Only valid when generating compact EH frames (i.e.
4681 with @code{.cfi_sections eh_frame_entry}.
4683 @subsection @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4684 @code{.cfi_lsda} defines LSDA and its encoding.
4685 @var{encoding} must be a constant determining how the LSDA
4686 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), the second
4687 argument is not present, otherwise the second argument should be a constant
4688 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4689 meaning that no LSDA is present.
4691 @subsection @code{.cfi_inline_lsda} [@var{align}]
4692 @code{.cfi_inline_lsda} marks the start of a LSDA data section and
4693 switches to the corresponding @code{.gnu.extab} section.
4694 Must be preceded by a CFI block containing a @code{.cfi_lsda} directive.
4695 Only valid when generating compact EH frames (i.e.
4696 with @code{.cfi_sections eh_frame_entry}.
4698 The table header and unwinding opcodes will be generated at this point,
4699 so that they are immediately followed by the LSDA data. The symbol
4700 referenced by the @code{.cfi_lsda} directive should still be defined
4701 in case a fallback FDE based encoding is used. The LSDA data is terminated
4702 by a section directive.
4704 The optional @var{align} argument specifies the alignment required.
4705 The alignment is specified as a power of two, as with the
4706 @code{.p2align} directive.
4708 @subsection @code{.cfi_def_cfa @var{register}, @var{offset}}
4709 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4710 address from @var{register} and add @var{offset} to it}.
4712 @subsection @code{.cfi_def_cfa_register @var{register}}
4713 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4714 now on @var{register} will be used instead of the old one. Offset
4717 @subsection @code{.cfi_def_cfa_offset @var{offset}}
4718 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4719 remains the same, but @var{offset} is new. Note that it is the
4720 absolute offset that will be added to a defined register to compute
4723 @subsection @code{.cfi_adjust_cfa_offset @var{offset}}
4724 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4725 value that is added/substracted from the previous offset.
4727 @subsection @code{.cfi_offset @var{register}, @var{offset}}
4728 Previous value of @var{register} is saved at offset @var{offset} from
4731 @subsection @code{.cfi_rel_offset @var{register}, @var{offset}}
4732 Previous value of @var{register} is saved at offset @var{offset} from
4733 the current CFA register. This is transformed to @code{.cfi_offset}
4734 using the known displacement of the CFA register from the CFA.
4735 This is often easier to use, because the number will match the
4736 code it's annotating.
4738 @subsection @code{.cfi_register @var{register1}, @var{register2}}
4739 Previous value of @var{register1} is saved in register @var{register2}.
4741 @subsection @code{.cfi_restore @var{register}}
4742 @code{.cfi_restore} says that the rule for @var{register} is now the
4743 same as it was at the beginning of the function, after all initial
4744 instruction added by @code{.cfi_startproc} were executed.
4746 @subsection @code{.cfi_undefined @var{register}}
4747 From now on the previous value of @var{register} can't be restored anymore.
4749 @subsection @code{.cfi_same_value @var{register}}
4750 Current value of @var{register} is the same like in the previous frame,
4751 i.e. no restoration needed.
4753 @subsection @code{.cfi_remember_state},
4754 First save all current rules for all registers by @code{.cfi_remember_state},
4755 then totally screw them up by subsequent @code{.cfi_*} directives and when
4756 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4757 the previous saved state.
4759 @subsection @code{.cfi_return_column @var{register}}
4760 Change return column @var{register}, i.e. the return address is either
4761 directly in @var{register} or can be accessed by rules for @var{register}.
4763 @subsection @code{.cfi_signal_frame}
4764 Mark current function as signal trampoline.
4766 @subsection @code{.cfi_window_save}
4767 SPARC register window has been saved.
4769 @subsection @code{.cfi_escape} @var{expression}[, @dots{}]
4770 Allows the user to add arbitrary bytes to the unwind info. One
4771 might use this to add OS-specific CFI opcodes, or generic CFI
4772 opcodes that GAS does not yet support.
4774 @subsection @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4775 The current value of @var{register} is @var{label}. The value of @var{label}
4776 will be encoded in the output file according to @var{encoding}; see the
4777 description of @code{.cfi_personality} for details on this encoding.
4779 The usefulness of equating a register to a fixed label is probably
4780 limited to the return address register. Here, it can be useful to
4781 mark a code segment that has only one return address which is reached
4782 by a direct branch and no copy of the return address exists in memory
4783 or another register.
4786 @section @code{.comm @var{symbol} , @var{length} }
4788 @cindex @code{comm} directive
4789 @cindex symbol, common
4790 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4791 common symbol in one object file may be merged with a defined or common symbol
4792 of the same name in another object file. If @code{@value{LD}} does not see a
4793 definition for the symbol--just one or more common symbols--then it will
4794 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4795 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4796 the same name, and they do not all have the same size, it will allocate space
4797 using the largest size.
4800 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4801 an optional third argument. This is the desired alignment of the symbol,
4802 specified for ELF as a byte boundary (for example, an alignment of 16 means
4803 that the least significant 4 bits of the address should be zero), and for PE
4804 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4805 boundary). The alignment must be an absolute expression, and it must be a
4806 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4807 common symbol, it will use the alignment when placing the symbol. If no
4808 alignment is specified, @command{@value{AS}} will set the alignment to the
4809 largest power of two less than or equal to the size of the symbol, up to a
4810 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4811 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4812 @samp{--section-alignment} option; image file sections in PE are aligned to
4813 multiples of 4096, which is far too large an alignment for ordinary variables.
4814 It is rather the default alignment for (non-debug) sections within object
4815 (@samp{*.o}) files, which are less strictly aligned.}.
4819 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4820 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4824 @section @code{.data @var{subsection}}
4826 @cindex @code{data} directive
4827 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4828 end of the data subsection numbered @var{subsection} (which is an
4829 absolute expression). If @var{subsection} is omitted, it defaults
4834 @section @code{.def @var{name}}
4836 @cindex @code{def} directive
4837 @cindex COFF symbols, debugging
4838 @cindex debugging COFF symbols
4839 Begin defining debugging information for a symbol @var{name}; the
4840 definition extends until the @code{.endef} directive is encountered.
4843 This directive is only observed when @command{@value{AS}} is configured for COFF
4844 format output; when producing @code{b.out}, @samp{.def} is recognized,
4851 @section @code{.desc @var{symbol}, @var{abs-expression}}
4853 @cindex @code{desc} directive
4854 @cindex COFF symbol descriptor
4855 @cindex symbol descriptor, COFF
4856 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4857 to the low 16 bits of an absolute expression.
4860 The @samp{.desc} directive is not available when @command{@value{AS}} is
4861 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4862 object format. For the sake of compatibility, @command{@value{AS}} accepts
4863 it, but produces no output, when configured for COFF.
4869 @section @code{.dim}
4871 @cindex @code{dim} directive
4872 @cindex COFF auxiliary symbol information
4873 @cindex auxiliary symbol information, COFF
4874 This directive is generated by compilers to include auxiliary debugging
4875 information in the symbol table. It is only permitted inside
4876 @code{.def}/@code{.endef} pairs.
4879 @samp{.dim} is only meaningful when generating COFF format output; when
4880 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4886 @section @code{.double @var{flonums}}
4888 @cindex @code{double} directive
4889 @cindex floating point numbers (double)
4890 @code{.double} expects zero or more flonums, separated by commas. It
4891 assembles floating point numbers.
4893 The exact kind of floating point numbers emitted depends on how
4894 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4898 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4899 in @sc{ieee} format.
4904 @section @code{.eject}
4906 @cindex @code{eject} directive
4907 @cindex new page, in listings
4908 @cindex page, in listings
4909 @cindex listing control: new page
4910 Force a page break at this point, when generating assembly listings.
4913 @section @code{.else}
4915 @cindex @code{else} directive
4916 @code{.else} is part of the @command{@value{AS}} support for conditional
4917 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4918 of code to be assembled if the condition for the preceding @code{.if}
4922 @section @code{.elseif}
4924 @cindex @code{elseif} directive
4925 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4926 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4927 @code{.if} block that would otherwise fill the entire @code{.else} section.
4930 @section @code{.end}
4932 @cindex @code{end} directive
4933 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4934 process anything in the file past the @code{.end} directive.
4938 @section @code{.endef}
4940 @cindex @code{endef} directive
4941 This directive flags the end of a symbol definition begun with
4945 @samp{.endef} is only meaningful when generating COFF format output; if
4946 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4947 directive but ignores it.
4952 @section @code{.endfunc}
4953 @cindex @code{endfunc} directive
4954 @code{.endfunc} marks the end of a function specified with @code{.func}.
4957 @section @code{.endif}
4959 @cindex @code{endif} directive
4960 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4961 it marks the end of a block of code that is only assembled
4962 conditionally. @xref{If,,@code{.if}}.
4965 @section @code{.equ @var{symbol}, @var{expression}}
4967 @cindex @code{equ} directive
4968 @cindex assigning values to symbols
4969 @cindex symbols, assigning values to
4970 This directive sets the value of @var{symbol} to @var{expression}.
4971 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4974 The syntax for @code{equ} on the HPPA is
4975 @samp{@var{symbol} .equ @var{expression}}.
4979 The syntax for @code{equ} on the Z80 is
4980 @samp{@var{symbol} equ @var{expression}}.
4981 On the Z80 it is an eror if @var{symbol} is already defined,
4982 but the symbol is not protected from later redefinition.
4983 Compare @ref{Equiv}.
4987 @section @code{.equiv @var{symbol}, @var{expression}}
4988 @cindex @code{equiv} directive
4989 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4990 the assembler will signal an error if @var{symbol} is already defined. Note a
4991 symbol which has been referenced but not actually defined is considered to be
4994 Except for the contents of the error message, this is roughly equivalent to
5001 plus it protects the symbol from later redefinition.
5004 @section @code{.eqv @var{symbol}, @var{expression}}
5005 @cindex @code{eqv} directive
5006 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
5007 evaluate the expression or any part of it immediately. Instead each time
5008 the resulting symbol is used in an expression, a snapshot of its current
5012 @section @code{.err}
5013 @cindex @code{err} directive
5014 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
5015 message and, unless the @option{-Z} option was used, it will not generate an
5016 object file. This can be used to signal an error in conditionally compiled code.
5019 @section @code{.error "@var{string}"}
5020 @cindex error directive
5022 Similarly to @code{.err}, this directive emits an error, but you can specify a
5023 string that will be emitted as the error message. If you don't specify the
5024 message, it defaults to @code{".error directive invoked in source file"}.
5025 @xref{Errors, ,Error and Warning Messages}.
5028 .error "This code has not been assembled and tested."
5032 @section @code{.exitm}
5033 Exit early from the current macro definition. @xref{Macro}.
5036 @section @code{.extern}
5038 @cindex @code{extern} directive
5039 @code{.extern} is accepted in the source program---for compatibility
5040 with other assemblers---but it is ignored. @command{@value{AS}} treats
5041 all undefined symbols as external.
5044 @section @code{.fail @var{expression}}
5046 @cindex @code{fail} directive
5047 Generates an error or a warning. If the value of the @var{expression} is 500
5048 or more, @command{@value{AS}} will print a warning message. If the value is less
5049 than 500, @command{@value{AS}} will print an error message. The message will
5050 include the value of @var{expression}. This can occasionally be useful inside
5051 complex nested macros or conditional assembly.
5054 @section @code{.file}
5055 @cindex @code{file} directive
5057 @ifclear no-file-dir
5058 There are two different versions of the @code{.file} directive. Targets
5059 that support DWARF2 line number information use the DWARF2 version of
5060 @code{.file}. Other targets use the default version.
5062 @subheading Default Version
5064 @cindex logical file name
5065 @cindex file name, logical
5066 This version of the @code{.file} directive tells @command{@value{AS}} that we
5067 are about to start a new logical file. The syntax is:
5073 @var{string} is the new file name. In general, the filename is
5074 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
5075 to specify an empty file name, you must give the quotes--@code{""}. This
5076 statement may go away in future: it is only recognized to be compatible with
5077 old @command{@value{AS}} programs.
5079 @subheading DWARF2 Version
5082 When emitting DWARF2 line number information, @code{.file} assigns filenames
5083 to the @code{.debug_line} file name table. The syntax is:
5086 .file @var{fileno} @var{filename}
5089 The @var{fileno} operand should be a unique positive integer to use as the
5090 index of the entry in the table. The @var{filename} operand is a C string
5093 The detail of filename indices is exposed to the user because the filename
5094 table is shared with the @code{.debug_info} section of the DWARF2 debugging
5095 information, and thus the user must know the exact indices that table
5099 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
5101 @cindex @code{fill} directive
5102 @cindex writing patterns in memory
5103 @cindex patterns, writing in memory
5104 @var{repeat}, @var{size} and @var{value} are absolute expressions.
5105 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
5106 may be zero or more. @var{Size} may be zero or more, but if it is
5107 more than 8, then it is deemed to have the value 8, compatible with
5108 other people's assemblers. The contents of each @var{repeat} bytes
5109 is taken from an 8-byte number. The highest order 4 bytes are
5110 zero. The lowest order 4 bytes are @var{value} rendered in the
5111 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
5112 Each @var{size} bytes in a repetition is taken from the lowest order
5113 @var{size} bytes of this number. Again, this bizarre behavior is
5114 compatible with other people's assemblers.
5116 @var{size} and @var{value} are optional.
5117 If the second comma and @var{value} are absent, @var{value} is
5118 assumed zero. If the first comma and following tokens are absent,
5119 @var{size} is assumed to be 1.
5122 @section @code{.float @var{flonums}}
5124 @cindex floating point numbers (single)
5125 @cindex @code{float} directive
5126 This directive assembles zero or more flonums, separated by commas. It
5127 has the same effect as @code{.single}.
5129 The exact kind of floating point numbers emitted depends on how
5130 @command{@value{AS}} is configured.
5131 @xref{Machine Dependencies}.
5135 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5136 in @sc{ieee} format.
5141 @section @code{.func @var{name}[,@var{label}]}
5142 @cindex @code{func} directive
5143 @code{.func} emits debugging information to denote function @var{name}, and
5144 is ignored unless the file is assembled with debugging enabled.
5145 Only @samp{--gstabs[+]} is currently supported.
5146 @var{label} is the entry point of the function and if omitted @var{name}
5147 prepended with the @samp{leading char} is used.
5148 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5149 All functions are currently defined to have @code{void} return type.
5150 The function must be terminated with @code{.endfunc}.
5153 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5155 @cindex @code{global} directive
5156 @cindex symbol, making visible to linker
5157 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5158 @var{symbol} in your partial program, its value is made available to
5159 other partial programs that are linked with it. Otherwise,
5160 @var{symbol} takes its attributes from a symbol of the same name
5161 from another file linked into the same program.
5163 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5164 compatibility with other assemblers.
5167 On the HPPA, @code{.global} is not always enough to make it accessible to other
5168 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5169 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5174 @section @code{.gnu_attribute @var{tag},@var{value}}
5175 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5178 @section @code{.hidden @var{names}}
5180 @cindex @code{hidden} directive
5182 This is one of the ELF visibility directives. The other two are
5183 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5184 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5186 This directive overrides the named symbols default visibility (which is set by
5187 their binding: local, global or weak). The directive sets the visibility to
5188 @code{hidden} which means that the symbols are not visible to other components.
5189 Such symbols are always considered to be @code{protected} as well.
5193 @section @code{.hword @var{expressions}}
5195 @cindex @code{hword} directive
5196 @cindex integers, 16-bit
5197 @cindex numbers, 16-bit
5198 @cindex sixteen bit integers
5199 This expects zero or more @var{expressions}, and emits
5200 a 16 bit number for each.
5203 This directive is a synonym for @samp{.short}; depending on the target
5204 architecture, it may also be a synonym for @samp{.word}.
5208 This directive is a synonym for @samp{.short}.
5211 This directive is a synonym for both @samp{.short} and @samp{.word}.
5216 @section @code{.ident}
5218 @cindex @code{ident} directive
5220 This directive is used by some assemblers to place tags in object files. The
5221 behavior of this directive varies depending on the target. When using the
5222 a.out object file format, @command{@value{AS}} simply accepts the directive for
5223 source-file compatibility with existing assemblers, but does not emit anything
5224 for it. When using COFF, comments are emitted to the @code{.comment} or
5225 @code{.rdata} section, depending on the target. When using ELF, comments are
5226 emitted to the @code{.comment} section.
5229 @section @code{.if @var{absolute expression}}
5231 @cindex conditional assembly
5232 @cindex @code{if} directive
5233 @code{.if} marks the beginning of a section of code which is only
5234 considered part of the source program being assembled if the argument
5235 (which must be an @var{absolute expression}) is non-zero. The end of
5236 the conditional section of code must be marked by @code{.endif}
5237 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5238 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5239 If you have several conditions to check, @code{.elseif} may be used to avoid
5240 nesting blocks if/else within each subsequent @code{.else} block.
5242 The following variants of @code{.if} are also supported:
5244 @cindex @code{ifdef} directive
5245 @item .ifdef @var{symbol}
5246 Assembles the following section of code if the specified @var{symbol}
5247 has been defined. Note a symbol which has been referenced but not yet defined
5248 is considered to be undefined.
5250 @cindex @code{ifb} directive
5251 @item .ifb @var{text}
5252 Assembles the following section of code if the operand is blank (empty).
5254 @cindex @code{ifc} directive
5255 @item .ifc @var{string1},@var{string2}
5256 Assembles the following section of code if the two strings are the same. The
5257 strings may be optionally quoted with single quotes. If they are not quoted,
5258 the first string stops at the first comma, and the second string stops at the
5259 end of the line. Strings which contain whitespace should be quoted. The
5260 string comparison is case sensitive.
5262 @cindex @code{ifeq} directive
5263 @item .ifeq @var{absolute expression}
5264 Assembles the following section of code if the argument is zero.
5266 @cindex @code{ifeqs} directive
5267 @item .ifeqs @var{string1},@var{string2}
5268 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5270 @cindex @code{ifge} directive
5271 @item .ifge @var{absolute expression}
5272 Assembles the following section of code if the argument is greater than or
5275 @cindex @code{ifgt} directive
5276 @item .ifgt @var{absolute expression}
5277 Assembles the following section of code if the argument is greater than zero.
5279 @cindex @code{ifle} directive
5280 @item .ifle @var{absolute expression}
5281 Assembles the following section of code if the argument is less than or equal
5284 @cindex @code{iflt} directive
5285 @item .iflt @var{absolute expression}
5286 Assembles the following section of code if the argument is less than zero.
5288 @cindex @code{ifnb} directive
5289 @item .ifnb @var{text}
5290 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5291 following section of code if the operand is non-blank (non-empty).
5293 @cindex @code{ifnc} directive
5294 @item .ifnc @var{string1},@var{string2}.
5295 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5296 following section of code if the two strings are not the same.
5298 @cindex @code{ifndef} directive
5299 @cindex @code{ifnotdef} directive
5300 @item .ifndef @var{symbol}
5301 @itemx .ifnotdef @var{symbol}
5302 Assembles the following section of code if the specified @var{symbol}
5303 has not been defined. Both spelling variants are equivalent. Note a symbol
5304 which has been referenced but not yet defined is considered to be undefined.
5306 @cindex @code{ifne} directive
5307 @item .ifne @var{absolute expression}
5308 Assembles the following section of code if the argument is not equal to zero
5309 (in other words, this is equivalent to @code{.if}).
5311 @cindex @code{ifnes} directive
5312 @item .ifnes @var{string1},@var{string2}
5313 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5314 following section of code if the two strings are not the same.
5318 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5320 @cindex @code{incbin} directive
5321 @cindex binary files, including
5322 The @code{incbin} directive includes @var{file} verbatim at the current
5323 location. You can control the search paths used with the @samp{-I} command-line
5324 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5327 The @var{skip} argument skips a number of bytes from the start of the
5328 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5329 read. Note that the data is not aligned in any way, so it is the user's
5330 responsibility to make sure that proper alignment is provided both before and
5331 after the @code{incbin} directive.
5334 @section @code{.include "@var{file}"}
5336 @cindex @code{include} directive
5337 @cindex supporting files, including
5338 @cindex files, including
5339 This directive provides a way to include supporting files at specified
5340 points in your source program. The code from @var{file} is assembled as
5341 if it followed the point of the @code{.include}; when the end of the
5342 included file is reached, assembly of the original file continues. You
5343 can control the search paths used with the @samp{-I} command-line option
5344 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5348 @section @code{.int @var{expressions}}
5350 @cindex @code{int} directive
5351 @cindex integers, 32-bit
5352 Expect zero or more @var{expressions}, of any section, separated by commas.
5353 For each expression, emit a number that, at run time, is the value of that
5354 expression. The byte order and bit size of the number depends on what kind
5355 of target the assembly is for.
5359 On most forms of the H8/300, @code{.int} emits 16-bit
5360 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5367 @section @code{.internal @var{names}}
5369 @cindex @code{internal} directive
5371 This is one of the ELF visibility directives. The other two are
5372 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5373 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5375 This directive overrides the named symbols default visibility (which is set by
5376 their binding: local, global or weak). The directive sets the visibility to
5377 @code{internal} which means that the symbols are considered to be @code{hidden}
5378 (i.e., not visible to other components), and that some extra, processor specific
5379 processing must also be performed upon the symbols as well.
5383 @section @code{.irp @var{symbol},@var{values}}@dots{}
5385 @cindex @code{irp} directive
5386 Evaluate a sequence of statements assigning different values to @var{symbol}.
5387 The sequence of statements starts at the @code{.irp} directive, and is
5388 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5389 set to @var{value}, and the sequence of statements is assembled. If no
5390 @var{value} is listed, the sequence of statements is assembled once, with
5391 @var{symbol} set to the null string. To refer to @var{symbol} within the
5392 sequence of statements, use @var{\symbol}.
5394 For example, assembling
5402 is equivalent to assembling
5410 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5413 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5415 @cindex @code{irpc} directive
5416 Evaluate a sequence of statements assigning different values to @var{symbol}.
5417 The sequence of statements starts at the @code{.irpc} directive, and is
5418 terminated by an @code{.endr} directive. For each character in @var{value},
5419 @var{symbol} is set to the character, and the sequence of statements is
5420 assembled. If no @var{value} is listed, the sequence of statements is
5421 assembled once, with @var{symbol} set to the null string. To refer to
5422 @var{symbol} within the sequence of statements, use @var{\symbol}.
5424 For example, assembling
5432 is equivalent to assembling
5440 For some caveats with the spelling of @var{symbol}, see also the discussion
5444 @section @code{.lcomm @var{symbol} , @var{length}}
5446 @cindex @code{lcomm} directive
5447 @cindex local common symbols
5448 @cindex symbols, local common
5449 Reserve @var{length} (an absolute expression) bytes for a local common
5450 denoted by @var{symbol}. The section and value of @var{symbol} are
5451 those of the new local common. The addresses are allocated in the bss
5452 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5453 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5454 not visible to @code{@value{LD}}.
5457 Some targets permit a third argument to be used with @code{.lcomm}. This
5458 argument specifies the desired alignment of the symbol in the bss section.
5462 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5463 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5467 @section @code{.lflags}
5469 @cindex @code{lflags} directive (ignored)
5470 @command{@value{AS}} accepts this directive, for compatibility with other
5471 assemblers, but ignores it.
5473 @ifclear no-line-dir
5475 @section @code{.line @var{line-number}}
5477 @cindex @code{line} directive
5478 @cindex logical line number
5480 Change the logical line number. @var{line-number} must be an absolute
5481 expression. The next line has that logical line number. Therefore any other
5482 statements on the current line (after a statement separator character) are
5483 reported as on logical line number @var{line-number} @minus{} 1. One day
5484 @command{@value{AS}} will no longer support this directive: it is recognized only
5485 for compatibility with existing assembler programs.
5488 Even though this is a directive associated with the @code{a.out} or
5489 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5490 when producing COFF output, and treats @samp{.line} as though it
5491 were the COFF @samp{.ln} @emph{if} it is found outside a
5492 @code{.def}/@code{.endef} pair.
5494 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5495 used by compilers to generate auxiliary symbol information for
5500 @section @code{.linkonce [@var{type}]}
5502 @cindex @code{linkonce} directive
5503 @cindex common sections
5504 Mark the current section so that the linker only includes a single copy of it.
5505 This may be used to include the same section in several different object files,
5506 but ensure that the linker will only include it once in the final output file.
5507 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5508 Duplicate sections are detected based on the section name, so it should be
5511 This directive is only supported by a few object file formats; as of this
5512 writing, the only object file format which supports it is the Portable
5513 Executable format used on Windows NT.
5515 The @var{type} argument is optional. If specified, it must be one of the
5516 following strings. For example:
5520 Not all types may be supported on all object file formats.
5524 Silently discard duplicate sections. This is the default.
5527 Warn if there are duplicate sections, but still keep only one copy.
5530 Warn if any of the duplicates have different sizes.
5533 Warn if any of the duplicates do not have exactly the same contents.
5537 @section @code{.list}
5539 @cindex @code{list} directive
5540 @cindex listing control, turning on
5541 Control (in conjunction with the @code{.nolist} directive) whether or
5542 not assembly listings are generated. These two directives maintain an
5543 internal counter (which is zero initially). @code{.list} increments the
5544 counter, and @code{.nolist} decrements it. Assembly listings are
5545 generated whenever the counter is greater than zero.
5547 By default, listings are disabled. When you enable them (with the
5548 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5549 the initial value of the listing counter is one.
5552 @section @code{.ln @var{line-number}}
5554 @cindex @code{ln} directive
5555 @ifclear no-line-dir
5556 @samp{.ln} is a synonym for @samp{.line}.
5559 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5560 must be an absolute expression. The next line has that logical
5561 line number, so any other statements on the current line (after a
5562 statement separator character @code{;}) are reported as on logical
5563 line number @var{line-number} @minus{} 1.
5566 This directive is accepted, but ignored, when @command{@value{AS}} is
5567 configured for @code{b.out}; its effect is only associated with COFF
5573 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5574 @cindex @code{loc} directive
5575 When emitting DWARF2 line number information,
5576 the @code{.loc} directive will add a row to the @code{.debug_line} line
5577 number matrix corresponding to the immediately following assembly
5578 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5579 arguments will be applied to the @code{.debug_line} state machine before
5582 The @var{options} are a sequence of the following tokens in any order:
5586 This option will set the @code{basic_block} register in the
5587 @code{.debug_line} state machine to @code{true}.
5590 This option will set the @code{prologue_end} register in the
5591 @code{.debug_line} state machine to @code{true}.
5593 @item epilogue_begin
5594 This option will set the @code{epilogue_begin} register in the
5595 @code{.debug_line} state machine to @code{true}.
5597 @item is_stmt @var{value}
5598 This option will set the @code{is_stmt} register in the
5599 @code{.debug_line} state machine to @code{value}, which must be
5602 @item isa @var{value}
5603 This directive will set the @code{isa} register in the @code{.debug_line}
5604 state machine to @var{value}, which must be an unsigned integer.
5606 @item discriminator @var{value}
5607 This directive will set the @code{discriminator} register in the @code{.debug_line}
5608 state machine to @var{value}, which must be an unsigned integer.
5612 @node Loc_mark_labels
5613 @section @code{.loc_mark_labels @var{enable}}
5614 @cindex @code{loc_mark_labels} directive
5615 When emitting DWARF2 line number information,
5616 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5617 to the @code{.debug_line} line number matrix with the @code{basic_block}
5618 register in the state machine set whenever a code label is seen.
5619 The @var{enable} argument should be either 1 or 0, to enable or disable
5620 this function respectively.
5624 @section @code{.local @var{names}}
5626 @cindex @code{local} directive
5627 This directive, which is available for ELF targets, marks each symbol in
5628 the comma-separated list of @code{names} as a local symbol so that it
5629 will not be externally visible. If the symbols do not already exist,
5630 they will be created.
5632 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5633 accept an alignment argument, which is the case for most ELF targets,
5634 the @code{.local} directive can be used in combination with @code{.comm}
5635 (@pxref{Comm}) to define aligned local common data.
5639 @section @code{.long @var{expressions}}
5641 @cindex @code{long} directive
5642 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5645 @c no one seems to know what this is for or whether this description is
5646 @c what it really ought to do
5648 @section @code{.lsym @var{symbol}, @var{expression}}
5650 @cindex @code{lsym} directive
5651 @cindex symbol, not referenced in assembly
5652 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5653 the hash table, ensuring it cannot be referenced by name during the
5654 rest of the assembly. This sets the attributes of the symbol to be
5655 the same as the expression value:
5657 @var{other} = @var{descriptor} = 0
5658 @var{type} = @r{(section of @var{expression})}
5659 @var{value} = @var{expression}
5662 The new symbol is not flagged as external.
5666 @section @code{.macro}
5669 The commands @code{.macro} and @code{.endm} allow you to define macros that
5670 generate assembly output. For example, this definition specifies a macro
5671 @code{sum} that puts a sequence of numbers into memory:
5674 .macro sum from=0, to=5
5683 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5695 @item .macro @var{macname}
5696 @itemx .macro @var{macname} @var{macargs} @dots{}
5697 @cindex @code{macro} directive
5698 Begin the definition of a macro called @var{macname}. If your macro
5699 definition requires arguments, specify their names after the macro name,
5700 separated by commas or spaces. You can qualify the macro argument to
5701 indicate whether all invocations must specify a non-blank value (through
5702 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5703 (through @samp{:@code{vararg}}). You can supply a default value for any
5704 macro argument by following the name with @samp{=@var{deflt}}. You
5705 cannot define two macros with the same @var{macname} unless it has been
5706 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5707 definitions. For example, these are all valid @code{.macro} statements:
5711 Begin the definition of a macro called @code{comm}, which takes no
5714 @item .macro plus1 p, p1
5715 @itemx .macro plus1 p p1
5716 Either statement begins the definition of a macro called @code{plus1},
5717 which takes two arguments; within the macro definition, write
5718 @samp{\p} or @samp{\p1} to evaluate the arguments.
5720 @item .macro reserve_str p1=0 p2
5721 Begin the definition of a macro called @code{reserve_str}, with two
5722 arguments. The first argument has a default value, but not the second.
5723 After the definition is complete, you can call the macro either as
5724 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5725 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5726 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5727 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5729 @item .macro m p1:req, p2=0, p3:vararg
5730 Begin the definition of a macro called @code{m}, with at least three
5731 arguments. The first argument must always have a value specified, but
5732 not the second, which instead has a default value. The third formal
5733 will get assigned all remaining arguments specified at invocation time.
5735 When you call a macro, you can specify the argument values either by
5736 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5737 @samp{sum to=17, from=9}.
5741 Note that since each of the @var{macargs} can be an identifier exactly
5742 as any other one permitted by the target architecture, there may be
5743 occasional problems if the target hand-crafts special meanings to certain
5744 characters when they occur in a special position. For example, if the colon
5745 (@code{:}) is generally permitted to be part of a symbol name, but the
5746 architecture specific code special-cases it when occurring as the final
5747 character of a symbol (to denote a label), then the macro parameter
5748 replacement code will have no way of knowing that and consider the whole
5749 construct (including the colon) an identifier, and check only this
5750 identifier for being the subject to parameter substitution. So for example
5751 this macro definition:
5759 might not work as expected. Invoking @samp{label foo} might not create a label
5760 called @samp{foo} but instead just insert the text @samp{\l:} into the
5761 assembler source, probably generating an error about an unrecognised
5764 Similarly problems might occur with the period character (@samp{.})
5765 which is often allowed inside opcode names (and hence identifier names). So
5766 for example constructing a macro to build an opcode from a base name and a
5767 length specifier like this:
5770 .macro opcode base length
5775 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5776 instruction but instead generate some kind of error as the assembler tries to
5777 interpret the text @samp{\base.\length}.
5779 There are several possible ways around this problem:
5782 @item Insert white space
5783 If it is possible to use white space characters then this is the simplest
5792 @item Use @samp{\()}
5793 The string @samp{\()} can be used to separate the end of a macro argument from
5794 the following text. eg:
5797 .macro opcode base length
5802 @item Use the alternate macro syntax mode
5803 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5804 used as a separator. eg:
5814 Note: this problem of correctly identifying string parameters to pseudo ops
5815 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5816 and @code{.irpc} (@pxref{Irpc}) as well.
5819 @cindex @code{endm} directive
5820 Mark the end of a macro definition.
5823 @cindex @code{exitm} directive
5824 Exit early from the current macro definition.
5826 @cindex number of macros executed
5827 @cindex macros, count executed
5829 @command{@value{AS}} maintains a counter of how many macros it has
5830 executed in this pseudo-variable; you can copy that number to your
5831 output with @samp{\@@}, but @emph{only within a macro definition}.
5833 @item LOCAL @var{name} [ , @dots{} ]
5834 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5835 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5836 @xref{Altmacro,,@code{.altmacro}}.
5840 @section @code{.mri @var{val}}
5842 @cindex @code{mri} directive
5843 @cindex MRI mode, temporarily
5844 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5845 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5846 affects code assembled until the next @code{.mri} directive, or until the end
5847 of the file. @xref{M, MRI mode, MRI mode}.
5850 @section @code{.noaltmacro}
5851 Disable alternate macro mode. @xref{Altmacro}.
5854 @section @code{.nolist}
5856 @cindex @code{nolist} directive
5857 @cindex listing control, turning off
5858 Control (in conjunction with the @code{.list} directive) whether or
5859 not assembly listings are generated. These two directives maintain an
5860 internal counter (which is zero initially). @code{.list} increments the
5861 counter, and @code{.nolist} decrements it. Assembly listings are
5862 generated whenever the counter is greater than zero.
5865 @section @code{.octa @var{bignums}}
5867 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5868 @cindex @code{octa} directive
5869 @cindex integer, 16-byte
5870 @cindex sixteen byte integer
5871 This directive expects zero or more bignums, separated by commas. For each
5872 bignum, it emits a 16-byte integer.
5874 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5875 hence @emph{octa}-word for 16 bytes.
5878 @section @code{.offset @var{loc}}
5880 @cindex @code{offset} directive
5881 Set the location counter to @var{loc} in the absolute section. @var{loc} must
5882 be an absolute expression. This directive may be useful for defining
5883 symbols with absolute values. Do not confuse it with the @code{.org}
5887 @section @code{.org @var{new-lc} , @var{fill}}
5889 @cindex @code{org} directive
5890 @cindex location counter, advancing
5891 @cindex advancing location counter
5892 @cindex current address, advancing
5893 Advance the location counter of the current section to
5894 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5895 expression with the same section as the current subsection. That is,
5896 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5897 wrong section, the @code{.org} directive is ignored. To be compatible
5898 with former assemblers, if the section of @var{new-lc} is absolute,
5899 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5900 is the same as the current subsection.
5902 @code{.org} may only increase the location counter, or leave it
5903 unchanged; you cannot use @code{.org} to move the location counter
5906 @c double negative used below "not undefined" because this is a specific
5907 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5908 @c section. doc@cygnus.com 18feb91
5909 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5910 may not be undefined. If you really detest this restriction we eagerly await
5911 a chance to share your improved assembler.
5913 Beware that the origin is relative to the start of the section, not
5914 to the start of the subsection. This is compatible with other
5915 people's assemblers.
5917 When the location counter (of the current subsection) is advanced, the
5918 intervening bytes are filled with @var{fill} which should be an
5919 absolute expression. If the comma and @var{fill} are omitted,
5920 @var{fill} defaults to zero.
5923 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5925 @cindex padding the location counter given a power of two
5926 @cindex @code{p2align} directive
5927 Pad the location counter (in the current subsection) to a particular
5928 storage boundary. The first expression (which must be absolute) is the
5929 number of low-order zero bits the location counter must have after
5930 advancement. For example @samp{.p2align 3} advances the location
5931 counter until it a multiple of 8. If the location counter is already a
5932 multiple of 8, no change is needed.
5934 The second expression (also absolute) gives the fill value to be stored in the
5935 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5936 padding bytes are normally zero. However, on some systems, if the section is
5937 marked as containing code and the fill value is omitted, the space is filled
5938 with no-op instructions.
5940 The third expression is also absolute, and is also optional. If it is present,
5941 it is the maximum number of bytes that should be skipped by this alignment
5942 directive. If doing the alignment would require skipping more bytes than the
5943 specified maximum, then the alignment is not done at all. You can omit the
5944 fill value (the second argument) entirely by simply using two commas after the
5945 required alignment; this can be useful if you want the alignment to be filled
5946 with no-op instructions when appropriate.
5948 @cindex @code{p2alignw} directive
5949 @cindex @code{p2alignl} directive
5950 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5951 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5952 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5953 fill pattern as a four byte longword value. For example, @code{.p2alignw
5954 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5955 filled in with the value 0x368d (the exact placement of the bytes depends upon
5956 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5961 @section @code{.popsection}
5963 @cindex @code{popsection} directive
5964 @cindex Section Stack
5965 This is one of the ELF section stack manipulation directives. The others are
5966 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5967 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5970 This directive replaces the current section (and subsection) with the top
5971 section (and subsection) on the section stack. This section is popped off the
5977 @section @code{.previous}
5979 @cindex @code{previous} directive
5980 @cindex Section Stack
5981 This is one of the ELF section stack manipulation directives. The others are
5982 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5983 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5984 (@pxref{PopSection}).
5986 This directive swaps the current section (and subsection) with most recently
5987 referenced section/subsection pair prior to this one. Multiple
5988 @code{.previous} directives in a row will flip between two sections (and their
5989 subsections). For example:
6001 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
6007 # Now in section A subsection 1
6011 # Now in section B subsection 0
6014 # Now in section B subsection 1
6017 # Now in section B subsection 0
6021 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
6022 section B and 0x9abc into subsection 1 of section B.
6024 In terms of the section stack, this directive swaps the current section with
6025 the top section on the section stack.
6029 @section @code{.print @var{string}}
6031 @cindex @code{print} directive
6032 @command{@value{AS}} will print @var{string} on the standard output during
6033 assembly. You must put @var{string} in double quotes.
6037 @section @code{.protected @var{names}}
6039 @cindex @code{protected} directive
6041 This is one of the ELF visibility directives. The other two are
6042 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
6044 This directive overrides the named symbols default visibility (which is set by
6045 their binding: local, global or weak). The directive sets the visibility to
6046 @code{protected} which means that any references to the symbols from within the
6047 components that defines them must be resolved to the definition in that
6048 component, even if a definition in another component would normally preempt
6053 @section @code{.psize @var{lines} , @var{columns}}
6055 @cindex @code{psize} directive
6056 @cindex listing control: paper size
6057 @cindex paper size, for listings
6058 Use this directive to declare the number of lines---and, optionally, the
6059 number of columns---to use for each page, when generating listings.
6061 If you do not use @code{.psize}, listings use a default line-count
6062 of 60. You may omit the comma and @var{columns} specification; the
6063 default width is 200 columns.
6065 @command{@value{AS}} generates formfeeds whenever the specified number of
6066 lines is exceeded (or whenever you explicitly request one, using
6069 If you specify @var{lines} as @code{0}, no formfeeds are generated save
6070 those explicitly specified with @code{.eject}.
6073 @section @code{.purgem @var{name}}
6075 @cindex @code{purgem} directive
6076 Undefine the macro @var{name}, so that later uses of the string will not be
6077 expanded. @xref{Macro}.
6081 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
6083 @cindex @code{pushsection} directive
6084 @cindex Section Stack
6085 This is one of the ELF section stack manipulation directives. The others are
6086 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6087 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6090 This directive pushes the current section (and subsection) onto the
6091 top of the section stack, and then replaces the current section and
6092 subsection with @code{name} and @code{subsection}. The optional
6093 @code{flags}, @code{type} and @code{arguments} are treated the same
6094 as in the @code{.section} (@pxref{Section}) directive.
6098 @section @code{.quad @var{bignums}}
6100 @cindex @code{quad} directive
6101 @code{.quad} expects zero or more bignums, separated by commas. For
6102 each bignum, it emits
6104 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
6105 warning message; and just takes the lowest order 8 bytes of the bignum.
6106 @cindex eight-byte integer
6107 @cindex integer, 8-byte
6109 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
6110 hence @emph{quad}-word for 8 bytes.
6113 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
6114 warning message; and just takes the lowest order 16 bytes of the bignum.
6115 @cindex sixteen-byte integer
6116 @cindex integer, 16-byte
6120 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6122 @cindex @code{reloc} directive
6123 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6124 @var{expression}. If @var{offset} is a number, the relocation is generated in
6125 the current section. If @var{offset} is an expression that resolves to a
6126 symbol plus offset, the relocation is generated in the given symbol's section.
6127 @var{expression}, if present, must resolve to a symbol plus addend or to an
6128 absolute value, but note that not all targets support an addend. e.g. ELF REL
6129 targets such as i386 store an addend in the section contents rather than in the
6130 relocation. This low level interface does not support addends stored in the
6134 @section @code{.rept @var{count}}
6136 @cindex @code{rept} directive
6137 Repeat the sequence of lines between the @code{.rept} directive and the next
6138 @code{.endr} directive @var{count} times.
6140 For example, assembling
6148 is equivalent to assembling
6157 @section @code{.sbttl "@var{subheading}"}
6159 @cindex @code{sbttl} directive
6160 @cindex subtitles for listings
6161 @cindex listing control: subtitle
6162 Use @var{subheading} as the title (third line, immediately after the
6163 title line) when generating assembly listings.
6165 This directive affects subsequent pages, as well as the current page if
6166 it appears within ten lines of the top of a page.
6170 @section @code{.scl @var{class}}
6172 @cindex @code{scl} directive
6173 @cindex symbol storage class (COFF)
6174 @cindex COFF symbol storage class
6175 Set the storage-class value for a symbol. This directive may only be
6176 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6177 whether a symbol is static or external, or it may record further
6178 symbolic debugging information.
6181 The @samp{.scl} directive is primarily associated with COFF output; when
6182 configured to generate @code{b.out} output format, @command{@value{AS}}
6183 accepts this directive but ignores it.
6189 @section @code{.section @var{name}}
6191 @cindex named section
6192 Use the @code{.section} directive to assemble the following code into a section
6195 This directive is only supported for targets that actually support arbitrarily
6196 named sections; on @code{a.out} targets, for example, it is not accepted, even
6197 with a standard @code{a.out} section name.
6201 @c only print the extra heading if both COFF and ELF are set
6202 @subheading COFF Version
6205 @cindex @code{section} directive (COFF version)
6206 For COFF targets, the @code{.section} directive is used in one of the following
6210 .section @var{name}[, "@var{flags}"]
6211 .section @var{name}[, @var{subsection}]
6214 If the optional argument is quoted, it is taken as flags to use for the
6215 section. Each flag is a single character. The following flags are recognized:
6218 bss section (uninitialized data)
6220 section is not loaded
6226 exclude section from linking
6232 shared section (meaningful for PE targets)
6234 ignored. (For compatibility with the ELF version)
6236 section is not readable (meaningful for PE targets)
6238 single-digit power-of-two section alignment (GNU extension)
6241 If no flags are specified, the default flags depend upon the section name. If
6242 the section name is not recognized, the default will be for the section to be
6243 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6244 from the section, rather than adding them, so if they are used on their own it
6245 will be as if no flags had been specified at all.
6247 If the optional argument to the @code{.section} directive is not quoted, it is
6248 taken as a subsection number (@pxref{Sub-Sections}).
6253 @c only print the extra heading if both COFF and ELF are set
6254 @subheading ELF Version
6257 @cindex Section Stack
6258 This is one of the ELF section stack manipulation directives. The others are
6259 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6260 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6261 @code{.previous} (@pxref{Previous}).
6263 @cindex @code{section} directive (ELF version)
6264 For ELF targets, the @code{.section} directive is used like this:
6267 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6270 @anchor{Section Name Substitutions}
6271 @kindex --sectname-subst
6272 @cindex section name substitution
6273 If the @samp{--sectname-subst} command-line option is provided, the @var{name}
6274 argument may contain a substitution sequence. Only @code{%S} is supported
6275 at the moment, and substitutes the current section name. For example:
6278 .macro exception_code
6279 .section %S.exception
6280 [exception code here]
6295 The two @code{exception_code} invocations above would create the
6296 @code{.text.exception} and @code{.init.exception} sections respectively.
6297 This is useful e.g. to discriminate between anciliary sections that are
6298 tied to setup code to be discarded after use from anciliary sections that
6299 need to stay resident without having to define multiple @code{exception_code}
6300 macros just for that purpose.
6302 The optional @var{flags} argument is a quoted string which may contain any
6303 combination of the following characters:
6306 section is allocatable
6308 section is excluded from executable and shared library.
6312 section is executable
6314 section is mergeable
6316 section contains zero terminated strings
6318 section is a member of a section group
6320 section is used for thread-local-storage
6322 section is a member of the previously-current section's group, if any
6325 The optional @var{type} argument may contain one of the following constants:
6328 section contains data
6330 section does not contain data (i.e., section only occupies space)
6332 section contains data which is used by things other than the program
6334 section contains an array of pointers to init functions
6336 section contains an array of pointers to finish functions
6337 @item @@preinit_array
6338 section contains an array of pointers to pre-init functions
6341 Many targets only support the first three section types.
6343 Note on targets where the @code{@@} character is the start of a comment (eg
6344 ARM) then another character is used instead. For example the ARM port uses the
6347 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6348 be specified as well as an extra argument---@var{entsize}---like this:
6351 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6354 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6355 constants, each @var{entsize} octets long. Sections with both @code{M} and
6356 @code{S} must contain zero terminated strings where each character is
6357 @var{entsize} bytes long. The linker may remove duplicates within sections with
6358 the same name, same entity size and same flags. @var{entsize} must be an
6359 absolute expression. For sections with both @code{M} and @code{S}, a string
6360 which is a suffix of a larger string is considered a duplicate. Thus
6361 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6362 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6364 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6365 be present along with an additional field like this:
6368 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6371 The @var{GroupName} field specifies the name of the section group to which this
6372 particular section belongs. The optional linkage field can contain:
6375 indicates that only one copy of this section should be retained
6380 Note: if both the @var{M} and @var{G} flags are present then the fields for
6381 the Merge flag should come first, like this:
6384 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6387 If @var{flags} contains the @code{?} symbol then it may not also contain the
6388 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6389 present. Instead, @code{?} says to consider the section that's current before
6390 this directive. If that section used @code{G}, then the new section will use
6391 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6392 If not, then the @code{?} symbol has no effect.
6394 If no flags are specified, the default flags depend upon the section name. If
6395 the section name is not recognized, the default will be for the section to have
6396 none of the above flags: it will not be allocated in memory, nor writable, nor
6397 executable. The section will contain data.
6399 For ELF targets, the assembler supports another type of @code{.section}
6400 directive for compatibility with the Solaris assembler:
6403 .section "@var{name}"[, @var{flags}...]
6406 Note that the section name is quoted. There may be a sequence of comma
6410 section is allocatable
6414 section is executable
6416 section is excluded from executable and shared library.
6418 section is used for thread local storage
6421 This directive replaces the current section and subsection. See the
6422 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6423 some examples of how this directive and the other section stack directives
6429 @section @code{.set @var{symbol}, @var{expression}}
6431 @cindex @code{set} directive
6432 @cindex symbol value, setting
6433 Set the value of @var{symbol} to @var{expression}. This
6434 changes @var{symbol}'s value and type to conform to
6435 @var{expression}. If @var{symbol} was flagged as external, it remains
6436 flagged (@pxref{Symbol Attributes}).
6438 You may @code{.set} a symbol many times in the same assembly provided that the
6439 values given to the symbol are constants. Values that are based on expressions
6440 involving other symbols are allowed, but some targets may restrict this to only
6441 being done once per assembly. This is because those targets do not set the
6442 addresses of symbols at assembly time, but rather delay the assignment until a
6443 final link is performed. This allows the linker a chance to change the code in
6444 the files, changing the location of, and the relative distance between, various
6447 If you @code{.set} a global symbol, the value stored in the object
6448 file is the last value stored into it.
6451 On Z80 @code{set} is a real instruction, use
6452 @samp{@var{symbol} defl @var{expression}} instead.
6456 @section @code{.short @var{expressions}}
6458 @cindex @code{short} directive
6460 @code{.short} is normally the same as @samp{.word}.
6461 @xref{Word,,@code{.word}}.
6463 In some configurations, however, @code{.short} and @code{.word} generate
6464 numbers of different lengths. @xref{Machine Dependencies}.
6468 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6471 This expects zero or more @var{expressions}, and emits
6472 a 16 bit number for each.
6477 @section @code{.single @var{flonums}}
6479 @cindex @code{single} directive
6480 @cindex floating point numbers (single)
6481 This directive assembles zero or more flonums, separated by commas. It
6482 has the same effect as @code{.float}.
6484 The exact kind of floating point numbers emitted depends on how
6485 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6489 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6490 numbers in @sc{ieee} format.
6496 @section @code{.size}
6498 This directive is used to set the size associated with a symbol.
6502 @c only print the extra heading if both COFF and ELF are set
6503 @subheading COFF Version
6506 @cindex @code{size} directive (COFF version)
6507 For COFF targets, the @code{.size} directive is only permitted inside
6508 @code{.def}/@code{.endef} pairs. It is used like this:
6511 .size @var{expression}
6515 @samp{.size} is only meaningful when generating COFF format output; when
6516 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6523 @c only print the extra heading if both COFF and ELF are set
6524 @subheading ELF Version
6527 @cindex @code{size} directive (ELF version)
6528 For ELF targets, the @code{.size} directive is used like this:
6531 .size @var{name} , @var{expression}
6534 This directive sets the size associated with a symbol @var{name}.
6535 The size in bytes is computed from @var{expression} which can make use of label
6536 arithmetic. This directive is typically used to set the size of function
6541 @ifclear no-space-dir
6543 @section @code{.skip @var{size} , @var{fill}}
6545 @cindex @code{skip} directive
6546 @cindex filling memory
6547 This directive emits @var{size} bytes, each of value @var{fill}. Both
6548 @var{size} and @var{fill} are absolute expressions. If the comma and
6549 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6554 @section @code{.sleb128 @var{expressions}}
6556 @cindex @code{sleb128} directive
6557 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6558 compact, variable length representation of numbers used by the DWARF
6559 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6561 @ifclear no-space-dir
6563 @section @code{.space @var{size} , @var{fill}}
6565 @cindex @code{space} directive
6566 @cindex filling memory
6567 This directive emits @var{size} bytes, each of value @var{fill}. Both
6568 @var{size} and @var{fill} are absolute expressions. If the comma
6569 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6574 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6575 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6576 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6577 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6585 @section @code{.stabd, .stabn, .stabs}
6587 @cindex symbolic debuggers, information for
6588 @cindex @code{stab@var{x}} directives
6589 There are three directives that begin @samp{.stab}.
6590 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6591 The symbols are not entered in the @command{@value{AS}} hash table: they
6592 cannot be referenced elsewhere in the source file.
6593 Up to five fields are required:
6597 This is the symbol's name. It may contain any character except
6598 @samp{\000}, so is more general than ordinary symbol names. Some
6599 debuggers used to code arbitrarily complex structures into symbol names
6603 An absolute expression. The symbol's type is set to the low 8 bits of
6604 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6605 and debuggers choke on silly bit patterns.
6608 An absolute expression. The symbol's ``other'' attribute is set to the
6609 low 8 bits of this expression.
6612 An absolute expression. The symbol's descriptor is set to the low 16
6613 bits of this expression.
6616 An absolute expression which becomes the symbol's value.
6619 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6620 or @code{.stabs} statement, the symbol has probably already been created;
6621 you get a half-formed symbol in your object file. This is
6622 compatible with earlier assemblers!
6625 @cindex @code{stabd} directive
6626 @item .stabd @var{type} , @var{other} , @var{desc}
6628 The ``name'' of the symbol generated is not even an empty string.
6629 It is a null pointer, for compatibility. Older assemblers used a
6630 null pointer so they didn't waste space in object files with empty
6633 The symbol's value is set to the location counter,
6634 relocatably. When your program is linked, the value of this symbol
6635 is the address of the location counter when the @code{.stabd} was
6638 @cindex @code{stabn} directive
6639 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6640 The name of the symbol is set to the empty string @code{""}.
6642 @cindex @code{stabs} directive
6643 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6644 All five fields are specified.
6650 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6651 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6653 @cindex string, copying to object file
6654 @cindex string8, copying to object file
6655 @cindex string16, copying to object file
6656 @cindex string32, copying to object file
6657 @cindex string64, copying to object file
6658 @cindex @code{string} directive
6659 @cindex @code{string8} directive
6660 @cindex @code{string16} directive
6661 @cindex @code{string32} directive
6662 @cindex @code{string64} directive
6664 Copy the characters in @var{str} to the object file. You may specify more than
6665 one string to copy, separated by commas. Unless otherwise specified for a
6666 particular machine, the assembler marks the end of each string with a 0 byte.
6667 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6669 The variants @code{string16}, @code{string32} and @code{string64} differ from
6670 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6671 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6672 are stored in target endianness byte order.
6678 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6679 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6684 @section @code{.struct @var{expression}}
6686 @cindex @code{struct} directive
6687 Switch to the absolute section, and set the section offset to @var{expression},
6688 which must be an absolute expression. You might use this as follows:
6697 This would define the symbol @code{field1} to have the value 0, the symbol
6698 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6699 value 8. Assembly would be left in the absolute section, and you would need to
6700 use a @code{.section} directive of some sort to change to some other section
6701 before further assembly.
6705 @section @code{.subsection @var{name}}
6707 @cindex @code{subsection} directive
6708 @cindex Section Stack
6709 This is one of the ELF section stack manipulation directives. The others are
6710 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6711 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6714 This directive replaces the current subsection with @code{name}. The current
6715 section is not changed. The replaced subsection is put onto the section stack
6716 in place of the then current top of stack subsection.
6721 @section @code{.symver}
6722 @cindex @code{symver} directive
6723 @cindex symbol versioning
6724 @cindex versions of symbols
6725 Use the @code{.symver} directive to bind symbols to specific version nodes
6726 within a source file. This is only supported on ELF platforms, and is
6727 typically used when assembling files to be linked into a shared library.
6728 There are cases where it may make sense to use this in objects to be bound
6729 into an application itself so as to override a versioned symbol from a
6732 For ELF targets, the @code{.symver} directive can be used like this:
6734 .symver @var{name}, @var{name2@@nodename}
6736 If the symbol @var{name} is defined within the file
6737 being assembled, the @code{.symver} directive effectively creates a symbol
6738 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6739 just don't try and create a regular alias is that the @var{@@} character isn't
6740 permitted in symbol names. The @var{name2} part of the name is the actual name
6741 of the symbol by which it will be externally referenced. The name @var{name}
6742 itself is merely a name of convenience that is used so that it is possible to
6743 have definitions for multiple versions of a function within a single source
6744 file, and so that the compiler can unambiguously know which version of a
6745 function is being mentioned. The @var{nodename} portion of the alias should be
6746 the name of a node specified in the version script supplied to the linker when
6747 building a shared library. If you are attempting to override a versioned
6748 symbol from a shared library, then @var{nodename} should correspond to the
6749 nodename of the symbol you are trying to override.
6751 If the symbol @var{name} is not defined within the file being assembled, all
6752 references to @var{name} will be changed to @var{name2@@nodename}. If no
6753 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6756 Another usage of the @code{.symver} directive is:
6758 .symver @var{name}, @var{name2@@@@nodename}
6760 In this case, the symbol @var{name} must exist and be defined within
6761 the file being assembled. It is similar to @var{name2@@nodename}. The
6762 difference is @var{name2@@@@nodename} will also be used to resolve
6763 references to @var{name2} by the linker.
6765 The third usage of the @code{.symver} directive is:
6767 .symver @var{name}, @var{name2@@@@@@nodename}
6769 When @var{name} is not defined within the
6770 file being assembled, it is treated as @var{name2@@nodename}. When
6771 @var{name} is defined within the file being assembled, the symbol
6772 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6777 @section @code{.tag @var{structname}}
6779 @cindex COFF structure debugging
6780 @cindex structure debugging, COFF
6781 @cindex @code{tag} directive
6782 This directive is generated by compilers to include auxiliary debugging
6783 information in the symbol table. It is only permitted inside
6784 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6785 definitions in the symbol table with instances of those structures.
6788 @samp{.tag} is only used when generating COFF format output; when
6789 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6795 @section @code{.text @var{subsection}}
6797 @cindex @code{text} directive
6798 Tells @command{@value{AS}} to assemble the following statements onto the end of
6799 the text subsection numbered @var{subsection}, which is an absolute
6800 expression. If @var{subsection} is omitted, subsection number zero
6804 @section @code{.title "@var{heading}"}
6806 @cindex @code{title} directive
6807 @cindex listing control: title line
6808 Use @var{heading} as the title (second line, immediately after the
6809 source file name and pagenumber) when generating assembly listings.
6811 This directive affects subsequent pages, as well as the current page if
6812 it appears within ten lines of the top of a page.
6816 @section @code{.type}
6818 This directive is used to set the type of a symbol.
6822 @c only print the extra heading if both COFF and ELF are set
6823 @subheading COFF Version
6826 @cindex COFF symbol type
6827 @cindex symbol type, COFF
6828 @cindex @code{type} directive (COFF version)
6829 For COFF targets, this directive is permitted only within
6830 @code{.def}/@code{.endef} pairs. It is used like this:
6836 This records the integer @var{int} as the type attribute of a symbol table
6840 @samp{.type} is associated only with COFF format output; when
6841 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6842 directive but ignores it.
6848 @c only print the extra heading if both COFF and ELF are set
6849 @subheading ELF Version
6852 @cindex ELF symbol type
6853 @cindex symbol type, ELF
6854 @cindex @code{type} directive (ELF version)
6855 For ELF targets, the @code{.type} directive is used like this:
6858 .type @var{name} , @var{type description}
6861 This sets the type of symbol @var{name} to be either a
6862 function symbol or an object symbol. There are five different syntaxes
6863 supported for the @var{type description} field, in order to provide
6864 compatibility with various other assemblers.
6866 Because some of the characters used in these syntaxes (such as @samp{@@} and
6867 @samp{#}) are comment characters for some architectures, some of the syntaxes
6868 below do not work on all architectures. The first variant will be accepted by
6869 the GNU assembler on all architectures so that variant should be used for
6870 maximum portability, if you do not need to assemble your code with other
6873 The syntaxes supported are:
6876 .type <name> STT_<TYPE_IN_UPPER_CASE>
6877 .type <name>,#<type>
6878 .type <name>,@@<type>
6879 .type <name>,%<type>
6880 .type <name>,"<type>"
6883 The types supported are:
6888 Mark the symbol as being a function name.
6891 @itemx gnu_indirect_function
6892 Mark the symbol as an indirect function when evaluated during reloc
6893 processing. (This is only supported on assemblers targeting GNU systems).
6897 Mark the symbol as being a data object.
6901 Mark the symbol as being a thead-local data object.
6905 Mark the symbol as being a common data object.
6909 Does not mark the symbol in any way. It is supported just for completeness.
6911 @item gnu_unique_object
6912 Marks the symbol as being a globally unique data object. The dynamic linker
6913 will make sure that in the entire process there is just one symbol with this
6914 name and type in use. (This is only supported on assemblers targeting GNU
6919 Note: Some targets support extra types in addition to those listed above.
6925 @section @code{.uleb128 @var{expressions}}
6927 @cindex @code{uleb128} directive
6928 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6929 compact, variable length representation of numbers used by the DWARF
6930 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6934 @section @code{.val @var{addr}}
6936 @cindex @code{val} directive
6937 @cindex COFF value attribute
6938 @cindex value attribute, COFF
6939 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6940 records the address @var{addr} as the value attribute of a symbol table
6944 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6945 configured for @code{b.out}, it accepts this directive but ignores it.
6951 @section @code{.version "@var{string}"}
6953 @cindex @code{version} directive
6954 This directive creates a @code{.note} section and places into it an ELF
6955 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6960 @section @code{.vtable_entry @var{table}, @var{offset}}
6962 @cindex @code{vtable_entry} directive
6963 This directive finds or creates a symbol @code{table} and creates a
6964 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6967 @section @code{.vtable_inherit @var{child}, @var{parent}}
6969 @cindex @code{vtable_inherit} directive
6970 This directive finds the symbol @code{child} and finds or creates the symbol
6971 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6972 parent whose addend is the value of the child symbol. As a special case the
6973 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6977 @section @code{.warning "@var{string}"}
6978 @cindex warning directive
6979 Similar to the directive @code{.error}
6980 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6983 @section @code{.weak @var{names}}
6985 @cindex @code{weak} directive
6986 This directive sets the weak attribute on the comma separated list of symbol
6987 @code{names}. If the symbols do not already exist, they will be created.
6989 On COFF targets other than PE, weak symbols are a GNU extension. This
6990 directive sets the weak attribute on the comma separated list of symbol
6991 @code{names}. If the symbols do not already exist, they will be created.
6993 On the PE target, weak symbols are supported natively as weak aliases.
6994 When a weak symbol is created that is not an alias, GAS creates an
6995 alternate symbol to hold the default value.
6998 @section @code{.weakref @var{alias}, @var{target}}
7000 @cindex @code{weakref} directive
7001 This directive creates an alias to the target symbol that enables the symbol to
7002 be referenced with weak-symbol semantics, but without actually making it weak.
7003 If direct references or definitions of the symbol are present, then the symbol
7004 will not be weak, but if all references to it are through weak references, the
7005 symbol will be marked as weak in the symbol table.
7007 The effect is equivalent to moving all references to the alias to a separate
7008 assembly source file, renaming the alias to the symbol in it, declaring the
7009 symbol as weak there, and running a reloadable link to merge the object files
7010 resulting from the assembly of the new source file and the old source file that
7011 had the references to the alias removed.
7013 The alias itself never makes to the symbol table, and is entirely handled
7014 within the assembler.
7017 @section @code{.word @var{expressions}}
7019 @cindex @code{word} directive
7020 This directive expects zero or more @var{expressions}, of any section,
7021 separated by commas.
7024 For each expression, @command{@value{AS}} emits a 32-bit number.
7027 For each expression, @command{@value{AS}} emits a 16-bit number.
7032 The size of the number emitted, and its byte order,
7033 depend on what target computer the assembly is for.
7036 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
7037 @c happen---32-bit addressability, period; no long/short jumps.
7038 @ifset DIFF-TBL-KLUGE
7039 @cindex difference tables altered
7040 @cindex altered difference tables
7042 @emph{Warning: Special Treatment to support Compilers}
7046 Machines with a 32-bit address space, but that do less than 32-bit
7047 addressing, require the following special treatment. If the machine of
7048 interest to you does 32-bit addressing (or doesn't require it;
7049 @pxref{Machine Dependencies}), you can ignore this issue.
7052 In order to assemble compiler output into something that works,
7053 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
7054 Directives of the form @samp{.word sym1-sym2} are often emitted by
7055 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
7056 directive of the form @samp{.word sym1-sym2}, and the difference between
7057 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
7058 creates a @dfn{secondary jump table}, immediately before the next label.
7059 This secondary jump table is preceded by a short-jump to the
7060 first byte after the secondary table. This short-jump prevents the flow
7061 of control from accidentally falling into the new table. Inside the
7062 table is a long-jump to @code{sym2}. The original @samp{.word}
7063 contains @code{sym1} minus the address of the long-jump to
7066 If there were several occurrences of @samp{.word sym1-sym2} before the
7067 secondary jump table, all of them are adjusted. If there was a
7068 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
7069 long-jump to @code{sym4} is included in the secondary jump table,
7070 and the @code{.word} directives are adjusted to contain @code{sym3}
7071 minus the address of the long-jump to @code{sym4}; and so on, for as many
7072 entries in the original jump table as necessary.
7075 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
7076 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
7077 assembly language programmers.
7080 @c end DIFF-TBL-KLUGE
7082 @ifclear no-space-dir
7084 @section @code{.zero @var{size}}
7086 @cindex @code{zero} directive
7087 @cindex filling memory with zero bytes
7088 This directive emits @var{size} 0-valued bytes. @var{size} must be an absolute
7089 expression. This directive is actually an alias for the @samp{.skip} directive
7090 so in can take an optional second argument of the value to store in the bytes
7091 instead of zero. Using @samp{.zero} in this way would be confusing however.
7095 @section Deprecated Directives
7097 @cindex deprecated directives
7098 @cindex obsolescent directives
7099 One day these directives won't work.
7100 They are included for compatibility with older assemblers.
7107 @node Object Attributes
7108 @chapter Object Attributes
7109 @cindex object attributes
7111 @command{@value{AS}} assembles source files written for a specific architecture
7112 into object files for that architecture. But not all object files are alike.
7113 Many architectures support incompatible variations. For instance, floating
7114 point arguments might be passed in floating point registers if the object file
7115 requires hardware floating point support---or floating point arguments might be
7116 passed in integer registers if the object file supports processors with no
7117 hardware floating point unit. Or, if two objects are built for different
7118 generations of the same architecture, the combination may require the
7119 newer generation at run-time.
7121 This information is useful during and after linking. At link time,
7122 @command{@value{LD}} can warn about incompatible object files. After link
7123 time, tools like @command{gdb} can use it to process the linked file
7126 Compatibility information is recorded as a series of object attributes. Each
7127 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
7128 string, and indicates who sets the meaning of the tag. The tag is an integer,
7129 and indicates what property the attribute describes. The value may be a string
7130 or an integer, and indicates how the property affects this object. Missing
7131 attributes are the same as attributes with a zero value or empty string value.
7133 Object attributes were developed as part of the ABI for the ARM Architecture.
7134 The file format is documented in @cite{ELF for the ARM Architecture}.
7137 * GNU Object Attributes:: @sc{gnu} Object Attributes
7138 * Defining New Object Attributes:: Defining New Object Attributes
7141 @node GNU Object Attributes
7142 @section @sc{gnu} Object Attributes
7144 The @code{.gnu_attribute} directive records an object attribute
7145 with vendor @samp{gnu}.
7147 Except for @samp{Tag_compatibility}, which has both an integer and a string for
7148 its value, @sc{gnu} attributes have a string value if the tag number is odd and
7149 an integer value if the tag number is even. The second bit (@code{@var{tag} &
7150 2} is set for architecture-independent attributes and clear for
7151 architecture-dependent ones.
7153 @subsection Common @sc{gnu} attributes
7155 These attributes are valid on all architectures.
7158 @item Tag_compatibility (32)
7159 The compatibility attribute takes an integer flag value and a vendor name. If
7160 the flag value is 0, the file is compatible with other toolchains. If it is 1,
7161 then the file is only compatible with the named toolchain. If it is greater
7162 than 1, the file can only be processed by other toolchains under some private
7163 arrangement indicated by the flag value and the vendor name.
7166 @subsection MIPS Attributes
7169 @item Tag_GNU_MIPS_ABI_FP (4)
7170 The floating-point ABI used by this object file. The value will be:
7174 0 for files not affected by the floating-point ABI.
7176 1 for files using the hardware floating-point ABI with a standard
7177 double-precision FPU.
7179 2 for files using the hardware floating-point ABI with a single-precision FPU.
7181 3 for files using the software floating-point ABI.
7183 4 for files using the deprecated hardware floating-point ABI which used 64-bit
7184 floating-point registers, 32-bit general-purpose registers and increased the
7185 number of callee-saved floating-point registers.
7187 5 for files using the hardware floating-point ABI with a double-precision FPU
7188 with either 32-bit or 64-bit floating-point registers and 32-bit
7189 general-purpose registers.
7191 6 for files using the hardware floating-point ABI with 64-bit floating-point
7192 registers and 32-bit general-purpose registers.
7194 7 for files using the hardware floating-point ABI with 64-bit floating-point
7195 registers, 32-bit general-purpose registers and a rule that forbids the
7196 direct use of odd-numbered single-precision floating-point registers.
7200 @subsection PowerPC Attributes
7203 @item Tag_GNU_Power_ABI_FP (4)
7204 The floating-point ABI used by this object file. The value will be:
7208 0 for files not affected by the floating-point ABI.
7210 1 for files using double-precision hardware floating-point ABI.
7212 2 for files using the software floating-point ABI.
7214 3 for files using single-precision hardware floating-point ABI.
7217 @item Tag_GNU_Power_ABI_Vector (8)
7218 The vector ABI used by this object file. The value will be:
7222 0 for files not affected by the vector ABI.
7224 1 for files using general purpose registers to pass vectors.
7226 2 for files using AltiVec registers to pass vectors.
7228 3 for files using SPE registers to pass vectors.
7232 @subsection IBM z Systems Attributes
7235 @item Tag_GNU_S390_ABI_Vector (8)
7236 The vector ABI used by this object file. The value will be:
7240 0 for files not affected by the vector ABI.
7242 1 for files using software vector ABI.
7244 2 for files using hardware vector ABI.
7248 @node Defining New Object Attributes
7249 @section Defining New Object Attributes
7251 If you want to define a new @sc{gnu} object attribute, here are the places you
7252 will need to modify. New attributes should be discussed on the @samp{binutils}
7257 This manual, which is the official register of attributes.
7259 The header for your architecture @file{include/elf}, to define the tag.
7261 The @file{bfd} support file for your architecture, to merge the attribute
7262 and issue any appropriate link warnings.
7264 Test cases in @file{ld/testsuite} for merging and link warnings.
7266 @file{binutils/readelf.c} to display your attribute.
7268 GCC, if you want the compiler to mark the attribute automatically.
7274 @node Machine Dependencies
7275 @chapter Machine Dependent Features
7277 @cindex machine dependencies
7278 The machine instruction sets are (almost by definition) different on
7279 each machine where @command{@value{AS}} runs. Floating point representations
7280 vary as well, and @command{@value{AS}} often supports a few additional
7281 directives or command-line options for compatibility with other
7282 assemblers on a particular platform. Finally, some versions of
7283 @command{@value{AS}} support special pseudo-instructions for branch
7286 This chapter discusses most of these differences, though it does not
7287 include details on any machine's instruction set. For details on that
7288 subject, see the hardware manufacturer's manual.
7292 * AArch64-Dependent:: AArch64 Dependent Features
7295 * Alpha-Dependent:: Alpha Dependent Features
7298 * ARC-Dependent:: ARC Dependent Features
7301 * ARM-Dependent:: ARM Dependent Features
7304 * AVR-Dependent:: AVR Dependent Features
7307 * Blackfin-Dependent:: Blackfin Dependent Features
7310 * CR16-Dependent:: CR16 Dependent Features
7313 * CRIS-Dependent:: CRIS Dependent Features
7316 * D10V-Dependent:: D10V Dependent Features
7319 * D30V-Dependent:: D30V Dependent Features
7322 * Epiphany-Dependent:: EPIPHANY Dependent Features
7325 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7328 * HPPA-Dependent:: HPPA Dependent Features
7331 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
7334 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7337 * i860-Dependent:: Intel 80860 Dependent Features
7340 * i960-Dependent:: Intel 80960 Dependent Features
7343 * IA-64-Dependent:: Intel IA-64 Dependent Features
7346 * IP2K-Dependent:: IP2K Dependent Features
7349 * LM32-Dependent:: LM32 Dependent Features
7352 * M32C-Dependent:: M32C Dependent Features
7355 * M32R-Dependent:: M32R Dependent Features
7358 * M68K-Dependent:: M680x0 Dependent Features
7361 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7364 * Meta-Dependent :: Meta Dependent Features
7367 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7370 * MIPS-Dependent:: MIPS Dependent Features
7373 * MMIX-Dependent:: MMIX Dependent Features
7376 * MSP430-Dependent:: MSP430 Dependent Features
7379 * NDS32-Dependent:: Andes NDS32 Dependent Features
7382 * NiosII-Dependent:: Altera Nios II Dependent Features
7385 * NS32K-Dependent:: NS32K Dependent Features
7388 * PDP-11-Dependent:: PDP-11 Dependent Features
7391 * PJ-Dependent:: picoJava Dependent Features
7394 * PPC-Dependent:: PowerPC Dependent Features
7397 * RL78-Dependent:: RL78 Dependent Features
7400 * RX-Dependent:: RX Dependent Features
7403 * S/390-Dependent:: IBM S/390 Dependent Features
7406 * SCORE-Dependent:: SCORE Dependent Features
7409 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7410 * SH64-Dependent:: SuperH SH64 Dependent Features
7413 * Sparc-Dependent:: SPARC Dependent Features
7416 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7419 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7422 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7425 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7428 * V850-Dependent:: V850 Dependent Features
7431 * Vax-Dependent:: VAX Dependent Features
7434 * Visium-Dependent:: Visium Dependent Features
7437 * XGATE-Dependent:: XGATE Features
7440 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7443 * Xtensa-Dependent:: Xtensa Dependent Features
7446 * Z80-Dependent:: Z80 Dependent Features
7449 * Z8000-Dependent:: Z8000 Dependent Features
7456 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7457 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7458 @c peculiarity: to preserve cross-references, there must be a node called
7459 @c "Machine Dependencies". Hence the conditional nodenames in each
7460 @c major node below. Node defaulting in makeinfo requires adjacency of
7461 @c node and sectioning commands; hence the repetition of @chapter BLAH
7462 @c in both conditional blocks.
7465 @include c-aarch64.texi
7469 @include c-alpha.texi
7485 @include c-bfin.texi
7489 @include c-cr16.texi
7493 @include c-cris.texi
7498 @node Machine Dependencies
7499 @chapter Machine Dependent Features
7501 The machine instruction sets are different on each Renesas chip family,
7502 and there are also some syntax differences among the families. This
7503 chapter describes the specific @command{@value{AS}} features for each
7507 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7508 * SH-Dependent:: Renesas SH Dependent Features
7515 @include c-d10v.texi
7519 @include c-d30v.texi
7523 @include c-epiphany.texi
7527 @include c-h8300.texi
7531 @include c-hppa.texi
7535 @include c-i370.texi
7539 @include c-i386.texi
7543 @include c-i860.texi
7547 @include c-i960.texi
7551 @include c-ia64.texi
7555 @include c-ip2k.texi
7559 @include c-lm32.texi
7563 @include c-m32c.texi
7567 @include c-m32r.texi
7571 @include c-m68k.texi
7575 @include c-m68hc11.texi
7579 @include c-metag.texi
7583 @include c-microblaze.texi
7587 @include c-mips.texi
7591 @include c-mmix.texi
7595 @include c-msp430.texi
7599 @include c-nds32.texi
7603 @include c-nios2.texi
7607 @include c-ns32k.texi
7611 @include c-pdp11.texi
7623 @include c-rl78.texi
7631 @include c-s390.texi
7635 @include c-score.texi
7640 @include c-sh64.texi
7644 @include c-sparc.texi
7648 @include c-tic54x.texi
7652 @include c-tic6x.texi
7656 @include c-tilegx.texi
7660 @include c-tilepro.texi
7664 @include c-v850.texi
7672 @include c-visium.texi
7676 @include c-xgate.texi
7680 @include c-xstormy16.texi
7684 @include c-xtensa.texi
7696 @c reverse effect of @down at top of generic Machine-Dep chapter
7700 @node Reporting Bugs
7701 @chapter Reporting Bugs
7702 @cindex bugs in assembler
7703 @cindex reporting bugs in assembler
7705 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7707 Reporting a bug may help you by bringing a solution to your problem, or it may
7708 not. But in any case the principal function of a bug report is to help the
7709 entire community by making the next version of @command{@value{AS}} work better.
7710 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7712 In order for a bug report to serve its purpose, you must include the
7713 information that enables us to fix the bug.
7716 * Bug Criteria:: Have you found a bug?
7717 * Bug Reporting:: How to report bugs
7721 @section Have You Found a Bug?
7722 @cindex bug criteria
7724 If you are not sure whether you have found a bug, here are some guidelines:
7727 @cindex fatal signal
7728 @cindex assembler crash
7729 @cindex crash of assembler
7731 If the assembler gets a fatal signal, for any input whatever, that is a
7732 @command{@value{AS}} bug. Reliable assemblers never crash.
7734 @cindex error on valid input
7736 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7738 @cindex invalid input
7740 If @command{@value{AS}} does not produce an error message for invalid input, that
7741 is a bug. However, you should note that your idea of ``invalid input'' might
7742 be our idea of ``an extension'' or ``support for traditional practice''.
7745 If you are an experienced user of assemblers, your suggestions for improvement
7746 of @command{@value{AS}} are welcome in any case.
7750 @section How to Report Bugs
7752 @cindex assembler bugs, reporting
7754 A number of companies and individuals offer support for @sc{gnu} products. If
7755 you obtained @command{@value{AS}} from a support organization, we recommend you
7756 contact that organization first.
7758 You can find contact information for many support companies and
7759 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7763 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7767 The fundamental principle of reporting bugs usefully is this:
7768 @strong{report all the facts}. If you are not sure whether to state a
7769 fact or leave it out, state it!
7771 Often people omit facts because they think they know what causes the problem
7772 and assume that some details do not matter. Thus, you might assume that the
7773 name of a symbol you use in an example does not matter. Well, probably it does
7774 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7775 happens to fetch from the location where that name is stored in memory;
7776 perhaps, if the name were different, the contents of that location would fool
7777 the assembler into doing the right thing despite the bug. Play it safe and
7778 give a specific, complete example. That is the easiest thing for you to do,
7779 and the most helpful.
7781 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7782 it is new to us. Therefore, always write your bug reports on the assumption
7783 that the bug has not been reported previously.
7785 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7786 bell?'' This cannot help us fix a bug, so it is basically useless. We
7787 respond by asking for enough details to enable us to investigate.
7788 You might as well expedite matters by sending them to begin with.
7790 To enable us to fix the bug, you should include all these things:
7794 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
7795 it with the @samp{--version} argument.
7797 Without this, we will not know whether there is any point in looking for
7798 the bug in the current version of @command{@value{AS}}.
7801 Any patches you may have applied to the @command{@value{AS}} source.
7804 The type of machine you are using, and the operating system name and
7808 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
7812 The command arguments you gave the assembler to assemble your example and
7813 observe the bug. To guarantee you will not omit something important, list them
7814 all. A copy of the Makefile (or the output from make) is sufficient.
7816 If we were to try to guess the arguments, we would probably guess wrong
7817 and then we might not encounter the bug.
7820 A complete input file that will reproduce the bug. If the bug is observed when
7821 the assembler is invoked via a compiler, send the assembler source, not the
7822 high level language source. Most compilers will produce the assembler source
7823 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
7824 the options @samp{-v --save-temps}; this will save the assembler source in a
7825 file with an extension of @file{.s}, and also show you exactly how
7826 @command{@value{AS}} is being run.
7829 A description of what behavior you observe that you believe is
7830 incorrect. For example, ``It gets a fatal signal.''
7832 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
7833 will certainly notice it. But if the bug is incorrect output, we might not
7834 notice unless it is glaringly wrong. You might as well not give us a chance to
7837 Even if the problem you experience is a fatal signal, you should still say so
7838 explicitly. Suppose something strange is going on, such as, your copy of
7839 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
7840 library on your system. (This has happened!) Your copy might crash and ours
7841 would not. If you told us to expect a crash, then when ours fails to crash, we
7842 would know that the bug was not happening for us. If you had not told us to
7843 expect a crash, then we would not be able to draw any conclusion from our
7847 If you wish to suggest changes to the @command{@value{AS}} source, send us context
7848 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
7849 option. Always send diffs from the old file to the new file. If you even
7850 discuss something in the @command{@value{AS}} source, refer to it by context, not
7853 The line numbers in our development sources will not match those in your
7854 sources. Your line numbers would convey no useful information to us.
7857 Here are some things that are not necessary:
7861 A description of the envelope of the bug.
7863 Often people who encounter a bug spend a lot of time investigating
7864 which changes to the input file will make the bug go away and which
7865 changes will not affect it.
7867 This is often time consuming and not very useful, because the way we
7868 will find the bug is by running a single example under the debugger
7869 with breakpoints, not by pure deduction from a series of examples.
7870 We recommend that you save your time for something else.
7872 Of course, if you can find a simpler example to report @emph{instead}
7873 of the original one, that is a convenience for us. Errors in the
7874 output will be easier to spot, running under the debugger will take
7875 less time, and so on.
7877 However, simplification is not vital; if you do not want to do this,
7878 report the bug anyway and send us the entire test case you used.
7881 A patch for the bug.
7883 A patch for the bug does help us if it is a good one. But do not omit
7884 the necessary information, such as the test case, on the assumption that
7885 a patch is all we need. We might see problems with your patch and decide
7886 to fix the problem another way, or we might not understand it at all.
7888 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
7889 construct an example that will make the program follow a certain path through
7890 the code. If you do not send us the example, we will not be able to construct
7891 one, so we will not be able to verify that the bug is fixed.
7893 And if we cannot understand what bug you are trying to fix, or why your
7894 patch should be an improvement, we will not install it. A test case will
7895 help us to understand.
7898 A guess about what the bug is or what it depends on.
7900 Such guesses are usually wrong. Even we cannot guess right about such
7901 things without first using the debugger to find the facts.
7904 @node Acknowledgements
7905 @chapter Acknowledgements
7907 If you have contributed to GAS and your name isn't listed here,
7908 it is not meant as a slight. We just don't know about it. Send mail to the
7909 maintainer, and we'll correct the situation. Currently
7911 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
7913 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
7916 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
7917 information and the 68k series machines, most of the preprocessing pass, and
7918 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
7920 K. Richard Pixley maintained GAS for a while, adding various enhancements and
7921 many bug fixes, including merging support for several processors, breaking GAS
7922 up to handle multiple object file format back ends (including heavy rewrite,
7923 testing, an integration of the coff and b.out back ends), adding configuration
7924 including heavy testing and verification of cross assemblers and file splits
7925 and renaming, converted GAS to strictly ANSI C including full prototypes, added
7926 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
7927 port (including considerable amounts of reverse engineering), a SPARC opcode
7928 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
7929 assertions and made them work, much other reorganization, cleanup, and lint.
7931 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
7932 in format-specific I/O modules.
7934 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
7935 has done much work with it since.
7937 The Intel 80386 machine description was written by Eliot Dresselhaus.
7939 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
7941 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
7942 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
7944 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
7945 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
7946 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
7947 support a.out format.
7949 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
7950 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
7951 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
7952 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
7955 John Gilmore built the AMD 29000 support, added @code{.include} support, and
7956 simplified the configuration of which versions accept which directives. He
7957 updated the 68k machine description so that Motorola's opcodes always produced
7958 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
7959 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
7960 cross-compilation support, and one bug in relaxation that took a week and
7961 required the proverbial one-bit fix.
7963 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
7964 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
7965 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
7966 PowerPC assembler, and made a few other minor patches.
7968 Steve Chamberlain made GAS able to generate listings.
7970 Hewlett-Packard contributed support for the HP9000/300.
7972 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
7973 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
7974 formats). This work was supported by both the Center for Software Science at
7975 the University of Utah and Cygnus Support.
7977 Support for ELF format files has been worked on by Mark Eichin of Cygnus
7978 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
7979 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
7980 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
7981 and some initial 64-bit support).
7983 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
7985 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
7986 support for openVMS/Alpha.
7988 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
7991 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
7992 Inc.@: added support for Xtensa processors.
7994 Several engineers at Cygnus Support have also provided many small bug fixes and
7995 configuration enhancements.
7997 Jon Beniston added support for the Lattice Mico32 architecture.
7999 Many others have contributed large or small bugfixes and enhancements. If
8000 you have contributed significant work and are not mentioned on this list, and
8001 want to be, let us know. Some of the history has been lost; we are not
8002 intentionally leaving anyone out.
8004 @node GNU Free Documentation License
8005 @appendix GNU Free Documentation License
8009 @unnumbered AS Index