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{--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 Print the maximum space (in bytes) and total time (in seconds) used by
773 @item --strip-local-absolute
774 Remove local absolute symbols from the outgoing symbol table.
778 Print the @command{as} version.
781 Print the @command{as} version and exit.
785 Suppress warning messages.
787 @item --fatal-warnings
788 Treat warnings as errors.
791 Don't suppress warning messages or treat them as errors.
800 Generate an object file even after errors.
802 @item -- | @var{files} @dots{}
803 Standard input, or source files to assemble.
811 @xref{AArch64 Options}, for the options available when @value{AS} is configured
812 for the 64-bit mode of the ARM Architecture (AArch64).
817 The following options are available when @value{AS} is configured for the
818 64-bit mode of the ARM Architecture (AArch64).
821 @include c-aarch64.texi
822 @c ended inside the included file
830 @xref{Alpha Options}, for the options available when @value{AS} is configured
831 for an Alpha processor.
836 The following options are available when @value{AS} is configured for an Alpha
840 @include c-alpha.texi
841 @c ended inside the included file
848 The following options are available when @value{AS} is configured for
853 This option selects the core processor variant.
855 Select either big-endian (-EB) or little-endian (-EL) output.
860 The following options are available when @value{AS} is configured for the ARM
864 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
865 Specify which ARM processor variant is the target.
866 @item -march=@var{architecture}[+@var{extension}@dots{}]
867 Specify which ARM architecture variant is used by the target.
868 @item -mfpu=@var{floating-point-format}
869 Select which Floating Point architecture is the target.
870 @item -mfloat-abi=@var{abi}
871 Select which floating point ABI is in use.
873 Enable Thumb only instruction decoding.
874 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
875 Select which procedure calling convention is in use.
877 Select either big-endian (-EB) or little-endian (-EL) output.
878 @item -mthumb-interwork
879 Specify that the code has been generated with interworking between Thumb and
882 Turns on CodeComposer Studio assembly syntax compatibility mode.
884 Specify that PIC code has been generated.
892 @xref{Blackfin Options}, for the options available when @value{AS} is
893 configured for the Blackfin processor family.
898 The following options are available when @value{AS} is configured for
899 the Blackfin processor family.
903 @c ended inside the included file
910 See the info pages for documentation of the CRIS-specific options.
914 The following options are available when @value{AS} is configured for
917 @cindex D10V optimization
918 @cindex optimization, D10V
920 Optimize output by parallelizing instructions.
925 The following options are available when @value{AS} is configured for a D30V
928 @cindex D30V optimization
929 @cindex optimization, D30V
931 Optimize output by parallelizing instructions.
935 Warn when nops are generated.
937 @cindex D30V nops after 32-bit multiply
939 Warn when a nop after a 32-bit multiply instruction is generated.
945 The following options are available when @value{AS} is configured for the
946 Adapteva EPIPHANY series.
949 @xref{Epiphany Options}, for the options available when @value{AS} is
950 configured for an Epiphany processor.
955 The following options are available when @value{AS} is configured for
956 an Epiphany processor.
959 @include c-epiphany.texi
960 @c ended inside the included file
968 @xref{H8/300 Options}, for the options available when @value{AS} is configured
969 for an H8/300 processor.
974 The following options are available when @value{AS} is configured for an H8/300
978 @include c-h8300.texi
979 @c ended inside the included file
987 @xref{i386-Options}, for the options available when @value{AS} is
988 configured for an i386 processor.
993 The following options are available when @value{AS} is configured for
998 @c ended inside the included file
1003 @c man begin OPTIONS
1005 The following options are available when @value{AS} is configured for the
1006 Intel 80960 processor.
1009 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
1010 Specify which variant of the 960 architecture is the target.
1013 Add code to collect statistics about branches taken.
1016 Do not alter compare-and-branch instructions for long displacements;
1023 The following options are available when @value{AS} is configured for the
1029 Specifies that the extended IP2022 instructions are allowed.
1032 Restores the default behaviour, which restricts the permitted instructions to
1033 just the basic IP2022 ones.
1039 The following options are available when @value{AS} is configured for the
1040 Renesas M32C and M16C processors.
1045 Assemble M32C instructions.
1048 Assemble M16C instructions (the default).
1051 Enable support for link-time relaxations.
1054 Support H'00 style hex constants in addition to 0x00 style.
1060 The following options are available when @value{AS} is configured for the
1061 Renesas M32R (formerly Mitsubishi M32R) series.
1066 Specify which processor in the M32R family is the target. The default
1067 is normally the M32R, but this option changes it to the M32RX.
1069 @item --warn-explicit-parallel-conflicts or --Wp
1070 Produce warning messages when questionable parallel constructs are
1073 @item --no-warn-explicit-parallel-conflicts or --Wnp
1074 Do not produce warning messages when questionable parallel constructs are
1081 The following options are available when @value{AS} is configured for the
1082 Motorola 68000 series.
1087 Shorten references to undefined symbols, to one word instead of two.
1089 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1090 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1091 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1092 Specify what processor in the 68000 family is the target. The default
1093 is normally the 68020, but this can be changed at configuration time.
1095 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1096 The target machine does (or does not) have a floating-point coprocessor.
1097 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1098 the basic 68000 is not compatible with the 68881, a combination of the
1099 two can be specified, since it's possible to do emulation of the
1100 coprocessor instructions with the main processor.
1102 @item -m68851 | -mno-68851
1103 The target machine does (or does not) have a memory-management
1104 unit coprocessor. The default is to assume an MMU for 68020 and up.
1112 @xref{Nios II Options}, for the options available when @value{AS} is configured
1113 for an Altera Nios II processor.
1117 @c man begin OPTIONS
1118 The following options are available when @value{AS} is configured for an
1119 Altera Nios II processor.
1121 @c man begin INCLUDE
1122 @include c-nios2.texi
1123 @c ended inside the included file
1129 For details about the PDP-11 machine dependent features options,
1130 see @ref{PDP-11-Options}.
1133 @item -mpic | -mno-pic
1134 Generate position-independent (or position-dependent) code. The
1135 default is @option{-mpic}.
1138 @itemx -mall-extensions
1139 Enable all instruction set extensions. This is the default.
1141 @item -mno-extensions
1142 Disable all instruction set extensions.
1144 @item -m@var{extension} | -mno-@var{extension}
1145 Enable (or disable) a particular instruction set extension.
1148 Enable the instruction set extensions supported by a particular CPU, and
1149 disable all other extensions.
1151 @item -m@var{machine}
1152 Enable the instruction set extensions supported by a particular machine
1153 model, and disable all other extensions.
1159 The following options are available when @value{AS} is configured for
1160 a picoJava processor.
1164 @cindex PJ endianness
1165 @cindex endianness, PJ
1166 @cindex big endian output, PJ
1168 Generate ``big endian'' format output.
1170 @cindex little endian output, PJ
1172 Generate ``little endian'' format output.
1178 The following options are available when @value{AS} is configured for the
1179 Motorola 68HC11 or 68HC12 series.
1183 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1184 Specify what processor is the target. The default is
1185 defined by the configuration option when building the assembler.
1187 @item --xgate-ramoffset
1188 Instruct the linker to offset RAM addresses from S12X address space into
1189 XGATE address space.
1192 Specify to use the 16-bit integer ABI.
1195 Specify to use the 32-bit integer ABI.
1197 @item -mshort-double
1198 Specify to use the 32-bit double ABI.
1201 Specify to use the 64-bit double ABI.
1203 @item --force-long-branches
1204 Relative branches are turned into absolute ones. This concerns
1205 conditional branches, unconditional branches and branches to a
1208 @item -S | --short-branches
1209 Do not turn relative branches into absolute ones
1210 when the offset is out of range.
1212 @item --strict-direct-mode
1213 Do not turn the direct addressing mode into extended addressing mode
1214 when the instruction does not support direct addressing mode.
1216 @item --print-insn-syntax
1217 Print the syntax of instruction in case of error.
1219 @item --print-opcodes
1220 Print the list of instructions with syntax and then exit.
1222 @item --generate-example
1223 Print an example of instruction for each possible instruction and then exit.
1224 This option is only useful for testing @command{@value{AS}}.
1230 The following options are available when @command{@value{AS}} is configured
1231 for the SPARC architecture:
1234 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1235 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1236 Explicitly select a variant of the SPARC architecture.
1238 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1239 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1241 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1242 UltraSPARC extensions.
1244 @item -xarch=v8plus | -xarch=v8plusa
1245 For compatibility with the Solaris v9 assembler. These options are
1246 equivalent to -Av8plus and -Av8plusa, respectively.
1249 Warn when the assembler switches to another architecture.
1254 The following options are available when @value{AS} is configured for the 'c54x
1259 Enable extended addressing mode. All addresses and relocations will assume
1260 extended addressing (usually 23 bits).
1261 @item -mcpu=@var{CPU_VERSION}
1262 Sets the CPU version being compiled for.
1263 @item -merrors-to-file @var{FILENAME}
1264 Redirect error output to a file, for broken systems which don't support such
1265 behaviour in the shell.
1270 The following options are available when @value{AS} is configured for
1275 This option sets the largest size of an object that can be referenced
1276 implicitly with the @code{gp} register. It is only accepted for targets that
1277 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1279 @cindex MIPS endianness
1280 @cindex endianness, MIPS
1281 @cindex big endian output, MIPS
1283 Generate ``big endian'' format output.
1285 @cindex little endian output, MIPS
1287 Generate ``little endian'' format output.
1305 Generate code for a particular MIPS Instruction Set Architecture level.
1306 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1307 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1308 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1309 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1310 @samp{-mips32r5}, @samp{-mips32r6}, @samp{-mips64}, @samp{-mips64r2},
1311 @samp{-mips64r3}, @samp{-mips64r5}, and @samp{-mips64r6} correspond to generic
1312 MIPS V, MIPS32, MIPS32 Release 2, MIPS32 Release 3, MIPS32 Release 5, MIPS32
1313 Release 6, MIPS64, MIPS64 Release 2, MIPS64 Release 3, MIPS64 Release 5, and
1314 MIPS64 Release 6 ISA processors, respectively.
1316 @item -march=@var{cpu}
1317 Generate code for a particular MIPS CPU.
1319 @item -mtune=@var{cpu}
1320 Schedule and tune for a particular MIPS CPU.
1324 Cause nops to be inserted if the read of the destination register
1325 of an mfhi or mflo instruction occurs in the following two instructions.
1328 @itemx -mno-fix-rm7000
1329 Cause nops to be inserted if a dmult or dmultu instruction is
1330 followed by a load instruction.
1334 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1335 section instead of the standard ELF .stabs sections.
1339 Control generation of @code{.pdr} sections.
1343 The register sizes are normally inferred from the ISA and ABI, but these
1344 flags force a certain group of registers to be treated as 32 bits wide at
1345 all times. @samp{-mgp32} controls the size of general-purpose registers
1346 and @samp{-mfp32} controls the size of floating-point registers.
1350 The register sizes are normally inferred from the ISA and ABI, but these
1351 flags force a certain group of registers to be treated as 64 bits wide at
1352 all times. @samp{-mgp64} controls the size of general-purpose registers
1353 and @samp{-mfp64} controls the size of floating-point registers.
1356 The register sizes are normally inferred from the ISA and ABI, but using
1357 this flag in combination with @samp{-mabi=32} enables an ABI variant
1358 which will operate correctly with floating-point registers which are
1362 @itemx -mno-odd-spreg
1363 Enable use of floating-point operations on odd-numbered single-precision
1364 registers when supported by the ISA. @samp{-mfpxx} implies
1365 @samp{-mno-odd-spreg}, otherwise the default is @samp{-modd-spreg}.
1369 Generate code for the MIPS 16 processor. This is equivalent to putting
1370 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1371 turns off this option.
1374 @itemx -mno-micromips
1375 Generate code for the microMIPS processor. This is equivalent to putting
1376 @code{.set micromips} at the start of the assembly file. @samp{-mno-micromips}
1377 turns off this option. This is equivalent to putting @code{.set nomicromips}
1378 at the start of the assembly file.
1381 @itemx -mno-smartmips
1382 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1383 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1384 @samp{-mno-smartmips} turns off this option.
1388 Generate code for the MIPS-3D Application Specific Extension.
1389 This tells the assembler to accept MIPS-3D instructions.
1390 @samp{-no-mips3d} turns off this option.
1394 Generate code for the MDMX Application Specific Extension.
1395 This tells the assembler to accept MDMX instructions.
1396 @samp{-no-mdmx} turns off this option.
1400 Generate code for the DSP Release 1 Application Specific Extension.
1401 This tells the assembler to accept DSP Release 1 instructions.
1402 @samp{-mno-dsp} turns off this option.
1406 Generate code for the DSP Release 2 Application Specific Extension.
1407 This option implies -mdsp.
1408 This tells the assembler to accept DSP Release 2 instructions.
1409 @samp{-mno-dspr2} turns off this option.
1413 Generate code for the MIPS SIMD Architecture Extension.
1414 This tells the assembler to accept MSA instructions.
1415 @samp{-mno-msa} turns off this option.
1419 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1420 This tells the assembler to accept XPA instructions.
1421 @samp{-mno-xpa} turns off this option.
1425 Generate code for the MT Application Specific Extension.
1426 This tells the assembler to accept MT instructions.
1427 @samp{-mno-mt} turns off this option.
1431 Generate code for the MCU Application Specific Extension.
1432 This tells the assembler to accept MCU instructions.
1433 @samp{-mno-mcu} turns off this option.
1437 Only use 32-bit instruction encodings when generating code for the
1438 microMIPS processor. This option inhibits the use of any 16-bit
1439 instructions. This is equivalent to putting @code{.set insn32} at
1440 the start of the assembly file. @samp{-mno-insn32} turns off this
1441 option. This is equivalent to putting @code{.set noinsn32} at the
1442 start of the assembly file. By default @samp{-mno-insn32} is
1443 selected, allowing all instructions to be used.
1445 @item --construct-floats
1446 @itemx --no-construct-floats
1447 The @samp{--no-construct-floats} option disables the construction of
1448 double width floating point constants by loading the two halves of the
1449 value into the two single width floating point registers that make up
1450 the double width register. By default @samp{--construct-floats} is
1451 selected, allowing construction of these floating point constants.
1453 @item --relax-branch
1454 @itemx --no-relax-branch
1455 The @samp{--relax-branch} option enables the relaxation of out-of-range
1456 branches. By default @samp{--no-relax-branch} is selected, causing any
1457 out-of-range branches to produce an error.
1459 @item -mnan=@var{encoding}
1460 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1461 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1464 @item --emulation=@var{name}
1465 This option was formerly used to switch between ELF and ECOFF output
1466 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1467 removed in GAS 2.24, so the option now serves little purpose.
1468 It is retained for backwards compatibility.
1470 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1471 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1472 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1473 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1474 preferred options instead.
1477 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1484 Control how to deal with multiplication overflow and division by zero.
1485 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1486 (and only work for Instruction Set Architecture level 2 and higher);
1487 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1491 When this option is used, @command{@value{AS}} will issue a warning every
1492 time it generates a nop instruction from a macro.
1497 The following options are available when @value{AS} is configured for
1503 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1504 The command line option @samp{-nojsri2bsr} can be used to disable it.
1508 Enable or disable the silicon filter behaviour. By default this is disabled.
1509 The default can be overridden by the @samp{-sifilter} command line option.
1512 Alter jump instructions for long displacements.
1514 @item -mcpu=[210|340]
1515 Select the cpu type on the target hardware. This controls which instructions
1519 Assemble for a big endian target.
1522 Assemble for a little endian target.
1531 @xref{Meta Options}, for the options available when @value{AS} is configured
1532 for a Meta processor.
1536 @c man begin OPTIONS
1537 The following options are available when @value{AS} is configured for a
1540 @c man begin INCLUDE
1541 @include c-metag.texi
1542 @c ended inside the included file
1547 @c man begin OPTIONS
1549 See the info pages for documentation of the MMIX-specific options.
1555 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1556 for a NDS32 processor.
1558 @c ended inside the included file
1562 @c man begin OPTIONS
1563 The following options are available when @value{AS} is configured for a
1566 @c man begin INCLUDE
1567 @include c-nds32.texi
1568 @c ended inside the included file
1575 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1576 for a PowerPC processor.
1580 @c man begin OPTIONS
1581 The following options are available when @value{AS} is configured for a
1584 @c man begin INCLUDE
1586 @c ended inside the included file
1591 @c man begin OPTIONS
1593 See the info pages for documentation of the RX-specific options.
1597 The following options are available when @value{AS} is configured for the s390
1603 Select the word size, either 31/32 bits or 64 bits.
1606 Select the architecture mode, either the Enterprise System
1607 Architecture (esa) or the z/Architecture mode (zarch).
1608 @item -march=@var{processor}
1609 Specify which s390 processor variant is the target, @samp{g6}, @samp{g6},
1610 @samp{z900}, @samp{z990}, @samp{z9-109}, @samp{z9-ec}, @samp{z10},
1611 @samp{z196}, @samp{zEC12}, or @samp{z13}.
1613 @itemx -mno-regnames
1614 Allow or disallow symbolic names for registers.
1615 @item -mwarn-areg-zero
1616 Warn whenever the operand for a base or index register has been specified
1617 but evaluates to zero.
1625 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1626 for a TMS320C6000 processor.
1630 @c man begin OPTIONS
1631 The following options are available when @value{AS} is configured for a
1632 TMS320C6000 processor.
1634 @c man begin INCLUDE
1635 @include c-tic6x.texi
1636 @c ended inside the included file
1644 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1645 for a TILE-Gx processor.
1649 @c man begin OPTIONS
1650 The following options are available when @value{AS} is configured for a TILE-Gx
1653 @c man begin INCLUDE
1654 @include c-tilegx.texi
1655 @c ended inside the included file
1663 @xref{Visium Options}, for the options available when @value{AS} is configured
1664 for a Visium processor.
1668 @c man begin OPTIONS
1669 The following option is available when @value{AS} is configured for a Visium
1672 @c man begin INCLUDE
1673 @include c-visium.texi
1674 @c ended inside the included file
1682 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1683 for an Xtensa processor.
1687 @c man begin OPTIONS
1688 The following options are available when @value{AS} is configured for an
1691 @c man begin INCLUDE
1692 @include c-xtensa.texi
1693 @c ended inside the included file
1698 @c man begin OPTIONS
1701 The following options are available when @value{AS} is configured for
1702 a Z80 family processor.
1705 Assemble for Z80 processor.
1707 Assemble for R800 processor.
1708 @item -ignore-undocumented-instructions
1710 Assemble undocumented Z80 instructions that also work on R800 without warning.
1711 @item -ignore-unportable-instructions
1713 Assemble all undocumented Z80 instructions without warning.
1714 @item -warn-undocumented-instructions
1716 Issue a warning for undocumented Z80 instructions that also work on R800.
1717 @item -warn-unportable-instructions
1719 Issue a warning for undocumented Z80 instructions that do not work on R800.
1720 @item -forbid-undocumented-instructions
1722 Treat all undocumented instructions as errors.
1723 @item -forbid-unportable-instructions
1725 Treat undocumented Z80 instructions that do not work on R800 as errors.
1732 * Manual:: Structure of this Manual
1733 * GNU Assembler:: The GNU Assembler
1734 * Object Formats:: Object File Formats
1735 * Command Line:: Command Line
1736 * Input Files:: Input Files
1737 * Object:: Output (Object) File
1738 * Errors:: Error and Warning Messages
1742 @section Structure of this Manual
1744 @cindex manual, structure and purpose
1745 This manual is intended to describe what you need to know to use
1746 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1747 notation for symbols, constants, and expressions; the directives that
1748 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1751 We also cover special features in the @value{TARGET}
1752 configuration of @command{@value{AS}}, including assembler directives.
1755 This manual also describes some of the machine-dependent features of
1756 various flavors of the assembler.
1759 @cindex machine instructions (not covered)
1760 On the other hand, this manual is @emph{not} intended as an introduction
1761 to programming in assembly language---let alone programming in general!
1762 In a similar vein, we make no attempt to introduce the machine
1763 architecture; we do @emph{not} describe the instruction set, standard
1764 mnemonics, registers or addressing modes that are standard to a
1765 particular architecture.
1767 You may want to consult the manufacturer's
1768 machine architecture manual for this information.
1772 For information on the H8/300 machine instruction set, see @cite{H8/300
1773 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1774 Programming Manual} (Renesas).
1777 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1778 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1779 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1780 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1783 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1787 @c I think this is premature---doc@cygnus.com, 17jan1991
1789 Throughout this manual, we assume that you are running @dfn{GNU},
1790 the portable operating system from the @dfn{Free Software
1791 Foundation, Inc.}. This restricts our attention to certain kinds of
1792 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1793 once this assumption is granted examples and definitions need less
1796 @command{@value{AS}} is part of a team of programs that turn a high-level
1797 human-readable series of instructions into a low-level
1798 computer-readable series of instructions. Different versions of
1799 @command{@value{AS}} are used for different kinds of computer.
1802 @c There used to be a section "Terminology" here, which defined
1803 @c "contents", "byte", "word", and "long". Defining "word" to any
1804 @c particular size is confusing when the .word directive may generate 16
1805 @c bits on one machine and 32 bits on another; in general, for the user
1806 @c version of this manual, none of these terms seem essential to define.
1807 @c They were used very little even in the former draft of the manual;
1808 @c this draft makes an effort to avoid them (except in names of
1812 @section The GNU Assembler
1814 @c man begin DESCRIPTION
1816 @sc{gnu} @command{as} is really a family of assemblers.
1818 This manual describes @command{@value{AS}}, a member of that family which is
1819 configured for the @value{TARGET} architectures.
1821 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1822 should find a fairly similar environment when you use it on another
1823 architecture. Each version has much in common with the others,
1824 including object file formats, most assembler directives (often called
1825 @dfn{pseudo-ops}) and assembler syntax.@refill
1827 @cindex purpose of @sc{gnu} assembler
1828 @command{@value{AS}} is primarily intended to assemble the output of the
1829 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1830 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1831 assemble correctly everything that other assemblers for the same
1832 machine would assemble.
1834 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1837 @c This remark should appear in generic version of manual; assumption
1838 @c here is that generic version sets M680x0.
1839 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1840 assembler for the same architecture; for example, we know of several
1841 incompatible versions of 680x0 assembly language syntax.
1846 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1847 program in one pass of the source file. This has a subtle impact on the
1848 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1850 @node Object Formats
1851 @section Object File Formats
1853 @cindex object file format
1854 The @sc{gnu} assembler can be configured to produce several alternative
1855 object file formats. For the most part, this does not affect how you
1856 write assembly language programs; but directives for debugging symbols
1857 are typically different in different file formats. @xref{Symbol
1858 Attributes,,Symbol Attributes}.
1861 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1862 @value{OBJ-NAME} format object files.
1864 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1866 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1867 @code{b.out} or COFF format object files.
1870 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1871 SOM or ELF format object files.
1876 @section Command Line
1878 @cindex command line conventions
1880 After the program name @command{@value{AS}}, the command line may contain
1881 options and file names. Options may appear in any order, and may be
1882 before, after, or between file names. The order of file names is
1885 @cindex standard input, as input file
1887 @file{--} (two hyphens) by itself names the standard input file
1888 explicitly, as one of the files for @command{@value{AS}} to assemble.
1890 @cindex options, command line
1891 Except for @samp{--} any command line argument that begins with a
1892 hyphen (@samp{-}) is an option. Each option changes the behavior of
1893 @command{@value{AS}}. No option changes the way another option works. An
1894 option is a @samp{-} followed by one or more letters; the case of
1895 the letter is important. All options are optional.
1897 Some options expect exactly one file name to follow them. The file
1898 name may either immediately follow the option's letter (compatible
1899 with older assemblers) or it may be the next command argument (@sc{gnu}
1900 standard). These two command lines are equivalent:
1903 @value{AS} -o my-object-file.o mumble.s
1904 @value{AS} -omy-object-file.o mumble.s
1908 @section Input Files
1911 @cindex source program
1912 @cindex files, input
1913 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1914 describe the program input to one run of @command{@value{AS}}. The program may
1915 be in one or more files; how the source is partitioned into files
1916 doesn't change the meaning of the source.
1918 @c I added "con" prefix to "catenation" just to prove I can overcome my
1919 @c APL training... doc@cygnus.com
1920 The source program is a concatenation of the text in all the files, in the
1923 @c man begin DESCRIPTION
1924 Each time you run @command{@value{AS}} it assembles exactly one source
1925 program. The source program is made up of one or more files.
1926 (The standard input is also a file.)
1928 You give @command{@value{AS}} a command line that has zero or more input file
1929 names. The input files are read (from left file name to right). A
1930 command line argument (in any position) that has no special meaning
1931 is taken to be an input file name.
1933 If you give @command{@value{AS}} no file names it attempts to read one input file
1934 from the @command{@value{AS}} standard input, which is normally your terminal. You
1935 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1938 Use @samp{--} if you need to explicitly name the standard input file
1939 in your command line.
1941 If the source is empty, @command{@value{AS}} produces a small, empty object
1946 @subheading Filenames and Line-numbers
1948 @cindex input file linenumbers
1949 @cindex line numbers, in input files
1950 There are two ways of locating a line in the input file (or files) and
1951 either may be used in reporting error messages. One way refers to a line
1952 number in a physical file; the other refers to a line number in a
1953 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1955 @dfn{Physical files} are those files named in the command line given
1956 to @command{@value{AS}}.
1958 @dfn{Logical files} are simply names declared explicitly by assembler
1959 directives; they bear no relation to physical files. Logical file names help
1960 error messages reflect the original source file, when @command{@value{AS}} source
1961 is itself synthesized from other files. @command{@value{AS}} understands the
1962 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1963 @ref{File,,@code{.file}}.
1966 @section Output (Object) File
1972 Every time you run @command{@value{AS}} it produces an output file, which is
1973 your assembly language program translated into numbers. This file
1974 is the object file. Its default name is
1982 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1984 You can give it another name by using the @option{-o} option. Conventionally,
1985 object file names end with @file{.o}. The default name is used for historical
1986 reasons: older assemblers were capable of assembling self-contained programs
1987 directly into a runnable program. (For some formats, this isn't currently
1988 possible, but it can be done for the @code{a.out} format.)
1992 The object file is meant for input to the linker @code{@value{LD}}. It contains
1993 assembled program code, information to help @code{@value{LD}} integrate
1994 the assembled program into a runnable file, and (optionally) symbolic
1995 information for the debugger.
1997 @c link above to some info file(s) like the description of a.out.
1998 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
2001 @section Error and Warning Messages
2003 @c man begin DESCRIPTION
2005 @cindex error messages
2006 @cindex warning messages
2007 @cindex messages from assembler
2008 @command{@value{AS}} may write warnings and error messages to the standard error
2009 file (usually your terminal). This should not happen when a compiler
2010 runs @command{@value{AS}} automatically. Warnings report an assumption made so
2011 that @command{@value{AS}} could keep assembling a flawed program; errors report a
2012 grave problem that stops the assembly.
2016 @cindex format of warning messages
2017 Warning messages have the format
2020 file_name:@b{NNN}:Warning Message Text
2024 @cindex line numbers, in warnings/errors
2025 (where @b{NNN} is a line number). If a logical file name has been given
2026 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
2027 the current input file is used. If a logical line number was given
2029 (@pxref{Line,,@code{.line}})
2031 then it is used to calculate the number printed,
2032 otherwise the actual line in the current source file is printed. The
2033 message text is intended to be self explanatory (in the grand Unix
2036 @cindex format of error messages
2037 Error messages have the format
2039 file_name:@b{NNN}:FATAL:Error Message Text
2041 The file name and line number are derived as for warning
2042 messages. The actual message text may be rather less explanatory
2043 because many of them aren't supposed to happen.
2046 @chapter Command-Line Options
2048 @cindex options, all versions of assembler
2049 This chapter describes command-line options available in @emph{all}
2050 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
2051 for options specific
2053 to the @value{TARGET} target.
2056 to particular machine architectures.
2059 @c man begin DESCRIPTION
2061 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2062 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2063 The assembler arguments must be separated from each other (and the @samp{-Wa})
2064 by commas. For example:
2067 gcc -c -g -O -Wa,-alh,-L file.c
2071 This passes two options to the assembler: @samp{-alh} (emit a listing to
2072 standard output with high-level and assembly source) and @samp{-L} (retain
2073 local symbols in the symbol table).
2075 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2076 command-line options are automatically passed to the assembler by the compiler.
2077 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2078 precisely what options it passes to each compilation pass, including the
2084 * a:: -a[cdghlns] enable listings
2085 * alternate:: --alternate enable alternate macro syntax
2086 * D:: -D for compatibility
2087 * f:: -f to work faster
2088 * I:: -I for .include search path
2089 @ifclear DIFF-TBL-KLUGE
2090 * K:: -K for compatibility
2092 @ifset DIFF-TBL-KLUGE
2093 * K:: -K for difference tables
2096 * L:: -L to retain local symbols
2097 * listing:: --listing-XXX to configure listing output
2098 * M:: -M or --mri to assemble in MRI compatibility mode
2099 * MD:: --MD for dependency tracking
2100 * o:: -o to name the object file
2101 * R:: -R to join data and text sections
2102 * statistics:: --statistics to see statistics about assembly
2103 * traditional-format:: --traditional-format for compatible output
2104 * v:: -v to announce version
2105 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2106 * Z:: -Z to make object file even after errors
2110 @section Enable Listings: @option{-a[cdghlns]}
2120 @cindex listings, enabling
2121 @cindex assembly listings, enabling
2123 These options enable listing output from the assembler. By itself,
2124 @samp{-a} requests high-level, assembly, and symbols listing.
2125 You can use other letters to select specific options for the list:
2126 @samp{-ah} requests a high-level language listing,
2127 @samp{-al} requests an output-program assembly listing, and
2128 @samp{-as} requests a symbol table listing.
2129 High-level listings require that a compiler debugging option like
2130 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2133 Use the @samp{-ag} option to print a first section with general assembly
2134 information, like @value{AS} version, switches passed, or time stamp.
2136 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2137 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2138 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2139 omitted from the listing.
2141 Use the @samp{-ad} option to omit debugging directives from the
2144 Once you have specified one of these options, you can further control
2145 listing output and its appearance using the directives @code{.list},
2146 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2148 The @samp{-an} option turns off all forms processing.
2149 If you do not request listing output with one of the @samp{-a} options, the
2150 listing-control directives have no effect.
2152 The letters after @samp{-a} may be combined into one option,
2153 @emph{e.g.}, @samp{-aln}.
2155 Note if the assembler source is coming from the standard input (e.g.,
2157 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
2158 is being used) then the listing will not contain any comments or preprocessor
2159 directives. This is because the listing code buffers input source lines from
2160 stdin only after they have been preprocessed by the assembler. This reduces
2161 memory usage and makes the code more efficient.
2164 @section @option{--alternate}
2167 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2170 @section @option{-D}
2173 This option has no effect whatsoever, but it is accepted to make it more
2174 likely that scripts written for other assemblers also work with
2175 @command{@value{AS}}.
2178 @section Work Faster: @option{-f}
2181 @cindex trusted compiler
2182 @cindex faster processing (@option{-f})
2183 @samp{-f} should only be used when assembling programs written by a
2184 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2185 and comment preprocessing on
2186 the input file(s) before assembling them. @xref{Preprocessing,
2190 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2191 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2196 @section @code{.include} Search Path: @option{-I} @var{path}
2198 @kindex -I @var{path}
2199 @cindex paths for @code{.include}
2200 @cindex search path for @code{.include}
2201 @cindex @code{include} directive search path
2202 Use this option to add a @var{path} to the list of directories
2203 @command{@value{AS}} searches for files specified in @code{.include}
2204 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2205 many times as necessary to include a variety of paths. The current
2206 working directory is always searched first; after that, @command{@value{AS}}
2207 searches any @samp{-I} directories in the same order as they were
2208 specified (left to right) on the command line.
2211 @section Difference Tables: @option{-K}
2214 @ifclear DIFF-TBL-KLUGE
2215 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2216 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2217 where it can be used to warn when the assembler alters the machine code
2218 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2219 family does not have the addressing limitations that sometimes lead to this
2220 alteration on other platforms.
2223 @ifset DIFF-TBL-KLUGE
2224 @cindex difference tables, warning
2225 @cindex warning for altered difference tables
2226 @command{@value{AS}} sometimes alters the code emitted for directives of the
2227 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2228 You can use the @samp{-K} option if you want a warning issued when this
2233 @section Include Local Symbols: @option{-L}
2236 @cindex local symbols, retaining in output
2237 Symbols beginning with system-specific local label prefixes, typically
2238 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2239 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2240 such symbols when debugging, because they are intended for the use of
2241 programs (like compilers) that compose assembler programs, not for your
2242 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2243 such symbols, so you do not normally debug with them.
2245 This option tells @command{@value{AS}} to retain those local symbols
2246 in the object file. Usually if you do this you also tell the linker
2247 @code{@value{LD}} to preserve those symbols.
2250 @section Configuring listing output: @option{--listing}
2252 The listing feature of the assembler can be enabled via the command line switch
2253 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2254 hex dump of the corresponding locations in the output object file, and displays
2255 them as a listing file. The format of this listing can be controlled by
2256 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2257 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2258 @code{.psize} (@pxref{Psize}), and
2259 @code{.eject} (@pxref{Eject}) and also by the following switches:
2262 @item --listing-lhs-width=@samp{number}
2263 @kindex --listing-lhs-width
2264 @cindex Width of first line disassembly output
2265 Sets the maximum width, in words, of the first line of the hex byte dump. This
2266 dump appears on the left hand side of the listing output.
2268 @item --listing-lhs-width2=@samp{number}
2269 @kindex --listing-lhs-width2
2270 @cindex Width of continuation lines of disassembly output
2271 Sets the maximum width, in words, of any further lines of the hex byte dump for
2272 a given input source line. If this value is not specified, it defaults to being
2273 the same as the value specified for @samp{--listing-lhs-width}. If neither
2274 switch is used the default is to one.
2276 @item --listing-rhs-width=@samp{number}
2277 @kindex --listing-rhs-width
2278 @cindex Width of source line output
2279 Sets the maximum width, in characters, of the source line that is displayed
2280 alongside the hex dump. The default value for this parameter is 100. The
2281 source line is displayed on the right hand side of the listing output.
2283 @item --listing-cont-lines=@samp{number}
2284 @kindex --listing-cont-lines
2285 @cindex Maximum number of continuation lines
2286 Sets the maximum number of continuation lines of hex dump that will be
2287 displayed for a given single line of source input. The default value is 4.
2291 @section Assemble in MRI Compatibility Mode: @option{-M}
2294 @cindex MRI compatibility mode
2295 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2296 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2297 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
2298 configured target) assembler from Microtec Research. The exact nature of the
2299 MRI syntax will not be documented here; see the MRI manuals for more
2300 information. Note in particular that the handling of macros and macro
2301 arguments is somewhat different. The purpose of this option is to permit
2302 assembling existing MRI assembler code using @command{@value{AS}}.
2304 The MRI compatibility is not complete. Certain operations of the MRI assembler
2305 depend upon its object file format, and can not be supported using other object
2306 file formats. Supporting these would require enhancing each object file format
2307 individually. These are:
2310 @item global symbols in common section
2312 The m68k MRI assembler supports common sections which are merged by the linker.
2313 Other object file formats do not support this. @command{@value{AS}} handles
2314 common sections by treating them as a single common symbol. It permits local
2315 symbols to be defined within a common section, but it can not support global
2316 symbols, since it has no way to describe them.
2318 @item complex relocations
2320 The MRI assemblers support relocations against a negated section address, and
2321 relocations which combine the start addresses of two or more sections. These
2322 are not support by other object file formats.
2324 @item @code{END} pseudo-op specifying start address
2326 The MRI @code{END} pseudo-op permits the specification of a start address.
2327 This is not supported by other object file formats. The start address may
2328 instead be specified using the @option{-e} option to the linker, or in a linker
2331 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2333 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2334 name to the output file. This is not supported by other object file formats.
2336 @item @code{ORG} pseudo-op
2338 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2339 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2340 which changes the location within the current section. Absolute sections are
2341 not supported by other object file formats. The address of a section may be
2342 assigned within a linker script.
2345 There are some other features of the MRI assembler which are not supported by
2346 @command{@value{AS}}, typically either because they are difficult or because they
2347 seem of little consequence. Some of these may be supported in future releases.
2351 @item EBCDIC strings
2353 EBCDIC strings are not supported.
2355 @item packed binary coded decimal
2357 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2358 and @code{DCB.P} pseudo-ops are not supported.
2360 @item @code{FEQU} pseudo-op
2362 The m68k @code{FEQU} pseudo-op is not supported.
2364 @item @code{NOOBJ} pseudo-op
2366 The m68k @code{NOOBJ} pseudo-op is not supported.
2368 @item @code{OPT} branch control options
2370 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2371 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2372 relaxes all branches, whether forward or backward, to an appropriate size, so
2373 these options serve no purpose.
2375 @item @code{OPT} list control options
2377 The following m68k @code{OPT} list control options are ignored: @code{C},
2378 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2379 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2381 @item other @code{OPT} options
2383 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2384 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2386 @item @code{OPT} @code{D} option is default
2388 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2389 @code{OPT NOD} may be used to turn it off.
2391 @item @code{XREF} pseudo-op.
2393 The m68k @code{XREF} pseudo-op is ignored.
2395 @item @code{.debug} pseudo-op
2397 The i960 @code{.debug} pseudo-op is not supported.
2399 @item @code{.extended} pseudo-op
2401 The i960 @code{.extended} pseudo-op is not supported.
2403 @item @code{.list} pseudo-op.
2405 The various options of the i960 @code{.list} pseudo-op are not supported.
2407 @item @code{.optimize} pseudo-op
2409 The i960 @code{.optimize} pseudo-op is not supported.
2411 @item @code{.output} pseudo-op
2413 The i960 @code{.output} pseudo-op is not supported.
2415 @item @code{.setreal} pseudo-op
2417 The i960 @code{.setreal} pseudo-op is not supported.
2422 @section Dependency Tracking: @option{--MD}
2425 @cindex dependency tracking
2428 @command{@value{AS}} can generate a dependency file for the file it creates. This
2429 file consists of a single rule suitable for @code{make} describing the
2430 dependencies of the main source file.
2432 The rule is written to the file named in its argument.
2434 This feature is used in the automatic updating of makefiles.
2437 @section Name the Object File: @option{-o}
2440 @cindex naming object file
2441 @cindex object file name
2442 There is always one object file output when you run @command{@value{AS}}. By
2443 default it has the name
2446 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2460 You use this option (which takes exactly one filename) to give the
2461 object file a different name.
2463 Whatever the object file is called, @command{@value{AS}} overwrites any
2464 existing file of the same name.
2467 @section Join Data and Text Sections: @option{-R}
2470 @cindex data and text sections, joining
2471 @cindex text and data sections, joining
2472 @cindex joining text and data sections
2473 @cindex merging text and data sections
2474 @option{-R} tells @command{@value{AS}} to write the object file as if all
2475 data-section data lives in the text section. This is only done at
2476 the very last moment: your binary data are the same, but data
2477 section parts are relocated differently. The data section part of
2478 your object file is zero bytes long because all its bytes are
2479 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2481 When you specify @option{-R} it would be possible to generate shorter
2482 address displacements (because we do not have to cross between text and
2483 data section). We refrain from doing this simply for compatibility with
2484 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2487 When @command{@value{AS}} is configured for COFF or ELF output,
2488 this option is only useful if you use sections named @samp{.text} and
2493 @option{-R} is not supported for any of the HPPA targets. Using
2494 @option{-R} generates a warning from @command{@value{AS}}.
2498 @section Display Assembly Statistics: @option{--statistics}
2500 @kindex --statistics
2501 @cindex statistics, about assembly
2502 @cindex time, total for assembly
2503 @cindex space used, maximum for assembly
2504 Use @samp{--statistics} to display two statistics about the resources used by
2505 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2506 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2509 @node traditional-format
2510 @section Compatible Output: @option{--traditional-format}
2512 @kindex --traditional-format
2513 For some targets, the output of @command{@value{AS}} is different in some ways
2514 from the output of some existing assembler. This switch requests
2515 @command{@value{AS}} to use the traditional format instead.
2517 For example, it disables the exception frame optimizations which
2518 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2521 @section Announce Version: @option{-v}
2525 @cindex assembler version
2526 @cindex version of assembler
2527 You can find out what version of as is running by including the
2528 option @samp{-v} (which you can also spell as @samp{-version}) on the
2532 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2534 @command{@value{AS}} should never give a warning or error message when
2535 assembling compiler output. But programs written by people often
2536 cause @command{@value{AS}} to give a warning that a particular assumption was
2537 made. All such warnings are directed to the standard error file.
2541 @cindex suppressing warnings
2542 @cindex warnings, suppressing
2543 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2544 This only affects the warning messages: it does not change any particular of
2545 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2548 @kindex --fatal-warnings
2549 @cindex errors, caused by warnings
2550 @cindex warnings, causing error
2551 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2552 files that generate warnings to be in error.
2555 @cindex warnings, switching on
2556 You can switch these options off again by specifying @option{--warn}, which
2557 causes warnings to be output as usual.
2560 @section Generate Object File in Spite of Errors: @option{-Z}
2561 @cindex object file, after errors
2562 @cindex errors, continuing after
2563 After an error message, @command{@value{AS}} normally produces no output. If for
2564 some reason you are interested in object file output even after
2565 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2566 option. If there are any errors, @command{@value{AS}} continues anyways, and
2567 writes an object file after a final warning message of the form @samp{@var{n}
2568 errors, @var{m} warnings, generating bad object file.}
2573 @cindex machine-independent syntax
2574 @cindex syntax, machine-independent
2575 This chapter describes the machine-independent syntax allowed in a
2576 source file. @command{@value{AS}} syntax is similar to what many other
2577 assemblers use; it is inspired by the BSD 4.2
2582 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2586 * Preprocessing:: Preprocessing
2587 * Whitespace:: Whitespace
2588 * Comments:: Comments
2589 * Symbol Intro:: Symbols
2590 * Statements:: Statements
2591 * Constants:: Constants
2595 @section Preprocessing
2597 @cindex preprocessing
2598 The @command{@value{AS}} internal preprocessor:
2600 @cindex whitespace, removed by preprocessor
2602 adjusts and removes extra whitespace. It leaves one space or tab before
2603 the keywords on a line, and turns any other whitespace on the line into
2606 @cindex comments, removed by preprocessor
2608 removes all comments, replacing them with a single space, or an
2609 appropriate number of newlines.
2611 @cindex constants, converted by preprocessor
2613 converts character constants into the appropriate numeric values.
2616 It does not do macro processing, include file handling, or
2617 anything else you may get from your C compiler's preprocessor. You can
2618 do include file processing with the @code{.include} directive
2619 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2620 to get other ``CPP'' style preprocessing by giving the input file a
2621 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2622 Output, gcc.info, Using GNU CC}.
2624 Excess whitespace, comments, and character constants
2625 cannot be used in the portions of the input text that are not
2628 @cindex turning preprocessing on and off
2629 @cindex preprocessing, turning on and off
2632 If the first line of an input file is @code{#NO_APP} or if you use the
2633 @samp{-f} option, whitespace and comments are not removed from the input file.
2634 Within an input file, you can ask for whitespace and comment removal in
2635 specific portions of the by putting a line that says @code{#APP} before the
2636 text that may contain whitespace or comments, and putting a line that says
2637 @code{#NO_APP} after this text. This feature is mainly intend to support
2638 @code{asm} statements in compilers whose output is otherwise free of comments
2645 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2646 Whitespace is used to separate symbols, and to make programs neater for
2647 people to read. Unless within character constants
2648 (@pxref{Characters,,Character Constants}), any whitespace means the same
2649 as exactly one space.
2655 There are two ways of rendering comments to @command{@value{AS}}. In both
2656 cases the comment is equivalent to one space.
2658 Anything from @samp{/*} through the next @samp{*/} is a comment.
2659 This means you may not nest these comments.
2663 The only way to include a newline ('\n') in a comment
2664 is to use this sort of comment.
2667 /* This sort of comment does not nest. */
2670 @cindex line comment character
2671 Anything from a @dfn{line comment} character up to the next newline is
2672 considered a comment and is ignored. The line comment character is target
2673 specific, and some targets multiple comment characters. Some targets also have
2674 line comment characters that only work if they are the first character on a
2675 line. Some targets use a sequence of two characters to introduce a line
2676 comment. Some targets can also change their line comment characters depending
2677 upon command line options that have been used. For more details see the
2678 @emph{Syntax} section in the documentation for individual targets.
2680 If the line comment character is the hash sign (@samp{#}) then it still has the
2681 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2682 to specify logical line numbers:
2685 @cindex lines starting with @code{#}
2686 @cindex logical line numbers
2687 To be compatible with past assemblers, lines that begin with @samp{#} have a
2688 special interpretation. Following the @samp{#} should be an absolute
2689 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2690 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2691 new logical file name. The rest of the line, if any, should be whitespace.
2693 If the first non-whitespace characters on the line are not numeric,
2694 the line is ignored. (Just like a comment.)
2697 # This is an ordinary comment.
2698 # 42-6 "new_file_name" # New logical file name
2699 # This is logical line # 36.
2701 This feature is deprecated, and may disappear from future versions
2702 of @command{@value{AS}}.
2707 @cindex characters used in symbols
2708 @ifclear SPECIAL-SYMS
2709 A @dfn{symbol} is one or more characters chosen from the set of all
2710 letters (both upper and lower case), digits and the three characters
2716 A @dfn{symbol} is one or more characters chosen from the set of all
2717 letters (both upper and lower case), digits and the three characters
2718 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2724 On most machines, you can also use @code{$} in symbol names; exceptions
2725 are noted in @ref{Machine Dependencies}.
2727 No symbol may begin with a digit. Case is significant.
2728 There is no length limit: all characters are significant. Multibyte characters
2729 are supported. Symbols are delimited by characters not in that set, or by the
2730 beginning of a file (since the source program must end with a newline, the end
2731 of a file is not a possible symbol delimiter). @xref{Symbols}.
2732 @cindex length of symbols
2737 @cindex statements, structure of
2738 @cindex line separator character
2739 @cindex statement separator character
2741 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2742 @dfn{line separator character}. The line separator character is target
2743 specific and described in the @emph{Syntax} section of each
2744 target's documentation. Not all targets support a line separator character.
2745 The newline or line separator character is considered to be part of the
2746 preceding statement. Newlines and separators within character constants are an
2747 exception: they do not end statements.
2749 @cindex newline, required at file end
2750 @cindex EOF, newline must precede
2751 It is an error to end any statement with end-of-file: the last
2752 character of any input file should be a newline.@refill
2754 An empty statement is allowed, and may include whitespace. It is ignored.
2756 @cindex instructions and directives
2757 @cindex directives and instructions
2758 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2759 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2761 A statement begins with zero or more labels, optionally followed by a
2762 key symbol which determines what kind of statement it is. The key
2763 symbol determines the syntax of the rest of the statement. If the
2764 symbol begins with a dot @samp{.} then the statement is an assembler
2765 directive: typically valid for any computer. If the symbol begins with
2766 a letter the statement is an assembly language @dfn{instruction}: it
2767 assembles into a machine language instruction.
2769 Different versions of @command{@value{AS}} for different computers
2770 recognize different instructions. In fact, the same symbol may
2771 represent a different instruction in a different computer's assembly
2775 @cindex @code{:} (label)
2776 @cindex label (@code{:})
2777 A label is a symbol immediately followed by a colon (@code{:}).
2778 Whitespace before a label or after a colon is permitted, but you may not
2779 have whitespace between a label's symbol and its colon. @xref{Labels}.
2782 For HPPA targets, labels need not be immediately followed by a colon, but
2783 the definition of a label must begin in column zero. This also implies that
2784 only one label may be defined on each line.
2788 label: .directive followed by something
2789 another_label: # This is an empty statement.
2790 instruction operand_1, operand_2, @dots{}
2797 A constant is a number, written so that its value is known by
2798 inspection, without knowing any context. Like this:
2801 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2802 .ascii "Ring the bell\7" # A string constant.
2803 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2804 .float 0f-314159265358979323846264338327\
2805 95028841971.693993751E-40 # - pi, a flonum.
2810 * Characters:: Character Constants
2811 * Numbers:: Number Constants
2815 @subsection Character Constants
2817 @cindex character constants
2818 @cindex constants, character
2819 There are two kinds of character constants. A @dfn{character} stands
2820 for one character in one byte and its value may be used in
2821 numeric expressions. String constants (properly called string
2822 @emph{literals}) are potentially many bytes and their values may not be
2823 used in arithmetic expressions.
2827 * Chars:: Characters
2831 @subsubsection Strings
2833 @cindex string constants
2834 @cindex constants, string
2835 A @dfn{string} is written between double-quotes. It may contain
2836 double-quotes or null characters. The way to get special characters
2837 into a string is to @dfn{escape} these characters: precede them with
2838 a backslash @samp{\} character. For example @samp{\\} represents
2839 one backslash: the first @code{\} is an escape which tells
2840 @command{@value{AS}} to interpret the second character literally as a backslash
2841 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2842 escape character). The complete list of escapes follows.
2844 @cindex escape codes, character
2845 @cindex character escape codes
2848 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2850 @cindex @code{\b} (backspace character)
2851 @cindex backspace (@code{\b})
2853 Mnemonic for backspace; for ASCII this is octal code 010.
2856 @c Mnemonic for EOText; for ASCII this is octal code 004.
2858 @cindex @code{\f} (formfeed character)
2859 @cindex formfeed (@code{\f})
2861 Mnemonic for FormFeed; for ASCII this is octal code 014.
2863 @cindex @code{\n} (newline character)
2864 @cindex newline (@code{\n})
2866 Mnemonic for newline; for ASCII this is octal code 012.
2869 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2871 @cindex @code{\r} (carriage return character)
2872 @cindex carriage return (@code{\r})
2874 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2877 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2878 @c other assemblers.
2880 @cindex @code{\t} (tab)
2881 @cindex tab (@code{\t})
2883 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2886 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2887 @c @item \x @var{digit} @var{digit} @var{digit}
2888 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2890 @cindex @code{\@var{ddd}} (octal character code)
2891 @cindex octal character code (@code{\@var{ddd}})
2892 @item \ @var{digit} @var{digit} @var{digit}
2893 An octal character code. The numeric code is 3 octal digits.
2894 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2895 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2897 @cindex @code{\@var{xd...}} (hex character code)
2898 @cindex hex character code (@code{\@var{xd...}})
2899 @item \@code{x} @var{hex-digits...}
2900 A hex character code. All trailing hex digits are combined. Either upper or
2901 lower case @code{x} works.
2903 @cindex @code{\\} (@samp{\} character)
2904 @cindex backslash (@code{\\})
2906 Represents one @samp{\} character.
2909 @c Represents one @samp{'} (accent acute) character.
2910 @c This is needed in single character literals
2911 @c (@xref{Characters,,Character Constants}.) to represent
2914 @cindex @code{\"} (doublequote character)
2915 @cindex doublequote (@code{\"})
2917 Represents one @samp{"} character. Needed in strings to represent
2918 this character, because an unescaped @samp{"} would end the string.
2920 @item \ @var{anything-else}
2921 Any other character when escaped by @kbd{\} gives a warning, but
2922 assembles as if the @samp{\} was not present. The idea is that if
2923 you used an escape sequence you clearly didn't want the literal
2924 interpretation of the following character. However @command{@value{AS}} has no
2925 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2926 code and warns you of the fact.
2929 Which characters are escapable, and what those escapes represent,
2930 varies widely among assemblers. The current set is what we think
2931 the BSD 4.2 assembler recognizes, and is a subset of what most C
2932 compilers recognize. If you are in doubt, do not use an escape
2936 @subsubsection Characters
2938 @cindex single character constant
2939 @cindex character, single
2940 @cindex constant, single character
2941 A single character may be written as a single quote immediately
2942 followed by that character. The same escapes apply to characters as
2943 to strings. So if you want to write the character backslash, you
2944 must write @kbd{'\\} where the first @code{\} escapes the second
2945 @code{\}. As you can see, the quote is an acute accent, not a
2946 grave accent. A newline
2948 @ifclear abnormal-separator
2949 (or semicolon @samp{;})
2951 @ifset abnormal-separator
2953 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2958 immediately following an acute accent is taken as a literal character
2959 and does not count as the end of a statement. The value of a character
2960 constant in a numeric expression is the machine's byte-wide code for
2961 that character. @command{@value{AS}} assumes your character code is ASCII:
2962 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2965 @subsection Number Constants
2967 @cindex constants, number
2968 @cindex number constants
2969 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2970 are stored in the target machine. @emph{Integers} are numbers that
2971 would fit into an @code{int} in the C language. @emph{Bignums} are
2972 integers, but they are stored in more than 32 bits. @emph{Flonums}
2973 are floating point numbers, described below.
2976 * Integers:: Integers
2981 * Bit Fields:: Bit Fields
2987 @subsubsection Integers
2989 @cindex constants, integer
2991 @cindex binary integers
2992 @cindex integers, binary
2993 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2994 the binary digits @samp{01}.
2996 @cindex octal integers
2997 @cindex integers, octal
2998 An octal integer is @samp{0} followed by zero or more of the octal
2999 digits (@samp{01234567}).
3001 @cindex decimal integers
3002 @cindex integers, decimal
3003 A decimal integer starts with a non-zero digit followed by zero or
3004 more digits (@samp{0123456789}).
3006 @cindex hexadecimal integers
3007 @cindex integers, hexadecimal
3008 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
3009 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
3011 Integers have the usual values. To denote a negative integer, use
3012 the prefix operator @samp{-} discussed under expressions
3013 (@pxref{Prefix Ops,,Prefix Operators}).
3016 @subsubsection Bignums
3019 @cindex constants, bignum
3020 A @dfn{bignum} has the same syntax and semantics as an integer
3021 except that the number (or its negative) takes more than 32 bits to
3022 represent in binary. The distinction is made because in some places
3023 integers are permitted while bignums are not.
3026 @subsubsection Flonums
3028 @cindex floating point numbers
3029 @cindex constants, floating point
3031 @cindex precision, floating point
3032 A @dfn{flonum} represents a floating point number. The translation is
3033 indirect: a decimal floating point number from the text is converted by
3034 @command{@value{AS}} to a generic binary floating point number of more than
3035 sufficient precision. This generic floating point number is converted
3036 to a particular computer's floating point format (or formats) by a
3037 portion of @command{@value{AS}} specialized to that computer.
3039 A flonum is written by writing (in order)
3044 (@samp{0} is optional on the HPPA.)
3048 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
3050 @kbd{e} is recommended. Case is not important.
3052 @c FIXME: verify if flonum syntax really this vague for most cases
3053 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
3054 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
3057 On the H8/300, Renesas / SuperH SH,
3058 and AMD 29K architectures, the letter must be
3059 one of the letters @samp{DFPRSX} (in upper or lower case).
3061 On the ARC, the letter must be one of the letters @samp{DFRS}
3062 (in upper or lower case).
3064 On the Intel 960 architecture, the letter must be
3065 one of the letters @samp{DFT} (in upper or lower case).
3067 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3071 One of the letters @samp{DFRS} (in upper or lower case).
3074 One of the letters @samp{DFPRSX} (in upper or lower case).
3077 The letter @samp{E} (upper case only).
3080 One of the letters @samp{DFT} (in upper or lower case).
3085 An optional sign: either @samp{+} or @samp{-}.
3088 An optional @dfn{integer part}: zero or more decimal digits.
3091 An optional @dfn{fractional part}: @samp{.} followed by zero
3092 or more decimal digits.
3095 An optional exponent, consisting of:
3099 An @samp{E} or @samp{e}.
3100 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3101 @c principle this can perfectly well be different on different targets.
3103 Optional sign: either @samp{+} or @samp{-}.
3105 One or more decimal digits.
3110 At least one of the integer part or the fractional part must be
3111 present. The floating point number has the usual base-10 value.
3113 @command{@value{AS}} does all processing using integers. Flonums are computed
3114 independently of any floating point hardware in the computer running
3115 @command{@value{AS}}.
3119 @c Bit fields are written as a general facility but are also controlled
3120 @c by a conditional-compilation flag---which is as of now (21mar91)
3121 @c turned on only by the i960 config of GAS.
3123 @subsubsection Bit Fields
3126 @cindex constants, bit field
3127 You can also define numeric constants as @dfn{bit fields}.
3128 Specify two numbers separated by a colon---
3130 @var{mask}:@var{value}
3133 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
3136 The resulting number is then packed
3138 @c this conditional paren in case bit fields turned on elsewhere than 960
3139 (in host-dependent byte order)
3141 into a field whose width depends on which assembler directive has the
3142 bit-field as its argument. Overflow (a result from the bitwise and
3143 requiring more binary digits to represent) is not an error; instead,
3144 more constants are generated, of the specified width, beginning with the
3145 least significant digits.@refill
3147 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
3148 @code{.short}, and @code{.word} accept bit-field arguments.
3153 @chapter Sections and Relocation
3158 * Secs Background:: Background
3159 * Ld Sections:: Linker Sections
3160 * As Sections:: Assembler Internal Sections
3161 * Sub-Sections:: Sub-Sections
3165 @node Secs Background
3168 Roughly, a section is a range of addresses, with no gaps; all data
3169 ``in'' those addresses is treated the same for some particular purpose.
3170 For example there may be a ``read only'' section.
3172 @cindex linker, and assembler
3173 @cindex assembler, and linker
3174 The linker @code{@value{LD}} reads many object files (partial programs) and
3175 combines their contents to form a runnable program. When @command{@value{AS}}
3176 emits an object file, the partial program is assumed to start at address 0.
3177 @code{@value{LD}} assigns the final addresses for the partial program, so that
3178 different partial programs do not overlap. This is actually an
3179 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3182 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3183 addresses. These blocks slide to their run-time addresses as rigid
3184 units; their length does not change and neither does the order of bytes
3185 within them. Such a rigid unit is called a @emph{section}. Assigning
3186 run-time addresses to sections is called @dfn{relocation}. It includes
3187 the task of adjusting mentions of object-file addresses so they refer to
3188 the proper run-time addresses.
3190 For the H8/300, and for the Renesas / SuperH SH,
3191 @command{@value{AS}} pads sections if needed to
3192 ensure they end on a word (sixteen bit) boundary.
3195 @cindex standard assembler sections
3196 An object file written by @command{@value{AS}} has at least three sections, any
3197 of which may be empty. These are named @dfn{text}, @dfn{data} and
3202 When it generates COFF or ELF output,
3204 @command{@value{AS}} can also generate whatever other named sections you specify
3205 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3206 If you do not use any directives that place output in the @samp{.text}
3207 or @samp{.data} sections, these sections still exist, but are empty.
3212 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3214 @command{@value{AS}} can also generate whatever other named sections you
3215 specify using the @samp{.space} and @samp{.subspace} directives. See
3216 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3217 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3218 assembler directives.
3221 Additionally, @command{@value{AS}} uses different names for the standard
3222 text, data, and bss sections when generating SOM output. Program text
3223 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3224 BSS into @samp{$BSS$}.
3228 Within the object file, the text section starts at address @code{0}, the
3229 data section follows, and the bss section follows the data section.
3232 When generating either SOM or ELF output files on the HPPA, the text
3233 section starts at address @code{0}, the data section at address
3234 @code{0x4000000}, and the bss section follows the data section.
3237 To let @code{@value{LD}} know which data changes when the sections are
3238 relocated, and how to change that data, @command{@value{AS}} also writes to the
3239 object file details of the relocation needed. To perform relocation
3240 @code{@value{LD}} must know, each time an address in the object
3244 Where in the object file is the beginning of this reference to
3247 How long (in bytes) is this reference?
3249 Which section does the address refer to? What is the numeric value of
3251 (@var{address}) @minus{} (@var{start-address of section})?
3254 Is the reference to an address ``Program-Counter relative''?
3257 @cindex addresses, format of
3258 @cindex section-relative addressing
3259 In fact, every address @command{@value{AS}} ever uses is expressed as
3261 (@var{section}) + (@var{offset into section})
3264 Further, most expressions @command{@value{AS}} computes have this section-relative
3267 (For some object formats, such as SOM for the HPPA, some expressions are
3268 symbol-relative instead.)
3271 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3272 @var{N} into section @var{secname}.''
3274 Apart from text, data and bss sections you need to know about the
3275 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3276 addresses in the absolute section remain unchanged. For example, address
3277 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3278 @code{@value{LD}}. Although the linker never arranges two partial programs'
3279 data sections with overlapping addresses after linking, @emph{by definition}
3280 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3281 part of a program is always the same address when the program is running as
3282 address @code{@{absolute@ 239@}} in any other part of the program.
3284 The idea of sections is extended to the @dfn{undefined} section. Any
3285 address whose section is unknown at assembly time is by definition
3286 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3287 Since numbers are always defined, the only way to generate an undefined
3288 address is to mention an undefined symbol. A reference to a named
3289 common block would be such a symbol: its value is unknown at assembly
3290 time so it has section @emph{undefined}.
3292 By analogy the word @emph{section} is used to describe groups of sections in
3293 the linked program. @code{@value{LD}} puts all partial programs' text
3294 sections in contiguous addresses in the linked program. It is
3295 customary to refer to the @emph{text section} of a program, meaning all
3296 the addresses of all partial programs' text sections. Likewise for
3297 data and bss sections.
3299 Some sections are manipulated by @code{@value{LD}}; others are invented for
3300 use of @command{@value{AS}} and have no meaning except during assembly.
3303 @section Linker Sections
3304 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3309 @cindex named sections
3310 @cindex sections, named
3311 @item named sections
3314 @cindex text section
3315 @cindex data section
3319 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3320 separate but equal sections. Anything you can say of one section is
3323 When the program is running, however, it is
3324 customary for the text section to be unalterable. The
3325 text section is often shared among processes: it contains
3326 instructions, constants and the like. The data section of a running
3327 program is usually alterable: for example, C variables would be stored
3328 in the data section.
3333 This section contains zeroed bytes when your program begins running. It
3334 is used to hold uninitialized variables or common storage. The length of
3335 each partial program's bss section is important, but because it starts
3336 out containing zeroed bytes there is no need to store explicit zero
3337 bytes in the object file. The bss section was invented to eliminate
3338 those explicit zeros from object files.
3340 @cindex absolute section
3341 @item absolute section
3342 Address 0 of this section is always ``relocated'' to runtime address 0.
3343 This is useful if you want to refer to an address that @code{@value{LD}} must
3344 not change when relocating. In this sense we speak of absolute
3345 addresses being ``unrelocatable'': they do not change during relocation.
3347 @cindex undefined section
3348 @item undefined section
3349 This ``section'' is a catch-all for address references to objects not in
3350 the preceding sections.
3351 @c FIXME: ref to some other doc on obj-file formats could go here.
3354 @cindex relocation example
3355 An idealized example of three relocatable sections follows.
3357 The example uses the traditional section names @samp{.text} and @samp{.data}.
3359 Memory addresses are on the horizontal axis.
3363 @c END TEXI2ROFF-KILL
3366 partial program # 1: |ttttt|dddd|00|
3373 partial program # 2: |TTT|DDD|000|
3376 +--+---+-----+--+----+---+-----+~~
3377 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3378 +--+---+-----+--+----+---+-----+~~
3380 addresses: 0 @dots{}
3387 \line{\it Partial program \#1: \hfil}
3388 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3389 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3391 \line{\it Partial program \#2: \hfil}
3392 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3393 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3395 \line{\it linked program: \hfil}
3396 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3397 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3398 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3399 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3401 \line{\it addresses: \hfil}
3405 @c END TEXI2ROFF-KILL
3408 @section Assembler Internal Sections
3410 @cindex internal assembler sections
3411 @cindex sections in messages, internal
3412 These sections are meant only for the internal use of @command{@value{AS}}. They
3413 have no meaning at run-time. You do not really need to know about these
3414 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3415 warning messages, so it might be helpful to have an idea of their
3416 meanings to @command{@value{AS}}. These sections are used to permit the
3417 value of every expression in your assembly language program to be a
3418 section-relative address.
3421 @cindex assembler internal logic error
3422 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3423 An internal assembler logic error has been found. This means there is a
3424 bug in the assembler.
3426 @cindex expr (internal section)
3428 The assembler stores complex expression internally as combinations of
3429 symbols. When it needs to represent an expression as a symbol, it puts
3430 it in the expr section.
3432 @c FIXME item transfer[t] vector preload
3433 @c FIXME item transfer[t] vector postload
3434 @c FIXME item register
3438 @section Sub-Sections
3440 @cindex numbered subsections
3441 @cindex grouping data
3447 fall into two sections: text and data.
3449 You may have separate groups of
3451 data in named sections
3455 data in named sections
3461 that you want to end up near to each other in the object file, even though they
3462 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3463 use @dfn{subsections} for this purpose. Within each section, there can be
3464 numbered subsections with values from 0 to 8192. Objects assembled into the
3465 same subsection go into the object file together with other objects in the same
3466 subsection. For example, a compiler might want to store constants in the text
3467 section, but might not want to have them interspersed with the program being
3468 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3469 section of code being output, and a @samp{.text 1} before each group of
3470 constants being output.
3472 Subsections are optional. If you do not use subsections, everything
3473 goes in subsection number zero.
3476 Each subsection is zero-padded up to a multiple of four bytes.
3477 (Subsections may be padded a different amount on different flavors
3478 of @command{@value{AS}}.)
3482 On the H8/300 platform, each subsection is zero-padded to a word
3483 boundary (two bytes).
3484 The same is true on the Renesas SH.
3487 @c FIXME section padding (alignment)?
3488 @c Rich Pixley says padding here depends on target obj code format; that
3489 @c doesn't seem particularly useful to say without further elaboration,
3490 @c so for now I say nothing about it. If this is a generic BFD issue,
3491 @c these paragraphs might need to vanish from this manual, and be
3492 @c discussed in BFD chapter of binutils (or some such).
3496 Subsections appear in your object file in numeric order, lowest numbered
3497 to highest. (All this to be compatible with other people's assemblers.)
3498 The object file contains no representation of subsections; @code{@value{LD}} and
3499 other programs that manipulate object files see no trace of them.
3500 They just see all your text subsections as a text section, and all your
3501 data subsections as a data section.
3503 To specify which subsection you want subsequent statements assembled
3504 into, use a numeric argument to specify it, in a @samp{.text
3505 @var{expression}} or a @samp{.data @var{expression}} statement.
3508 When generating COFF output, you
3513 can also use an extra subsection
3514 argument with arbitrary named sections: @samp{.section @var{name},
3519 When generating ELF output, you
3524 can also use the @code{.subsection} directive (@pxref{SubSection})
3525 to specify a subsection: @samp{.subsection @var{expression}}.
3527 @var{Expression} should be an absolute expression
3528 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3529 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3530 begins in @code{text 0}. For instance:
3532 .text 0 # The default subsection is text 0 anyway.
3533 .ascii "This lives in the first text subsection. *"
3535 .ascii "But this lives in the second text subsection."
3537 .ascii "This lives in the data section,"
3538 .ascii "in the first data subsection."
3540 .ascii "This lives in the first text section,"
3541 .ascii "immediately following the asterisk (*)."
3544 Each section has a @dfn{location counter} incremented by one for every byte
3545 assembled into that section. Because subsections are merely a convenience
3546 restricted to @command{@value{AS}} there is no concept of a subsection location
3547 counter. There is no way to directly manipulate a location counter---but the
3548 @code{.align} directive changes it, and any label definition captures its
3549 current value. The location counter of the section where statements are being
3550 assembled is said to be the @dfn{active} location counter.
3553 @section bss Section
3556 @cindex common variable storage
3557 The bss section is used for local common variable storage.
3558 You may allocate address space in the bss section, but you may
3559 not dictate data to load into it before your program executes. When
3560 your program starts running, all the contents of the bss
3561 section are zeroed bytes.
3563 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3564 @ref{Lcomm,,@code{.lcomm}}.
3566 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3567 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3570 When assembling for a target which supports multiple sections, such as ELF or
3571 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3572 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3573 section. Typically the section will only contain symbol definitions and
3574 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3581 Symbols are a central concept: the programmer uses symbols to name
3582 things, the linker uses symbols to link, and the debugger uses symbols
3586 @cindex debuggers, and symbol order
3587 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3588 the same order they were declared. This may break some debuggers.
3593 * Setting Symbols:: Giving Symbols Other Values
3594 * Symbol Names:: Symbol Names
3595 * Dot:: The Special Dot Symbol
3596 * Symbol Attributes:: Symbol Attributes
3603 A @dfn{label} is written as a symbol immediately followed by a colon
3604 @samp{:}. The symbol then represents the current value of the
3605 active location counter, and is, for example, a suitable instruction
3606 operand. You are warned if you use the same symbol to represent two
3607 different locations: the first definition overrides any other
3611 On the HPPA, the usual form for a label need not be immediately followed by a
3612 colon, but instead must start in column zero. Only one label may be defined on
3613 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3614 provides a special directive @code{.label} for defining labels more flexibly.
3617 @node Setting Symbols
3618 @section Giving Symbols Other Values
3620 @cindex assigning values to symbols
3621 @cindex symbol values, assigning
3622 A symbol can be given an arbitrary value by writing a symbol, followed
3623 by an equals sign @samp{=}, followed by an expression
3624 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3625 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3626 equals sign @samp{=}@samp{=} here represents an equivalent of the
3627 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3630 Blackfin does not support symbol assignment with @samp{=}.
3634 @section Symbol Names
3636 @cindex symbol names
3637 @cindex names, symbol
3638 @ifclear SPECIAL-SYMS
3639 Symbol names begin with a letter or with one of @samp{._}. On most
3640 machines, you can also use @code{$} in symbol names; exceptions are
3641 noted in @ref{Machine Dependencies}. That character may be followed by any
3642 string of digits, letters, dollar signs (unless otherwise noted for a
3643 particular target machine), and underscores.
3647 Symbol names begin with a letter or with one of @samp{._}. On the
3648 Renesas SH you can also use @code{$} in symbol names. That
3649 character may be followed by any string of digits, letters, dollar signs (save
3650 on the H8/300), and underscores.
3654 Case of letters is significant: @code{foo} is a different symbol name
3657 Multibyte characters are supported. To generate a symbol name containing
3658 multibyte characters enclose it within double quotes and use escape codes. cf
3659 @xref{Strings}. Generating a multibyte symbol name from a label is not
3660 currently supported.
3662 Each symbol has exactly one name. Each name in an assembly language program
3663 refers to exactly one symbol. You may use that symbol name any number of times
3666 @subheading Local Symbol Names
3668 @cindex local symbol names
3669 @cindex symbol names, local
3670 A local symbol is any symbol beginning with certain local label prefixes.
3671 By default, the local label prefix is @samp{.L} for ELF systems or
3672 @samp{L} for traditional a.out systems, but each target may have its own
3673 set of local label prefixes.
3675 On the HPPA local symbols begin with @samp{L$}.
3678 Local symbols are defined and used within the assembler, but they are
3679 normally not saved in object files. Thus, they are not visible when debugging.
3680 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols})
3681 to retain the local symbols in the object files.
3683 @subheading Local Labels
3685 @cindex local labels
3686 @cindex temporary symbol names
3687 @cindex symbol names, temporary
3688 Local labels help compilers and programmers use names temporarily.
3689 They create symbols which are guaranteed to be unique over the entire scope of
3690 the input source code and which can be referred to by a simple notation.
3691 To define a local label, write a label of the form @samp{@b{N}:} (where @b{N}
3692 represents any positive integer). To refer to the most recent previous
3693 definition of that label write @samp{@b{N}b}, using the same number as when
3694 you defined the label. To refer to the next definition of a local label, write
3695 @samp{@b{N}f}---the @samp{b} stands for ``backwards'' and the @samp{f} stands
3698 There is no restriction on how you can use these labels, and you can reuse them
3699 too. So that it is possible to repeatedly define the same local label (using
3700 the same number @samp{@b{N}}), although you can only refer to the most recently
3701 defined local label of that number (for a backwards reference) or the next
3702 definition of a specific local label for a forward reference. It is also worth
3703 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3704 implemented in a slightly more efficient manner than the others.
3715 Which is the equivalent of:
3718 label_1: branch label_3
3719 label_2: branch label_1
3720 label_3: branch label_4
3721 label_4: branch label_3
3724 Local label names are only a notational device. They are immediately
3725 transformed into more conventional symbol names before the assembler uses them.
3726 The symbol names are stored in the symbol table, appear in error messages, and
3727 are optionally emitted to the object file. The names are constructed using
3731 @item @emph{local label prefix}
3732 All local symbols begin with the system-specific local label prefix.
3733 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3734 that start with the local label prefix. These labels are
3735 used for symbols you are never intended to see. If you use the
3736 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3737 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3738 you may use them in debugging.
3741 This is the number that was used in the local label definition. So if the
3742 label is written @samp{55:} then the number is @samp{55}.
3745 This unusual character is included so you do not accidentally invent a symbol
3746 of the same name. The character has ASCII value of @samp{\002} (control-B).
3748 @item @emph{ordinal number}
3749 This is a serial number to keep the labels distinct. The first definition of
3750 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3751 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3752 the number @samp{1} and its 15th definition gets @samp{15} as well.
3755 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3756 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3758 @subheading Dollar Local Labels
3759 @cindex dollar local symbols
3761 @code{@value{AS}} also supports an even more local form of local labels called
3762 dollar labels. These labels go out of scope (i.e., they become undefined) as
3763 soon as a non-local label is defined. Thus they remain valid for only a small
3764 region of the input source code. Normal local labels, by contrast, remain in
3765 scope for the entire file, or until they are redefined by another occurrence of
3766 the same local label.
3768 Dollar labels are defined in exactly the same way as ordinary local labels,
3769 except that they have a dollar sign suffix to their numeric value, e.g.,
3772 They can also be distinguished from ordinary local labels by their transformed
3773 names which use ASCII character @samp{\001} (control-A) as the magic character
3774 to distinguish them from ordinary labels. For example, the fifth definition of
3775 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3778 @section The Special Dot Symbol
3780 @cindex dot (symbol)
3781 @cindex @code{.} (symbol)
3782 @cindex current address
3783 @cindex location counter
3784 The special symbol @samp{.} refers to the current address that
3785 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3786 .long .} defines @code{melvin} to contain its own address.
3787 Assigning a value to @code{.} is treated the same as a @code{.org}
3789 @ifclear no-space-dir
3790 Thus, the expression @samp{.=.+4} is the same as saying
3794 @node Symbol Attributes
3795 @section Symbol Attributes
3797 @cindex symbol attributes
3798 @cindex attributes, symbol
3799 Every symbol has, as well as its name, the attributes ``Value'' and
3800 ``Type''. Depending on output format, symbols can also have auxiliary
3803 The detailed definitions are in @file{a.out.h}.
3806 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3807 all these attributes, and probably won't warn you. This makes the
3808 symbol an externally defined symbol, which is generally what you
3812 * Symbol Value:: Value
3813 * Symbol Type:: Type
3816 * a.out Symbols:: Symbol Attributes: @code{a.out}
3820 * a.out Symbols:: Symbol Attributes: @code{a.out}
3823 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3828 * COFF Symbols:: Symbol Attributes for COFF
3831 * SOM Symbols:: Symbol Attributes for SOM
3838 @cindex value of a symbol
3839 @cindex symbol value
3840 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3841 location in the text, data, bss or absolute sections the value is the
3842 number of addresses from the start of that section to the label.
3843 Naturally for text, data and bss sections the value of a symbol changes
3844 as @code{@value{LD}} changes section base addresses during linking. Absolute
3845 symbols' values do not change during linking: that is why they are
3848 The value of an undefined symbol is treated in a special way. If it is
3849 0 then the symbol is not defined in this assembler source file, and
3850 @code{@value{LD}} tries to determine its value from other files linked into the
3851 same program. You make this kind of symbol simply by mentioning a symbol
3852 name without defining it. A non-zero value represents a @code{.comm}
3853 common declaration. The value is how much common storage to reserve, in
3854 bytes (addresses). The symbol refers to the first address of the
3860 @cindex type of a symbol
3862 The type attribute of a symbol contains relocation (section)
3863 information, any flag settings indicating that a symbol is external, and
3864 (optionally), other information for linkers and debuggers. The exact
3865 format depends on the object-code output format in use.
3870 @c The following avoids a "widow" subsection title. @group would be
3871 @c better if it were available outside examples.
3874 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3876 @cindex @code{b.out} symbol attributes
3877 @cindex symbol attributes, @code{b.out}
3878 These symbol attributes appear only when @command{@value{AS}} is configured for
3879 one of the Berkeley-descended object output formats---@code{a.out} or
3885 @subsection Symbol Attributes: @code{a.out}
3887 @cindex @code{a.out} symbol attributes
3888 @cindex symbol attributes, @code{a.out}
3894 @subsection Symbol Attributes: @code{a.out}
3896 @cindex @code{a.out} symbol attributes
3897 @cindex symbol attributes, @code{a.out}
3901 * Symbol Desc:: Descriptor
3902 * Symbol Other:: Other
3906 @subsubsection Descriptor
3908 @cindex descriptor, of @code{a.out} symbol
3909 This is an arbitrary 16-bit value. You may establish a symbol's
3910 descriptor value by using a @code{.desc} statement
3911 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3912 @command{@value{AS}}.
3915 @subsubsection Other
3917 @cindex other attribute, of @code{a.out} symbol
3918 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3923 @subsection Symbol Attributes for COFF
3925 @cindex COFF symbol attributes
3926 @cindex symbol attributes, COFF
3928 The COFF format supports a multitude of auxiliary symbol attributes;
3929 like the primary symbol attributes, they are set between @code{.def} and
3930 @code{.endef} directives.
3932 @subsubsection Primary Attributes
3934 @cindex primary attributes, COFF symbols
3935 The symbol name is set with @code{.def}; the value and type,
3936 respectively, with @code{.val} and @code{.type}.
3938 @subsubsection Auxiliary Attributes
3940 @cindex auxiliary attributes, COFF symbols
3941 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3942 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3943 table information for COFF.
3948 @subsection Symbol Attributes for SOM
3950 @cindex SOM symbol attributes
3951 @cindex symbol attributes, SOM
3953 The SOM format for the HPPA supports a multitude of symbol attributes set with
3954 the @code{.EXPORT} and @code{.IMPORT} directives.
3956 The attributes are described in @cite{HP9000 Series 800 Assembly
3957 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3958 @code{EXPORT} assembler directive documentation.
3962 @chapter Expressions
3966 @cindex numeric values
3967 An @dfn{expression} specifies an address or numeric value.
3968 Whitespace may precede and/or follow an expression.
3970 The result of an expression must be an absolute number, or else an offset into
3971 a particular section. If an expression is not absolute, and there is not
3972 enough information when @command{@value{AS}} sees the expression to know its
3973 section, a second pass over the source program might be necessary to interpret
3974 the expression---but the second pass is currently not implemented.
3975 @command{@value{AS}} aborts with an error message in this situation.
3978 * Empty Exprs:: Empty Expressions
3979 * Integer Exprs:: Integer Expressions
3983 @section Empty Expressions
3985 @cindex empty expressions
3986 @cindex expressions, empty
3987 An empty expression has no value: it is just whitespace or null.
3988 Wherever an absolute expression is required, you may omit the
3989 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3990 is compatible with other assemblers.
3993 @section Integer Expressions
3995 @cindex integer expressions
3996 @cindex expressions, integer
3997 An @dfn{integer expression} is one or more @emph{arguments} delimited
3998 by @emph{operators}.
4001 * Arguments:: Arguments
4002 * Operators:: Operators
4003 * Prefix Ops:: Prefix Operators
4004 * Infix Ops:: Infix Operators
4008 @subsection Arguments
4010 @cindex expression arguments
4011 @cindex arguments in expressions
4012 @cindex operands in expressions
4013 @cindex arithmetic operands
4014 @dfn{Arguments} are symbols, numbers or subexpressions. In other
4015 contexts arguments are sometimes called ``arithmetic operands''. In
4016 this manual, to avoid confusing them with the ``instruction operands'' of
4017 the machine language, we use the term ``argument'' to refer to parts of
4018 expressions only, reserving the word ``operand'' to refer only to machine
4019 instruction operands.
4021 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
4022 @var{section} is one of text, data, bss, absolute,
4023 or undefined. @var{NNN} is a signed, 2's complement 32 bit
4026 Numbers are usually integers.
4028 A number can be a flonum or bignum. In this case, you are warned
4029 that only the low order 32 bits are used, and @command{@value{AS}} pretends
4030 these 32 bits are an integer. You may write integer-manipulating
4031 instructions that act on exotic constants, compatible with other
4034 @cindex subexpressions
4035 Subexpressions are a left parenthesis @samp{(} followed by an integer
4036 expression, followed by a right parenthesis @samp{)}; or a prefix
4037 operator followed by an argument.
4040 @subsection Operators
4042 @cindex operators, in expressions
4043 @cindex arithmetic functions
4044 @cindex functions, in expressions
4045 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
4046 operators are followed by an argument. Infix operators appear
4047 between their arguments. Operators may be preceded and/or followed by
4051 @subsection Prefix Operator
4053 @cindex prefix operators
4054 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
4055 one argument, which must be absolute.
4057 @c the tex/end tex stuff surrounding this small table is meant to make
4058 @c it align, on the printed page, with the similar table in the next
4059 @c section (which is inside an enumerate).
4061 \global\advance\leftskip by \itemindent
4066 @dfn{Negation}. Two's complement negation.
4068 @dfn{Complementation}. Bitwise not.
4072 \global\advance\leftskip by -\itemindent
4076 @subsection Infix Operators
4078 @cindex infix operators
4079 @cindex operators, permitted arguments
4080 @dfn{Infix operators} take two arguments, one on either side. Operators
4081 have precedence, but operations with equal precedence are performed left
4082 to right. Apart from @code{+} or @option{-}, both arguments must be
4083 absolute, and the result is absolute.
4086 @cindex operator precedence
4087 @cindex precedence of operators
4094 @dfn{Multiplication}.
4097 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4103 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4106 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4110 Intermediate precedence
4115 @dfn{Bitwise Inclusive Or}.
4121 @dfn{Bitwise Exclusive Or}.
4124 @dfn{Bitwise Or Not}.
4131 @cindex addition, permitted arguments
4132 @cindex plus, permitted arguments
4133 @cindex arguments for addition
4135 @dfn{Addition}. If either argument is absolute, the result has the section of
4136 the other argument. You may not add together arguments from different
4139 @cindex subtraction, permitted arguments
4140 @cindex minus, permitted arguments
4141 @cindex arguments for subtraction
4143 @dfn{Subtraction}. If the right argument is absolute, the
4144 result has the section of the left argument.
4145 If both arguments are in the same section, the result is absolute.
4146 You may not subtract arguments from different sections.
4147 @c FIXME is there still something useful to say about undefined - undefined ?
4149 @cindex comparison expressions
4150 @cindex expressions, comparison
4155 @dfn{Is Not Equal To}
4159 @dfn{Is Greater Than}
4161 @dfn{Is Greater Than Or Equal To}
4163 @dfn{Is Less Than Or Equal To}
4165 The comparison operators can be used as infix operators. A true results has a
4166 value of -1 whereas a false result has a value of 0. Note, these operators
4167 perform signed comparisons.
4170 @item Lowest Precedence
4179 These two logical operations can be used to combine the results of sub
4180 expressions. Note, unlike the comparison operators a true result returns a
4181 value of 1 but a false results does still return 0. Also note that the logical
4182 or operator has a slightly lower precedence than logical and.
4187 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4188 address; you can only have a defined section in one of the two arguments.
4191 @chapter Assembler Directives
4193 @cindex directives, machine independent
4194 @cindex pseudo-ops, machine independent
4195 @cindex machine independent directives
4196 All assembler directives have names that begin with a period (@samp{.}).
4197 The rest of the name is letters, usually in lower case.
4199 This chapter discusses directives that are available regardless of the
4200 target machine configuration for the @sc{gnu} assembler.
4202 Some machine configurations provide additional directives.
4203 @xref{Machine Dependencies}.
4206 @ifset machine-directives
4207 @xref{Machine Dependencies}, for additional directives.
4212 * Abort:: @code{.abort}
4214 * ABORT (COFF):: @code{.ABORT}
4217 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4218 * Altmacro:: @code{.altmacro}
4219 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4220 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4221 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4222 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, etc
4223 * Byte:: @code{.byte @var{expressions}}
4224 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4225 * Comm:: @code{.comm @var{symbol} , @var{length} }
4226 * Data:: @code{.data @var{subsection}}
4228 * Def:: @code{.def @var{name}}
4231 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4237 * Double:: @code{.double @var{flonums}}
4238 * Eject:: @code{.eject}
4239 * Else:: @code{.else}
4240 * Elseif:: @code{.elseif}
4243 * Endef:: @code{.endef}
4246 * Endfunc:: @code{.endfunc}
4247 * Endif:: @code{.endif}
4248 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4249 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4250 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4252 * Error:: @code{.error @var{string}}
4253 * Exitm:: @code{.exitm}
4254 * Extern:: @code{.extern}
4255 * Fail:: @code{.fail}
4256 * File:: @code{.file}
4257 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4258 * Float:: @code{.float @var{flonums}}
4259 * Func:: @code{.func}
4260 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4262 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4263 * Hidden:: @code{.hidden @var{names}}
4266 * hword:: @code{.hword @var{expressions}}
4267 * Ident:: @code{.ident}
4268 * If:: @code{.if @var{absolute expression}}
4269 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4270 * Include:: @code{.include "@var{file}"}
4271 * Int:: @code{.int @var{expressions}}
4273 * Internal:: @code{.internal @var{names}}
4276 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4277 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4278 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4279 * Lflags:: @code{.lflags}
4280 @ifclear no-line-dir
4281 * Line:: @code{.line @var{line-number}}
4284 * Linkonce:: @code{.linkonce [@var{type}]}
4285 * List:: @code{.list}
4286 * Ln:: @code{.ln @var{line-number}}
4287 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4288 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4290 * Local:: @code{.local @var{names}}
4293 * Long:: @code{.long @var{expressions}}
4295 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4298 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4299 * MRI:: @code{.mri @var{val}}
4300 * Noaltmacro:: @code{.noaltmacro}
4301 * Nolist:: @code{.nolist}
4302 * Octa:: @code{.octa @var{bignums}}
4303 * Offset:: @code{.offset @var{loc}}
4304 * Org:: @code{.org @var{new-lc}, @var{fill}}
4305 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4307 * PopSection:: @code{.popsection}
4308 * Previous:: @code{.previous}
4311 * Print:: @code{.print @var{string}}
4313 * Protected:: @code{.protected @var{names}}
4316 * Psize:: @code{.psize @var{lines}, @var{columns}}
4317 * Purgem:: @code{.purgem @var{name}}
4319 * PushSection:: @code{.pushsection @var{name}}
4322 * Quad:: @code{.quad @var{bignums}}
4323 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4324 * Rept:: @code{.rept @var{count}}
4325 * Sbttl:: @code{.sbttl "@var{subheading}"}
4327 * Scl:: @code{.scl @var{class}}
4330 * Section:: @code{.section @var{name}[, @var{flags}]}
4333 * Set:: @code{.set @var{symbol}, @var{expression}}
4334 * Short:: @code{.short @var{expressions}}
4335 * Single:: @code{.single @var{flonums}}
4337 * Size:: @code{.size [@var{name} , @var{expression}]}
4339 @ifclear no-space-dir
4340 * Skip:: @code{.skip @var{size} , @var{fill}}
4343 * Sleb128:: @code{.sleb128 @var{expressions}}
4344 @ifclear no-space-dir
4345 * Space:: @code{.space @var{size} , @var{fill}}
4348 * Stab:: @code{.stabd, .stabn, .stabs}
4351 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4352 * Struct:: @code{.struct @var{expression}}
4354 * SubSection:: @code{.subsection}
4355 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4359 * Tag:: @code{.tag @var{structname}}
4362 * Text:: @code{.text @var{subsection}}
4363 * Title:: @code{.title "@var{heading}"}
4365 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4368 * Uleb128:: @code{.uleb128 @var{expressions}}
4370 * Val:: @code{.val @var{addr}}
4374 * Version:: @code{.version "@var{string}"}
4375 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4376 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4379 * Warning:: @code{.warning @var{string}}
4380 * Weak:: @code{.weak @var{names}}
4381 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4382 * Word:: @code{.word @var{expressions}}
4383 * Deprecated:: Deprecated Directives
4387 @section @code{.abort}
4389 @cindex @code{abort} directive
4390 @cindex stopping the assembly
4391 This directive stops the assembly immediately. It is for
4392 compatibility with other assemblers. The original idea was that the
4393 assembly language source would be piped into the assembler. If the sender
4394 of the source quit, it could use this directive tells @command{@value{AS}} to
4395 quit also. One day @code{.abort} will not be supported.
4399 @section @code{.ABORT} (COFF)
4401 @cindex @code{ABORT} directive
4402 When producing COFF output, @command{@value{AS}} accepts this directive as a
4403 synonym for @samp{.abort}.
4406 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4412 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4414 @cindex padding the location counter
4415 @cindex @code{align} directive
4416 Pad the location counter (in the current subsection) to a particular storage
4417 boundary. The first expression (which must be absolute) is the alignment
4418 required, as described below.
4420 The second expression (also absolute) gives the fill value to be stored in the
4421 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4422 padding bytes are normally zero. However, on some systems, if the section is
4423 marked as containing code and the fill value is omitted, the space is filled
4424 with no-op instructions.
4426 The third expression is also absolute, and is also optional. If it is present,
4427 it is the maximum number of bytes that should be skipped by this alignment
4428 directive. If doing the alignment would require skipping more bytes than the
4429 specified maximum, then the alignment is not done at all. You can omit the
4430 fill value (the second argument) entirely by simply using two commas after the
4431 required alignment; this can be useful if you want the alignment to be filled
4432 with no-op instructions when appropriate.
4434 The way the required alignment is specified varies from system to system.
4435 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or1k,
4436 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4437 alignment request in bytes. For example @samp{.align 8} advances
4438 the location counter until it is a multiple of 8. If the location counter
4439 is already a multiple of 8, no change is needed. For the tic54x, the
4440 first expression is the alignment request in words.
4442 For other systems, including ppc, i386 using a.out format, arm and
4443 strongarm, it is the
4444 number of low-order zero bits the location counter must have after
4445 advancement. For example @samp{.align 3} advances the location
4446 counter until it a multiple of 8. If the location counter is already a
4447 multiple of 8, no change is needed.
4449 This inconsistency is due to the different behaviors of the various
4450 native assemblers for these systems which GAS must emulate.
4451 GAS also provides @code{.balign} and @code{.p2align} directives,
4452 described later, which have a consistent behavior across all
4453 architectures (but are specific to GAS).
4456 @section @code{.altmacro}
4457 Enable alternate macro mode, enabling:
4460 @item LOCAL @var{name} [ , @dots{} ]
4461 One additional directive, @code{LOCAL}, is available. It is used to
4462 generate a string replacement for each of the @var{name} arguments, and
4463 replace any instances of @var{name} in each macro expansion. The
4464 replacement string is unique in the assembly, and different for each
4465 separate macro expansion. @code{LOCAL} allows you to write macros that
4466 define symbols, without fear of conflict between separate macro expansions.
4468 @item String delimiters
4469 You can write strings delimited in these other ways besides
4470 @code{"@var{string}"}:
4473 @item '@var{string}'
4474 You can delimit strings with single-quote characters.
4476 @item <@var{string}>
4477 You can delimit strings with matching angle brackets.
4480 @item single-character string escape
4481 To include any single character literally in a string (even if the
4482 character would otherwise have some special meaning), you can prefix the
4483 character with @samp{!} (an exclamation mark). For example, you can
4484 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4486 @item Expression results as strings
4487 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4488 and use the result as a string.
4492 @section @code{.ascii "@var{string}"}@dots{}
4494 @cindex @code{ascii} directive
4495 @cindex string literals
4496 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4497 separated by commas. It assembles each string (with no automatic
4498 trailing zero byte) into consecutive addresses.
4501 @section @code{.asciz "@var{string}"}@dots{}
4503 @cindex @code{asciz} directive
4504 @cindex zero-terminated strings
4505 @cindex null-terminated strings
4506 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4507 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4510 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4512 @cindex padding the location counter given number of bytes
4513 @cindex @code{balign} directive
4514 Pad the location counter (in the current subsection) to a particular
4515 storage boundary. The first expression (which must be absolute) is the
4516 alignment request in bytes. For example @samp{.balign 8} advances
4517 the location counter until it is a multiple of 8. If the location counter
4518 is already a multiple of 8, no change is needed.
4520 The second expression (also absolute) gives the fill value to be stored in the
4521 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4522 padding bytes are normally zero. However, on some systems, if the section is
4523 marked as containing code and the fill value is omitted, the space is filled
4524 with no-op instructions.
4526 The third expression is also absolute, and is also optional. If it is present,
4527 it is the maximum number of bytes that should be skipped by this alignment
4528 directive. If doing the alignment would require skipping more bytes than the
4529 specified maximum, then the alignment is not done at all. You can omit the
4530 fill value (the second argument) entirely by simply using two commas after the
4531 required alignment; this can be useful if you want the alignment to be filled
4532 with no-op instructions when appropriate.
4534 @cindex @code{balignw} directive
4535 @cindex @code{balignl} directive
4536 The @code{.balignw} and @code{.balignl} directives are variants of the
4537 @code{.balign} directive. The @code{.balignw} directive treats the fill
4538 pattern as a two byte word value. The @code{.balignl} directives treats the
4539 fill pattern as a four byte longword value. For example, @code{.balignw
4540 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4541 filled in with the value 0x368d (the exact placement of the bytes depends upon
4542 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4545 @node Bundle directives
4546 @section Bundle directives
4547 @subsection @code{.bundle_align_mode @var{abs-expr}}
4548 @cindex @code{bundle_align_mode} directive
4550 @cindex instruction bundle
4551 @cindex aligned instruction bundle
4552 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4553 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4554 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4555 disabled (which is the default state). If the argument it not zero, it
4556 gives the size of an instruction bundle as a power of two (as for the
4557 @code{.p2align} directive, @pxref{P2align}).
4559 For some targets, it's an ABI requirement that no instruction may span a
4560 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4561 instructions that starts on an aligned boundary. For example, if
4562 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4563 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4564 effect, no single instruction may span a boundary between bundles. If an
4565 instruction would start too close to the end of a bundle for the length of
4566 that particular instruction to fit within the bundle, then the space at the
4567 end of that bundle is filled with no-op instructions so the instruction
4568 starts in the next bundle. As a corollary, it's an error if any single
4569 instruction's encoding is longer than the bundle size.
4571 @subsection @code{.bundle_lock} and @code{.bundle_unlock}
4572 @cindex @code{bundle_lock} directive
4573 @cindex @code{bundle_unlock} directive
4574 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4575 allow explicit control over instruction bundle padding. These directives
4576 are only valid when @code{.bundle_align_mode} has been used to enable
4577 aligned instruction bundle mode. It's an error if they appear when
4578 @code{.bundle_align_mode} has not been used at all, or when the last
4579 directive was @w{@code{.bundle_align_mode 0}}.
4581 @cindex bundle-locked
4582 For some targets, it's an ABI requirement that certain instructions may
4583 appear only as part of specified permissible sequences of multiple
4584 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4585 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4586 instruction sequence. For purposes of aligned instruction bundle mode, a
4587 sequence starting with @code{.bundle_lock} and ending with
4588 @code{.bundle_unlock} is treated as a single instruction. That is, the
4589 entire sequence must fit into a single bundle and may not span a bundle
4590 boundary. If necessary, no-op instructions will be inserted before the
4591 first instruction of the sequence so that the whole sequence starts on an
4592 aligned bundle boundary. It's an error if the sequence is longer than the
4595 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4596 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4597 nested. That is, a second @code{.bundle_lock} directive before the next
4598 @code{.bundle_unlock} directive has no effect except that it must be
4599 matched by another closing @code{.bundle_unlock} so that there is the
4600 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4603 @section @code{.byte @var{expressions}}
4605 @cindex @code{byte} directive
4606 @cindex integers, one byte
4607 @code{.byte} expects zero or more expressions, separated by commas.
4608 Each expression is assembled into the next byte.
4610 @node CFI directives
4611 @section CFI directives
4612 @subsection @code{.cfi_sections @var{section_list}}
4613 @cindex @code{cfi_sections} directive
4614 @code{.cfi_sections} may be used to specify whether CFI directives
4615 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4616 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4617 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4618 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4619 directive is not used is @code{.cfi_sections .eh_frame}.
4621 @subsection @code{.cfi_startproc [simple]}
4622 @cindex @code{cfi_startproc} directive
4623 @code{.cfi_startproc} is used at the beginning of each function that
4624 should have an entry in @code{.eh_frame}. It initializes some internal
4625 data structures. Don't forget to close the function by
4626 @code{.cfi_endproc}.
4628 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4629 it also emits some architecture dependent initial CFI instructions.
4631 @subsection @code{.cfi_endproc}
4632 @cindex @code{cfi_endproc} directive
4633 @code{.cfi_endproc} is used at the end of a function where it closes its
4634 unwind entry previously opened by
4635 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4637 @subsection @code{.cfi_personality @var{encoding} [, @var{exp}]}
4638 @code{.cfi_personality} defines personality routine and its encoding.
4639 @var{encoding} must be a constant determining how the personality
4640 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4641 argument is not present, otherwise second argument should be
4642 a constant or a symbol name. When using indirect encodings,
4643 the symbol provided should be the location where personality
4644 can be loaded from, not the personality routine itself.
4645 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4646 no personality routine.
4648 @subsection @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4649 @code{.cfi_lsda} defines LSDA and its encoding.
4650 @var{encoding} must be a constant determining how the LSDA
4651 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4652 argument is not present, otherwise second argument should be a constant
4653 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4656 @subsection @code{.cfi_def_cfa @var{register}, @var{offset}}
4657 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4658 address from @var{register} and add @var{offset} to it}.
4660 @subsection @code{.cfi_def_cfa_register @var{register}}
4661 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4662 now on @var{register} will be used instead of the old one. Offset
4665 @subsection @code{.cfi_def_cfa_offset @var{offset}}
4666 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4667 remains the same, but @var{offset} is new. Note that it is the
4668 absolute offset that will be added to a defined register to compute
4671 @subsection @code{.cfi_adjust_cfa_offset @var{offset}}
4672 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4673 value that is added/substracted from the previous offset.
4675 @subsection @code{.cfi_offset @var{register}, @var{offset}}
4676 Previous value of @var{register} is saved at offset @var{offset} from
4679 @subsection @code{.cfi_rel_offset @var{register}, @var{offset}}
4680 Previous value of @var{register} is saved at offset @var{offset} from
4681 the current CFA register. This is transformed to @code{.cfi_offset}
4682 using the known displacement of the CFA register from the CFA.
4683 This is often easier to use, because the number will match the
4684 code it's annotating.
4686 @subsection @code{.cfi_register @var{register1}, @var{register2}}
4687 Previous value of @var{register1} is saved in register @var{register2}.
4689 @subsection @code{.cfi_restore @var{register}}
4690 @code{.cfi_restore} says that the rule for @var{register} is now the
4691 same as it was at the beginning of the function, after all initial
4692 instruction added by @code{.cfi_startproc} were executed.
4694 @subsection @code{.cfi_undefined @var{register}}
4695 From now on the previous value of @var{register} can't be restored anymore.
4697 @subsection @code{.cfi_same_value @var{register}}
4698 Current value of @var{register} is the same like in the previous frame,
4699 i.e. no restoration needed.
4701 @subsection @code{.cfi_remember_state},
4702 First save all current rules for all registers by @code{.cfi_remember_state},
4703 then totally screw them up by subsequent @code{.cfi_*} directives and when
4704 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4705 the previous saved state.
4707 @subsection @code{.cfi_return_column @var{register}}
4708 Change return column @var{register}, i.e. the return address is either
4709 directly in @var{register} or can be accessed by rules for @var{register}.
4711 @subsection @code{.cfi_signal_frame}
4712 Mark current function as signal trampoline.
4714 @subsection @code{.cfi_window_save}
4715 SPARC register window has been saved.
4717 @subsection @code{.cfi_escape} @var{expression}[, @dots{}]
4718 Allows the user to add arbitrary bytes to the unwind info. One
4719 might use this to add OS-specific CFI opcodes, or generic CFI
4720 opcodes that GAS does not yet support.
4722 @subsection @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4723 The current value of @var{register} is @var{label}. The value of @var{label}
4724 will be encoded in the output file according to @var{encoding}; see the
4725 description of @code{.cfi_personality} for details on this encoding.
4727 The usefulness of equating a register to a fixed label is probably
4728 limited to the return address register. Here, it can be useful to
4729 mark a code segment that has only one return address which is reached
4730 by a direct branch and no copy of the return address exists in memory
4731 or another register.
4734 @section @code{.comm @var{symbol} , @var{length} }
4736 @cindex @code{comm} directive
4737 @cindex symbol, common
4738 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4739 common symbol in one object file may be merged with a defined or common symbol
4740 of the same name in another object file. If @code{@value{LD}} does not see a
4741 definition for the symbol--just one or more common symbols--then it will
4742 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4743 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4744 the same name, and they do not all have the same size, it will allocate space
4745 using the largest size.
4748 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4749 an optional third argument. This is the desired alignment of the symbol,
4750 specified for ELF as a byte boundary (for example, an alignment of 16 means
4751 that the least significant 4 bits of the address should be zero), and for PE
4752 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4753 boundary). The alignment must be an absolute expression, and it must be a
4754 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4755 common symbol, it will use the alignment when placing the symbol. If no
4756 alignment is specified, @command{@value{AS}} will set the alignment to the
4757 largest power of two less than or equal to the size of the symbol, up to a
4758 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4759 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4760 @samp{--section-alignment} option; image file sections in PE are aligned to
4761 multiples of 4096, which is far too large an alignment for ordinary variables.
4762 It is rather the default alignment for (non-debug) sections within object
4763 (@samp{*.o}) files, which are less strictly aligned.}.
4767 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4768 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4772 @section @code{.data @var{subsection}}
4774 @cindex @code{data} directive
4775 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4776 end of the data subsection numbered @var{subsection} (which is an
4777 absolute expression). If @var{subsection} is omitted, it defaults
4782 @section @code{.def @var{name}}
4784 @cindex @code{def} directive
4785 @cindex COFF symbols, debugging
4786 @cindex debugging COFF symbols
4787 Begin defining debugging information for a symbol @var{name}; the
4788 definition extends until the @code{.endef} directive is encountered.
4791 This directive is only observed when @command{@value{AS}} is configured for COFF
4792 format output; when producing @code{b.out}, @samp{.def} is recognized,
4799 @section @code{.desc @var{symbol}, @var{abs-expression}}
4801 @cindex @code{desc} directive
4802 @cindex COFF symbol descriptor
4803 @cindex symbol descriptor, COFF
4804 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4805 to the low 16 bits of an absolute expression.
4808 The @samp{.desc} directive is not available when @command{@value{AS}} is
4809 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4810 object format. For the sake of compatibility, @command{@value{AS}} accepts
4811 it, but produces no output, when configured for COFF.
4817 @section @code{.dim}
4819 @cindex @code{dim} directive
4820 @cindex COFF auxiliary symbol information
4821 @cindex auxiliary symbol information, COFF
4822 This directive is generated by compilers to include auxiliary debugging
4823 information in the symbol table. It is only permitted inside
4824 @code{.def}/@code{.endef} pairs.
4827 @samp{.dim} is only meaningful when generating COFF format output; when
4828 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4834 @section @code{.double @var{flonums}}
4836 @cindex @code{double} directive
4837 @cindex floating point numbers (double)
4838 @code{.double} expects zero or more flonums, separated by commas. It
4839 assembles floating point numbers.
4841 The exact kind of floating point numbers emitted depends on how
4842 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4846 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4847 in @sc{ieee} format.
4852 @section @code{.eject}
4854 @cindex @code{eject} directive
4855 @cindex new page, in listings
4856 @cindex page, in listings
4857 @cindex listing control: new page
4858 Force a page break at this point, when generating assembly listings.
4861 @section @code{.else}
4863 @cindex @code{else} directive
4864 @code{.else} is part of the @command{@value{AS}} support for conditional
4865 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4866 of code to be assembled if the condition for the preceding @code{.if}
4870 @section @code{.elseif}
4872 @cindex @code{elseif} directive
4873 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4874 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4875 @code{.if} block that would otherwise fill the entire @code{.else} section.
4878 @section @code{.end}
4880 @cindex @code{end} directive
4881 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4882 process anything in the file past the @code{.end} directive.
4886 @section @code{.endef}
4888 @cindex @code{endef} directive
4889 This directive flags the end of a symbol definition begun with
4893 @samp{.endef} is only meaningful when generating COFF format output; if
4894 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4895 directive but ignores it.
4900 @section @code{.endfunc}
4901 @cindex @code{endfunc} directive
4902 @code{.endfunc} marks the end of a function specified with @code{.func}.
4905 @section @code{.endif}
4907 @cindex @code{endif} directive
4908 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4909 it marks the end of a block of code that is only assembled
4910 conditionally. @xref{If,,@code{.if}}.
4913 @section @code{.equ @var{symbol}, @var{expression}}
4915 @cindex @code{equ} directive
4916 @cindex assigning values to symbols
4917 @cindex symbols, assigning values to
4918 This directive sets the value of @var{symbol} to @var{expression}.
4919 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4922 The syntax for @code{equ} on the HPPA is
4923 @samp{@var{symbol} .equ @var{expression}}.
4927 The syntax for @code{equ} on the Z80 is
4928 @samp{@var{symbol} equ @var{expression}}.
4929 On the Z80 it is an eror if @var{symbol} is already defined,
4930 but the symbol is not protected from later redefinition.
4931 Compare @ref{Equiv}.
4935 @section @code{.equiv @var{symbol}, @var{expression}}
4936 @cindex @code{equiv} directive
4937 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4938 the assembler will signal an error if @var{symbol} is already defined. Note a
4939 symbol which has been referenced but not actually defined is considered to be
4942 Except for the contents of the error message, this is roughly equivalent to
4949 plus it protects the symbol from later redefinition.
4952 @section @code{.eqv @var{symbol}, @var{expression}}
4953 @cindex @code{eqv} directive
4954 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
4955 evaluate the expression or any part of it immediately. Instead each time
4956 the resulting symbol is used in an expression, a snapshot of its current
4960 @section @code{.err}
4961 @cindex @code{err} directive
4962 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4963 message and, unless the @option{-Z} option was used, it will not generate an
4964 object file. This can be used to signal an error in conditionally compiled code.
4967 @section @code{.error "@var{string}"}
4968 @cindex error directive
4970 Similarly to @code{.err}, this directive emits an error, but you can specify a
4971 string that will be emitted as the error message. If you don't specify the
4972 message, it defaults to @code{".error directive invoked in source file"}.
4973 @xref{Errors, ,Error and Warning Messages}.
4976 .error "This code has not been assembled and tested."
4980 @section @code{.exitm}
4981 Exit early from the current macro definition. @xref{Macro}.
4984 @section @code{.extern}
4986 @cindex @code{extern} directive
4987 @code{.extern} is accepted in the source program---for compatibility
4988 with other assemblers---but it is ignored. @command{@value{AS}} treats
4989 all undefined symbols as external.
4992 @section @code{.fail @var{expression}}
4994 @cindex @code{fail} directive
4995 Generates an error or a warning. If the value of the @var{expression} is 500
4996 or more, @command{@value{AS}} will print a warning message. If the value is less
4997 than 500, @command{@value{AS}} will print an error message. The message will
4998 include the value of @var{expression}. This can occasionally be useful inside
4999 complex nested macros or conditional assembly.
5002 @section @code{.file}
5003 @cindex @code{file} directive
5005 @ifclear no-file-dir
5006 There are two different versions of the @code{.file} directive. Targets
5007 that support DWARF2 line number information use the DWARF2 version of
5008 @code{.file}. Other targets use the default version.
5010 @subheading Default Version
5012 @cindex logical file name
5013 @cindex file name, logical
5014 This version of the @code{.file} directive tells @command{@value{AS}} that we
5015 are about to start a new logical file. The syntax is:
5021 @var{string} is the new file name. In general, the filename is
5022 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
5023 to specify an empty file name, you must give the quotes--@code{""}. This
5024 statement may go away in future: it is only recognized to be compatible with
5025 old @command{@value{AS}} programs.
5027 @subheading DWARF2 Version
5030 When emitting DWARF2 line number information, @code{.file} assigns filenames
5031 to the @code{.debug_line} file name table. The syntax is:
5034 .file @var{fileno} @var{filename}
5037 The @var{fileno} operand should be a unique positive integer to use as the
5038 index of the entry in the table. The @var{filename} operand is a C string
5041 The detail of filename indices is exposed to the user because the filename
5042 table is shared with the @code{.debug_info} section of the DWARF2 debugging
5043 information, and thus the user must know the exact indices that table
5047 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
5049 @cindex @code{fill} directive
5050 @cindex writing patterns in memory
5051 @cindex patterns, writing in memory
5052 @var{repeat}, @var{size} and @var{value} are absolute expressions.
5053 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
5054 may be zero or more. @var{Size} may be zero or more, but if it is
5055 more than 8, then it is deemed to have the value 8, compatible with
5056 other people's assemblers. The contents of each @var{repeat} bytes
5057 is taken from an 8-byte number. The highest order 4 bytes are
5058 zero. The lowest order 4 bytes are @var{value} rendered in the
5059 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
5060 Each @var{size} bytes in a repetition is taken from the lowest order
5061 @var{size} bytes of this number. Again, this bizarre behavior is
5062 compatible with other people's assemblers.
5064 @var{size} and @var{value} are optional.
5065 If the second comma and @var{value} are absent, @var{value} is
5066 assumed zero. If the first comma and following tokens are absent,
5067 @var{size} is assumed to be 1.
5070 @section @code{.float @var{flonums}}
5072 @cindex floating point numbers (single)
5073 @cindex @code{float} directive
5074 This directive assembles zero or more flonums, separated by commas. It
5075 has the same effect as @code{.single}.
5077 The exact kind of floating point numbers emitted depends on how
5078 @command{@value{AS}} is configured.
5079 @xref{Machine Dependencies}.
5083 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5084 in @sc{ieee} format.
5089 @section @code{.func @var{name}[,@var{label}]}
5090 @cindex @code{func} directive
5091 @code{.func} emits debugging information to denote function @var{name}, and
5092 is ignored unless the file is assembled with debugging enabled.
5093 Only @samp{--gstabs[+]} is currently supported.
5094 @var{label} is the entry point of the function and if omitted @var{name}
5095 prepended with the @samp{leading char} is used.
5096 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5097 All functions are currently defined to have @code{void} return type.
5098 The function must be terminated with @code{.endfunc}.
5101 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5103 @cindex @code{global} directive
5104 @cindex symbol, making visible to linker
5105 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5106 @var{symbol} in your partial program, its value is made available to
5107 other partial programs that are linked with it. Otherwise,
5108 @var{symbol} takes its attributes from a symbol of the same name
5109 from another file linked into the same program.
5111 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5112 compatibility with other assemblers.
5115 On the HPPA, @code{.global} is not always enough to make it accessible to other
5116 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5117 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5122 @section @code{.gnu_attribute @var{tag},@var{value}}
5123 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5126 @section @code{.hidden @var{names}}
5128 @cindex @code{hidden} directive
5130 This is one of the ELF visibility directives. The other two are
5131 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5132 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5134 This directive overrides the named symbols default visibility (which is set by
5135 their binding: local, global or weak). The directive sets the visibility to
5136 @code{hidden} which means that the symbols are not visible to other components.
5137 Such symbols are always considered to be @code{protected} as well.
5141 @section @code{.hword @var{expressions}}
5143 @cindex @code{hword} directive
5144 @cindex integers, 16-bit
5145 @cindex numbers, 16-bit
5146 @cindex sixteen bit integers
5147 This expects zero or more @var{expressions}, and emits
5148 a 16 bit number for each.
5151 This directive is a synonym for @samp{.short}; depending on the target
5152 architecture, it may also be a synonym for @samp{.word}.
5156 This directive is a synonym for @samp{.short}.
5159 This directive is a synonym for both @samp{.short} and @samp{.word}.
5164 @section @code{.ident}
5166 @cindex @code{ident} directive
5168 This directive is used by some assemblers to place tags in object files. The
5169 behavior of this directive varies depending on the target. When using the
5170 a.out object file format, @command{@value{AS}} simply accepts the directive for
5171 source-file compatibility with existing assemblers, but does not emit anything
5172 for it. When using COFF, comments are emitted to the @code{.comment} or
5173 @code{.rdata} section, depending on the target. When using ELF, comments are
5174 emitted to the @code{.comment} section.
5177 @section @code{.if @var{absolute expression}}
5179 @cindex conditional assembly
5180 @cindex @code{if} directive
5181 @code{.if} marks the beginning of a section of code which is only
5182 considered part of the source program being assembled if the argument
5183 (which must be an @var{absolute expression}) is non-zero. The end of
5184 the conditional section of code must be marked by @code{.endif}
5185 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5186 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5187 If you have several conditions to check, @code{.elseif} may be used to avoid
5188 nesting blocks if/else within each subsequent @code{.else} block.
5190 The following variants of @code{.if} are also supported:
5192 @cindex @code{ifdef} directive
5193 @item .ifdef @var{symbol}
5194 Assembles the following section of code if the specified @var{symbol}
5195 has been defined. Note a symbol which has been referenced but not yet defined
5196 is considered to be undefined.
5198 @cindex @code{ifb} directive
5199 @item .ifb @var{text}
5200 Assembles the following section of code if the operand is blank (empty).
5202 @cindex @code{ifc} directive
5203 @item .ifc @var{string1},@var{string2}
5204 Assembles the following section of code if the two strings are the same. The
5205 strings may be optionally quoted with single quotes. If they are not quoted,
5206 the first string stops at the first comma, and the second string stops at the
5207 end of the line. Strings which contain whitespace should be quoted. The
5208 string comparison is case sensitive.
5210 @cindex @code{ifeq} directive
5211 @item .ifeq @var{absolute expression}
5212 Assembles the following section of code if the argument is zero.
5214 @cindex @code{ifeqs} directive
5215 @item .ifeqs @var{string1},@var{string2}
5216 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5218 @cindex @code{ifge} directive
5219 @item .ifge @var{absolute expression}
5220 Assembles the following section of code if the argument is greater than or
5223 @cindex @code{ifgt} directive
5224 @item .ifgt @var{absolute expression}
5225 Assembles the following section of code if the argument is greater than zero.
5227 @cindex @code{ifle} directive
5228 @item .ifle @var{absolute expression}
5229 Assembles the following section of code if the argument is less than or equal
5232 @cindex @code{iflt} directive
5233 @item .iflt @var{absolute expression}
5234 Assembles the following section of code if the argument is less than zero.
5236 @cindex @code{ifnb} directive
5237 @item .ifnb @var{text}
5238 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5239 following section of code if the operand is non-blank (non-empty).
5241 @cindex @code{ifnc} directive
5242 @item .ifnc @var{string1},@var{string2}.
5243 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5244 following section of code if the two strings are not the same.
5246 @cindex @code{ifndef} directive
5247 @cindex @code{ifnotdef} directive
5248 @item .ifndef @var{symbol}
5249 @itemx .ifnotdef @var{symbol}
5250 Assembles the following section of code if the specified @var{symbol}
5251 has not been defined. Both spelling variants are equivalent. Note a symbol
5252 which has been referenced but not yet defined is considered to be undefined.
5254 @cindex @code{ifne} directive
5255 @item .ifne @var{absolute expression}
5256 Assembles the following section of code if the argument is not equal to zero
5257 (in other words, this is equivalent to @code{.if}).
5259 @cindex @code{ifnes} directive
5260 @item .ifnes @var{string1},@var{string2}
5261 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5262 following section of code if the two strings are not the same.
5266 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5268 @cindex @code{incbin} directive
5269 @cindex binary files, including
5270 The @code{incbin} directive includes @var{file} verbatim at the current
5271 location. You can control the search paths used with the @samp{-I} command-line
5272 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5275 The @var{skip} argument skips a number of bytes from the start of the
5276 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5277 read. Note that the data is not aligned in any way, so it is the user's
5278 responsibility to make sure that proper alignment is provided both before and
5279 after the @code{incbin} directive.
5282 @section @code{.include "@var{file}"}
5284 @cindex @code{include} directive
5285 @cindex supporting files, including
5286 @cindex files, including
5287 This directive provides a way to include supporting files at specified
5288 points in your source program. The code from @var{file} is assembled as
5289 if it followed the point of the @code{.include}; when the end of the
5290 included file is reached, assembly of the original file continues. You
5291 can control the search paths used with the @samp{-I} command-line option
5292 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5296 @section @code{.int @var{expressions}}
5298 @cindex @code{int} directive
5299 @cindex integers, 32-bit
5300 Expect zero or more @var{expressions}, of any section, separated by commas.
5301 For each expression, emit a number that, at run time, is the value of that
5302 expression. The byte order and bit size of the number depends on what kind
5303 of target the assembly is for.
5307 On most forms of the H8/300, @code{.int} emits 16-bit
5308 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5315 @section @code{.internal @var{names}}
5317 @cindex @code{internal} directive
5319 This is one of the ELF visibility directives. The other two are
5320 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5321 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5323 This directive overrides the named symbols default visibility (which is set by
5324 their binding: local, global or weak). The directive sets the visibility to
5325 @code{internal} which means that the symbols are considered to be @code{hidden}
5326 (i.e., not visible to other components), and that some extra, processor specific
5327 processing must also be performed upon the symbols as well.
5331 @section @code{.irp @var{symbol},@var{values}}@dots{}
5333 @cindex @code{irp} directive
5334 Evaluate a sequence of statements assigning different values to @var{symbol}.
5335 The sequence of statements starts at the @code{.irp} directive, and is
5336 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5337 set to @var{value}, and the sequence of statements is assembled. If no
5338 @var{value} is listed, the sequence of statements is assembled once, with
5339 @var{symbol} set to the null string. To refer to @var{symbol} within the
5340 sequence of statements, use @var{\symbol}.
5342 For example, assembling
5350 is equivalent to assembling
5358 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5361 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5363 @cindex @code{irpc} directive
5364 Evaluate a sequence of statements assigning different values to @var{symbol}.
5365 The sequence of statements starts at the @code{.irpc} directive, and is
5366 terminated by an @code{.endr} directive. For each character in @var{value},
5367 @var{symbol} is set to the character, and the sequence of statements is
5368 assembled. If no @var{value} is listed, the sequence of statements is
5369 assembled once, with @var{symbol} set to the null string. To refer to
5370 @var{symbol} within the sequence of statements, use @var{\symbol}.
5372 For example, assembling
5380 is equivalent to assembling
5388 For some caveats with the spelling of @var{symbol}, see also the discussion
5392 @section @code{.lcomm @var{symbol} , @var{length}}
5394 @cindex @code{lcomm} directive
5395 @cindex local common symbols
5396 @cindex symbols, local common
5397 Reserve @var{length} (an absolute expression) bytes for a local common
5398 denoted by @var{symbol}. The section and value of @var{symbol} are
5399 those of the new local common. The addresses are allocated in the bss
5400 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5401 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5402 not visible to @code{@value{LD}}.
5405 Some targets permit a third argument to be used with @code{.lcomm}. This
5406 argument specifies the desired alignment of the symbol in the bss section.
5410 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5411 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5415 @section @code{.lflags}
5417 @cindex @code{lflags} directive (ignored)
5418 @command{@value{AS}} accepts this directive, for compatibility with other
5419 assemblers, but ignores it.
5421 @ifclear no-line-dir
5423 @section @code{.line @var{line-number}}
5425 @cindex @code{line} directive
5426 @cindex logical line number
5428 Change the logical line number. @var{line-number} must be an absolute
5429 expression. The next line has that logical line number. Therefore any other
5430 statements on the current line (after a statement separator character) are
5431 reported as on logical line number @var{line-number} @minus{} 1. One day
5432 @command{@value{AS}} will no longer support this directive: it is recognized only
5433 for compatibility with existing assembler programs.
5436 Even though this is a directive associated with the @code{a.out} or
5437 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5438 when producing COFF output, and treats @samp{.line} as though it
5439 were the COFF @samp{.ln} @emph{if} it is found outside a
5440 @code{.def}/@code{.endef} pair.
5442 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5443 used by compilers to generate auxiliary symbol information for
5448 @section @code{.linkonce [@var{type}]}
5450 @cindex @code{linkonce} directive
5451 @cindex common sections
5452 Mark the current section so that the linker only includes a single copy of it.
5453 This may be used to include the same section in several different object files,
5454 but ensure that the linker will only include it once in the final output file.
5455 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5456 Duplicate sections are detected based on the section name, so it should be
5459 This directive is only supported by a few object file formats; as of this
5460 writing, the only object file format which supports it is the Portable
5461 Executable format used on Windows NT.
5463 The @var{type} argument is optional. If specified, it must be one of the
5464 following strings. For example:
5468 Not all types may be supported on all object file formats.
5472 Silently discard duplicate sections. This is the default.
5475 Warn if there are duplicate sections, but still keep only one copy.
5478 Warn if any of the duplicates have different sizes.
5481 Warn if any of the duplicates do not have exactly the same contents.
5485 @section @code{.list}
5487 @cindex @code{list} directive
5488 @cindex listing control, turning on
5489 Control (in conjunction with the @code{.nolist} directive) whether or
5490 not assembly listings are generated. These two directives maintain an
5491 internal counter (which is zero initially). @code{.list} increments the
5492 counter, and @code{.nolist} decrements it. Assembly listings are
5493 generated whenever the counter is greater than zero.
5495 By default, listings are disabled. When you enable them (with the
5496 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5497 the initial value of the listing counter is one.
5500 @section @code{.ln @var{line-number}}
5502 @cindex @code{ln} directive
5503 @ifclear no-line-dir
5504 @samp{.ln} is a synonym for @samp{.line}.
5507 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5508 must be an absolute expression. The next line has that logical
5509 line number, so any other statements on the current line (after a
5510 statement separator character @code{;}) are reported as on logical
5511 line number @var{line-number} @minus{} 1.
5514 This directive is accepted, but ignored, when @command{@value{AS}} is
5515 configured for @code{b.out}; its effect is only associated with COFF
5521 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5522 @cindex @code{loc} directive
5523 When emitting DWARF2 line number information,
5524 the @code{.loc} directive will add a row to the @code{.debug_line} line
5525 number matrix corresponding to the immediately following assembly
5526 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5527 arguments will be applied to the @code{.debug_line} state machine before
5530 The @var{options} are a sequence of the following tokens in any order:
5534 This option will set the @code{basic_block} register in the
5535 @code{.debug_line} state machine to @code{true}.
5538 This option will set the @code{prologue_end} register in the
5539 @code{.debug_line} state machine to @code{true}.
5541 @item epilogue_begin
5542 This option will set the @code{epilogue_begin} register in the
5543 @code{.debug_line} state machine to @code{true}.
5545 @item is_stmt @var{value}
5546 This option will set the @code{is_stmt} register in the
5547 @code{.debug_line} state machine to @code{value}, which must be
5550 @item isa @var{value}
5551 This directive will set the @code{isa} register in the @code{.debug_line}
5552 state machine to @var{value}, which must be an unsigned integer.
5554 @item discriminator @var{value}
5555 This directive will set the @code{discriminator} register in the @code{.debug_line}
5556 state machine to @var{value}, which must be an unsigned integer.
5560 @node Loc_mark_labels
5561 @section @code{.loc_mark_labels @var{enable}}
5562 @cindex @code{loc_mark_labels} directive
5563 When emitting DWARF2 line number information,
5564 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5565 to the @code{.debug_line} line number matrix with the @code{basic_block}
5566 register in the state machine set whenever a code label is seen.
5567 The @var{enable} argument should be either 1 or 0, to enable or disable
5568 this function respectively.
5572 @section @code{.local @var{names}}
5574 @cindex @code{local} directive
5575 This directive, which is available for ELF targets, marks each symbol in
5576 the comma-separated list of @code{names} as a local symbol so that it
5577 will not be externally visible. If the symbols do not already exist,
5578 they will be created.
5580 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5581 accept an alignment argument, which is the case for most ELF targets,
5582 the @code{.local} directive can be used in combination with @code{.comm}
5583 (@pxref{Comm}) to define aligned local common data.
5587 @section @code{.long @var{expressions}}
5589 @cindex @code{long} directive
5590 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5593 @c no one seems to know what this is for or whether this description is
5594 @c what it really ought to do
5596 @section @code{.lsym @var{symbol}, @var{expression}}
5598 @cindex @code{lsym} directive
5599 @cindex symbol, not referenced in assembly
5600 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5601 the hash table, ensuring it cannot be referenced by name during the
5602 rest of the assembly. This sets the attributes of the symbol to be
5603 the same as the expression value:
5605 @var{other} = @var{descriptor} = 0
5606 @var{type} = @r{(section of @var{expression})}
5607 @var{value} = @var{expression}
5610 The new symbol is not flagged as external.
5614 @section @code{.macro}
5617 The commands @code{.macro} and @code{.endm} allow you to define macros that
5618 generate assembly output. For example, this definition specifies a macro
5619 @code{sum} that puts a sequence of numbers into memory:
5622 .macro sum from=0, to=5
5631 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5643 @item .macro @var{macname}
5644 @itemx .macro @var{macname} @var{macargs} @dots{}
5645 @cindex @code{macro} directive
5646 Begin the definition of a macro called @var{macname}. If your macro
5647 definition requires arguments, specify their names after the macro name,
5648 separated by commas or spaces. You can qualify the macro argument to
5649 indicate whether all invocations must specify a non-blank value (through
5650 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5651 (through @samp{:@code{vararg}}). You can supply a default value for any
5652 macro argument by following the name with @samp{=@var{deflt}}. You
5653 cannot define two macros with the same @var{macname} unless it has been
5654 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5655 definitions. For example, these are all valid @code{.macro} statements:
5659 Begin the definition of a macro called @code{comm}, which takes no
5662 @item .macro plus1 p, p1
5663 @itemx .macro plus1 p p1
5664 Either statement begins the definition of a macro called @code{plus1},
5665 which takes two arguments; within the macro definition, write
5666 @samp{\p} or @samp{\p1} to evaluate the arguments.
5668 @item .macro reserve_str p1=0 p2
5669 Begin the definition of a macro called @code{reserve_str}, with two
5670 arguments. The first argument has a default value, but not the second.
5671 After the definition is complete, you can call the macro either as
5672 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5673 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5674 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5675 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5677 @item .macro m p1:req, p2=0, p3:vararg
5678 Begin the definition of a macro called @code{m}, with at least three
5679 arguments. The first argument must always have a value specified, but
5680 not the second, which instead has a default value. The third formal
5681 will get assigned all remaining arguments specified at invocation time.
5683 When you call a macro, you can specify the argument values either by
5684 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5685 @samp{sum to=17, from=9}.
5689 Note that since each of the @var{macargs} can be an identifier exactly
5690 as any other one permitted by the target architecture, there may be
5691 occasional problems if the target hand-crafts special meanings to certain
5692 characters when they occur in a special position. For example, if the colon
5693 (@code{:}) is generally permitted to be part of a symbol name, but the
5694 architecture specific code special-cases it when occurring as the final
5695 character of a symbol (to denote a label), then the macro parameter
5696 replacement code will have no way of knowing that and consider the whole
5697 construct (including the colon) an identifier, and check only this
5698 identifier for being the subject to parameter substitution. So for example
5699 this macro definition:
5707 might not work as expected. Invoking @samp{label foo} might not create a label
5708 called @samp{foo} but instead just insert the text @samp{\l:} into the
5709 assembler source, probably generating an error about an unrecognised
5712 Similarly problems might occur with the period character (@samp{.})
5713 which is often allowed inside opcode names (and hence identifier names). So
5714 for example constructing a macro to build an opcode from a base name and a
5715 length specifier like this:
5718 .macro opcode base length
5723 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5724 instruction but instead generate some kind of error as the assembler tries to
5725 interpret the text @samp{\base.\length}.
5727 There are several possible ways around this problem:
5730 @item Insert white space
5731 If it is possible to use white space characters then this is the simplest
5740 @item Use @samp{\()}
5741 The string @samp{\()} can be used to separate the end of a macro argument from
5742 the following text. eg:
5745 .macro opcode base length
5750 @item Use the alternate macro syntax mode
5751 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5752 used as a separator. eg:
5762 Note: this problem of correctly identifying string parameters to pseudo ops
5763 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5764 and @code{.irpc} (@pxref{Irpc}) as well.
5767 @cindex @code{endm} directive
5768 Mark the end of a macro definition.
5771 @cindex @code{exitm} directive
5772 Exit early from the current macro definition.
5774 @cindex number of macros executed
5775 @cindex macros, count executed
5777 @command{@value{AS}} maintains a counter of how many macros it has
5778 executed in this pseudo-variable; you can copy that number to your
5779 output with @samp{\@@}, but @emph{only within a macro definition}.
5781 @item LOCAL @var{name} [ , @dots{} ]
5782 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5783 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5784 @xref{Altmacro,,@code{.altmacro}}.
5788 @section @code{.mri @var{val}}
5790 @cindex @code{mri} directive
5791 @cindex MRI mode, temporarily
5792 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5793 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5794 affects code assembled until the next @code{.mri} directive, or until the end
5795 of the file. @xref{M, MRI mode, MRI mode}.
5798 @section @code{.noaltmacro}
5799 Disable alternate macro mode. @xref{Altmacro}.
5802 @section @code{.nolist}
5804 @cindex @code{nolist} directive
5805 @cindex listing control, turning off
5806 Control (in conjunction with the @code{.list} directive) whether or
5807 not assembly listings are generated. These two directives maintain an
5808 internal counter (which is zero initially). @code{.list} increments the
5809 counter, and @code{.nolist} decrements it. Assembly listings are
5810 generated whenever the counter is greater than zero.
5813 @section @code{.octa @var{bignums}}
5815 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5816 @cindex @code{octa} directive
5817 @cindex integer, 16-byte
5818 @cindex sixteen byte integer
5819 This directive expects zero or more bignums, separated by commas. For each
5820 bignum, it emits a 16-byte integer.
5822 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5823 hence @emph{octa}-word for 16 bytes.
5826 @section @code{.offset @var{loc}}
5828 @cindex @code{offset} directive
5829 Set the location counter to @var{loc} in the absolute section. @var{loc} must
5830 be an absolute expression. This directive may be useful for defining
5831 symbols with absolute values. Do not confuse it with the @code{.org}
5835 @section @code{.org @var{new-lc} , @var{fill}}
5837 @cindex @code{org} directive
5838 @cindex location counter, advancing
5839 @cindex advancing location counter
5840 @cindex current address, advancing
5841 Advance the location counter of the current section to
5842 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5843 expression with the same section as the current subsection. That is,
5844 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5845 wrong section, the @code{.org} directive is ignored. To be compatible
5846 with former assemblers, if the section of @var{new-lc} is absolute,
5847 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5848 is the same as the current subsection.
5850 @code{.org} may only increase the location counter, or leave it
5851 unchanged; you cannot use @code{.org} to move the location counter
5854 @c double negative used below "not undefined" because this is a specific
5855 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5856 @c section. doc@cygnus.com 18feb91
5857 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5858 may not be undefined. If you really detest this restriction we eagerly await
5859 a chance to share your improved assembler.
5861 Beware that the origin is relative to the start of the section, not
5862 to the start of the subsection. This is compatible with other
5863 people's assemblers.
5865 When the location counter (of the current subsection) is advanced, the
5866 intervening bytes are filled with @var{fill} which should be an
5867 absolute expression. If the comma and @var{fill} are omitted,
5868 @var{fill} defaults to zero.
5871 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5873 @cindex padding the location counter given a power of two
5874 @cindex @code{p2align} directive
5875 Pad the location counter (in the current subsection) to a particular
5876 storage boundary. The first expression (which must be absolute) is the
5877 number of low-order zero bits the location counter must have after
5878 advancement. For example @samp{.p2align 3} advances the location
5879 counter until it a multiple of 8. If the location counter is already a
5880 multiple of 8, no change is needed.
5882 The second expression (also absolute) gives the fill value to be stored in the
5883 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5884 padding bytes are normally zero. However, on some systems, if the section is
5885 marked as containing code and the fill value is omitted, the space is filled
5886 with no-op instructions.
5888 The third expression is also absolute, and is also optional. If it is present,
5889 it is the maximum number of bytes that should be skipped by this alignment
5890 directive. If doing the alignment would require skipping more bytes than the
5891 specified maximum, then the alignment is not done at all. You can omit the
5892 fill value (the second argument) entirely by simply using two commas after the
5893 required alignment; this can be useful if you want the alignment to be filled
5894 with no-op instructions when appropriate.
5896 @cindex @code{p2alignw} directive
5897 @cindex @code{p2alignl} directive
5898 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5899 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5900 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5901 fill pattern as a four byte longword value. For example, @code{.p2alignw
5902 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5903 filled in with the value 0x368d (the exact placement of the bytes depends upon
5904 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5909 @section @code{.popsection}
5911 @cindex @code{popsection} directive
5912 @cindex Section Stack
5913 This is one of the ELF section stack manipulation directives. The others are
5914 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5915 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5918 This directive replaces the current section (and subsection) with the top
5919 section (and subsection) on the section stack. This section is popped off the
5925 @section @code{.previous}
5927 @cindex @code{previous} directive
5928 @cindex Section Stack
5929 This is one of the ELF section stack manipulation directives. The others are
5930 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5931 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5932 (@pxref{PopSection}).
5934 This directive swaps the current section (and subsection) with most recently
5935 referenced section/subsection pair prior to this one. Multiple
5936 @code{.previous} directives in a row will flip between two sections (and their
5937 subsections). For example:
5949 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
5955 # Now in section A subsection 1
5959 # Now in section B subsection 0
5962 # Now in section B subsection 1
5965 # Now in section B subsection 0
5969 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
5970 section B and 0x9abc into subsection 1 of section B.
5972 In terms of the section stack, this directive swaps the current section with
5973 the top section on the section stack.
5977 @section @code{.print @var{string}}
5979 @cindex @code{print} directive
5980 @command{@value{AS}} will print @var{string} on the standard output during
5981 assembly. You must put @var{string} in double quotes.
5985 @section @code{.protected @var{names}}
5987 @cindex @code{protected} directive
5989 This is one of the ELF visibility directives. The other two are
5990 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5992 This directive overrides the named symbols default visibility (which is set by
5993 their binding: local, global or weak). The directive sets the visibility to
5994 @code{protected} which means that any references to the symbols from within the
5995 components that defines them must be resolved to the definition in that
5996 component, even if a definition in another component would normally preempt
6001 @section @code{.psize @var{lines} , @var{columns}}
6003 @cindex @code{psize} directive
6004 @cindex listing control: paper size
6005 @cindex paper size, for listings
6006 Use this directive to declare the number of lines---and, optionally, the
6007 number of columns---to use for each page, when generating listings.
6009 If you do not use @code{.psize}, listings use a default line-count
6010 of 60. You may omit the comma and @var{columns} specification; the
6011 default width is 200 columns.
6013 @command{@value{AS}} generates formfeeds whenever the specified number of
6014 lines is exceeded (or whenever you explicitly request one, using
6017 If you specify @var{lines} as @code{0}, no formfeeds are generated save
6018 those explicitly specified with @code{.eject}.
6021 @section @code{.purgem @var{name}}
6023 @cindex @code{purgem} directive
6024 Undefine the macro @var{name}, so that later uses of the string will not be
6025 expanded. @xref{Macro}.
6029 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
6031 @cindex @code{pushsection} directive
6032 @cindex Section Stack
6033 This is one of the ELF section stack manipulation directives. The others are
6034 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6035 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6038 This directive pushes the current section (and subsection) onto the
6039 top of the section stack, and then replaces the current section and
6040 subsection with @code{name} and @code{subsection}. The optional
6041 @code{flags}, @code{type} and @code{arguments} are treated the same
6042 as in the @code{.section} (@pxref{Section}) directive.
6046 @section @code{.quad @var{bignums}}
6048 @cindex @code{quad} directive
6049 @code{.quad} expects zero or more bignums, separated by commas. For
6050 each bignum, it emits
6052 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
6053 warning message; and just takes the lowest order 8 bytes of the bignum.
6054 @cindex eight-byte integer
6055 @cindex integer, 8-byte
6057 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
6058 hence @emph{quad}-word for 8 bytes.
6061 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
6062 warning message; and just takes the lowest order 16 bytes of the bignum.
6063 @cindex sixteen-byte integer
6064 @cindex integer, 16-byte
6068 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6070 @cindex @code{reloc} directive
6071 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6072 @var{expression}. If @var{offset} is a number, the relocation is generated in
6073 the current section. If @var{offset} is an expression that resolves to a
6074 symbol plus offset, the relocation is generated in the given symbol's section.
6075 @var{expression}, if present, must resolve to a symbol plus addend or to an
6076 absolute value, but note that not all targets support an addend. e.g. ELF REL
6077 targets such as i386 store an addend in the section contents rather than in the
6078 relocation. This low level interface does not support addends stored in the
6082 @section @code{.rept @var{count}}
6084 @cindex @code{rept} directive
6085 Repeat the sequence of lines between the @code{.rept} directive and the next
6086 @code{.endr} directive @var{count} times.
6088 For example, assembling
6096 is equivalent to assembling
6105 @section @code{.sbttl "@var{subheading}"}
6107 @cindex @code{sbttl} directive
6108 @cindex subtitles for listings
6109 @cindex listing control: subtitle
6110 Use @var{subheading} as the title (third line, immediately after the
6111 title line) when generating assembly listings.
6113 This directive affects subsequent pages, as well as the current page if
6114 it appears within ten lines of the top of a page.
6118 @section @code{.scl @var{class}}
6120 @cindex @code{scl} directive
6121 @cindex symbol storage class (COFF)
6122 @cindex COFF symbol storage class
6123 Set the storage-class value for a symbol. This directive may only be
6124 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6125 whether a symbol is static or external, or it may record further
6126 symbolic debugging information.
6129 The @samp{.scl} directive is primarily associated with COFF output; when
6130 configured to generate @code{b.out} output format, @command{@value{AS}}
6131 accepts this directive but ignores it.
6137 @section @code{.section @var{name}}
6139 @cindex named section
6140 Use the @code{.section} directive to assemble the following code into a section
6143 This directive is only supported for targets that actually support arbitrarily
6144 named sections; on @code{a.out} targets, for example, it is not accepted, even
6145 with a standard @code{a.out} section name.
6149 @c only print the extra heading if both COFF and ELF are set
6150 @subheading COFF Version
6153 @cindex @code{section} directive (COFF version)
6154 For COFF targets, the @code{.section} directive is used in one of the following
6158 .section @var{name}[, "@var{flags}"]
6159 .section @var{name}[, @var{subsection}]
6162 If the optional argument is quoted, it is taken as flags to use for the
6163 section. Each flag is a single character. The following flags are recognized:
6166 bss section (uninitialized data)
6168 section is not loaded
6174 exclude section from linking
6180 shared section (meaningful for PE targets)
6182 ignored. (For compatibility with the ELF version)
6184 section is not readable (meaningful for PE targets)
6186 single-digit power-of-two section alignment (GNU extension)
6189 If no flags are specified, the default flags depend upon the section name. If
6190 the section name is not recognized, the default will be for the section to be
6191 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6192 from the section, rather than adding them, so if they are used on their own it
6193 will be as if no flags had been specified at all.
6195 If the optional argument to the @code{.section} directive is not quoted, it is
6196 taken as a subsection number (@pxref{Sub-Sections}).
6201 @c only print the extra heading if both COFF and ELF are set
6202 @subheading ELF Version
6205 @cindex Section Stack
6206 This is one of the ELF section stack manipulation directives. The others are
6207 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6208 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6209 @code{.previous} (@pxref{Previous}).
6211 @cindex @code{section} directive (ELF version)
6212 For ELF targets, the @code{.section} directive is used like this:
6215 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6218 The optional @var{flags} argument is a quoted string which may contain any
6219 combination of the following characters:
6222 section is allocatable
6224 section is excluded from executable and shared library.
6228 section is executable
6230 section is mergeable
6232 section contains zero terminated strings
6234 section is a member of a section group
6236 section is used for thread-local-storage
6238 section is a member of the previously-current section's group, if any
6241 The optional @var{type} argument may contain one of the following constants:
6244 section contains data
6246 section does not contain data (i.e., section only occupies space)
6248 section contains data which is used by things other than the program
6250 section contains an array of pointers to init functions
6252 section contains an array of pointers to finish functions
6253 @item @@preinit_array
6254 section contains an array of pointers to pre-init functions
6257 Many targets only support the first three section types.
6259 Note on targets where the @code{@@} character is the start of a comment (eg
6260 ARM) then another character is used instead. For example the ARM port uses the
6263 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6264 be specified as well as an extra argument---@var{entsize}---like this:
6267 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6270 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6271 constants, each @var{entsize} octets long. Sections with both @code{M} and
6272 @code{S} must contain zero terminated strings where each character is
6273 @var{entsize} bytes long. The linker may remove duplicates within sections with
6274 the same name, same entity size and same flags. @var{entsize} must be an
6275 absolute expression. For sections with both @code{M} and @code{S}, a string
6276 which is a suffix of a larger string is considered a duplicate. Thus
6277 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6278 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6280 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6281 be present along with an additional field like this:
6284 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6287 The @var{GroupName} field specifies the name of the section group to which this
6288 particular section belongs. The optional linkage field can contain:
6291 indicates that only one copy of this section should be retained
6296 Note: if both the @var{M} and @var{G} flags are present then the fields for
6297 the Merge flag should come first, like this:
6300 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6303 If @var{flags} contains the @code{?} symbol then it may not also contain the
6304 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6305 present. Instead, @code{?} says to consider the section that's current before
6306 this directive. If that section used @code{G}, then the new section will use
6307 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6308 If not, then the @code{?} symbol has no effect.
6310 If no flags are specified, the default flags depend upon the section name. If
6311 the section name is not recognized, the default will be for the section to have
6312 none of the above flags: it will not be allocated in memory, nor writable, nor
6313 executable. The section will contain data.
6315 For ELF targets, the assembler supports another type of @code{.section}
6316 directive for compatibility with the Solaris assembler:
6319 .section "@var{name}"[, @var{flags}...]
6322 Note that the section name is quoted. There may be a sequence of comma
6326 section is allocatable
6330 section is executable
6332 section is excluded from executable and shared library.
6334 section is used for thread local storage
6337 This directive replaces the current section and subsection. See the
6338 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6339 some examples of how this directive and the other section stack directives
6345 @section @code{.set @var{symbol}, @var{expression}}
6347 @cindex @code{set} directive
6348 @cindex symbol value, setting
6349 Set the value of @var{symbol} to @var{expression}. This
6350 changes @var{symbol}'s value and type to conform to
6351 @var{expression}. If @var{symbol} was flagged as external, it remains
6352 flagged (@pxref{Symbol Attributes}).
6354 You may @code{.set} a symbol many times in the same assembly provided that the
6355 values given to the symbol are constants. Values that are based on expressions
6356 involving other symbols are allowed, but some targets may restrict this to only
6357 being done once per assembly. This is because those targets do not set the
6358 addresses of symbols at assembly time, but rather delay the assignment until a
6359 final link is performed. This allows the linker a chance to change the code in
6360 the files, changing the location of, and the relative distance between, various
6363 If you @code{.set} a global symbol, the value stored in the object
6364 file is the last value stored into it.
6367 On Z80 @code{set} is a real instruction, use
6368 @samp{@var{symbol} defl @var{expression}} instead.
6372 @section @code{.short @var{expressions}}
6374 @cindex @code{short} directive
6376 @code{.short} is normally the same as @samp{.word}.
6377 @xref{Word,,@code{.word}}.
6379 In some configurations, however, @code{.short} and @code{.word} generate
6380 numbers of different lengths. @xref{Machine Dependencies}.
6384 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6387 This expects zero or more @var{expressions}, and emits
6388 a 16 bit number for each.
6393 @section @code{.single @var{flonums}}
6395 @cindex @code{single} directive
6396 @cindex floating point numbers (single)
6397 This directive assembles zero or more flonums, separated by commas. It
6398 has the same effect as @code{.float}.
6400 The exact kind of floating point numbers emitted depends on how
6401 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6405 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6406 numbers in @sc{ieee} format.
6412 @section @code{.size}
6414 This directive is used to set the size associated with a symbol.
6418 @c only print the extra heading if both COFF and ELF are set
6419 @subheading COFF Version
6422 @cindex @code{size} directive (COFF version)
6423 For COFF targets, the @code{.size} directive is only permitted inside
6424 @code{.def}/@code{.endef} pairs. It is used like this:
6427 .size @var{expression}
6431 @samp{.size} is only meaningful when generating COFF format output; when
6432 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6439 @c only print the extra heading if both COFF and ELF are set
6440 @subheading ELF Version
6443 @cindex @code{size} directive (ELF version)
6444 For ELF targets, the @code{.size} directive is used like this:
6447 .size @var{name} , @var{expression}
6450 This directive sets the size associated with a symbol @var{name}.
6451 The size in bytes is computed from @var{expression} which can make use of label
6452 arithmetic. This directive is typically used to set the size of function
6457 @ifclear no-space-dir
6459 @section @code{.skip @var{size} , @var{fill}}
6461 @cindex @code{skip} directive
6462 @cindex filling memory
6463 This directive emits @var{size} bytes, each of value @var{fill}. Both
6464 @var{size} and @var{fill} are absolute expressions. If the comma and
6465 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6470 @section @code{.sleb128 @var{expressions}}
6472 @cindex @code{sleb128} directive
6473 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6474 compact, variable length representation of numbers used by the DWARF
6475 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6477 @ifclear no-space-dir
6479 @section @code{.space @var{size} , @var{fill}}
6481 @cindex @code{space} directive
6482 @cindex filling memory
6483 This directive emits @var{size} bytes, each of value @var{fill}. Both
6484 @var{size} and @var{fill} are absolute expressions. If the comma
6485 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6490 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6491 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6492 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6493 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6501 @section @code{.stabd, .stabn, .stabs}
6503 @cindex symbolic debuggers, information for
6504 @cindex @code{stab@var{x}} directives
6505 There are three directives that begin @samp{.stab}.
6506 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6507 The symbols are not entered in the @command{@value{AS}} hash table: they
6508 cannot be referenced elsewhere in the source file.
6509 Up to five fields are required:
6513 This is the symbol's name. It may contain any character except
6514 @samp{\000}, so is more general than ordinary symbol names. Some
6515 debuggers used to code arbitrarily complex structures into symbol names
6519 An absolute expression. The symbol's type is set to the low 8 bits of
6520 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6521 and debuggers choke on silly bit patterns.
6524 An absolute expression. The symbol's ``other'' attribute is set to the
6525 low 8 bits of this expression.
6528 An absolute expression. The symbol's descriptor is set to the low 16
6529 bits of this expression.
6532 An absolute expression which becomes the symbol's value.
6535 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6536 or @code{.stabs} statement, the symbol has probably already been created;
6537 you get a half-formed symbol in your object file. This is
6538 compatible with earlier assemblers!
6541 @cindex @code{stabd} directive
6542 @item .stabd @var{type} , @var{other} , @var{desc}
6544 The ``name'' of the symbol generated is not even an empty string.
6545 It is a null pointer, for compatibility. Older assemblers used a
6546 null pointer so they didn't waste space in object files with empty
6549 The symbol's value is set to the location counter,
6550 relocatably. When your program is linked, the value of this symbol
6551 is the address of the location counter when the @code{.stabd} was
6554 @cindex @code{stabn} directive
6555 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6556 The name of the symbol is set to the empty string @code{""}.
6558 @cindex @code{stabs} directive
6559 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6560 All five fields are specified.
6566 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6567 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6569 @cindex string, copying to object file
6570 @cindex string8, copying to object file
6571 @cindex string16, copying to object file
6572 @cindex string32, copying to object file
6573 @cindex string64, copying to object file
6574 @cindex @code{string} directive
6575 @cindex @code{string8} directive
6576 @cindex @code{string16} directive
6577 @cindex @code{string32} directive
6578 @cindex @code{string64} directive
6580 Copy the characters in @var{str} to the object file. You may specify more than
6581 one string to copy, separated by commas. Unless otherwise specified for a
6582 particular machine, the assembler marks the end of each string with a 0 byte.
6583 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6585 The variants @code{string16}, @code{string32} and @code{string64} differ from
6586 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6587 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6588 are stored in target endianness byte order.
6594 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6595 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6600 @section @code{.struct @var{expression}}
6602 @cindex @code{struct} directive
6603 Switch to the absolute section, and set the section offset to @var{expression},
6604 which must be an absolute expression. You might use this as follows:
6613 This would define the symbol @code{field1} to have the value 0, the symbol
6614 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6615 value 8. Assembly would be left in the absolute section, and you would need to
6616 use a @code{.section} directive of some sort to change to some other section
6617 before further assembly.
6621 @section @code{.subsection @var{name}}
6623 @cindex @code{subsection} directive
6624 @cindex Section Stack
6625 This is one of the ELF section stack manipulation directives. The others are
6626 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6627 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6630 This directive replaces the current subsection with @code{name}. The current
6631 section is not changed. The replaced subsection is put onto the section stack
6632 in place of the then current top of stack subsection.
6637 @section @code{.symver}
6638 @cindex @code{symver} directive
6639 @cindex symbol versioning
6640 @cindex versions of symbols
6641 Use the @code{.symver} directive to bind symbols to specific version nodes
6642 within a source file. This is only supported on ELF platforms, and is
6643 typically used when assembling files to be linked into a shared library.
6644 There are cases where it may make sense to use this in objects to be bound
6645 into an application itself so as to override a versioned symbol from a
6648 For ELF targets, the @code{.symver} directive can be used like this:
6650 .symver @var{name}, @var{name2@@nodename}
6652 If the symbol @var{name} is defined within the file
6653 being assembled, the @code{.symver} directive effectively creates a symbol
6654 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6655 just don't try and create a regular alias is that the @var{@@} character isn't
6656 permitted in symbol names. The @var{name2} part of the name is the actual name
6657 of the symbol by which it will be externally referenced. The name @var{name}
6658 itself is merely a name of convenience that is used so that it is possible to
6659 have definitions for multiple versions of a function within a single source
6660 file, and so that the compiler can unambiguously know which version of a
6661 function is being mentioned. The @var{nodename} portion of the alias should be
6662 the name of a node specified in the version script supplied to the linker when
6663 building a shared library. If you are attempting to override a versioned
6664 symbol from a shared library, then @var{nodename} should correspond to the
6665 nodename of the symbol you are trying to override.
6667 If the symbol @var{name} is not defined within the file being assembled, all
6668 references to @var{name} will be changed to @var{name2@@nodename}. If no
6669 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6672 Another usage of the @code{.symver} directive is:
6674 .symver @var{name}, @var{name2@@@@nodename}
6676 In this case, the symbol @var{name} must exist and be defined within
6677 the file being assembled. It is similar to @var{name2@@nodename}. The
6678 difference is @var{name2@@@@nodename} will also be used to resolve
6679 references to @var{name2} by the linker.
6681 The third usage of the @code{.symver} directive is:
6683 .symver @var{name}, @var{name2@@@@@@nodename}
6685 When @var{name} is not defined within the
6686 file being assembled, it is treated as @var{name2@@nodename}. When
6687 @var{name} is defined within the file being assembled, the symbol
6688 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6693 @section @code{.tag @var{structname}}
6695 @cindex COFF structure debugging
6696 @cindex structure debugging, COFF
6697 @cindex @code{tag} directive
6698 This directive is generated by compilers to include auxiliary debugging
6699 information in the symbol table. It is only permitted inside
6700 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6701 definitions in the symbol table with instances of those structures.
6704 @samp{.tag} is only used when generating COFF format output; when
6705 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6711 @section @code{.text @var{subsection}}
6713 @cindex @code{text} directive
6714 Tells @command{@value{AS}} to assemble the following statements onto the end of
6715 the text subsection numbered @var{subsection}, which is an absolute
6716 expression. If @var{subsection} is omitted, subsection number zero
6720 @section @code{.title "@var{heading}"}
6722 @cindex @code{title} directive
6723 @cindex listing control: title line
6724 Use @var{heading} as the title (second line, immediately after the
6725 source file name and pagenumber) when generating assembly listings.
6727 This directive affects subsequent pages, as well as the current page if
6728 it appears within ten lines of the top of a page.
6732 @section @code{.type}
6734 This directive is used to set the type of a symbol.
6738 @c only print the extra heading if both COFF and ELF are set
6739 @subheading COFF Version
6742 @cindex COFF symbol type
6743 @cindex symbol type, COFF
6744 @cindex @code{type} directive (COFF version)
6745 For COFF targets, this directive is permitted only within
6746 @code{.def}/@code{.endef} pairs. It is used like this:
6752 This records the integer @var{int} as the type attribute of a symbol table
6756 @samp{.type} is associated only with COFF format output; when
6757 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6758 directive but ignores it.
6764 @c only print the extra heading if both COFF and ELF are set
6765 @subheading ELF Version
6768 @cindex ELF symbol type
6769 @cindex symbol type, ELF
6770 @cindex @code{type} directive (ELF version)
6771 For ELF targets, the @code{.type} directive is used like this:
6774 .type @var{name} , @var{type description}
6777 This sets the type of symbol @var{name} to be either a
6778 function symbol or an object symbol. There are five different syntaxes
6779 supported for the @var{type description} field, in order to provide
6780 compatibility with various other assemblers.
6782 Because some of the characters used in these syntaxes (such as @samp{@@} and
6783 @samp{#}) are comment characters for some architectures, some of the syntaxes
6784 below do not work on all architectures. The first variant will be accepted by
6785 the GNU assembler on all architectures so that variant should be used for
6786 maximum portability, if you do not need to assemble your code with other
6789 The syntaxes supported are:
6792 .type <name> STT_<TYPE_IN_UPPER_CASE>
6793 .type <name>,#<type>
6794 .type <name>,@@<type>
6795 .type <name>,%<type>
6796 .type <name>,"<type>"
6799 The types supported are:
6804 Mark the symbol as being a function name.
6807 @itemx gnu_indirect_function
6808 Mark the symbol as an indirect function when evaluated during reloc
6809 processing. (This is only supported on assemblers targeting GNU systems).
6813 Mark the symbol as being a data object.
6817 Mark the symbol as being a thead-local data object.
6821 Mark the symbol as being a common data object.
6825 Does not mark the symbol in any way. It is supported just for completeness.
6827 @item gnu_unique_object
6828 Marks the symbol as being a globally unique data object. The dynamic linker
6829 will make sure that in the entire process there is just one symbol with this
6830 name and type in use. (This is only supported on assemblers targeting GNU
6835 Note: Some targets support extra types in addition to those listed above.
6841 @section @code{.uleb128 @var{expressions}}
6843 @cindex @code{uleb128} directive
6844 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6845 compact, variable length representation of numbers used by the DWARF
6846 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6850 @section @code{.val @var{addr}}
6852 @cindex @code{val} directive
6853 @cindex COFF value attribute
6854 @cindex value attribute, COFF
6855 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6856 records the address @var{addr} as the value attribute of a symbol table
6860 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6861 configured for @code{b.out}, it accepts this directive but ignores it.
6867 @section @code{.version "@var{string}"}
6869 @cindex @code{version} directive
6870 This directive creates a @code{.note} section and places into it an ELF
6871 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6876 @section @code{.vtable_entry @var{table}, @var{offset}}
6878 @cindex @code{vtable_entry} directive
6879 This directive finds or creates a symbol @code{table} and creates a
6880 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6883 @section @code{.vtable_inherit @var{child}, @var{parent}}
6885 @cindex @code{vtable_inherit} directive
6886 This directive finds the symbol @code{child} and finds or creates the symbol
6887 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6888 parent whose addend is the value of the child symbol. As a special case the
6889 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6893 @section @code{.warning "@var{string}"}
6894 @cindex warning directive
6895 Similar to the directive @code{.error}
6896 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6899 @section @code{.weak @var{names}}
6901 @cindex @code{weak} directive
6902 This directive sets the weak attribute on the comma separated list of symbol
6903 @code{names}. If the symbols do not already exist, they will be created.
6905 On COFF targets other than PE, weak symbols are a GNU extension. This
6906 directive sets the weak attribute on the comma separated list of symbol
6907 @code{names}. If the symbols do not already exist, they will be created.
6909 On the PE target, weak symbols are supported natively as weak aliases.
6910 When a weak symbol is created that is not an alias, GAS creates an
6911 alternate symbol to hold the default value.
6914 @section @code{.weakref @var{alias}, @var{target}}
6916 @cindex @code{weakref} directive
6917 This directive creates an alias to the target symbol that enables the symbol to
6918 be referenced with weak-symbol semantics, but without actually making it weak.
6919 If direct references or definitions of the symbol are present, then the symbol
6920 will not be weak, but if all references to it are through weak references, the
6921 symbol will be marked as weak in the symbol table.
6923 The effect is equivalent to moving all references to the alias to a separate
6924 assembly source file, renaming the alias to the symbol in it, declaring the
6925 symbol as weak there, and running a reloadable link to merge the object files
6926 resulting from the assembly of the new source file and the old source file that
6927 had the references to the alias removed.
6929 The alias itself never makes to the symbol table, and is entirely handled
6930 within the assembler.
6933 @section @code{.word @var{expressions}}
6935 @cindex @code{word} directive
6936 This directive expects zero or more @var{expressions}, of any section,
6937 separated by commas.
6940 For each expression, @command{@value{AS}} emits a 32-bit number.
6943 For each expression, @command{@value{AS}} emits a 16-bit number.
6948 The size of the number emitted, and its byte order,
6949 depend on what target computer the assembly is for.
6952 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
6953 @c happen---32-bit addressability, period; no long/short jumps.
6954 @ifset DIFF-TBL-KLUGE
6955 @cindex difference tables altered
6956 @cindex altered difference tables
6958 @emph{Warning: Special Treatment to support Compilers}
6962 Machines with a 32-bit address space, but that do less than 32-bit
6963 addressing, require the following special treatment. If the machine of
6964 interest to you does 32-bit addressing (or doesn't require it;
6965 @pxref{Machine Dependencies}), you can ignore this issue.
6968 In order to assemble compiler output into something that works,
6969 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6970 Directives of the form @samp{.word sym1-sym2} are often emitted by
6971 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6972 directive of the form @samp{.word sym1-sym2}, and the difference between
6973 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6974 creates a @dfn{secondary jump table}, immediately before the next label.
6975 This secondary jump table is preceded by a short-jump to the
6976 first byte after the secondary table. This short-jump prevents the flow
6977 of control from accidentally falling into the new table. Inside the
6978 table is a long-jump to @code{sym2}. The original @samp{.word}
6979 contains @code{sym1} minus the address of the long-jump to
6982 If there were several occurrences of @samp{.word sym1-sym2} before the
6983 secondary jump table, all of them are adjusted. If there was a
6984 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6985 long-jump to @code{sym4} is included in the secondary jump table,
6986 and the @code{.word} directives are adjusted to contain @code{sym3}
6987 minus the address of the long-jump to @code{sym4}; and so on, for as many
6988 entries in the original jump table as necessary.
6991 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6992 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6993 assembly language programmers.
6996 @c end DIFF-TBL-KLUGE
6999 @section Deprecated Directives
7001 @cindex deprecated directives
7002 @cindex obsolescent directives
7003 One day these directives won't work.
7004 They are included for compatibility with older assemblers.
7011 @node Object Attributes
7012 @chapter Object Attributes
7013 @cindex object attributes
7015 @command{@value{AS}} assembles source files written for a specific architecture
7016 into object files for that architecture. But not all object files are alike.
7017 Many architectures support incompatible variations. For instance, floating
7018 point arguments might be passed in floating point registers if the object file
7019 requires hardware floating point support---or floating point arguments might be
7020 passed in integer registers if the object file supports processors with no
7021 hardware floating point unit. Or, if two objects are built for different
7022 generations of the same architecture, the combination may require the
7023 newer generation at run-time.
7025 This information is useful during and after linking. At link time,
7026 @command{@value{LD}} can warn about incompatible object files. After link
7027 time, tools like @command{gdb} can use it to process the linked file
7030 Compatibility information is recorded as a series of object attributes. Each
7031 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
7032 string, and indicates who sets the meaning of the tag. The tag is an integer,
7033 and indicates what property the attribute describes. The value may be a string
7034 or an integer, and indicates how the property affects this object. Missing
7035 attributes are the same as attributes with a zero value or empty string value.
7037 Object attributes were developed as part of the ABI for the ARM Architecture.
7038 The file format is documented in @cite{ELF for the ARM Architecture}.
7041 * GNU Object Attributes:: @sc{gnu} Object Attributes
7042 * Defining New Object Attributes:: Defining New Object Attributes
7045 @node GNU Object Attributes
7046 @section @sc{gnu} Object Attributes
7048 The @code{.gnu_attribute} directive records an object attribute
7049 with vendor @samp{gnu}.
7051 Except for @samp{Tag_compatibility}, which has both an integer and a string for
7052 its value, @sc{gnu} attributes have a string value if the tag number is odd and
7053 an integer value if the tag number is even. The second bit (@code{@var{tag} &
7054 2} is set for architecture-independent attributes and clear for
7055 architecture-dependent ones.
7057 @subsection Common @sc{gnu} attributes
7059 These attributes are valid on all architectures.
7062 @item Tag_compatibility (32)
7063 The compatibility attribute takes an integer flag value and a vendor name. If
7064 the flag value is 0, the file is compatible with other toolchains. If it is 1,
7065 then the file is only compatible with the named toolchain. If it is greater
7066 than 1, the file can only be processed by other toolchains under some private
7067 arrangement indicated by the flag value and the vendor name.
7070 @subsection MIPS Attributes
7073 @item Tag_GNU_MIPS_ABI_FP (4)
7074 The floating-point ABI used by this object file. The value will be:
7078 0 for files not affected by the floating-point ABI.
7080 1 for files using the hardware floating-point ABI with a standard
7081 double-precision FPU.
7083 2 for files using the hardware floating-point ABI with a single-precision FPU.
7085 3 for files using the software floating-point ABI.
7087 4 for files using the deprecated hardware floating-point ABI which used 64-bit
7088 floating-point registers, 32-bit general-purpose registers and increased the
7089 number of callee-saved floating-point registers.
7091 5 for files using the hardware floating-point ABI with a double-precision FPU
7092 with either 32-bit or 64-bit floating-point registers and 32-bit
7093 general-purpose registers.
7095 6 for files using the hardware floating-point ABI with 64-bit floating-point
7096 registers and 32-bit general-purpose registers.
7098 7 for files using the hardware floating-point ABI with 64-bit floating-point
7099 registers, 32-bit general-purpose registers and a rule that forbids the
7100 direct use of odd-numbered single-precision floating-point registers.
7104 @subsection PowerPC Attributes
7107 @item Tag_GNU_Power_ABI_FP (4)
7108 The floating-point ABI used by this object file. The value will be:
7112 0 for files not affected by the floating-point ABI.
7114 1 for files using double-precision hardware floating-point ABI.
7116 2 for files using the software floating-point ABI.
7118 3 for files using single-precision hardware floating-point ABI.
7121 @item Tag_GNU_Power_ABI_Vector (8)
7122 The vector ABI used by this object file. The value will be:
7126 0 for files not affected by the vector ABI.
7128 1 for files using general purpose registers to pass vectors.
7130 2 for files using AltiVec registers to pass vectors.
7132 3 for files using SPE registers to pass vectors.
7136 @subsection IBM z Systems Attributes
7139 @item Tag_GNU_S390_ABI_Vector (8)
7140 The vector ABI used by this object file. The value will be:
7144 0 for files not affected by the vector ABI.
7146 1 for files using software vector ABI.
7148 2 for files using hardware vector ABI.
7152 @node Defining New Object Attributes
7153 @section Defining New Object Attributes
7155 If you want to define a new @sc{gnu} object attribute, here are the places you
7156 will need to modify. New attributes should be discussed on the @samp{binutils}
7161 This manual, which is the official register of attributes.
7163 The header for your architecture @file{include/elf}, to define the tag.
7165 The @file{bfd} support file for your architecture, to merge the attribute
7166 and issue any appropriate link warnings.
7168 Test cases in @file{ld/testsuite} for merging and link warnings.
7170 @file{binutils/readelf.c} to display your attribute.
7172 GCC, if you want the compiler to mark the attribute automatically.
7178 @node Machine Dependencies
7179 @chapter Machine Dependent Features
7181 @cindex machine dependencies
7182 The machine instruction sets are (almost by definition) different on
7183 each machine where @command{@value{AS}} runs. Floating point representations
7184 vary as well, and @command{@value{AS}} often supports a few additional
7185 directives or command-line options for compatibility with other
7186 assemblers on a particular platform. Finally, some versions of
7187 @command{@value{AS}} support special pseudo-instructions for branch
7190 This chapter discusses most of these differences, though it does not
7191 include details on any machine's instruction set. For details on that
7192 subject, see the hardware manufacturer's manual.
7196 * AArch64-Dependent:: AArch64 Dependent Features
7199 * Alpha-Dependent:: Alpha Dependent Features
7202 * ARC-Dependent:: ARC Dependent Features
7205 * ARM-Dependent:: ARM Dependent Features
7208 * AVR-Dependent:: AVR Dependent Features
7211 * Blackfin-Dependent:: Blackfin Dependent Features
7214 * CR16-Dependent:: CR16 Dependent Features
7217 * CRIS-Dependent:: CRIS Dependent Features
7220 * D10V-Dependent:: D10V Dependent Features
7223 * D30V-Dependent:: D30V Dependent Features
7226 * Epiphany-Dependent:: EPIPHANY Dependent Features
7229 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7232 * HPPA-Dependent:: HPPA Dependent Features
7235 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
7238 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7241 * i860-Dependent:: Intel 80860 Dependent Features
7244 * i960-Dependent:: Intel 80960 Dependent Features
7247 * IA-64-Dependent:: Intel IA-64 Dependent Features
7250 * IP2K-Dependent:: IP2K Dependent Features
7253 * LM32-Dependent:: LM32 Dependent Features
7256 * M32C-Dependent:: M32C Dependent Features
7259 * M32R-Dependent:: M32R Dependent Features
7262 * M68K-Dependent:: M680x0 Dependent Features
7265 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7268 * Meta-Dependent :: Meta Dependent Features
7271 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7274 * MIPS-Dependent:: MIPS Dependent Features
7277 * MMIX-Dependent:: MMIX Dependent Features
7280 * MSP430-Dependent:: MSP430 Dependent Features
7283 * NDS32-Dependent:: Andes NDS32 Dependent Features
7286 * NiosII-Dependent:: Altera Nios II Dependent Features
7289 * NS32K-Dependent:: NS32K Dependent Features
7292 * PDP-11-Dependent:: PDP-11 Dependent Features
7295 * PJ-Dependent:: picoJava Dependent Features
7298 * PPC-Dependent:: PowerPC Dependent Features
7301 * RL78-Dependent:: RL78 Dependent Features
7304 * RX-Dependent:: RX Dependent Features
7307 * S/390-Dependent:: IBM S/390 Dependent Features
7310 * SCORE-Dependent:: SCORE Dependent Features
7313 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7314 * SH64-Dependent:: SuperH SH64 Dependent Features
7317 * Sparc-Dependent:: SPARC Dependent Features
7320 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7323 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7326 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7329 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7332 * V850-Dependent:: V850 Dependent Features
7335 * Vax-Dependent:: VAX Dependent Features
7338 * Visium-Dependent:: Visium Dependent Features
7341 * XGATE-Dependent:: XGATE Features
7344 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7347 * Xtensa-Dependent:: Xtensa Dependent Features
7350 * Z80-Dependent:: Z80 Dependent Features
7353 * Z8000-Dependent:: Z8000 Dependent Features
7360 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7361 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7362 @c peculiarity: to preserve cross-references, there must be a node called
7363 @c "Machine Dependencies". Hence the conditional nodenames in each
7364 @c major node below. Node defaulting in makeinfo requires adjacency of
7365 @c node and sectioning commands; hence the repetition of @chapter BLAH
7366 @c in both conditional blocks.
7369 @include c-aarch64.texi
7373 @include c-alpha.texi
7389 @include c-bfin.texi
7393 @include c-cr16.texi
7397 @include c-cris.texi
7402 @node Machine Dependencies
7403 @chapter Machine Dependent Features
7405 The machine instruction sets are different on each Renesas chip family,
7406 and there are also some syntax differences among the families. This
7407 chapter describes the specific @command{@value{AS}} features for each
7411 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7412 * SH-Dependent:: Renesas SH Dependent Features
7419 @include c-d10v.texi
7423 @include c-d30v.texi
7427 @include c-epiphany.texi
7431 @include c-h8300.texi
7435 @include c-hppa.texi
7439 @include c-i370.texi
7443 @include c-i386.texi
7447 @include c-i860.texi
7451 @include c-i960.texi
7455 @include c-ia64.texi
7459 @include c-ip2k.texi
7463 @include c-lm32.texi
7467 @include c-m32c.texi
7471 @include c-m32r.texi
7475 @include c-m68k.texi
7479 @include c-m68hc11.texi
7483 @include c-metag.texi
7487 @include c-microblaze.texi
7491 @include c-mips.texi
7495 @include c-mmix.texi
7499 @include c-msp430.texi
7503 @include c-nds32.texi
7507 @include c-nios2.texi
7511 @include c-ns32k.texi
7515 @include c-pdp11.texi
7527 @include c-rl78.texi
7535 @include c-s390.texi
7539 @include c-score.texi
7544 @include c-sh64.texi
7548 @include c-sparc.texi
7552 @include c-tic54x.texi
7556 @include c-tic6x.texi
7560 @include c-tilegx.texi
7564 @include c-tilepro.texi
7568 @include c-v850.texi
7576 @include c-visium.texi
7580 @include c-xgate.texi
7584 @include c-xstormy16.texi
7588 @include c-xtensa.texi
7600 @c reverse effect of @down at top of generic Machine-Dep chapter
7604 @node Reporting Bugs
7605 @chapter Reporting Bugs
7606 @cindex bugs in assembler
7607 @cindex reporting bugs in assembler
7609 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7611 Reporting a bug may help you by bringing a solution to your problem, or it may
7612 not. But in any case the principal function of a bug report is to help the
7613 entire community by making the next version of @command{@value{AS}} work better.
7614 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7616 In order for a bug report to serve its purpose, you must include the
7617 information that enables us to fix the bug.
7620 * Bug Criteria:: Have you found a bug?
7621 * Bug Reporting:: How to report bugs
7625 @section Have You Found a Bug?
7626 @cindex bug criteria
7628 If you are not sure whether you have found a bug, here are some guidelines:
7631 @cindex fatal signal
7632 @cindex assembler crash
7633 @cindex crash of assembler
7635 If the assembler gets a fatal signal, for any input whatever, that is a
7636 @command{@value{AS}} bug. Reliable assemblers never crash.
7638 @cindex error on valid input
7640 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7642 @cindex invalid input
7644 If @command{@value{AS}} does not produce an error message for invalid input, that
7645 is a bug. However, you should note that your idea of ``invalid input'' might
7646 be our idea of ``an extension'' or ``support for traditional practice''.
7649 If you are an experienced user of assemblers, your suggestions for improvement
7650 of @command{@value{AS}} are welcome in any case.
7654 @section How to Report Bugs
7656 @cindex assembler bugs, reporting
7658 A number of companies and individuals offer support for @sc{gnu} products. If
7659 you obtained @command{@value{AS}} from a support organization, we recommend you
7660 contact that organization first.
7662 You can find contact information for many support companies and
7663 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7667 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7671 The fundamental principle of reporting bugs usefully is this:
7672 @strong{report all the facts}. If you are not sure whether to state a
7673 fact or leave it out, state it!
7675 Often people omit facts because they think they know what causes the problem
7676 and assume that some details do not matter. Thus, you might assume that the
7677 name of a symbol you use in an example does not matter. Well, probably it does
7678 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7679 happens to fetch from the location where that name is stored in memory;
7680 perhaps, if the name were different, the contents of that location would fool
7681 the assembler into doing the right thing despite the bug. Play it safe and
7682 give a specific, complete example. That is the easiest thing for you to do,
7683 and the most helpful.
7685 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7686 it is new to us. Therefore, always write your bug reports on the assumption
7687 that the bug has not been reported previously.
7689 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7690 bell?'' This cannot help us fix a bug, so it is basically useless. We
7691 respond by asking for enough details to enable us to investigate.
7692 You might as well expedite matters by sending them to begin with.
7694 To enable us to fix the bug, you should include all these things:
7698 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
7699 it with the @samp{--version} argument.
7701 Without this, we will not know whether there is any point in looking for
7702 the bug in the current version of @command{@value{AS}}.
7705 Any patches you may have applied to the @command{@value{AS}} source.
7708 The type of machine you are using, and the operating system name and
7712 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
7716 The command arguments you gave the assembler to assemble your example and
7717 observe the bug. To guarantee you will not omit something important, list them
7718 all. A copy of the Makefile (or the output from make) is sufficient.
7720 If we were to try to guess the arguments, we would probably guess wrong
7721 and then we might not encounter the bug.
7724 A complete input file that will reproduce the bug. If the bug is observed when
7725 the assembler is invoked via a compiler, send the assembler source, not the
7726 high level language source. Most compilers will produce the assembler source
7727 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
7728 the options @samp{-v --save-temps}; this will save the assembler source in a
7729 file with an extension of @file{.s}, and also show you exactly how
7730 @command{@value{AS}} is being run.
7733 A description of what behavior you observe that you believe is
7734 incorrect. For example, ``It gets a fatal signal.''
7736 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
7737 will certainly notice it. But if the bug is incorrect output, we might not
7738 notice unless it is glaringly wrong. You might as well not give us a chance to
7741 Even if the problem you experience is a fatal signal, you should still say so
7742 explicitly. Suppose something strange is going on, such as, your copy of
7743 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
7744 library on your system. (This has happened!) Your copy might crash and ours
7745 would not. If you told us to expect a crash, then when ours fails to crash, we
7746 would know that the bug was not happening for us. If you had not told us to
7747 expect a crash, then we would not be able to draw any conclusion from our
7751 If you wish to suggest changes to the @command{@value{AS}} source, send us context
7752 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
7753 option. Always send diffs from the old file to the new file. If you even
7754 discuss something in the @command{@value{AS}} source, refer to it by context, not
7757 The line numbers in our development sources will not match those in your
7758 sources. Your line numbers would convey no useful information to us.
7761 Here are some things that are not necessary:
7765 A description of the envelope of the bug.
7767 Often people who encounter a bug spend a lot of time investigating
7768 which changes to the input file will make the bug go away and which
7769 changes will not affect it.
7771 This is often time consuming and not very useful, because the way we
7772 will find the bug is by running a single example under the debugger
7773 with breakpoints, not by pure deduction from a series of examples.
7774 We recommend that you save your time for something else.
7776 Of course, if you can find a simpler example to report @emph{instead}
7777 of the original one, that is a convenience for us. Errors in the
7778 output will be easier to spot, running under the debugger will take
7779 less time, and so on.
7781 However, simplification is not vital; if you do not want to do this,
7782 report the bug anyway and send us the entire test case you used.
7785 A patch for the bug.
7787 A patch for the bug does help us if it is a good one. But do not omit
7788 the necessary information, such as the test case, on the assumption that
7789 a patch is all we need. We might see problems with your patch and decide
7790 to fix the problem another way, or we might not understand it at all.
7792 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
7793 construct an example that will make the program follow a certain path through
7794 the code. If you do not send us the example, we will not be able to construct
7795 one, so we will not be able to verify that the bug is fixed.
7797 And if we cannot understand what bug you are trying to fix, or why your
7798 patch should be an improvement, we will not install it. A test case will
7799 help us to understand.
7802 A guess about what the bug is or what it depends on.
7804 Such guesses are usually wrong. Even we cannot guess right about such
7805 things without first using the debugger to find the facts.
7808 @node Acknowledgements
7809 @chapter Acknowledgements
7811 If you have contributed to GAS and your name isn't listed here,
7812 it is not meant as a slight. We just don't know about it. Send mail to the
7813 maintainer, and we'll correct the situation. Currently
7815 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
7817 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
7820 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
7821 information and the 68k series machines, most of the preprocessing pass, and
7822 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
7824 K. Richard Pixley maintained GAS for a while, adding various enhancements and
7825 many bug fixes, including merging support for several processors, breaking GAS
7826 up to handle multiple object file format back ends (including heavy rewrite,
7827 testing, an integration of the coff and b.out back ends), adding configuration
7828 including heavy testing and verification of cross assemblers and file splits
7829 and renaming, converted GAS to strictly ANSI C including full prototypes, added
7830 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
7831 port (including considerable amounts of reverse engineering), a SPARC opcode
7832 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
7833 assertions and made them work, much other reorganization, cleanup, and lint.
7835 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
7836 in format-specific I/O modules.
7838 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
7839 has done much work with it since.
7841 The Intel 80386 machine description was written by Eliot Dresselhaus.
7843 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
7845 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
7846 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
7848 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
7849 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
7850 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
7851 support a.out format.
7853 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
7854 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
7855 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
7856 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
7859 John Gilmore built the AMD 29000 support, added @code{.include} support, and
7860 simplified the configuration of which versions accept which directives. He
7861 updated the 68k machine description so that Motorola's opcodes always produced
7862 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
7863 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
7864 cross-compilation support, and one bug in relaxation that took a week and
7865 required the proverbial one-bit fix.
7867 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
7868 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
7869 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
7870 PowerPC assembler, and made a few other minor patches.
7872 Steve Chamberlain made GAS able to generate listings.
7874 Hewlett-Packard contributed support for the HP9000/300.
7876 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
7877 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
7878 formats). This work was supported by both the Center for Software Science at
7879 the University of Utah and Cygnus Support.
7881 Support for ELF format files has been worked on by Mark Eichin of Cygnus
7882 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
7883 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
7884 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
7885 and some initial 64-bit support).
7887 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
7889 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
7890 support for openVMS/Alpha.
7892 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
7895 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
7896 Inc.@: added support for Xtensa processors.
7898 Several engineers at Cygnus Support have also provided many small bug fixes and
7899 configuration enhancements.
7901 Jon Beniston added support for the Lattice Mico32 architecture.
7903 Many others have contributed large or small bugfixes and enhancements. If
7904 you have contributed significant work and are not mentioned on this list, and
7905 want to be, let us know. Some of the history has been lost; we are not
7906 intentionally leaving anyone out.
7908 @node GNU Free Documentation License
7909 @appendix GNU Free Documentation License
7913 @unnumbered AS Index