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.
632 @item --nocompress-debug-sections
633 Do not compress DWARF debug sections. This is the default.
636 Ignored. This option is accepted for script compatibility with calls to
639 @item --debug-prefix-map @var{old}=@var{new}
640 When assembling files in directory @file{@var{old}}, record debugging
641 information describing them as in @file{@var{new}} instead.
643 @item --defsym @var{sym}=@var{value}
644 Define the symbol @var{sym} to be @var{value} before assembling the input file.
645 @var{value} must be an integer constant. As in C, a leading @samp{0x}
646 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
647 value. The value of the symbol can be overridden inside a source file via the
648 use of a @code{.set} pseudo-op.
651 ``fast''---skip whitespace and comment preprocessing (assume source is
656 Generate debugging information for each assembler source line using whichever
657 debug format is preferred by the target. This currently means either STABS,
661 Generate stabs debugging information for each assembler line. This
662 may help debugging assembler code, if the debugger can handle it.
665 Generate stabs debugging information for each assembler line, with GNU
666 extensions that probably only gdb can handle, and that could make other
667 debuggers crash or refuse to read your program. This
668 may help debugging assembler code. Currently the only GNU extension is
669 the location of the current working directory at assembling time.
672 Generate DWARF2 debugging information for each assembler line. This
673 may help debugging assembler code, if the debugger can handle it. Note---this
674 option is only supported by some targets, not all of them.
676 @item --gdwarf-sections
677 Instead of creating a .debug_line section, create a series of
678 .debug_line.@var{foo} sections where @var{foo} is the name of the
679 corresponding code section. For example a code section called @var{.text.func}
680 will have its dwarf line number information placed into a section called
681 @var{.debug_line.text.func}. If the code section is just called @var{.text}
682 then debug line section will still be called just @var{.debug_line} without any
685 @item --size-check=error
686 @itemx --size-check=warning
687 Issue an error or warning for invalid ELF .size directive.
690 Print a summary of the command line options and exit.
693 Print a summary of all target specific options and exit.
696 Add directory @var{dir} to the search list for @code{.include} directives.
699 Don't warn about signed overflow.
702 @ifclear DIFF-TBL-KLUGE
703 This option is accepted but has no effect on the @value{TARGET} family.
705 @ifset DIFF-TBL-KLUGE
706 Issue warnings when difference tables altered for long displacements.
711 Keep (in the symbol table) local symbols. These symbols start with
712 system-specific local label prefixes, typically @samp{.L} for ELF systems
713 or @samp{L} for traditional a.out systems.
718 @item --listing-lhs-width=@var{number}
719 Set the maximum width, in words, of the output data column for an assembler
720 listing to @var{number}.
722 @item --listing-lhs-width2=@var{number}
723 Set the maximum width, in words, of the output data column for continuation
724 lines in an assembler listing to @var{number}.
726 @item --listing-rhs-width=@var{number}
727 Set the maximum width of an input source line, as displayed in a listing, to
730 @item --listing-cont-lines=@var{number}
731 Set the maximum number of lines printed in a listing for a single line of input
734 @item -o @var{objfile}
735 Name the object-file output from @command{@value{AS}} @var{objfile}.
738 Fold the data section into the text section.
740 @kindex --hash-size=@var{number}
741 Set the default size of GAS's hash tables to a prime number close to
742 @var{number}. Increasing this value can reduce the length of time it takes the
743 assembler to perform its tasks, at the expense of increasing the assembler's
744 memory requirements. Similarly reducing this value can reduce the memory
745 requirements at the expense of speed.
747 @item --reduce-memory-overheads
748 This option reduces GAS's memory requirements, at the expense of making the
749 assembly processes slower. Currently this switch is a synonym for
750 @samp{--hash-size=4051}, but in the future it may have other effects as well.
753 Print the maximum space (in bytes) and total time (in seconds) used by
756 @item --strip-local-absolute
757 Remove local absolute symbols from the outgoing symbol table.
761 Print the @command{as} version.
764 Print the @command{as} version and exit.
768 Suppress warning messages.
770 @item --fatal-warnings
771 Treat warnings as errors.
774 Don't suppress warning messages or treat them as errors.
783 Generate an object file even after errors.
785 @item -- | @var{files} @dots{}
786 Standard input, or source files to assemble.
794 @xref{AArch64 Options}, for the options available when @value{AS} is configured
795 for the 64-bit mode of the ARM Architecture (AArch64).
800 The following options are available when @value{AS} is configured for the
801 64-bit mode of the ARM Architecture (AArch64).
804 @include c-aarch64.texi
805 @c ended inside the included file
813 @xref{Alpha Options}, for the options available when @value{AS} is configured
814 for an Alpha processor.
819 The following options are available when @value{AS} is configured for an Alpha
823 @include c-alpha.texi
824 @c ended inside the included file
831 The following options are available when @value{AS} is configured for
836 This option selects the core processor variant.
838 Select either big-endian (-EB) or little-endian (-EL) output.
843 The following options are available when @value{AS} is configured for the ARM
847 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
848 Specify which ARM processor variant is the target.
849 @item -march=@var{architecture}[+@var{extension}@dots{}]
850 Specify which ARM architecture variant is used by the target.
851 @item -mfpu=@var{floating-point-format}
852 Select which Floating Point architecture is the target.
853 @item -mfloat-abi=@var{abi}
854 Select which floating point ABI is in use.
856 Enable Thumb only instruction decoding.
857 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
858 Select which procedure calling convention is in use.
860 Select either big-endian (-EB) or little-endian (-EL) output.
861 @item -mthumb-interwork
862 Specify that the code has been generated with interworking between Thumb and
865 Turns on CodeComposer Studio assembly syntax compatibility mode.
867 Specify that PIC code has been generated.
875 @xref{Blackfin Options}, for the options available when @value{AS} is
876 configured for the Blackfin processor family.
881 The following options are available when @value{AS} is configured for
882 the Blackfin processor family.
886 @c ended inside the included file
893 See the info pages for documentation of the CRIS-specific options.
897 The following options are available when @value{AS} is configured for
900 @cindex D10V optimization
901 @cindex optimization, D10V
903 Optimize output by parallelizing instructions.
908 The following options are available when @value{AS} is configured for a D30V
911 @cindex D30V optimization
912 @cindex optimization, D30V
914 Optimize output by parallelizing instructions.
918 Warn when nops are generated.
920 @cindex D30V nops after 32-bit multiply
922 Warn when a nop after a 32-bit multiply instruction is generated.
928 The following options are available when @value{AS} is configured for the
929 Adapteva EPIPHANY series.
932 @xref{Epiphany Options}, for the options available when @value{AS} is
933 configured for an Epiphany processor.
938 The following options are available when @value{AS} is configured for
939 an Epiphany processor.
942 @include c-epiphany.texi
943 @c ended inside the included file
951 @xref{H8/300 Options}, for the options available when @value{AS} is configured
952 for an H8/300 processor.
957 The following options are available when @value{AS} is configured for an H8/300
961 @include c-h8300.texi
962 @c ended inside the included file
970 @xref{i386-Options}, for the options available when @value{AS} is
971 configured for an i386 processor.
976 The following options are available when @value{AS} is configured for
981 @c ended inside the included file
988 The following options are available when @value{AS} is configured for the
989 Intel 80960 processor.
992 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
993 Specify which variant of the 960 architecture is the target.
996 Add code to collect statistics about branches taken.
999 Do not alter compare-and-branch instructions for long displacements;
1006 The following options are available when @value{AS} is configured for the
1012 Specifies that the extended IP2022 instructions are allowed.
1015 Restores the default behaviour, which restricts the permitted instructions to
1016 just the basic IP2022 ones.
1022 The following options are available when @value{AS} is configured for the
1023 Renesas M32C and M16C processors.
1028 Assemble M32C instructions.
1031 Assemble M16C instructions (the default).
1034 Enable support for link-time relaxations.
1037 Support H'00 style hex constants in addition to 0x00 style.
1043 The following options are available when @value{AS} is configured for the
1044 Renesas M32R (formerly Mitsubishi M32R) series.
1049 Specify which processor in the M32R family is the target. The default
1050 is normally the M32R, but this option changes it to the M32RX.
1052 @item --warn-explicit-parallel-conflicts or --Wp
1053 Produce warning messages when questionable parallel constructs are
1056 @item --no-warn-explicit-parallel-conflicts or --Wnp
1057 Do not produce warning messages when questionable parallel constructs are
1064 The following options are available when @value{AS} is configured for the
1065 Motorola 68000 series.
1070 Shorten references to undefined symbols, to one word instead of two.
1072 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1073 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1074 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1075 Specify what processor in the 68000 family is the target. The default
1076 is normally the 68020, but this can be changed at configuration time.
1078 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1079 The target machine does (or does not) have a floating-point coprocessor.
1080 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1081 the basic 68000 is not compatible with the 68881, a combination of the
1082 two can be specified, since it's possible to do emulation of the
1083 coprocessor instructions with the main processor.
1085 @item -m68851 | -mno-68851
1086 The target machine does (or does not) have a memory-management
1087 unit coprocessor. The default is to assume an MMU for 68020 and up.
1095 @xref{Nios II Options}, for the options available when @value{AS} is configured
1096 for an Altera Nios II processor.
1100 @c man begin OPTIONS
1101 The following options are available when @value{AS} is configured for an
1102 Altera Nios II processor.
1104 @c man begin INCLUDE
1105 @include c-nios2.texi
1106 @c ended inside the included file
1112 For details about the PDP-11 machine dependent features options,
1113 see @ref{PDP-11-Options}.
1116 @item -mpic | -mno-pic
1117 Generate position-independent (or position-dependent) code. The
1118 default is @option{-mpic}.
1121 @itemx -mall-extensions
1122 Enable all instruction set extensions. This is the default.
1124 @item -mno-extensions
1125 Disable all instruction set extensions.
1127 @item -m@var{extension} | -mno-@var{extension}
1128 Enable (or disable) a particular instruction set extension.
1131 Enable the instruction set extensions supported by a particular CPU, and
1132 disable all other extensions.
1134 @item -m@var{machine}
1135 Enable the instruction set extensions supported by a particular machine
1136 model, and disable all other extensions.
1142 The following options are available when @value{AS} is configured for
1143 a picoJava processor.
1147 @cindex PJ endianness
1148 @cindex endianness, PJ
1149 @cindex big endian output, PJ
1151 Generate ``big endian'' format output.
1153 @cindex little endian output, PJ
1155 Generate ``little endian'' format output.
1161 The following options are available when @value{AS} is configured for the
1162 Motorola 68HC11 or 68HC12 series.
1166 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1167 Specify what processor is the target. The default is
1168 defined by the configuration option when building the assembler.
1170 @item --xgate-ramoffset
1171 Instruct the linker to offset RAM addresses from S12X address space into
1172 XGATE address space.
1175 Specify to use the 16-bit integer ABI.
1178 Specify to use the 32-bit integer ABI.
1180 @item -mshort-double
1181 Specify to use the 32-bit double ABI.
1184 Specify to use the 64-bit double ABI.
1186 @item --force-long-branches
1187 Relative branches are turned into absolute ones. This concerns
1188 conditional branches, unconditional branches and branches to a
1191 @item -S | --short-branches
1192 Do not turn relative branches into absolute ones
1193 when the offset is out of range.
1195 @item --strict-direct-mode
1196 Do not turn the direct addressing mode into extended addressing mode
1197 when the instruction does not support direct addressing mode.
1199 @item --print-insn-syntax
1200 Print the syntax of instruction in case of error.
1202 @item --print-opcodes
1203 Print the list of instructions with syntax and then exit.
1205 @item --generate-example
1206 Print an example of instruction for each possible instruction and then exit.
1207 This option is only useful for testing @command{@value{AS}}.
1213 The following options are available when @command{@value{AS}} is configured
1214 for the SPARC architecture:
1217 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1218 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1219 Explicitly select a variant of the SPARC architecture.
1221 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1222 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1224 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1225 UltraSPARC extensions.
1227 @item -xarch=v8plus | -xarch=v8plusa
1228 For compatibility with the Solaris v9 assembler. These options are
1229 equivalent to -Av8plus and -Av8plusa, respectively.
1232 Warn when the assembler switches to another architecture.
1237 The following options are available when @value{AS} is configured for the 'c54x
1242 Enable extended addressing mode. All addresses and relocations will assume
1243 extended addressing (usually 23 bits).
1244 @item -mcpu=@var{CPU_VERSION}
1245 Sets the CPU version being compiled for.
1246 @item -merrors-to-file @var{FILENAME}
1247 Redirect error output to a file, for broken systems which don't support such
1248 behaviour in the shell.
1253 The following options are available when @value{AS} is configured for
1258 This option sets the largest size of an object that can be referenced
1259 implicitly with the @code{gp} register. It is only accepted for targets that
1260 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1262 @cindex MIPS endianness
1263 @cindex endianness, MIPS
1264 @cindex big endian output, MIPS
1266 Generate ``big endian'' format output.
1268 @cindex little endian output, MIPS
1270 Generate ``little endian'' format output.
1288 Generate code for a particular MIPS Instruction Set Architecture level.
1289 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1290 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1291 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1292 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1293 @samp{-mips32r5}, @samp{-mips32r6}, @samp{-mips64}, @samp{-mips64r2},
1294 @samp{-mips64r3}, @samp{-mips64r5}, and @samp{-mips64r6} correspond to generic
1295 MIPS V, MIPS32, MIPS32 Release 2, MIPS32 Release 3, MIPS32 Release 5, MIPS32
1296 Release 6, MIPS64, MIPS64 Release 2, MIPS64 Release 3, MIPS64 Release 5, and
1297 MIPS64 Release 6 ISA processors, respectively.
1299 @item -march=@var{cpu}
1300 Generate code for a particular MIPS CPU.
1302 @item -mtune=@var{cpu}
1303 Schedule and tune for a particular MIPS CPU.
1307 Cause nops to be inserted if the read of the destination register
1308 of an mfhi or mflo instruction occurs in the following two instructions.
1311 @itemx -mno-fix-rm7000
1312 Cause nops to be inserted if a dmult or dmultu instruction is
1313 followed by a load instruction.
1317 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1318 section instead of the standard ELF .stabs sections.
1322 Control generation of @code{.pdr} sections.
1326 The register sizes are normally inferred from the ISA and ABI, but these
1327 flags force a certain group of registers to be treated as 32 bits wide at
1328 all times. @samp{-mgp32} controls the size of general-purpose registers
1329 and @samp{-mfp32} controls the size of floating-point registers.
1333 The register sizes are normally inferred from the ISA and ABI, but these
1334 flags force a certain group of registers to be treated as 64 bits wide at
1335 all times. @samp{-mgp64} controls the size of general-purpose registers
1336 and @samp{-mfp64} controls the size of floating-point registers.
1339 The register sizes are normally inferred from the ISA and ABI, but using
1340 this flag in combination with @samp{-mabi=32} enables an ABI variant
1341 which will operate correctly with floating-point registers which are
1345 @itemx -mno-odd-spreg
1346 Enable use of floating-point operations on odd-numbered single-precision
1347 registers when supported by the ISA. @samp{-mfpxx} implies
1348 @samp{-mno-odd-spreg}, otherwise the default is @samp{-modd-spreg}.
1352 Generate code for the MIPS 16 processor. This is equivalent to putting
1353 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1354 turns off this option.
1357 @itemx -mno-micromips
1358 Generate code for the microMIPS processor. This is equivalent to putting
1359 @code{.set micromips} at the start of the assembly file. @samp{-mno-micromips}
1360 turns off this option. This is equivalent to putting @code{.set nomicromips}
1361 at the start of the assembly file.
1364 @itemx -mno-smartmips
1365 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1366 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1367 @samp{-mno-smartmips} turns off this option.
1371 Generate code for the MIPS-3D Application Specific Extension.
1372 This tells the assembler to accept MIPS-3D instructions.
1373 @samp{-no-mips3d} turns off this option.
1377 Generate code for the MDMX Application Specific Extension.
1378 This tells the assembler to accept MDMX instructions.
1379 @samp{-no-mdmx} turns off this option.
1383 Generate code for the DSP Release 1 Application Specific Extension.
1384 This tells the assembler to accept DSP Release 1 instructions.
1385 @samp{-mno-dsp} turns off this option.
1389 Generate code for the DSP Release 2 Application Specific Extension.
1390 This option implies -mdsp.
1391 This tells the assembler to accept DSP Release 2 instructions.
1392 @samp{-mno-dspr2} turns off this option.
1396 Generate code for the MIPS SIMD Architecture Extension.
1397 This tells the assembler to accept MSA instructions.
1398 @samp{-mno-msa} turns off this option.
1402 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1403 This tells the assembler to accept XPA instructions.
1404 @samp{-mno-xpa} turns off this option.
1408 Generate code for the MT Application Specific Extension.
1409 This tells the assembler to accept MT instructions.
1410 @samp{-mno-mt} turns off this option.
1414 Generate code for the MCU Application Specific Extension.
1415 This tells the assembler to accept MCU instructions.
1416 @samp{-mno-mcu} turns off this option.
1420 Only use 32-bit instruction encodings when generating code for the
1421 microMIPS processor. This option inhibits the use of any 16-bit
1422 instructions. This is equivalent to putting @code{.set insn32} at
1423 the start of the assembly file. @samp{-mno-insn32} turns off this
1424 option. This is equivalent to putting @code{.set noinsn32} at the
1425 start of the assembly file. By default @samp{-mno-insn32} is
1426 selected, allowing all instructions to be used.
1428 @item --construct-floats
1429 @itemx --no-construct-floats
1430 The @samp{--no-construct-floats} option disables the construction of
1431 double width floating point constants by loading the two halves of the
1432 value into the two single width floating point registers that make up
1433 the double width register. By default @samp{--construct-floats} is
1434 selected, allowing construction of these floating point constants.
1436 @item --relax-branch
1437 @itemx --no-relax-branch
1438 The @samp{--relax-branch} option enables the relaxation of out-of-range
1439 branches. By default @samp{--no-relax-branch} is selected, causing any
1440 out-of-range branches to produce an error.
1442 @item -mnan=@var{encoding}
1443 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1444 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1447 @item --emulation=@var{name}
1448 This option was formerly used to switch between ELF and ECOFF output
1449 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1450 removed in GAS 2.24, so the option now serves little purpose.
1451 It is retained for backwards compatibility.
1453 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1454 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1455 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1456 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1457 preferred options instead.
1460 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1467 Control how to deal with multiplication overflow and division by zero.
1468 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1469 (and only work for Instruction Set Architecture level 2 and higher);
1470 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1474 When this option is used, @command{@value{AS}} will issue a warning every
1475 time it generates a nop instruction from a macro.
1480 The following options are available when @value{AS} is configured for
1486 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1487 The command line option @samp{-nojsri2bsr} can be used to disable it.
1491 Enable or disable the silicon filter behaviour. By default this is disabled.
1492 The default can be overridden by the @samp{-sifilter} command line option.
1495 Alter jump instructions for long displacements.
1497 @item -mcpu=[210|340]
1498 Select the cpu type on the target hardware. This controls which instructions
1502 Assemble for a big endian target.
1505 Assemble for a little endian target.
1514 @xref{Meta Options}, for the options available when @value{AS} is configured
1515 for a Meta processor.
1519 @c man begin OPTIONS
1520 The following options are available when @value{AS} is configured for a
1523 @c man begin INCLUDE
1524 @include c-metag.texi
1525 @c ended inside the included file
1530 @c man begin OPTIONS
1532 See the info pages for documentation of the MMIX-specific options.
1538 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1539 for a NDS32 processor.
1541 @c ended inside the included file
1545 @c man begin OPTIONS
1546 The following options are available when @value{AS} is configured for a
1549 @c man begin INCLUDE
1550 @include c-nds32.texi
1551 @c ended inside the included file
1558 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1559 for a PowerPC processor.
1563 @c man begin OPTIONS
1564 The following options are available when @value{AS} is configured for a
1567 @c man begin INCLUDE
1569 @c ended inside the included file
1574 @c man begin OPTIONS
1576 See the info pages for documentation of the RX-specific options.
1580 The following options are available when @value{AS} is configured for the s390
1586 Select the word size, either 31/32 bits or 64 bits.
1589 Select the architecture mode, either the Enterprise System
1590 Architecture (esa) or the z/Architecture mode (zarch).
1591 @item -march=@var{processor}
1592 Specify which s390 processor variant is the target, @samp{g6}, @samp{g6},
1593 @samp{z900}, @samp{z990}, @samp{z9-109}, @samp{z9-ec}, @samp{z10},
1594 @samp{z196}, @samp{zEC12}, or @samp{z13}.
1596 @itemx -mno-regnames
1597 Allow or disallow symbolic names for registers.
1598 @item -mwarn-areg-zero
1599 Warn whenever the operand for a base or index register has been specified
1600 but evaluates to zero.
1608 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1609 for a TMS320C6000 processor.
1613 @c man begin OPTIONS
1614 The following options are available when @value{AS} is configured for a
1615 TMS320C6000 processor.
1617 @c man begin INCLUDE
1618 @include c-tic6x.texi
1619 @c ended inside the included file
1627 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1628 for a TILE-Gx processor.
1632 @c man begin OPTIONS
1633 The following options are available when @value{AS} is configured for a TILE-Gx
1636 @c man begin INCLUDE
1637 @include c-tilegx.texi
1638 @c ended inside the included file
1646 @xref{Visium Options}, for the options available when @value{AS} is configured
1647 for a Visium processor.
1651 @c man begin OPTIONS
1652 The following option is available when @value{AS} is configured for a Visium
1655 @c man begin INCLUDE
1656 @include c-visium.texi
1657 @c ended inside the included file
1665 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1666 for an Xtensa processor.
1670 @c man begin OPTIONS
1671 The following options are available when @value{AS} is configured for an
1674 @c man begin INCLUDE
1675 @include c-xtensa.texi
1676 @c ended inside the included file
1681 @c man begin OPTIONS
1684 The following options are available when @value{AS} is configured for
1685 a Z80 family processor.
1688 Assemble for Z80 processor.
1690 Assemble for R800 processor.
1691 @item -ignore-undocumented-instructions
1693 Assemble undocumented Z80 instructions that also work on R800 without warning.
1694 @item -ignore-unportable-instructions
1696 Assemble all undocumented Z80 instructions without warning.
1697 @item -warn-undocumented-instructions
1699 Issue a warning for undocumented Z80 instructions that also work on R800.
1700 @item -warn-unportable-instructions
1702 Issue a warning for undocumented Z80 instructions that do not work on R800.
1703 @item -forbid-undocumented-instructions
1705 Treat all undocumented instructions as errors.
1706 @item -forbid-unportable-instructions
1708 Treat undocumented Z80 instructions that do not work on R800 as errors.
1715 * Manual:: Structure of this Manual
1716 * GNU Assembler:: The GNU Assembler
1717 * Object Formats:: Object File Formats
1718 * Command Line:: Command Line
1719 * Input Files:: Input Files
1720 * Object:: Output (Object) File
1721 * Errors:: Error and Warning Messages
1725 @section Structure of this Manual
1727 @cindex manual, structure and purpose
1728 This manual is intended to describe what you need to know to use
1729 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1730 notation for symbols, constants, and expressions; the directives that
1731 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1734 We also cover special features in the @value{TARGET}
1735 configuration of @command{@value{AS}}, including assembler directives.
1738 This manual also describes some of the machine-dependent features of
1739 various flavors of the assembler.
1742 @cindex machine instructions (not covered)
1743 On the other hand, this manual is @emph{not} intended as an introduction
1744 to programming in assembly language---let alone programming in general!
1745 In a similar vein, we make no attempt to introduce the machine
1746 architecture; we do @emph{not} describe the instruction set, standard
1747 mnemonics, registers or addressing modes that are standard to a
1748 particular architecture.
1750 You may want to consult the manufacturer's
1751 machine architecture manual for this information.
1755 For information on the H8/300 machine instruction set, see @cite{H8/300
1756 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1757 Programming Manual} (Renesas).
1760 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1761 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1762 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1763 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1766 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1770 @c I think this is premature---doc@cygnus.com, 17jan1991
1772 Throughout this manual, we assume that you are running @dfn{GNU},
1773 the portable operating system from the @dfn{Free Software
1774 Foundation, Inc.}. This restricts our attention to certain kinds of
1775 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1776 once this assumption is granted examples and definitions need less
1779 @command{@value{AS}} is part of a team of programs that turn a high-level
1780 human-readable series of instructions into a low-level
1781 computer-readable series of instructions. Different versions of
1782 @command{@value{AS}} are used for different kinds of computer.
1785 @c There used to be a section "Terminology" here, which defined
1786 @c "contents", "byte", "word", and "long". Defining "word" to any
1787 @c particular size is confusing when the .word directive may generate 16
1788 @c bits on one machine and 32 bits on another; in general, for the user
1789 @c version of this manual, none of these terms seem essential to define.
1790 @c They were used very little even in the former draft of the manual;
1791 @c this draft makes an effort to avoid them (except in names of
1795 @section The GNU Assembler
1797 @c man begin DESCRIPTION
1799 @sc{gnu} @command{as} is really a family of assemblers.
1801 This manual describes @command{@value{AS}}, a member of that family which is
1802 configured for the @value{TARGET} architectures.
1804 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1805 should find a fairly similar environment when you use it on another
1806 architecture. Each version has much in common with the others,
1807 including object file formats, most assembler directives (often called
1808 @dfn{pseudo-ops}) and assembler syntax.@refill
1810 @cindex purpose of @sc{gnu} assembler
1811 @command{@value{AS}} is primarily intended to assemble the output of the
1812 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1813 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1814 assemble correctly everything that other assemblers for the same
1815 machine would assemble.
1817 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1820 @c This remark should appear in generic version of manual; assumption
1821 @c here is that generic version sets M680x0.
1822 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1823 assembler for the same architecture; for example, we know of several
1824 incompatible versions of 680x0 assembly language syntax.
1829 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1830 program in one pass of the source file. This has a subtle impact on the
1831 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1833 @node Object Formats
1834 @section Object File Formats
1836 @cindex object file format
1837 The @sc{gnu} assembler can be configured to produce several alternative
1838 object file formats. For the most part, this does not affect how you
1839 write assembly language programs; but directives for debugging symbols
1840 are typically different in different file formats. @xref{Symbol
1841 Attributes,,Symbol Attributes}.
1844 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1845 @value{OBJ-NAME} format object files.
1847 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1849 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1850 @code{b.out} or COFF format object files.
1853 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1854 SOM or ELF format object files.
1859 @section Command Line
1861 @cindex command line conventions
1863 After the program name @command{@value{AS}}, the command line may contain
1864 options and file names. Options may appear in any order, and may be
1865 before, after, or between file names. The order of file names is
1868 @cindex standard input, as input file
1870 @file{--} (two hyphens) by itself names the standard input file
1871 explicitly, as one of the files for @command{@value{AS}} to assemble.
1873 @cindex options, command line
1874 Except for @samp{--} any command line argument that begins with a
1875 hyphen (@samp{-}) is an option. Each option changes the behavior of
1876 @command{@value{AS}}. No option changes the way another option works. An
1877 option is a @samp{-} followed by one or more letters; the case of
1878 the letter is important. All options are optional.
1880 Some options expect exactly one file name to follow them. The file
1881 name may either immediately follow the option's letter (compatible
1882 with older assemblers) or it may be the next command argument (@sc{gnu}
1883 standard). These two command lines are equivalent:
1886 @value{AS} -o my-object-file.o mumble.s
1887 @value{AS} -omy-object-file.o mumble.s
1891 @section Input Files
1894 @cindex source program
1895 @cindex files, input
1896 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1897 describe the program input to one run of @command{@value{AS}}. The program may
1898 be in one or more files; how the source is partitioned into files
1899 doesn't change the meaning of the source.
1901 @c I added "con" prefix to "catenation" just to prove I can overcome my
1902 @c APL training... doc@cygnus.com
1903 The source program is a concatenation of the text in all the files, in the
1906 @c man begin DESCRIPTION
1907 Each time you run @command{@value{AS}} it assembles exactly one source
1908 program. The source program is made up of one or more files.
1909 (The standard input is also a file.)
1911 You give @command{@value{AS}} a command line that has zero or more input file
1912 names. The input files are read (from left file name to right). A
1913 command line argument (in any position) that has no special meaning
1914 is taken to be an input file name.
1916 If you give @command{@value{AS}} no file names it attempts to read one input file
1917 from the @command{@value{AS}} standard input, which is normally your terminal. You
1918 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1921 Use @samp{--} if you need to explicitly name the standard input file
1922 in your command line.
1924 If the source is empty, @command{@value{AS}} produces a small, empty object
1929 @subheading Filenames and Line-numbers
1931 @cindex input file linenumbers
1932 @cindex line numbers, in input files
1933 There are two ways of locating a line in the input file (or files) and
1934 either may be used in reporting error messages. One way refers to a line
1935 number in a physical file; the other refers to a line number in a
1936 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1938 @dfn{Physical files} are those files named in the command line given
1939 to @command{@value{AS}}.
1941 @dfn{Logical files} are simply names declared explicitly by assembler
1942 directives; they bear no relation to physical files. Logical file names help
1943 error messages reflect the original source file, when @command{@value{AS}} source
1944 is itself synthesized from other files. @command{@value{AS}} understands the
1945 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1946 @ref{File,,@code{.file}}.
1949 @section Output (Object) File
1955 Every time you run @command{@value{AS}} it produces an output file, which is
1956 your assembly language program translated into numbers. This file
1957 is the object file. Its default name is
1965 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1967 You can give it another name by using the @option{-o} option. Conventionally,
1968 object file names end with @file{.o}. The default name is used for historical
1969 reasons: older assemblers were capable of assembling self-contained programs
1970 directly into a runnable program. (For some formats, this isn't currently
1971 possible, but it can be done for the @code{a.out} format.)
1975 The object file is meant for input to the linker @code{@value{LD}}. It contains
1976 assembled program code, information to help @code{@value{LD}} integrate
1977 the assembled program into a runnable file, and (optionally) symbolic
1978 information for the debugger.
1980 @c link above to some info file(s) like the description of a.out.
1981 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1984 @section Error and Warning Messages
1986 @c man begin DESCRIPTION
1988 @cindex error messages
1989 @cindex warning messages
1990 @cindex messages from assembler
1991 @command{@value{AS}} may write warnings and error messages to the standard error
1992 file (usually your terminal). This should not happen when a compiler
1993 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1994 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1995 grave problem that stops the assembly.
1999 @cindex format of warning messages
2000 Warning messages have the format
2003 file_name:@b{NNN}:Warning Message Text
2007 @cindex line numbers, in warnings/errors
2008 (where @b{NNN} is a line number). If a logical file name has been given
2009 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
2010 the current input file is used. If a logical line number was given
2012 (@pxref{Line,,@code{.line}})
2014 then it is used to calculate the number printed,
2015 otherwise the actual line in the current source file is printed. The
2016 message text is intended to be self explanatory (in the grand Unix
2019 @cindex format of error messages
2020 Error messages have the format
2022 file_name:@b{NNN}:FATAL:Error Message Text
2024 The file name and line number are derived as for warning
2025 messages. The actual message text may be rather less explanatory
2026 because many of them aren't supposed to happen.
2029 @chapter Command-Line Options
2031 @cindex options, all versions of assembler
2032 This chapter describes command-line options available in @emph{all}
2033 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
2034 for options specific
2036 to the @value{TARGET} target.
2039 to particular machine architectures.
2042 @c man begin DESCRIPTION
2044 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2045 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2046 The assembler arguments must be separated from each other (and the @samp{-Wa})
2047 by commas. For example:
2050 gcc -c -g -O -Wa,-alh,-L file.c
2054 This passes two options to the assembler: @samp{-alh} (emit a listing to
2055 standard output with high-level and assembly source) and @samp{-L} (retain
2056 local symbols in the symbol table).
2058 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2059 command-line options are automatically passed to the assembler by the compiler.
2060 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2061 precisely what options it passes to each compilation pass, including the
2067 * a:: -a[cdghlns] enable listings
2068 * alternate:: --alternate enable alternate macro syntax
2069 * D:: -D for compatibility
2070 * f:: -f to work faster
2071 * I:: -I for .include search path
2072 @ifclear DIFF-TBL-KLUGE
2073 * K:: -K for compatibility
2075 @ifset DIFF-TBL-KLUGE
2076 * K:: -K for difference tables
2079 * L:: -L to retain local symbols
2080 * listing:: --listing-XXX to configure listing output
2081 * M:: -M or --mri to assemble in MRI compatibility mode
2082 * MD:: --MD for dependency tracking
2083 * o:: -o to name the object file
2084 * R:: -R to join data and text sections
2085 * statistics:: --statistics to see statistics about assembly
2086 * traditional-format:: --traditional-format for compatible output
2087 * v:: -v to announce version
2088 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2089 * Z:: -Z to make object file even after errors
2093 @section Enable Listings: @option{-a[cdghlns]}
2103 @cindex listings, enabling
2104 @cindex assembly listings, enabling
2106 These options enable listing output from the assembler. By itself,
2107 @samp{-a} requests high-level, assembly, and symbols listing.
2108 You can use other letters to select specific options for the list:
2109 @samp{-ah} requests a high-level language listing,
2110 @samp{-al} requests an output-program assembly listing, and
2111 @samp{-as} requests a symbol table listing.
2112 High-level listings require that a compiler debugging option like
2113 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2116 Use the @samp{-ag} option to print a first section with general assembly
2117 information, like @value{AS} version, switches passed, or time stamp.
2119 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2120 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2121 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2122 omitted from the listing.
2124 Use the @samp{-ad} option to omit debugging directives from the
2127 Once you have specified one of these options, you can further control
2128 listing output and its appearance using the directives @code{.list},
2129 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2131 The @samp{-an} option turns off all forms processing.
2132 If you do not request listing output with one of the @samp{-a} options, the
2133 listing-control directives have no effect.
2135 The letters after @samp{-a} may be combined into one option,
2136 @emph{e.g.}, @samp{-aln}.
2138 Note if the assembler source is coming from the standard input (e.g.,
2140 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
2141 is being used) then the listing will not contain any comments or preprocessor
2142 directives. This is because the listing code buffers input source lines from
2143 stdin only after they have been preprocessed by the assembler. This reduces
2144 memory usage and makes the code more efficient.
2147 @section @option{--alternate}
2150 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2153 @section @option{-D}
2156 This option has no effect whatsoever, but it is accepted to make it more
2157 likely that scripts written for other assemblers also work with
2158 @command{@value{AS}}.
2161 @section Work Faster: @option{-f}
2164 @cindex trusted compiler
2165 @cindex faster processing (@option{-f})
2166 @samp{-f} should only be used when assembling programs written by a
2167 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2168 and comment preprocessing on
2169 the input file(s) before assembling them. @xref{Preprocessing,
2173 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2174 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2179 @section @code{.include} Search Path: @option{-I} @var{path}
2181 @kindex -I @var{path}
2182 @cindex paths for @code{.include}
2183 @cindex search path for @code{.include}
2184 @cindex @code{include} directive search path
2185 Use this option to add a @var{path} to the list of directories
2186 @command{@value{AS}} searches for files specified in @code{.include}
2187 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2188 many times as necessary to include a variety of paths. The current
2189 working directory is always searched first; after that, @command{@value{AS}}
2190 searches any @samp{-I} directories in the same order as they were
2191 specified (left to right) on the command line.
2194 @section Difference Tables: @option{-K}
2197 @ifclear DIFF-TBL-KLUGE
2198 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2199 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2200 where it can be used to warn when the assembler alters the machine code
2201 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2202 family does not have the addressing limitations that sometimes lead to this
2203 alteration on other platforms.
2206 @ifset DIFF-TBL-KLUGE
2207 @cindex difference tables, warning
2208 @cindex warning for altered difference tables
2209 @command{@value{AS}} sometimes alters the code emitted for directives of the
2210 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2211 You can use the @samp{-K} option if you want a warning issued when this
2216 @section Include Local Symbols: @option{-L}
2219 @cindex local symbols, retaining in output
2220 Symbols beginning with system-specific local label prefixes, typically
2221 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2222 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2223 such symbols when debugging, because they are intended for the use of
2224 programs (like compilers) that compose assembler programs, not for your
2225 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2226 such symbols, so you do not normally debug with them.
2228 This option tells @command{@value{AS}} to retain those local symbols
2229 in the object file. Usually if you do this you also tell the linker
2230 @code{@value{LD}} to preserve those symbols.
2233 @section Configuring listing output: @option{--listing}
2235 The listing feature of the assembler can be enabled via the command line switch
2236 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2237 hex dump of the corresponding locations in the output object file, and displays
2238 them as a listing file. The format of this listing can be controlled by
2239 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2240 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2241 @code{.psize} (@pxref{Psize}), and
2242 @code{.eject} (@pxref{Eject}) and also by the following switches:
2245 @item --listing-lhs-width=@samp{number}
2246 @kindex --listing-lhs-width
2247 @cindex Width of first line disassembly output
2248 Sets the maximum width, in words, of the first line of the hex byte dump. This
2249 dump appears on the left hand side of the listing output.
2251 @item --listing-lhs-width2=@samp{number}
2252 @kindex --listing-lhs-width2
2253 @cindex Width of continuation lines of disassembly output
2254 Sets the maximum width, in words, of any further lines of the hex byte dump for
2255 a given input source line. If this value is not specified, it defaults to being
2256 the same as the value specified for @samp{--listing-lhs-width}. If neither
2257 switch is used the default is to one.
2259 @item --listing-rhs-width=@samp{number}
2260 @kindex --listing-rhs-width
2261 @cindex Width of source line output
2262 Sets the maximum width, in characters, of the source line that is displayed
2263 alongside the hex dump. The default value for this parameter is 100. The
2264 source line is displayed on the right hand side of the listing output.
2266 @item --listing-cont-lines=@samp{number}
2267 @kindex --listing-cont-lines
2268 @cindex Maximum number of continuation lines
2269 Sets the maximum number of continuation lines of hex dump that will be
2270 displayed for a given single line of source input. The default value is 4.
2274 @section Assemble in MRI Compatibility Mode: @option{-M}
2277 @cindex MRI compatibility mode
2278 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2279 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2280 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
2281 configured target) assembler from Microtec Research. The exact nature of the
2282 MRI syntax will not be documented here; see the MRI manuals for more
2283 information. Note in particular that the handling of macros and macro
2284 arguments is somewhat different. The purpose of this option is to permit
2285 assembling existing MRI assembler code using @command{@value{AS}}.
2287 The MRI compatibility is not complete. Certain operations of the MRI assembler
2288 depend upon its object file format, and can not be supported using other object
2289 file formats. Supporting these would require enhancing each object file format
2290 individually. These are:
2293 @item global symbols in common section
2295 The m68k MRI assembler supports common sections which are merged by the linker.
2296 Other object file formats do not support this. @command{@value{AS}} handles
2297 common sections by treating them as a single common symbol. It permits local
2298 symbols to be defined within a common section, but it can not support global
2299 symbols, since it has no way to describe them.
2301 @item complex relocations
2303 The MRI assemblers support relocations against a negated section address, and
2304 relocations which combine the start addresses of two or more sections. These
2305 are not support by other object file formats.
2307 @item @code{END} pseudo-op specifying start address
2309 The MRI @code{END} pseudo-op permits the specification of a start address.
2310 This is not supported by other object file formats. The start address may
2311 instead be specified using the @option{-e} option to the linker, or in a linker
2314 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2316 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2317 name to the output file. This is not supported by other object file formats.
2319 @item @code{ORG} pseudo-op
2321 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2322 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2323 which changes the location within the current section. Absolute sections are
2324 not supported by other object file formats. The address of a section may be
2325 assigned within a linker script.
2328 There are some other features of the MRI assembler which are not supported by
2329 @command{@value{AS}}, typically either because they are difficult or because they
2330 seem of little consequence. Some of these may be supported in future releases.
2334 @item EBCDIC strings
2336 EBCDIC strings are not supported.
2338 @item packed binary coded decimal
2340 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2341 and @code{DCB.P} pseudo-ops are not supported.
2343 @item @code{FEQU} pseudo-op
2345 The m68k @code{FEQU} pseudo-op is not supported.
2347 @item @code{NOOBJ} pseudo-op
2349 The m68k @code{NOOBJ} pseudo-op is not supported.
2351 @item @code{OPT} branch control options
2353 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2354 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2355 relaxes all branches, whether forward or backward, to an appropriate size, so
2356 these options serve no purpose.
2358 @item @code{OPT} list control options
2360 The following m68k @code{OPT} list control options are ignored: @code{C},
2361 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2362 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2364 @item other @code{OPT} options
2366 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2367 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2369 @item @code{OPT} @code{D} option is default
2371 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2372 @code{OPT NOD} may be used to turn it off.
2374 @item @code{XREF} pseudo-op.
2376 The m68k @code{XREF} pseudo-op is ignored.
2378 @item @code{.debug} pseudo-op
2380 The i960 @code{.debug} pseudo-op is not supported.
2382 @item @code{.extended} pseudo-op
2384 The i960 @code{.extended} pseudo-op is not supported.
2386 @item @code{.list} pseudo-op.
2388 The various options of the i960 @code{.list} pseudo-op are not supported.
2390 @item @code{.optimize} pseudo-op
2392 The i960 @code{.optimize} pseudo-op is not supported.
2394 @item @code{.output} pseudo-op
2396 The i960 @code{.output} pseudo-op is not supported.
2398 @item @code{.setreal} pseudo-op
2400 The i960 @code{.setreal} pseudo-op is not supported.
2405 @section Dependency Tracking: @option{--MD}
2408 @cindex dependency tracking
2411 @command{@value{AS}} can generate a dependency file for the file it creates. This
2412 file consists of a single rule suitable for @code{make} describing the
2413 dependencies of the main source file.
2415 The rule is written to the file named in its argument.
2417 This feature is used in the automatic updating of makefiles.
2420 @section Name the Object File: @option{-o}
2423 @cindex naming object file
2424 @cindex object file name
2425 There is always one object file output when you run @command{@value{AS}}. By
2426 default it has the name
2429 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2443 You use this option (which takes exactly one filename) to give the
2444 object file a different name.
2446 Whatever the object file is called, @command{@value{AS}} overwrites any
2447 existing file of the same name.
2450 @section Join Data and Text Sections: @option{-R}
2453 @cindex data and text sections, joining
2454 @cindex text and data sections, joining
2455 @cindex joining text and data sections
2456 @cindex merging text and data sections
2457 @option{-R} tells @command{@value{AS}} to write the object file as if all
2458 data-section data lives in the text section. This is only done at
2459 the very last moment: your binary data are the same, but data
2460 section parts are relocated differently. The data section part of
2461 your object file is zero bytes long because all its bytes are
2462 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2464 When you specify @option{-R} it would be possible to generate shorter
2465 address displacements (because we do not have to cross between text and
2466 data section). We refrain from doing this simply for compatibility with
2467 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2470 When @command{@value{AS}} is configured for COFF or ELF output,
2471 this option is only useful if you use sections named @samp{.text} and
2476 @option{-R} is not supported for any of the HPPA targets. Using
2477 @option{-R} generates a warning from @command{@value{AS}}.
2481 @section Display Assembly Statistics: @option{--statistics}
2483 @kindex --statistics
2484 @cindex statistics, about assembly
2485 @cindex time, total for assembly
2486 @cindex space used, maximum for assembly
2487 Use @samp{--statistics} to display two statistics about the resources used by
2488 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2489 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2492 @node traditional-format
2493 @section Compatible Output: @option{--traditional-format}
2495 @kindex --traditional-format
2496 For some targets, the output of @command{@value{AS}} is different in some ways
2497 from the output of some existing assembler. This switch requests
2498 @command{@value{AS}} to use the traditional format instead.
2500 For example, it disables the exception frame optimizations which
2501 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2504 @section Announce Version: @option{-v}
2508 @cindex assembler version
2509 @cindex version of assembler
2510 You can find out what version of as is running by including the
2511 option @samp{-v} (which you can also spell as @samp{-version}) on the
2515 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2517 @command{@value{AS}} should never give a warning or error message when
2518 assembling compiler output. But programs written by people often
2519 cause @command{@value{AS}} to give a warning that a particular assumption was
2520 made. All such warnings are directed to the standard error file.
2524 @cindex suppressing warnings
2525 @cindex warnings, suppressing
2526 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2527 This only affects the warning messages: it does not change any particular of
2528 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2531 @kindex --fatal-warnings
2532 @cindex errors, caused by warnings
2533 @cindex warnings, causing error
2534 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2535 files that generate warnings to be in error.
2538 @cindex warnings, switching on
2539 You can switch these options off again by specifying @option{--warn}, which
2540 causes warnings to be output as usual.
2543 @section Generate Object File in Spite of Errors: @option{-Z}
2544 @cindex object file, after errors
2545 @cindex errors, continuing after
2546 After an error message, @command{@value{AS}} normally produces no output. If for
2547 some reason you are interested in object file output even after
2548 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2549 option. If there are any errors, @command{@value{AS}} continues anyways, and
2550 writes an object file after a final warning message of the form @samp{@var{n}
2551 errors, @var{m} warnings, generating bad object file.}
2556 @cindex machine-independent syntax
2557 @cindex syntax, machine-independent
2558 This chapter describes the machine-independent syntax allowed in a
2559 source file. @command{@value{AS}} syntax is similar to what many other
2560 assemblers use; it is inspired by the BSD 4.2
2565 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2569 * Preprocessing:: Preprocessing
2570 * Whitespace:: Whitespace
2571 * Comments:: Comments
2572 * Symbol Intro:: Symbols
2573 * Statements:: Statements
2574 * Constants:: Constants
2578 @section Preprocessing
2580 @cindex preprocessing
2581 The @command{@value{AS}} internal preprocessor:
2583 @cindex whitespace, removed by preprocessor
2585 adjusts and removes extra whitespace. It leaves one space or tab before
2586 the keywords on a line, and turns any other whitespace on the line into
2589 @cindex comments, removed by preprocessor
2591 removes all comments, replacing them with a single space, or an
2592 appropriate number of newlines.
2594 @cindex constants, converted by preprocessor
2596 converts character constants into the appropriate numeric values.
2599 It does not do macro processing, include file handling, or
2600 anything else you may get from your C compiler's preprocessor. You can
2601 do include file processing with the @code{.include} directive
2602 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2603 to get other ``CPP'' style preprocessing by giving the input file a
2604 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2605 Output, gcc.info, Using GNU CC}.
2607 Excess whitespace, comments, and character constants
2608 cannot be used in the portions of the input text that are not
2611 @cindex turning preprocessing on and off
2612 @cindex preprocessing, turning on and off
2615 If the first line of an input file is @code{#NO_APP} or if you use the
2616 @samp{-f} option, whitespace and comments are not removed from the input file.
2617 Within an input file, you can ask for whitespace and comment removal in
2618 specific portions of the by putting a line that says @code{#APP} before the
2619 text that may contain whitespace or comments, and putting a line that says
2620 @code{#NO_APP} after this text. This feature is mainly intend to support
2621 @code{asm} statements in compilers whose output is otherwise free of comments
2628 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2629 Whitespace is used to separate symbols, and to make programs neater for
2630 people to read. Unless within character constants
2631 (@pxref{Characters,,Character Constants}), any whitespace means the same
2632 as exactly one space.
2638 There are two ways of rendering comments to @command{@value{AS}}. In both
2639 cases the comment is equivalent to one space.
2641 Anything from @samp{/*} through the next @samp{*/} is a comment.
2642 This means you may not nest these comments.
2646 The only way to include a newline ('\n') in a comment
2647 is to use this sort of comment.
2650 /* This sort of comment does not nest. */
2653 @cindex line comment character
2654 Anything from a @dfn{line comment} character up to the next newline is
2655 considered a comment and is ignored. The line comment character is target
2656 specific, and some targets multiple comment characters. Some targets also have
2657 line comment characters that only work if they are the first character on a
2658 line. Some targets use a sequence of two characters to introduce a line
2659 comment. Some targets can also change their line comment characters depending
2660 upon command line options that have been used. For more details see the
2661 @emph{Syntax} section in the documentation for individual targets.
2663 If the line comment character is the hash sign (@samp{#}) then it still has the
2664 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2665 to specify logical line numbers:
2668 @cindex lines starting with @code{#}
2669 @cindex logical line numbers
2670 To be compatible with past assemblers, lines that begin with @samp{#} have a
2671 special interpretation. Following the @samp{#} should be an absolute
2672 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2673 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2674 new logical file name. The rest of the line, if any, should be whitespace.
2676 If the first non-whitespace characters on the line are not numeric,
2677 the line is ignored. (Just like a comment.)
2680 # This is an ordinary comment.
2681 # 42-6 "new_file_name" # New logical file name
2682 # This is logical line # 36.
2684 This feature is deprecated, and may disappear from future versions
2685 of @command{@value{AS}}.
2690 @cindex characters used in symbols
2691 @ifclear SPECIAL-SYMS
2692 A @dfn{symbol} is one or more characters chosen from the set of all
2693 letters (both upper and lower case), digits and the three characters
2699 A @dfn{symbol} is one or more characters chosen from the set of all
2700 letters (both upper and lower case), digits and the three characters
2701 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2707 On most machines, you can also use @code{$} in symbol names; exceptions
2708 are noted in @ref{Machine Dependencies}.
2710 No symbol may begin with a digit. Case is significant.
2711 There is no length limit: all characters are significant. Multibyte characters
2712 are supported. Symbols are delimited by characters not in that set, or by the
2713 beginning of a file (since the source program must end with a newline, the end
2714 of a file is not a possible symbol delimiter). @xref{Symbols}.
2715 @cindex length of symbols
2720 @cindex statements, structure of
2721 @cindex line separator character
2722 @cindex statement separator character
2724 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2725 @dfn{line separator character}. The line separator character is target
2726 specific and described in the @emph{Syntax} section of each
2727 target's documentation. Not all targets support a line separator character.
2728 The newline or line separator character is considered to be part of the
2729 preceding statement. Newlines and separators within character constants are an
2730 exception: they do not end statements.
2732 @cindex newline, required at file end
2733 @cindex EOF, newline must precede
2734 It is an error to end any statement with end-of-file: the last
2735 character of any input file should be a newline.@refill
2737 An empty statement is allowed, and may include whitespace. It is ignored.
2739 @cindex instructions and directives
2740 @cindex directives and instructions
2741 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2742 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2744 A statement begins with zero or more labels, optionally followed by a
2745 key symbol which determines what kind of statement it is. The key
2746 symbol determines the syntax of the rest of the statement. If the
2747 symbol begins with a dot @samp{.} then the statement is an assembler
2748 directive: typically valid for any computer. If the symbol begins with
2749 a letter the statement is an assembly language @dfn{instruction}: it
2750 assembles into a machine language instruction.
2752 Different versions of @command{@value{AS}} for different computers
2753 recognize different instructions. In fact, the same symbol may
2754 represent a different instruction in a different computer's assembly
2758 @cindex @code{:} (label)
2759 @cindex label (@code{:})
2760 A label is a symbol immediately followed by a colon (@code{:}).
2761 Whitespace before a label or after a colon is permitted, but you may not
2762 have whitespace between a label's symbol and its colon. @xref{Labels}.
2765 For HPPA targets, labels need not be immediately followed by a colon, but
2766 the definition of a label must begin in column zero. This also implies that
2767 only one label may be defined on each line.
2771 label: .directive followed by something
2772 another_label: # This is an empty statement.
2773 instruction operand_1, operand_2, @dots{}
2780 A constant is a number, written so that its value is known by
2781 inspection, without knowing any context. Like this:
2784 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2785 .ascii "Ring the bell\7" # A string constant.
2786 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2787 .float 0f-314159265358979323846264338327\
2788 95028841971.693993751E-40 # - pi, a flonum.
2793 * Characters:: Character Constants
2794 * Numbers:: Number Constants
2798 @subsection Character Constants
2800 @cindex character constants
2801 @cindex constants, character
2802 There are two kinds of character constants. A @dfn{character} stands
2803 for one character in one byte and its value may be used in
2804 numeric expressions. String constants (properly called string
2805 @emph{literals}) are potentially many bytes and their values may not be
2806 used in arithmetic expressions.
2810 * Chars:: Characters
2814 @subsubsection Strings
2816 @cindex string constants
2817 @cindex constants, string
2818 A @dfn{string} is written between double-quotes. It may contain
2819 double-quotes or null characters. The way to get special characters
2820 into a string is to @dfn{escape} these characters: precede them with
2821 a backslash @samp{\} character. For example @samp{\\} represents
2822 one backslash: the first @code{\} is an escape which tells
2823 @command{@value{AS}} to interpret the second character literally as a backslash
2824 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2825 escape character). The complete list of escapes follows.
2827 @cindex escape codes, character
2828 @cindex character escape codes
2831 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2833 @cindex @code{\b} (backspace character)
2834 @cindex backspace (@code{\b})
2836 Mnemonic for backspace; for ASCII this is octal code 010.
2839 @c Mnemonic for EOText; for ASCII this is octal code 004.
2841 @cindex @code{\f} (formfeed character)
2842 @cindex formfeed (@code{\f})
2844 Mnemonic for FormFeed; for ASCII this is octal code 014.
2846 @cindex @code{\n} (newline character)
2847 @cindex newline (@code{\n})
2849 Mnemonic for newline; for ASCII this is octal code 012.
2852 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2854 @cindex @code{\r} (carriage return character)
2855 @cindex carriage return (@code{\r})
2857 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2860 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2861 @c other assemblers.
2863 @cindex @code{\t} (tab)
2864 @cindex tab (@code{\t})
2866 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2869 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2870 @c @item \x @var{digit} @var{digit} @var{digit}
2871 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2873 @cindex @code{\@var{ddd}} (octal character code)
2874 @cindex octal character code (@code{\@var{ddd}})
2875 @item \ @var{digit} @var{digit} @var{digit}
2876 An octal character code. The numeric code is 3 octal digits.
2877 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2878 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2880 @cindex @code{\@var{xd...}} (hex character code)
2881 @cindex hex character code (@code{\@var{xd...}})
2882 @item \@code{x} @var{hex-digits...}
2883 A hex character code. All trailing hex digits are combined. Either upper or
2884 lower case @code{x} works.
2886 @cindex @code{\\} (@samp{\} character)
2887 @cindex backslash (@code{\\})
2889 Represents one @samp{\} character.
2892 @c Represents one @samp{'} (accent acute) character.
2893 @c This is needed in single character literals
2894 @c (@xref{Characters,,Character Constants}.) to represent
2897 @cindex @code{\"} (doublequote character)
2898 @cindex doublequote (@code{\"})
2900 Represents one @samp{"} character. Needed in strings to represent
2901 this character, because an unescaped @samp{"} would end the string.
2903 @item \ @var{anything-else}
2904 Any other character when escaped by @kbd{\} gives a warning, but
2905 assembles as if the @samp{\} was not present. The idea is that if
2906 you used an escape sequence you clearly didn't want the literal
2907 interpretation of the following character. However @command{@value{AS}} has no
2908 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2909 code and warns you of the fact.
2912 Which characters are escapable, and what those escapes represent,
2913 varies widely among assemblers. The current set is what we think
2914 the BSD 4.2 assembler recognizes, and is a subset of what most C
2915 compilers recognize. If you are in doubt, do not use an escape
2919 @subsubsection Characters
2921 @cindex single character constant
2922 @cindex character, single
2923 @cindex constant, single character
2924 A single character may be written as a single quote immediately
2925 followed by that character. The same escapes apply to characters as
2926 to strings. So if you want to write the character backslash, you
2927 must write @kbd{'\\} where the first @code{\} escapes the second
2928 @code{\}. As you can see, the quote is an acute accent, not a
2929 grave accent. A newline
2931 @ifclear abnormal-separator
2932 (or semicolon @samp{;})
2934 @ifset abnormal-separator
2936 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2941 immediately following an acute accent is taken as a literal character
2942 and does not count as the end of a statement. The value of a character
2943 constant in a numeric expression is the machine's byte-wide code for
2944 that character. @command{@value{AS}} assumes your character code is ASCII:
2945 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2948 @subsection Number Constants
2950 @cindex constants, number
2951 @cindex number constants
2952 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2953 are stored in the target machine. @emph{Integers} are numbers that
2954 would fit into an @code{int} in the C language. @emph{Bignums} are
2955 integers, but they are stored in more than 32 bits. @emph{Flonums}
2956 are floating point numbers, described below.
2959 * Integers:: Integers
2964 * Bit Fields:: Bit Fields
2970 @subsubsection Integers
2972 @cindex constants, integer
2974 @cindex binary integers
2975 @cindex integers, binary
2976 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2977 the binary digits @samp{01}.
2979 @cindex octal integers
2980 @cindex integers, octal
2981 An octal integer is @samp{0} followed by zero or more of the octal
2982 digits (@samp{01234567}).
2984 @cindex decimal integers
2985 @cindex integers, decimal
2986 A decimal integer starts with a non-zero digit followed by zero or
2987 more digits (@samp{0123456789}).
2989 @cindex hexadecimal integers
2990 @cindex integers, hexadecimal
2991 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2992 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2994 Integers have the usual values. To denote a negative integer, use
2995 the prefix operator @samp{-} discussed under expressions
2996 (@pxref{Prefix Ops,,Prefix Operators}).
2999 @subsubsection Bignums
3002 @cindex constants, bignum
3003 A @dfn{bignum} has the same syntax and semantics as an integer
3004 except that the number (or its negative) takes more than 32 bits to
3005 represent in binary. The distinction is made because in some places
3006 integers are permitted while bignums are not.
3009 @subsubsection Flonums
3011 @cindex floating point numbers
3012 @cindex constants, floating point
3014 @cindex precision, floating point
3015 A @dfn{flonum} represents a floating point number. The translation is
3016 indirect: a decimal floating point number from the text is converted by
3017 @command{@value{AS}} to a generic binary floating point number of more than
3018 sufficient precision. This generic floating point number is converted
3019 to a particular computer's floating point format (or formats) by a
3020 portion of @command{@value{AS}} specialized to that computer.
3022 A flonum is written by writing (in order)
3027 (@samp{0} is optional on the HPPA.)
3031 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
3033 @kbd{e} is recommended. Case is not important.
3035 @c FIXME: verify if flonum syntax really this vague for most cases
3036 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
3037 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
3040 On the H8/300, Renesas / SuperH SH,
3041 and AMD 29K architectures, the letter must be
3042 one of the letters @samp{DFPRSX} (in upper or lower case).
3044 On the ARC, the letter must be one of the letters @samp{DFRS}
3045 (in upper or lower case).
3047 On the Intel 960 architecture, the letter must be
3048 one of the letters @samp{DFT} (in upper or lower case).
3050 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3054 One of the letters @samp{DFRS} (in upper or lower case).
3057 One of the letters @samp{DFPRSX} (in upper or lower case).
3060 The letter @samp{E} (upper case only).
3063 One of the letters @samp{DFT} (in upper or lower case).
3068 An optional sign: either @samp{+} or @samp{-}.
3071 An optional @dfn{integer part}: zero or more decimal digits.
3074 An optional @dfn{fractional part}: @samp{.} followed by zero
3075 or more decimal digits.
3078 An optional exponent, consisting of:
3082 An @samp{E} or @samp{e}.
3083 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3084 @c principle this can perfectly well be different on different targets.
3086 Optional sign: either @samp{+} or @samp{-}.
3088 One or more decimal digits.
3093 At least one of the integer part or the fractional part must be
3094 present. The floating point number has the usual base-10 value.
3096 @command{@value{AS}} does all processing using integers. Flonums are computed
3097 independently of any floating point hardware in the computer running
3098 @command{@value{AS}}.
3102 @c Bit fields are written as a general facility but are also controlled
3103 @c by a conditional-compilation flag---which is as of now (21mar91)
3104 @c turned on only by the i960 config of GAS.
3106 @subsubsection Bit Fields
3109 @cindex constants, bit field
3110 You can also define numeric constants as @dfn{bit fields}.
3111 Specify two numbers separated by a colon---
3113 @var{mask}:@var{value}
3116 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
3119 The resulting number is then packed
3121 @c this conditional paren in case bit fields turned on elsewhere than 960
3122 (in host-dependent byte order)
3124 into a field whose width depends on which assembler directive has the
3125 bit-field as its argument. Overflow (a result from the bitwise and
3126 requiring more binary digits to represent) is not an error; instead,
3127 more constants are generated, of the specified width, beginning with the
3128 least significant digits.@refill
3130 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
3131 @code{.short}, and @code{.word} accept bit-field arguments.
3136 @chapter Sections and Relocation
3141 * Secs Background:: Background
3142 * Ld Sections:: Linker Sections
3143 * As Sections:: Assembler Internal Sections
3144 * Sub-Sections:: Sub-Sections
3148 @node Secs Background
3151 Roughly, a section is a range of addresses, with no gaps; all data
3152 ``in'' those addresses is treated the same for some particular purpose.
3153 For example there may be a ``read only'' section.
3155 @cindex linker, and assembler
3156 @cindex assembler, and linker
3157 The linker @code{@value{LD}} reads many object files (partial programs) and
3158 combines their contents to form a runnable program. When @command{@value{AS}}
3159 emits an object file, the partial program is assumed to start at address 0.
3160 @code{@value{LD}} assigns the final addresses for the partial program, so that
3161 different partial programs do not overlap. This is actually an
3162 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3165 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3166 addresses. These blocks slide to their run-time addresses as rigid
3167 units; their length does not change and neither does the order of bytes
3168 within them. Such a rigid unit is called a @emph{section}. Assigning
3169 run-time addresses to sections is called @dfn{relocation}. It includes
3170 the task of adjusting mentions of object-file addresses so they refer to
3171 the proper run-time addresses.
3173 For the H8/300, and for the Renesas / SuperH SH,
3174 @command{@value{AS}} pads sections if needed to
3175 ensure they end on a word (sixteen bit) boundary.
3178 @cindex standard assembler sections
3179 An object file written by @command{@value{AS}} has at least three sections, any
3180 of which may be empty. These are named @dfn{text}, @dfn{data} and
3185 When it generates COFF or ELF output,
3187 @command{@value{AS}} can also generate whatever other named sections you specify
3188 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3189 If you do not use any directives that place output in the @samp{.text}
3190 or @samp{.data} sections, these sections still exist, but are empty.
3195 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3197 @command{@value{AS}} can also generate whatever other named sections you
3198 specify using the @samp{.space} and @samp{.subspace} directives. See
3199 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3200 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3201 assembler directives.
3204 Additionally, @command{@value{AS}} uses different names for the standard
3205 text, data, and bss sections when generating SOM output. Program text
3206 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3207 BSS into @samp{$BSS$}.
3211 Within the object file, the text section starts at address @code{0}, the
3212 data section follows, and the bss section follows the data section.
3215 When generating either SOM or ELF output files on the HPPA, the text
3216 section starts at address @code{0}, the data section at address
3217 @code{0x4000000}, and the bss section follows the data section.
3220 To let @code{@value{LD}} know which data changes when the sections are
3221 relocated, and how to change that data, @command{@value{AS}} also writes to the
3222 object file details of the relocation needed. To perform relocation
3223 @code{@value{LD}} must know, each time an address in the object
3227 Where in the object file is the beginning of this reference to
3230 How long (in bytes) is this reference?
3232 Which section does the address refer to? What is the numeric value of
3234 (@var{address}) @minus{} (@var{start-address of section})?
3237 Is the reference to an address ``Program-Counter relative''?
3240 @cindex addresses, format of
3241 @cindex section-relative addressing
3242 In fact, every address @command{@value{AS}} ever uses is expressed as
3244 (@var{section}) + (@var{offset into section})
3247 Further, most expressions @command{@value{AS}} computes have this section-relative
3250 (For some object formats, such as SOM for the HPPA, some expressions are
3251 symbol-relative instead.)
3254 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3255 @var{N} into section @var{secname}.''
3257 Apart from text, data and bss sections you need to know about the
3258 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3259 addresses in the absolute section remain unchanged. For example, address
3260 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3261 @code{@value{LD}}. Although the linker never arranges two partial programs'
3262 data sections with overlapping addresses after linking, @emph{by definition}
3263 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3264 part of a program is always the same address when the program is running as
3265 address @code{@{absolute@ 239@}} in any other part of the program.
3267 The idea of sections is extended to the @dfn{undefined} section. Any
3268 address whose section is unknown at assembly time is by definition
3269 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3270 Since numbers are always defined, the only way to generate an undefined
3271 address is to mention an undefined symbol. A reference to a named
3272 common block would be such a symbol: its value is unknown at assembly
3273 time so it has section @emph{undefined}.
3275 By analogy the word @emph{section} is used to describe groups of sections in
3276 the linked program. @code{@value{LD}} puts all partial programs' text
3277 sections in contiguous addresses in the linked program. It is
3278 customary to refer to the @emph{text section} of a program, meaning all
3279 the addresses of all partial programs' text sections. Likewise for
3280 data and bss sections.
3282 Some sections are manipulated by @code{@value{LD}}; others are invented for
3283 use of @command{@value{AS}} and have no meaning except during assembly.
3286 @section Linker Sections
3287 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3292 @cindex named sections
3293 @cindex sections, named
3294 @item named sections
3297 @cindex text section
3298 @cindex data section
3302 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3303 separate but equal sections. Anything you can say of one section is
3306 When the program is running, however, it is
3307 customary for the text section to be unalterable. The
3308 text section is often shared among processes: it contains
3309 instructions, constants and the like. The data section of a running
3310 program is usually alterable: for example, C variables would be stored
3311 in the data section.
3316 This section contains zeroed bytes when your program begins running. It
3317 is used to hold uninitialized variables or common storage. The length of
3318 each partial program's bss section is important, but because it starts
3319 out containing zeroed bytes there is no need to store explicit zero
3320 bytes in the object file. The bss section was invented to eliminate
3321 those explicit zeros from object files.
3323 @cindex absolute section
3324 @item absolute section
3325 Address 0 of this section is always ``relocated'' to runtime address 0.
3326 This is useful if you want to refer to an address that @code{@value{LD}} must
3327 not change when relocating. In this sense we speak of absolute
3328 addresses being ``unrelocatable'': they do not change during relocation.
3330 @cindex undefined section
3331 @item undefined section
3332 This ``section'' is a catch-all for address references to objects not in
3333 the preceding sections.
3334 @c FIXME: ref to some other doc on obj-file formats could go here.
3337 @cindex relocation example
3338 An idealized example of three relocatable sections follows.
3340 The example uses the traditional section names @samp{.text} and @samp{.data}.
3342 Memory addresses are on the horizontal axis.
3346 @c END TEXI2ROFF-KILL
3349 partial program # 1: |ttttt|dddd|00|
3356 partial program # 2: |TTT|DDD|000|
3359 +--+---+-----+--+----+---+-----+~~
3360 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3361 +--+---+-----+--+----+---+-----+~~
3363 addresses: 0 @dots{}
3370 \line{\it Partial program \#1: \hfil}
3371 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3372 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3374 \line{\it Partial program \#2: \hfil}
3375 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3376 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3378 \line{\it linked program: \hfil}
3379 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3380 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3381 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3382 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3384 \line{\it addresses: \hfil}
3388 @c END TEXI2ROFF-KILL
3391 @section Assembler Internal Sections
3393 @cindex internal assembler sections
3394 @cindex sections in messages, internal
3395 These sections are meant only for the internal use of @command{@value{AS}}. They
3396 have no meaning at run-time. You do not really need to know about these
3397 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3398 warning messages, so it might be helpful to have an idea of their
3399 meanings to @command{@value{AS}}. These sections are used to permit the
3400 value of every expression in your assembly language program to be a
3401 section-relative address.
3404 @cindex assembler internal logic error
3405 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3406 An internal assembler logic error has been found. This means there is a
3407 bug in the assembler.
3409 @cindex expr (internal section)
3411 The assembler stores complex expression internally as combinations of
3412 symbols. When it needs to represent an expression as a symbol, it puts
3413 it in the expr section.
3415 @c FIXME item transfer[t] vector preload
3416 @c FIXME item transfer[t] vector postload
3417 @c FIXME item register
3421 @section Sub-Sections
3423 @cindex numbered subsections
3424 @cindex grouping data
3430 fall into two sections: text and data.
3432 You may have separate groups of
3434 data in named sections
3438 data in named sections
3444 that you want to end up near to each other in the object file, even though they
3445 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3446 use @dfn{subsections} for this purpose. Within each section, there can be
3447 numbered subsections with values from 0 to 8192. Objects assembled into the
3448 same subsection go into the object file together with other objects in the same
3449 subsection. For example, a compiler might want to store constants in the text
3450 section, but might not want to have them interspersed with the program being
3451 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3452 section of code being output, and a @samp{.text 1} before each group of
3453 constants being output.
3455 Subsections are optional. If you do not use subsections, everything
3456 goes in subsection number zero.
3459 Each subsection is zero-padded up to a multiple of four bytes.
3460 (Subsections may be padded a different amount on different flavors
3461 of @command{@value{AS}}.)
3465 On the H8/300 platform, each subsection is zero-padded to a word
3466 boundary (two bytes).
3467 The same is true on the Renesas SH.
3470 @c FIXME section padding (alignment)?
3471 @c Rich Pixley says padding here depends on target obj code format; that
3472 @c doesn't seem particularly useful to say without further elaboration,
3473 @c so for now I say nothing about it. If this is a generic BFD issue,
3474 @c these paragraphs might need to vanish from this manual, and be
3475 @c discussed in BFD chapter of binutils (or some such).
3479 Subsections appear in your object file in numeric order, lowest numbered
3480 to highest. (All this to be compatible with other people's assemblers.)
3481 The object file contains no representation of subsections; @code{@value{LD}} and
3482 other programs that manipulate object files see no trace of them.
3483 They just see all your text subsections as a text section, and all your
3484 data subsections as a data section.
3486 To specify which subsection you want subsequent statements assembled
3487 into, use a numeric argument to specify it, in a @samp{.text
3488 @var{expression}} or a @samp{.data @var{expression}} statement.
3491 When generating COFF output, you
3496 can also use an extra subsection
3497 argument with arbitrary named sections: @samp{.section @var{name},
3502 When generating ELF output, you
3507 can also use the @code{.subsection} directive (@pxref{SubSection})
3508 to specify a subsection: @samp{.subsection @var{expression}}.
3510 @var{Expression} should be an absolute expression
3511 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3512 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3513 begins in @code{text 0}. For instance:
3515 .text 0 # The default subsection is text 0 anyway.
3516 .ascii "This lives in the first text subsection. *"
3518 .ascii "But this lives in the second text subsection."
3520 .ascii "This lives in the data section,"
3521 .ascii "in the first data subsection."
3523 .ascii "This lives in the first text section,"
3524 .ascii "immediately following the asterisk (*)."
3527 Each section has a @dfn{location counter} incremented by one for every byte
3528 assembled into that section. Because subsections are merely a convenience
3529 restricted to @command{@value{AS}} there is no concept of a subsection location
3530 counter. There is no way to directly manipulate a location counter---but the
3531 @code{.align} directive changes it, and any label definition captures its
3532 current value. The location counter of the section where statements are being
3533 assembled is said to be the @dfn{active} location counter.
3536 @section bss Section
3539 @cindex common variable storage
3540 The bss section is used for local common variable storage.
3541 You may allocate address space in the bss section, but you may
3542 not dictate data to load into it before your program executes. When
3543 your program starts running, all the contents of the bss
3544 section are zeroed bytes.
3546 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3547 @ref{Lcomm,,@code{.lcomm}}.
3549 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3550 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3553 When assembling for a target which supports multiple sections, such as ELF or
3554 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3555 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3556 section. Typically the section will only contain symbol definitions and
3557 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3564 Symbols are a central concept: the programmer uses symbols to name
3565 things, the linker uses symbols to link, and the debugger uses symbols
3569 @cindex debuggers, and symbol order
3570 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3571 the same order they were declared. This may break some debuggers.
3576 * Setting Symbols:: Giving Symbols Other Values
3577 * Symbol Names:: Symbol Names
3578 * Dot:: The Special Dot Symbol
3579 * Symbol Attributes:: Symbol Attributes
3586 A @dfn{label} is written as a symbol immediately followed by a colon
3587 @samp{:}. The symbol then represents the current value of the
3588 active location counter, and is, for example, a suitable instruction
3589 operand. You are warned if you use the same symbol to represent two
3590 different locations: the first definition overrides any other
3594 On the HPPA, the usual form for a label need not be immediately followed by a
3595 colon, but instead must start in column zero. Only one label may be defined on
3596 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3597 provides a special directive @code{.label} for defining labels more flexibly.
3600 @node Setting Symbols
3601 @section Giving Symbols Other Values
3603 @cindex assigning values to symbols
3604 @cindex symbol values, assigning
3605 A symbol can be given an arbitrary value by writing a symbol, followed
3606 by an equals sign @samp{=}, followed by an expression
3607 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3608 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3609 equals sign @samp{=}@samp{=} here represents an equivalent of the
3610 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3613 Blackfin does not support symbol assignment with @samp{=}.
3617 @section Symbol Names
3619 @cindex symbol names
3620 @cindex names, symbol
3621 @ifclear SPECIAL-SYMS
3622 Symbol names begin with a letter or with one of @samp{._}. On most
3623 machines, you can also use @code{$} in symbol names; exceptions are
3624 noted in @ref{Machine Dependencies}. That character may be followed by any
3625 string of digits, letters, dollar signs (unless otherwise noted for a
3626 particular target machine), and underscores.
3630 Symbol names begin with a letter or with one of @samp{._}. On the
3631 Renesas SH you can also use @code{$} in symbol names. That
3632 character may be followed by any string of digits, letters, dollar signs (save
3633 on the H8/300), and underscores.
3637 Case of letters is significant: @code{foo} is a different symbol name
3640 Multibyte characters are supported. To generate a symbol name containing
3641 multibyte characters enclose it within double quotes and use escape codes. cf
3642 @xref{Strings}. Generating a multibyte symbol name from a label is not
3643 currently supported.
3645 Each symbol has exactly one name. Each name in an assembly language program
3646 refers to exactly one symbol. You may use that symbol name any number of times
3649 @subheading Local Symbol Names
3651 @cindex local symbol names
3652 @cindex symbol names, local
3653 A local symbol is any symbol beginning with certain local label prefixes.
3654 By default, the local label prefix is @samp{.L} for ELF systems or
3655 @samp{L} for traditional a.out systems, but each target may have its own
3656 set of local label prefixes.
3658 On the HPPA local symbols begin with @samp{L$}.
3661 Local symbols are defined and used within the assembler, but they are
3662 normally not saved in object files. Thus, they are not visible when debugging.
3663 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols})
3664 to retain the local symbols in the object files.
3666 @subheading Local Labels
3668 @cindex local labels
3669 @cindex temporary symbol names
3670 @cindex symbol names, temporary
3671 Local labels help compilers and programmers use names temporarily.
3672 They create symbols which are guaranteed to be unique over the entire scope of
3673 the input source code and which can be referred to by a simple notation.
3674 To define a local label, write a label of the form @samp{@b{N}:} (where @b{N}
3675 represents any positive integer). To refer to the most recent previous
3676 definition of that label write @samp{@b{N}b}, using the same number as when
3677 you defined the label. To refer to the next definition of a local label, write
3678 @samp{@b{N}f}---the @samp{b} stands for ``backwards'' and the @samp{f} stands
3681 There is no restriction on how you can use these labels, and you can reuse them
3682 too. So that it is possible to repeatedly define the same local label (using
3683 the same number @samp{@b{N}}), although you can only refer to the most recently
3684 defined local label of that number (for a backwards reference) or the next
3685 definition of a specific local label for a forward reference. It is also worth
3686 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3687 implemented in a slightly more efficient manner than the others.
3698 Which is the equivalent of:
3701 label_1: branch label_3
3702 label_2: branch label_1
3703 label_3: branch label_4
3704 label_4: branch label_3
3707 Local label names are only a notational device. They are immediately
3708 transformed into more conventional symbol names before the assembler uses them.
3709 The symbol names are stored in the symbol table, appear in error messages, and
3710 are optionally emitted to the object file. The names are constructed using
3714 @item @emph{local label prefix}
3715 All local symbols begin with the system-specific local label prefix.
3716 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3717 that start with the local label prefix. These labels are
3718 used for symbols you are never intended to see. If you use the
3719 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3720 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3721 you may use them in debugging.
3724 This is the number that was used in the local label definition. So if the
3725 label is written @samp{55:} then the number is @samp{55}.
3728 This unusual character is included so you do not accidentally invent a symbol
3729 of the same name. The character has ASCII value of @samp{\002} (control-B).
3731 @item @emph{ordinal number}
3732 This is a serial number to keep the labels distinct. The first definition of
3733 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3734 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3735 the number @samp{1} and its 15th definition gets @samp{15} as well.
3738 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3739 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3741 @subheading Dollar Local Labels
3742 @cindex dollar local symbols
3744 @code{@value{AS}} also supports an even more local form of local labels called
3745 dollar labels. These labels go out of scope (i.e., they become undefined) as
3746 soon as a non-local label is defined. Thus they remain valid for only a small
3747 region of the input source code. Normal local labels, by contrast, remain in
3748 scope for the entire file, or until they are redefined by another occurrence of
3749 the same local label.
3751 Dollar labels are defined in exactly the same way as ordinary local labels,
3752 except that they have a dollar sign suffix to their numeric value, e.g.,
3755 They can also be distinguished from ordinary local labels by their transformed
3756 names which use ASCII character @samp{\001} (control-A) as the magic character
3757 to distinguish them from ordinary labels. For example, the fifth definition of
3758 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3761 @section The Special Dot Symbol
3763 @cindex dot (symbol)
3764 @cindex @code{.} (symbol)
3765 @cindex current address
3766 @cindex location counter
3767 The special symbol @samp{.} refers to the current address that
3768 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3769 .long .} defines @code{melvin} to contain its own address.
3770 Assigning a value to @code{.} is treated the same as a @code{.org}
3772 @ifclear no-space-dir
3773 Thus, the expression @samp{.=.+4} is the same as saying
3777 @node Symbol Attributes
3778 @section Symbol Attributes
3780 @cindex symbol attributes
3781 @cindex attributes, symbol
3782 Every symbol has, as well as its name, the attributes ``Value'' and
3783 ``Type''. Depending on output format, symbols can also have auxiliary
3786 The detailed definitions are in @file{a.out.h}.
3789 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3790 all these attributes, and probably won't warn you. This makes the
3791 symbol an externally defined symbol, which is generally what you
3795 * Symbol Value:: Value
3796 * Symbol Type:: Type
3799 * a.out Symbols:: Symbol Attributes: @code{a.out}
3803 * a.out Symbols:: Symbol Attributes: @code{a.out}
3806 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3811 * COFF Symbols:: Symbol Attributes for COFF
3814 * SOM Symbols:: Symbol Attributes for SOM
3821 @cindex value of a symbol
3822 @cindex symbol value
3823 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3824 location in the text, data, bss or absolute sections the value is the
3825 number of addresses from the start of that section to the label.
3826 Naturally for text, data and bss sections the value of a symbol changes
3827 as @code{@value{LD}} changes section base addresses during linking. Absolute
3828 symbols' values do not change during linking: that is why they are
3831 The value of an undefined symbol is treated in a special way. If it is
3832 0 then the symbol is not defined in this assembler source file, and
3833 @code{@value{LD}} tries to determine its value from other files linked into the
3834 same program. You make this kind of symbol simply by mentioning a symbol
3835 name without defining it. A non-zero value represents a @code{.comm}
3836 common declaration. The value is how much common storage to reserve, in
3837 bytes (addresses). The symbol refers to the first address of the
3843 @cindex type of a symbol
3845 The type attribute of a symbol contains relocation (section)
3846 information, any flag settings indicating that a symbol is external, and
3847 (optionally), other information for linkers and debuggers. The exact
3848 format depends on the object-code output format in use.
3853 @c The following avoids a "widow" subsection title. @group would be
3854 @c better if it were available outside examples.
3857 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3859 @cindex @code{b.out} symbol attributes
3860 @cindex symbol attributes, @code{b.out}
3861 These symbol attributes appear only when @command{@value{AS}} is configured for
3862 one of the Berkeley-descended object output formats---@code{a.out} or
3868 @subsection Symbol Attributes: @code{a.out}
3870 @cindex @code{a.out} symbol attributes
3871 @cindex symbol attributes, @code{a.out}
3877 @subsection Symbol Attributes: @code{a.out}
3879 @cindex @code{a.out} symbol attributes
3880 @cindex symbol attributes, @code{a.out}
3884 * Symbol Desc:: Descriptor
3885 * Symbol Other:: Other
3889 @subsubsection Descriptor
3891 @cindex descriptor, of @code{a.out} symbol
3892 This is an arbitrary 16-bit value. You may establish a symbol's
3893 descriptor value by using a @code{.desc} statement
3894 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3895 @command{@value{AS}}.
3898 @subsubsection Other
3900 @cindex other attribute, of @code{a.out} symbol
3901 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3906 @subsection Symbol Attributes for COFF
3908 @cindex COFF symbol attributes
3909 @cindex symbol attributes, COFF
3911 The COFF format supports a multitude of auxiliary symbol attributes;
3912 like the primary symbol attributes, they are set between @code{.def} and
3913 @code{.endef} directives.
3915 @subsubsection Primary Attributes
3917 @cindex primary attributes, COFF symbols
3918 The symbol name is set with @code{.def}; the value and type,
3919 respectively, with @code{.val} and @code{.type}.
3921 @subsubsection Auxiliary Attributes
3923 @cindex auxiliary attributes, COFF symbols
3924 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3925 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3926 table information for COFF.
3931 @subsection Symbol Attributes for SOM
3933 @cindex SOM symbol attributes
3934 @cindex symbol attributes, SOM
3936 The SOM format for the HPPA supports a multitude of symbol attributes set with
3937 the @code{.EXPORT} and @code{.IMPORT} directives.
3939 The attributes are described in @cite{HP9000 Series 800 Assembly
3940 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3941 @code{EXPORT} assembler directive documentation.
3945 @chapter Expressions
3949 @cindex numeric values
3950 An @dfn{expression} specifies an address or numeric value.
3951 Whitespace may precede and/or follow an expression.
3953 The result of an expression must be an absolute number, or else an offset into
3954 a particular section. If an expression is not absolute, and there is not
3955 enough information when @command{@value{AS}} sees the expression to know its
3956 section, a second pass over the source program might be necessary to interpret
3957 the expression---but the second pass is currently not implemented.
3958 @command{@value{AS}} aborts with an error message in this situation.
3961 * Empty Exprs:: Empty Expressions
3962 * Integer Exprs:: Integer Expressions
3966 @section Empty Expressions
3968 @cindex empty expressions
3969 @cindex expressions, empty
3970 An empty expression has no value: it is just whitespace or null.
3971 Wherever an absolute expression is required, you may omit the
3972 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3973 is compatible with other assemblers.
3976 @section Integer Expressions
3978 @cindex integer expressions
3979 @cindex expressions, integer
3980 An @dfn{integer expression} is one or more @emph{arguments} delimited
3981 by @emph{operators}.
3984 * Arguments:: Arguments
3985 * Operators:: Operators
3986 * Prefix Ops:: Prefix Operators
3987 * Infix Ops:: Infix Operators
3991 @subsection Arguments
3993 @cindex expression arguments
3994 @cindex arguments in expressions
3995 @cindex operands in expressions
3996 @cindex arithmetic operands
3997 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3998 contexts arguments are sometimes called ``arithmetic operands''. In
3999 this manual, to avoid confusing them with the ``instruction operands'' of
4000 the machine language, we use the term ``argument'' to refer to parts of
4001 expressions only, reserving the word ``operand'' to refer only to machine
4002 instruction operands.
4004 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
4005 @var{section} is one of text, data, bss, absolute,
4006 or undefined. @var{NNN} is a signed, 2's complement 32 bit
4009 Numbers are usually integers.
4011 A number can be a flonum or bignum. In this case, you are warned
4012 that only the low order 32 bits are used, and @command{@value{AS}} pretends
4013 these 32 bits are an integer. You may write integer-manipulating
4014 instructions that act on exotic constants, compatible with other
4017 @cindex subexpressions
4018 Subexpressions are a left parenthesis @samp{(} followed by an integer
4019 expression, followed by a right parenthesis @samp{)}; or a prefix
4020 operator followed by an argument.
4023 @subsection Operators
4025 @cindex operators, in expressions
4026 @cindex arithmetic functions
4027 @cindex functions, in expressions
4028 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
4029 operators are followed by an argument. Infix operators appear
4030 between their arguments. Operators may be preceded and/or followed by
4034 @subsection Prefix Operator
4036 @cindex prefix operators
4037 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
4038 one argument, which must be absolute.
4040 @c the tex/end tex stuff surrounding this small table is meant to make
4041 @c it align, on the printed page, with the similar table in the next
4042 @c section (which is inside an enumerate).
4044 \global\advance\leftskip by \itemindent
4049 @dfn{Negation}. Two's complement negation.
4051 @dfn{Complementation}. Bitwise not.
4055 \global\advance\leftskip by -\itemindent
4059 @subsection Infix Operators
4061 @cindex infix operators
4062 @cindex operators, permitted arguments
4063 @dfn{Infix operators} take two arguments, one on either side. Operators
4064 have precedence, but operations with equal precedence are performed left
4065 to right. Apart from @code{+} or @option{-}, both arguments must be
4066 absolute, and the result is absolute.
4069 @cindex operator precedence
4070 @cindex precedence of operators
4077 @dfn{Multiplication}.
4080 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4086 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4089 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4093 Intermediate precedence
4098 @dfn{Bitwise Inclusive Or}.
4104 @dfn{Bitwise Exclusive Or}.
4107 @dfn{Bitwise Or Not}.
4114 @cindex addition, permitted arguments
4115 @cindex plus, permitted arguments
4116 @cindex arguments for addition
4118 @dfn{Addition}. If either argument is absolute, the result has the section of
4119 the other argument. You may not add together arguments from different
4122 @cindex subtraction, permitted arguments
4123 @cindex minus, permitted arguments
4124 @cindex arguments for subtraction
4126 @dfn{Subtraction}. If the right argument is absolute, the
4127 result has the section of the left argument.
4128 If both arguments are in the same section, the result is absolute.
4129 You may not subtract arguments from different sections.
4130 @c FIXME is there still something useful to say about undefined - undefined ?
4132 @cindex comparison expressions
4133 @cindex expressions, comparison
4138 @dfn{Is Not Equal To}
4142 @dfn{Is Greater Than}
4144 @dfn{Is Greater Than Or Equal To}
4146 @dfn{Is Less Than Or Equal To}
4148 The comparison operators can be used as infix operators. A true results has a
4149 value of -1 whereas a false result has a value of 0. Note, these operators
4150 perform signed comparisons.
4153 @item Lowest Precedence
4162 These two logical operations can be used to combine the results of sub
4163 expressions. Note, unlike the comparison operators a true result returns a
4164 value of 1 but a false results does still return 0. Also note that the logical
4165 or operator has a slightly lower precedence than logical and.
4170 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4171 address; you can only have a defined section in one of the two arguments.
4174 @chapter Assembler Directives
4176 @cindex directives, machine independent
4177 @cindex pseudo-ops, machine independent
4178 @cindex machine independent directives
4179 All assembler directives have names that begin with a period (@samp{.}).
4180 The rest of the name is letters, usually in lower case.
4182 This chapter discusses directives that are available regardless of the
4183 target machine configuration for the @sc{gnu} assembler.
4185 Some machine configurations provide additional directives.
4186 @xref{Machine Dependencies}.
4189 @ifset machine-directives
4190 @xref{Machine Dependencies}, for additional directives.
4195 * Abort:: @code{.abort}
4197 * ABORT (COFF):: @code{.ABORT}
4200 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4201 * Altmacro:: @code{.altmacro}
4202 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4203 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4204 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4205 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, @code{.bundle_lock}, @code{.bundle_unlock}
4206 * Byte:: @code{.byte @var{expressions}}
4207 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4208 * Comm:: @code{.comm @var{symbol} , @var{length} }
4209 * Data:: @code{.data @var{subsection}}
4211 * Def:: @code{.def @var{name}}
4214 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4220 * Double:: @code{.double @var{flonums}}
4221 * Eject:: @code{.eject}
4222 * Else:: @code{.else}
4223 * Elseif:: @code{.elseif}
4226 * Endef:: @code{.endef}
4229 * Endfunc:: @code{.endfunc}
4230 * Endif:: @code{.endif}
4231 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4232 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4233 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4235 * Error:: @code{.error @var{string}}
4236 * Exitm:: @code{.exitm}
4237 * Extern:: @code{.extern}
4238 * Fail:: @code{.fail}
4239 * File:: @code{.file}
4240 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4241 * Float:: @code{.float @var{flonums}}
4242 * Func:: @code{.func}
4243 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4245 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4246 * Hidden:: @code{.hidden @var{names}}
4249 * hword:: @code{.hword @var{expressions}}
4250 * Ident:: @code{.ident}
4251 * If:: @code{.if @var{absolute expression}}
4252 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4253 * Include:: @code{.include "@var{file}"}
4254 * Int:: @code{.int @var{expressions}}
4256 * Internal:: @code{.internal @var{names}}
4259 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4260 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4261 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4262 * Lflags:: @code{.lflags}
4263 @ifclear no-line-dir
4264 * Line:: @code{.line @var{line-number}}
4267 * Linkonce:: @code{.linkonce [@var{type}]}
4268 * List:: @code{.list}
4269 * Ln:: @code{.ln @var{line-number}}
4270 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4271 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4273 * Local:: @code{.local @var{names}}
4276 * Long:: @code{.long @var{expressions}}
4278 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4281 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4282 * MRI:: @code{.mri @var{val}}
4283 * Noaltmacro:: @code{.noaltmacro}
4284 * Nolist:: @code{.nolist}
4285 * Octa:: @code{.octa @var{bignums}}
4286 * Offset:: @code{.offset @var{loc}}
4287 * Org:: @code{.org @var{new-lc}, @var{fill}}
4288 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4290 * PopSection:: @code{.popsection}
4291 * Previous:: @code{.previous}
4294 * Print:: @code{.print @var{string}}
4296 * Protected:: @code{.protected @var{names}}
4299 * Psize:: @code{.psize @var{lines}, @var{columns}}
4300 * Purgem:: @code{.purgem @var{name}}
4302 * PushSection:: @code{.pushsection @var{name}}
4305 * Quad:: @code{.quad @var{bignums}}
4306 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4307 * Rept:: @code{.rept @var{count}}
4308 * Sbttl:: @code{.sbttl "@var{subheading}"}
4310 * Scl:: @code{.scl @var{class}}
4313 * Section:: @code{.section @var{name}[, @var{flags}]}
4316 * Set:: @code{.set @var{symbol}, @var{expression}}
4317 * Short:: @code{.short @var{expressions}}
4318 * Single:: @code{.single @var{flonums}}
4320 * Size:: @code{.size [@var{name} , @var{expression}]}
4322 @ifclear no-space-dir
4323 * Skip:: @code{.skip @var{size} , @var{fill}}
4326 * Sleb128:: @code{.sleb128 @var{expressions}}
4327 @ifclear no-space-dir
4328 * Space:: @code{.space @var{size} , @var{fill}}
4331 * Stab:: @code{.stabd, .stabn, .stabs}
4334 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4335 * Struct:: @code{.struct @var{expression}}
4337 * SubSection:: @code{.subsection}
4338 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4342 * Tag:: @code{.tag @var{structname}}
4345 * Text:: @code{.text @var{subsection}}
4346 * Title:: @code{.title "@var{heading}"}
4348 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4351 * Uleb128:: @code{.uleb128 @var{expressions}}
4353 * Val:: @code{.val @var{addr}}
4357 * Version:: @code{.version "@var{string}"}
4358 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4359 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4362 * Warning:: @code{.warning @var{string}}
4363 * Weak:: @code{.weak @var{names}}
4364 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4365 * Word:: @code{.word @var{expressions}}
4366 * Deprecated:: Deprecated Directives
4370 @section @code{.abort}
4372 @cindex @code{abort} directive
4373 @cindex stopping the assembly
4374 This directive stops the assembly immediately. It is for
4375 compatibility with other assemblers. The original idea was that the
4376 assembly language source would be piped into the assembler. If the sender
4377 of the source quit, it could use this directive tells @command{@value{AS}} to
4378 quit also. One day @code{.abort} will not be supported.
4382 @section @code{.ABORT} (COFF)
4384 @cindex @code{ABORT} directive
4385 When producing COFF output, @command{@value{AS}} accepts this directive as a
4386 synonym for @samp{.abort}.
4389 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4395 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4397 @cindex padding the location counter
4398 @cindex @code{align} directive
4399 Pad the location counter (in the current subsection) to a particular storage
4400 boundary. The first expression (which must be absolute) is the alignment
4401 required, as described below.
4403 The second expression (also absolute) gives the fill value to be stored in the
4404 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4405 padding bytes are normally zero. However, on some systems, if the section is
4406 marked as containing code and the fill value is omitted, the space is filled
4407 with no-op instructions.
4409 The third expression is also absolute, and is also optional. If it is present,
4410 it is the maximum number of bytes that should be skipped by this alignment
4411 directive. If doing the alignment would require skipping more bytes than the
4412 specified maximum, then the alignment is not done at all. You can omit the
4413 fill value (the second argument) entirely by simply using two commas after the
4414 required alignment; this can be useful if you want the alignment to be filled
4415 with no-op instructions when appropriate.
4417 The way the required alignment is specified varies from system to system.
4418 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or1k,
4419 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4420 alignment request in bytes. For example @samp{.align 8} advances
4421 the location counter until it is a multiple of 8. If the location counter
4422 is already a multiple of 8, no change is needed. For the tic54x, the
4423 first expression is the alignment request in words.
4425 For other systems, including ppc, i386 using a.out format, arm and
4426 strongarm, it is the
4427 number of low-order zero bits the location counter must have after
4428 advancement. For example @samp{.align 3} advances the location
4429 counter until it a multiple of 8. If the location counter is already a
4430 multiple of 8, no change is needed.
4432 This inconsistency is due to the different behaviors of the various
4433 native assemblers for these systems which GAS must emulate.
4434 GAS also provides @code{.balign} and @code{.p2align} directives,
4435 described later, which have a consistent behavior across all
4436 architectures (but are specific to GAS).
4439 @section @code{.altmacro}
4440 Enable alternate macro mode, enabling:
4443 @item LOCAL @var{name} [ , @dots{} ]
4444 One additional directive, @code{LOCAL}, is available. It is used to
4445 generate a string replacement for each of the @var{name} arguments, and
4446 replace any instances of @var{name} in each macro expansion. The
4447 replacement string is unique in the assembly, and different for each
4448 separate macro expansion. @code{LOCAL} allows you to write macros that
4449 define symbols, without fear of conflict between separate macro expansions.
4451 @item String delimiters
4452 You can write strings delimited in these other ways besides
4453 @code{"@var{string}"}:
4456 @item '@var{string}'
4457 You can delimit strings with single-quote characters.
4459 @item <@var{string}>
4460 You can delimit strings with matching angle brackets.
4463 @item single-character string escape
4464 To include any single character literally in a string (even if the
4465 character would otherwise have some special meaning), you can prefix the
4466 character with @samp{!} (an exclamation mark). For example, you can
4467 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4469 @item Expression results as strings
4470 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4471 and use the result as a string.
4475 @section @code{.ascii "@var{string}"}@dots{}
4477 @cindex @code{ascii} directive
4478 @cindex string literals
4479 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4480 separated by commas. It assembles each string (with no automatic
4481 trailing zero byte) into consecutive addresses.
4484 @section @code{.asciz "@var{string}"}@dots{}
4486 @cindex @code{asciz} directive
4487 @cindex zero-terminated strings
4488 @cindex null-terminated strings
4489 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4490 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4493 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4495 @cindex padding the location counter given number of bytes
4496 @cindex @code{balign} directive
4497 Pad the location counter (in the current subsection) to a particular
4498 storage boundary. The first expression (which must be absolute) is the
4499 alignment request in bytes. For example @samp{.balign 8} advances
4500 the location counter until it is a multiple of 8. If the location counter
4501 is already a multiple of 8, no change is needed.
4503 The second expression (also absolute) gives the fill value to be stored in the
4504 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4505 padding bytes are normally zero. However, on some systems, if the section is
4506 marked as containing code and the fill value is omitted, the space is filled
4507 with no-op instructions.
4509 The third expression is also absolute, and is also optional. If it is present,
4510 it is the maximum number of bytes that should be skipped by this alignment
4511 directive. If doing the alignment would require skipping more bytes than the
4512 specified maximum, then the alignment is not done at all. You can omit the
4513 fill value (the second argument) entirely by simply using two commas after the
4514 required alignment; this can be useful if you want the alignment to be filled
4515 with no-op instructions when appropriate.
4517 @cindex @code{balignw} directive
4518 @cindex @code{balignl} directive
4519 The @code{.balignw} and @code{.balignl} directives are variants of the
4520 @code{.balign} directive. The @code{.balignw} directive treats the fill
4521 pattern as a two byte word value. The @code{.balignl} directives treats the
4522 fill pattern as a four byte longword value. For example, @code{.balignw
4523 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4524 filled in with the value 0x368d (the exact placement of the bytes depends upon
4525 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4528 @node Bundle directives
4529 @section @code{.bundle_align_mode @var{abs-expr}}
4530 @cindex @code{bundle_align_mode} directive
4532 @cindex instruction bundle
4533 @cindex aligned instruction bundle
4534 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4535 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4536 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4537 disabled (which is the default state). If the argument it not zero, it
4538 gives the size of an instruction bundle as a power of two (as for the
4539 @code{.p2align} directive, @pxref{P2align}).
4541 For some targets, it's an ABI requirement that no instruction may span a
4542 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4543 instructions that starts on an aligned boundary. For example, if
4544 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4545 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4546 effect, no single instruction may span a boundary between bundles. If an
4547 instruction would start too close to the end of a bundle for the length of
4548 that particular instruction to fit within the bundle, then the space at the
4549 end of that bundle is filled with no-op instructions so the instruction
4550 starts in the next bundle. As a corollary, it's an error if any single
4551 instruction's encoding is longer than the bundle size.
4553 @section @code{.bundle_lock} and @code{.bundle_unlock}
4554 @cindex @code{bundle_lock} directive
4555 @cindex @code{bundle_unlock} directive
4556 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4557 allow explicit control over instruction bundle padding. These directives
4558 are only valid when @code{.bundle_align_mode} has been used to enable
4559 aligned instruction bundle mode. It's an error if they appear when
4560 @code{.bundle_align_mode} has not been used at all, or when the last
4561 directive was @w{@code{.bundle_align_mode 0}}.
4563 @cindex bundle-locked
4564 For some targets, it's an ABI requirement that certain instructions may
4565 appear only as part of specified permissible sequences of multiple
4566 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4567 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4568 instruction sequence. For purposes of aligned instruction bundle mode, a
4569 sequence starting with @code{.bundle_lock} and ending with
4570 @code{.bundle_unlock} is treated as a single instruction. That is, the
4571 entire sequence must fit into a single bundle and may not span a bundle
4572 boundary. If necessary, no-op instructions will be inserted before the
4573 first instruction of the sequence so that the whole sequence starts on an
4574 aligned bundle boundary. It's an error if the sequence is longer than the
4577 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4578 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4579 nested. That is, a second @code{.bundle_lock} directive before the next
4580 @code{.bundle_unlock} directive has no effect except that it must be
4581 matched by another closing @code{.bundle_unlock} so that there is the
4582 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4585 @section @code{.byte @var{expressions}}
4587 @cindex @code{byte} directive
4588 @cindex integers, one byte
4589 @code{.byte} expects zero or more expressions, separated by commas.
4590 Each expression is assembled into the next byte.
4592 @node CFI directives
4593 @section @code{.cfi_sections @var{section_list}}
4594 @cindex @code{cfi_sections} directive
4595 @code{.cfi_sections} may be used to specify whether CFI directives
4596 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4597 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4598 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4599 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4600 directive is not used is @code{.cfi_sections .eh_frame}.
4602 @section @code{.cfi_startproc [simple]}
4603 @cindex @code{cfi_startproc} directive
4604 @code{.cfi_startproc} is used at the beginning of each function that
4605 should have an entry in @code{.eh_frame}. It initializes some internal
4606 data structures. Don't forget to close the function by
4607 @code{.cfi_endproc}.
4609 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4610 it also emits some architecture dependent initial CFI instructions.
4612 @section @code{.cfi_endproc}
4613 @cindex @code{cfi_endproc} directive
4614 @code{.cfi_endproc} is used at the end of a function where it closes its
4615 unwind entry previously opened by
4616 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4618 @section @code{.cfi_personality @var{encoding} [, @var{exp}]}
4619 @code{.cfi_personality} defines personality routine and its encoding.
4620 @var{encoding} must be a constant determining how the personality
4621 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4622 argument is not present, otherwise second argument should be
4623 a constant or a symbol name. When using indirect encodings,
4624 the symbol provided should be the location where personality
4625 can be loaded from, not the personality routine itself.
4626 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4627 no personality routine.
4629 @section @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4630 @code{.cfi_lsda} defines LSDA and its encoding.
4631 @var{encoding} must be a constant determining how the LSDA
4632 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4633 argument is not present, otherwise second argument should be a constant
4634 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4637 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4638 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4639 address from @var{register} and add @var{offset} to it}.
4641 @section @code{.cfi_def_cfa_register @var{register}}
4642 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4643 now on @var{register} will be used instead of the old one. Offset
4646 @section @code{.cfi_def_cfa_offset @var{offset}}
4647 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4648 remains the same, but @var{offset} is new. Note that it is the
4649 absolute offset that will be added to a defined register to compute
4652 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4653 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4654 value that is added/substracted from the previous offset.
4656 @section @code{.cfi_offset @var{register}, @var{offset}}
4657 Previous value of @var{register} is saved at offset @var{offset} from
4660 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4661 Previous value of @var{register} is saved at offset @var{offset} from
4662 the current CFA register. This is transformed to @code{.cfi_offset}
4663 using the known displacement of the CFA register from the CFA.
4664 This is often easier to use, because the number will match the
4665 code it's annotating.
4667 @section @code{.cfi_register @var{register1}, @var{register2}}
4668 Previous value of @var{register1} is saved in register @var{register2}.
4670 @section @code{.cfi_restore @var{register}}
4671 @code{.cfi_restore} says that the rule for @var{register} is now the
4672 same as it was at the beginning of the function, after all initial
4673 instruction added by @code{.cfi_startproc} were executed.
4675 @section @code{.cfi_undefined @var{register}}
4676 From now on the previous value of @var{register} can't be restored anymore.
4678 @section @code{.cfi_same_value @var{register}}
4679 Current value of @var{register} is the same like in the previous frame,
4680 i.e. no restoration needed.
4682 @section @code{.cfi_remember_state},
4683 First save all current rules for all registers by @code{.cfi_remember_state},
4684 then totally screw them up by subsequent @code{.cfi_*} directives and when
4685 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4686 the previous saved state.
4688 @section @code{.cfi_return_column @var{register}}
4689 Change return column @var{register}, i.e. the return address is either
4690 directly in @var{register} or can be accessed by rules for @var{register}.
4692 @section @code{.cfi_signal_frame}
4693 Mark current function as signal trampoline.
4695 @section @code{.cfi_window_save}
4696 SPARC register window has been saved.
4698 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4699 Allows the user to add arbitrary bytes to the unwind info. One
4700 might use this to add OS-specific CFI opcodes, or generic CFI
4701 opcodes that GAS does not yet support.
4703 @section @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4704 The current value of @var{register} is @var{label}. The value of @var{label}
4705 will be encoded in the output file according to @var{encoding}; see the
4706 description of @code{.cfi_personality} for details on this encoding.
4708 The usefulness of equating a register to a fixed label is probably
4709 limited to the return address register. Here, it can be useful to
4710 mark a code segment that has only one return address which is reached
4711 by a direct branch and no copy of the return address exists in memory
4712 or another register.
4715 @section @code{.comm @var{symbol} , @var{length} }
4717 @cindex @code{comm} directive
4718 @cindex symbol, common
4719 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4720 common symbol in one object file may be merged with a defined or common symbol
4721 of the same name in another object file. If @code{@value{LD}} does not see a
4722 definition for the symbol--just one or more common symbols--then it will
4723 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4724 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4725 the same name, and they do not all have the same size, it will allocate space
4726 using the largest size.
4729 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4730 an optional third argument. This is the desired alignment of the symbol,
4731 specified for ELF as a byte boundary (for example, an alignment of 16 means
4732 that the least significant 4 bits of the address should be zero), and for PE
4733 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4734 boundary). The alignment must be an absolute expression, and it must be a
4735 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4736 common symbol, it will use the alignment when placing the symbol. If no
4737 alignment is specified, @command{@value{AS}} will set the alignment to the
4738 largest power of two less than or equal to the size of the symbol, up to a
4739 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4740 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4741 @samp{--section-alignment} option; image file sections in PE are aligned to
4742 multiples of 4096, which is far too large an alignment for ordinary variables.
4743 It is rather the default alignment for (non-debug) sections within object
4744 (@samp{*.o}) files, which are less strictly aligned.}.
4748 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4749 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4753 @section @code{.data @var{subsection}}
4755 @cindex @code{data} directive
4756 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4757 end of the data subsection numbered @var{subsection} (which is an
4758 absolute expression). If @var{subsection} is omitted, it defaults
4763 @section @code{.def @var{name}}
4765 @cindex @code{def} directive
4766 @cindex COFF symbols, debugging
4767 @cindex debugging COFF symbols
4768 Begin defining debugging information for a symbol @var{name}; the
4769 definition extends until the @code{.endef} directive is encountered.
4772 This directive is only observed when @command{@value{AS}} is configured for COFF
4773 format output; when producing @code{b.out}, @samp{.def} is recognized,
4780 @section @code{.desc @var{symbol}, @var{abs-expression}}
4782 @cindex @code{desc} directive
4783 @cindex COFF symbol descriptor
4784 @cindex symbol descriptor, COFF
4785 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4786 to the low 16 bits of an absolute expression.
4789 The @samp{.desc} directive is not available when @command{@value{AS}} is
4790 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4791 object format. For the sake of compatibility, @command{@value{AS}} accepts
4792 it, but produces no output, when configured for COFF.
4798 @section @code{.dim}
4800 @cindex @code{dim} directive
4801 @cindex COFF auxiliary symbol information
4802 @cindex auxiliary symbol information, COFF
4803 This directive is generated by compilers to include auxiliary debugging
4804 information in the symbol table. It is only permitted inside
4805 @code{.def}/@code{.endef} pairs.
4808 @samp{.dim} is only meaningful when generating COFF format output; when
4809 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4815 @section @code{.double @var{flonums}}
4817 @cindex @code{double} directive
4818 @cindex floating point numbers (double)
4819 @code{.double} expects zero or more flonums, separated by commas. It
4820 assembles floating point numbers.
4822 The exact kind of floating point numbers emitted depends on how
4823 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4827 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4828 in @sc{ieee} format.
4833 @section @code{.eject}
4835 @cindex @code{eject} directive
4836 @cindex new page, in listings
4837 @cindex page, in listings
4838 @cindex listing control: new page
4839 Force a page break at this point, when generating assembly listings.
4842 @section @code{.else}
4844 @cindex @code{else} directive
4845 @code{.else} is part of the @command{@value{AS}} support for conditional
4846 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4847 of code to be assembled if the condition for the preceding @code{.if}
4851 @section @code{.elseif}
4853 @cindex @code{elseif} directive
4854 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4855 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4856 @code{.if} block that would otherwise fill the entire @code{.else} section.
4859 @section @code{.end}
4861 @cindex @code{end} directive
4862 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4863 process anything in the file past the @code{.end} directive.
4867 @section @code{.endef}
4869 @cindex @code{endef} directive
4870 This directive flags the end of a symbol definition begun with
4874 @samp{.endef} is only meaningful when generating COFF format output; if
4875 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4876 directive but ignores it.
4881 @section @code{.endfunc}
4882 @cindex @code{endfunc} directive
4883 @code{.endfunc} marks the end of a function specified with @code{.func}.
4886 @section @code{.endif}
4888 @cindex @code{endif} directive
4889 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4890 it marks the end of a block of code that is only assembled
4891 conditionally. @xref{If,,@code{.if}}.
4894 @section @code{.equ @var{symbol}, @var{expression}}
4896 @cindex @code{equ} directive
4897 @cindex assigning values to symbols
4898 @cindex symbols, assigning values to
4899 This directive sets the value of @var{symbol} to @var{expression}.
4900 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4903 The syntax for @code{equ} on the HPPA is
4904 @samp{@var{symbol} .equ @var{expression}}.
4908 The syntax for @code{equ} on the Z80 is
4909 @samp{@var{symbol} equ @var{expression}}.
4910 On the Z80 it is an eror if @var{symbol} is already defined,
4911 but the symbol is not protected from later redefinition.
4912 Compare @ref{Equiv}.
4916 @section @code{.equiv @var{symbol}, @var{expression}}
4917 @cindex @code{equiv} directive
4918 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4919 the assembler will signal an error if @var{symbol} is already defined. Note a
4920 symbol which has been referenced but not actually defined is considered to be
4923 Except for the contents of the error message, this is roughly equivalent to
4930 plus it protects the symbol from later redefinition.
4933 @section @code{.eqv @var{symbol}, @var{expression}}
4934 @cindex @code{eqv} directive
4935 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
4936 evaluate the expression or any part of it immediately. Instead each time
4937 the resulting symbol is used in an expression, a snapshot of its current
4941 @section @code{.err}
4942 @cindex @code{err} directive
4943 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4944 message and, unless the @option{-Z} option was used, it will not generate an
4945 object file. This can be used to signal an error in conditionally compiled code.
4948 @section @code{.error "@var{string}"}
4949 @cindex error directive
4951 Similarly to @code{.err}, this directive emits an error, but you can specify a
4952 string that will be emitted as the error message. If you don't specify the
4953 message, it defaults to @code{".error directive invoked in source file"}.
4954 @xref{Errors, ,Error and Warning Messages}.
4957 .error "This code has not been assembled and tested."
4961 @section @code{.exitm}
4962 Exit early from the current macro definition. @xref{Macro}.
4965 @section @code{.extern}
4967 @cindex @code{extern} directive
4968 @code{.extern} is accepted in the source program---for compatibility
4969 with other assemblers---but it is ignored. @command{@value{AS}} treats
4970 all undefined symbols as external.
4973 @section @code{.fail @var{expression}}
4975 @cindex @code{fail} directive
4976 Generates an error or a warning. If the value of the @var{expression} is 500
4977 or more, @command{@value{AS}} will print a warning message. If the value is less
4978 than 500, @command{@value{AS}} will print an error message. The message will
4979 include the value of @var{expression}. This can occasionally be useful inside
4980 complex nested macros or conditional assembly.
4983 @section @code{.file}
4984 @cindex @code{file} directive
4986 @ifclear no-file-dir
4987 There are two different versions of the @code{.file} directive. Targets
4988 that support DWARF2 line number information use the DWARF2 version of
4989 @code{.file}. Other targets use the default version.
4991 @subheading Default Version
4993 @cindex logical file name
4994 @cindex file name, logical
4995 This version of the @code{.file} directive tells @command{@value{AS}} that we
4996 are about to start a new logical file. The syntax is:
5002 @var{string} is the new file name. In general, the filename is
5003 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
5004 to specify an empty file name, you must give the quotes--@code{""}. This
5005 statement may go away in future: it is only recognized to be compatible with
5006 old @command{@value{AS}} programs.
5008 @subheading DWARF2 Version
5011 When emitting DWARF2 line number information, @code{.file} assigns filenames
5012 to the @code{.debug_line} file name table. The syntax is:
5015 .file @var{fileno} @var{filename}
5018 The @var{fileno} operand should be a unique positive integer to use as the
5019 index of the entry in the table. The @var{filename} operand is a C string
5022 The detail of filename indices is exposed to the user because the filename
5023 table is shared with the @code{.debug_info} section of the DWARF2 debugging
5024 information, and thus the user must know the exact indices that table
5028 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
5030 @cindex @code{fill} directive
5031 @cindex writing patterns in memory
5032 @cindex patterns, writing in memory
5033 @var{repeat}, @var{size} and @var{value} are absolute expressions.
5034 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
5035 may be zero or more. @var{Size} may be zero or more, but if it is
5036 more than 8, then it is deemed to have the value 8, compatible with
5037 other people's assemblers. The contents of each @var{repeat} bytes
5038 is taken from an 8-byte number. The highest order 4 bytes are
5039 zero. The lowest order 4 bytes are @var{value} rendered in the
5040 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
5041 Each @var{size} bytes in a repetition is taken from the lowest order
5042 @var{size} bytes of this number. Again, this bizarre behavior is
5043 compatible with other people's assemblers.
5045 @var{size} and @var{value} are optional.
5046 If the second comma and @var{value} are absent, @var{value} is
5047 assumed zero. If the first comma and following tokens are absent,
5048 @var{size} is assumed to be 1.
5051 @section @code{.float @var{flonums}}
5053 @cindex floating point numbers (single)
5054 @cindex @code{float} directive
5055 This directive assembles zero or more flonums, separated by commas. It
5056 has the same effect as @code{.single}.
5058 The exact kind of floating point numbers emitted depends on how
5059 @command{@value{AS}} is configured.
5060 @xref{Machine Dependencies}.
5064 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5065 in @sc{ieee} format.
5070 @section @code{.func @var{name}[,@var{label}]}
5071 @cindex @code{func} directive
5072 @code{.func} emits debugging information to denote function @var{name}, and
5073 is ignored unless the file is assembled with debugging enabled.
5074 Only @samp{--gstabs[+]} is currently supported.
5075 @var{label} is the entry point of the function and if omitted @var{name}
5076 prepended with the @samp{leading char} is used.
5077 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5078 All functions are currently defined to have @code{void} return type.
5079 The function must be terminated with @code{.endfunc}.
5082 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5084 @cindex @code{global} directive
5085 @cindex symbol, making visible to linker
5086 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5087 @var{symbol} in your partial program, its value is made available to
5088 other partial programs that are linked with it. Otherwise,
5089 @var{symbol} takes its attributes from a symbol of the same name
5090 from another file linked into the same program.
5092 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5093 compatibility with other assemblers.
5096 On the HPPA, @code{.global} is not always enough to make it accessible to other
5097 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5098 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5103 @section @code{.gnu_attribute @var{tag},@var{value}}
5104 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5107 @section @code{.hidden @var{names}}
5109 @cindex @code{hidden} directive
5111 This is one of the ELF visibility directives. The other two are
5112 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5113 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5115 This directive overrides the named symbols default visibility (which is set by
5116 their binding: local, global or weak). The directive sets the visibility to
5117 @code{hidden} which means that the symbols are not visible to other components.
5118 Such symbols are always considered to be @code{protected} as well.
5122 @section @code{.hword @var{expressions}}
5124 @cindex @code{hword} directive
5125 @cindex integers, 16-bit
5126 @cindex numbers, 16-bit
5127 @cindex sixteen bit integers
5128 This expects zero or more @var{expressions}, and emits
5129 a 16 bit number for each.
5132 This directive is a synonym for @samp{.short}; depending on the target
5133 architecture, it may also be a synonym for @samp{.word}.
5137 This directive is a synonym for @samp{.short}.
5140 This directive is a synonym for both @samp{.short} and @samp{.word}.
5145 @section @code{.ident}
5147 @cindex @code{ident} directive
5149 This directive is used by some assemblers to place tags in object files. The
5150 behavior of this directive varies depending on the target. When using the
5151 a.out object file format, @command{@value{AS}} simply accepts the directive for
5152 source-file compatibility with existing assemblers, but does not emit anything
5153 for it. When using COFF, comments are emitted to the @code{.comment} or
5154 @code{.rdata} section, depending on the target. When using ELF, comments are
5155 emitted to the @code{.comment} section.
5158 @section @code{.if @var{absolute expression}}
5160 @cindex conditional assembly
5161 @cindex @code{if} directive
5162 @code{.if} marks the beginning of a section of code which is only
5163 considered part of the source program being assembled if the argument
5164 (which must be an @var{absolute expression}) is non-zero. The end of
5165 the conditional section of code must be marked by @code{.endif}
5166 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5167 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5168 If you have several conditions to check, @code{.elseif} may be used to avoid
5169 nesting blocks if/else within each subsequent @code{.else} block.
5171 The following variants of @code{.if} are also supported:
5173 @cindex @code{ifdef} directive
5174 @item .ifdef @var{symbol}
5175 Assembles the following section of code if the specified @var{symbol}
5176 has been defined. Note a symbol which has been referenced but not yet defined
5177 is considered to be undefined.
5179 @cindex @code{ifb} directive
5180 @item .ifb @var{text}
5181 Assembles the following section of code if the operand is blank (empty).
5183 @cindex @code{ifc} directive
5184 @item .ifc @var{string1},@var{string2}
5185 Assembles the following section of code if the two strings are the same. The
5186 strings may be optionally quoted with single quotes. If they are not quoted,
5187 the first string stops at the first comma, and the second string stops at the
5188 end of the line. Strings which contain whitespace should be quoted. The
5189 string comparison is case sensitive.
5191 @cindex @code{ifeq} directive
5192 @item .ifeq @var{absolute expression}
5193 Assembles the following section of code if the argument is zero.
5195 @cindex @code{ifeqs} directive
5196 @item .ifeqs @var{string1},@var{string2}
5197 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5199 @cindex @code{ifge} directive
5200 @item .ifge @var{absolute expression}
5201 Assembles the following section of code if the argument is greater than or
5204 @cindex @code{ifgt} directive
5205 @item .ifgt @var{absolute expression}
5206 Assembles the following section of code if the argument is greater than zero.
5208 @cindex @code{ifle} directive
5209 @item .ifle @var{absolute expression}
5210 Assembles the following section of code if the argument is less than or equal
5213 @cindex @code{iflt} directive
5214 @item .iflt @var{absolute expression}
5215 Assembles the following section of code if the argument is less than zero.
5217 @cindex @code{ifnb} directive
5218 @item .ifnb @var{text}
5219 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5220 following section of code if the operand is non-blank (non-empty).
5222 @cindex @code{ifnc} directive
5223 @item .ifnc @var{string1},@var{string2}.
5224 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5225 following section of code if the two strings are not the same.
5227 @cindex @code{ifndef} directive
5228 @cindex @code{ifnotdef} directive
5229 @item .ifndef @var{symbol}
5230 @itemx .ifnotdef @var{symbol}
5231 Assembles the following section of code if the specified @var{symbol}
5232 has not been defined. Both spelling variants are equivalent. Note a symbol
5233 which has been referenced but not yet defined is considered to be undefined.
5235 @cindex @code{ifne} directive
5236 @item .ifne @var{absolute expression}
5237 Assembles the following section of code if the argument is not equal to zero
5238 (in other words, this is equivalent to @code{.if}).
5240 @cindex @code{ifnes} directive
5241 @item .ifnes @var{string1},@var{string2}
5242 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5243 following section of code if the two strings are not the same.
5247 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5249 @cindex @code{incbin} directive
5250 @cindex binary files, including
5251 The @code{incbin} directive includes @var{file} verbatim at the current
5252 location. You can control the search paths used with the @samp{-I} command-line
5253 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5256 The @var{skip} argument skips a number of bytes from the start of the
5257 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5258 read. Note that the data is not aligned in any way, so it is the user's
5259 responsibility to make sure that proper alignment is provided both before and
5260 after the @code{incbin} directive.
5263 @section @code{.include "@var{file}"}
5265 @cindex @code{include} directive
5266 @cindex supporting files, including
5267 @cindex files, including
5268 This directive provides a way to include supporting files at specified
5269 points in your source program. The code from @var{file} is assembled as
5270 if it followed the point of the @code{.include}; when the end of the
5271 included file is reached, assembly of the original file continues. You
5272 can control the search paths used with the @samp{-I} command-line option
5273 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5277 @section @code{.int @var{expressions}}
5279 @cindex @code{int} directive
5280 @cindex integers, 32-bit
5281 Expect zero or more @var{expressions}, of any section, separated by commas.
5282 For each expression, emit a number that, at run time, is the value of that
5283 expression. The byte order and bit size of the number depends on what kind
5284 of target the assembly is for.
5288 On most forms of the H8/300, @code{.int} emits 16-bit
5289 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5296 @section @code{.internal @var{names}}
5298 @cindex @code{internal} directive
5300 This is one of the ELF visibility directives. The other two are
5301 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5302 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5304 This directive overrides the named symbols default visibility (which is set by
5305 their binding: local, global or weak). The directive sets the visibility to
5306 @code{internal} which means that the symbols are considered to be @code{hidden}
5307 (i.e., not visible to other components), and that some extra, processor specific
5308 processing must also be performed upon the symbols as well.
5312 @section @code{.irp @var{symbol},@var{values}}@dots{}
5314 @cindex @code{irp} directive
5315 Evaluate a sequence of statements assigning different values to @var{symbol}.
5316 The sequence of statements starts at the @code{.irp} directive, and is
5317 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5318 set to @var{value}, and the sequence of statements is assembled. If no
5319 @var{value} is listed, the sequence of statements is assembled once, with
5320 @var{symbol} set to the null string. To refer to @var{symbol} within the
5321 sequence of statements, use @var{\symbol}.
5323 For example, assembling
5331 is equivalent to assembling
5339 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5342 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5344 @cindex @code{irpc} directive
5345 Evaluate a sequence of statements assigning different values to @var{symbol}.
5346 The sequence of statements starts at the @code{.irpc} directive, and is
5347 terminated by an @code{.endr} directive. For each character in @var{value},
5348 @var{symbol} is set to the character, and the sequence of statements is
5349 assembled. If no @var{value} is listed, the sequence of statements is
5350 assembled once, with @var{symbol} set to the null string. To refer to
5351 @var{symbol} within the sequence of statements, use @var{\symbol}.
5353 For example, assembling
5361 is equivalent to assembling
5369 For some caveats with the spelling of @var{symbol}, see also the discussion
5373 @section @code{.lcomm @var{symbol} , @var{length}}
5375 @cindex @code{lcomm} directive
5376 @cindex local common symbols
5377 @cindex symbols, local common
5378 Reserve @var{length} (an absolute expression) bytes for a local common
5379 denoted by @var{symbol}. The section and value of @var{symbol} are
5380 those of the new local common. The addresses are allocated in the bss
5381 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5382 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5383 not visible to @code{@value{LD}}.
5386 Some targets permit a third argument to be used with @code{.lcomm}. This
5387 argument specifies the desired alignment of the symbol in the bss section.
5391 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5392 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5396 @section @code{.lflags}
5398 @cindex @code{lflags} directive (ignored)
5399 @command{@value{AS}} accepts this directive, for compatibility with other
5400 assemblers, but ignores it.
5402 @ifclear no-line-dir
5404 @section @code{.line @var{line-number}}
5406 @cindex @code{line} directive
5407 @cindex logical line number
5409 Change the logical line number. @var{line-number} must be an absolute
5410 expression. The next line has that logical line number. Therefore any other
5411 statements on the current line (after a statement separator character) are
5412 reported as on logical line number @var{line-number} @minus{} 1. One day
5413 @command{@value{AS}} will no longer support this directive: it is recognized only
5414 for compatibility with existing assembler programs.
5417 Even though this is a directive associated with the @code{a.out} or
5418 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5419 when producing COFF output, and treats @samp{.line} as though it
5420 were the COFF @samp{.ln} @emph{if} it is found outside a
5421 @code{.def}/@code{.endef} pair.
5423 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5424 used by compilers to generate auxiliary symbol information for
5429 @section @code{.linkonce [@var{type}]}
5431 @cindex @code{linkonce} directive
5432 @cindex common sections
5433 Mark the current section so that the linker only includes a single copy of it.
5434 This may be used to include the same section in several different object files,
5435 but ensure that the linker will only include it once in the final output file.
5436 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5437 Duplicate sections are detected based on the section name, so it should be
5440 This directive is only supported by a few object file formats; as of this
5441 writing, the only object file format which supports it is the Portable
5442 Executable format used on Windows NT.
5444 The @var{type} argument is optional. If specified, it must be one of the
5445 following strings. For example:
5449 Not all types may be supported on all object file formats.
5453 Silently discard duplicate sections. This is the default.
5456 Warn if there are duplicate sections, but still keep only one copy.
5459 Warn if any of the duplicates have different sizes.
5462 Warn if any of the duplicates do not have exactly the same contents.
5466 @section @code{.list}
5468 @cindex @code{list} directive
5469 @cindex listing control, turning on
5470 Control (in conjunction with the @code{.nolist} directive) whether or
5471 not assembly listings are generated. These two directives maintain an
5472 internal counter (which is zero initially). @code{.list} increments the
5473 counter, and @code{.nolist} decrements it. Assembly listings are
5474 generated whenever the counter is greater than zero.
5476 By default, listings are disabled. When you enable them (with the
5477 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5478 the initial value of the listing counter is one.
5481 @section @code{.ln @var{line-number}}
5483 @cindex @code{ln} directive
5484 @ifclear no-line-dir
5485 @samp{.ln} is a synonym for @samp{.line}.
5488 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5489 must be an absolute expression. The next line has that logical
5490 line number, so any other statements on the current line (after a
5491 statement separator character @code{;}) are reported as on logical
5492 line number @var{line-number} @minus{} 1.
5495 This directive is accepted, but ignored, when @command{@value{AS}} is
5496 configured for @code{b.out}; its effect is only associated with COFF
5502 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5503 @cindex @code{loc} directive
5504 When emitting DWARF2 line number information,
5505 the @code{.loc} directive will add a row to the @code{.debug_line} line
5506 number matrix corresponding to the immediately following assembly
5507 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5508 arguments will be applied to the @code{.debug_line} state machine before
5511 The @var{options} are a sequence of the following tokens in any order:
5515 This option will set the @code{basic_block} register in the
5516 @code{.debug_line} state machine to @code{true}.
5519 This option will set the @code{prologue_end} register in the
5520 @code{.debug_line} state machine to @code{true}.
5522 @item epilogue_begin
5523 This option will set the @code{epilogue_begin} register in the
5524 @code{.debug_line} state machine to @code{true}.
5526 @item is_stmt @var{value}
5527 This option will set the @code{is_stmt} register in the
5528 @code{.debug_line} state machine to @code{value}, which must be
5531 @item isa @var{value}
5532 This directive will set the @code{isa} register in the @code{.debug_line}
5533 state machine to @var{value}, which must be an unsigned integer.
5535 @item discriminator @var{value}
5536 This directive will set the @code{discriminator} register in the @code{.debug_line}
5537 state machine to @var{value}, which must be an unsigned integer.
5541 @node Loc_mark_labels
5542 @section @code{.loc_mark_labels @var{enable}}
5543 @cindex @code{loc_mark_labels} directive
5544 When emitting DWARF2 line number information,
5545 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5546 to the @code{.debug_line} line number matrix with the @code{basic_block}
5547 register in the state machine set whenever a code label is seen.
5548 The @var{enable} argument should be either 1 or 0, to enable or disable
5549 this function respectively.
5553 @section @code{.local @var{names}}
5555 @cindex @code{local} directive
5556 This directive, which is available for ELF targets, marks each symbol in
5557 the comma-separated list of @code{names} as a local symbol so that it
5558 will not be externally visible. If the symbols do not already exist,
5559 they will be created.
5561 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5562 accept an alignment argument, which is the case for most ELF targets,
5563 the @code{.local} directive can be used in combination with @code{.comm}
5564 (@pxref{Comm}) to define aligned local common data.
5568 @section @code{.long @var{expressions}}
5570 @cindex @code{long} directive
5571 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5574 @c no one seems to know what this is for or whether this description is
5575 @c what it really ought to do
5577 @section @code{.lsym @var{symbol}, @var{expression}}
5579 @cindex @code{lsym} directive
5580 @cindex symbol, not referenced in assembly
5581 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5582 the hash table, ensuring it cannot be referenced by name during the
5583 rest of the assembly. This sets the attributes of the symbol to be
5584 the same as the expression value:
5586 @var{other} = @var{descriptor} = 0
5587 @var{type} = @r{(section of @var{expression})}
5588 @var{value} = @var{expression}
5591 The new symbol is not flagged as external.
5595 @section @code{.macro}
5598 The commands @code{.macro} and @code{.endm} allow you to define macros that
5599 generate assembly output. For example, this definition specifies a macro
5600 @code{sum} that puts a sequence of numbers into memory:
5603 .macro sum from=0, to=5
5612 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5624 @item .macro @var{macname}
5625 @itemx .macro @var{macname} @var{macargs} @dots{}
5626 @cindex @code{macro} directive
5627 Begin the definition of a macro called @var{macname}. If your macro
5628 definition requires arguments, specify their names after the macro name,
5629 separated by commas or spaces. You can qualify the macro argument to
5630 indicate whether all invocations must specify a non-blank value (through
5631 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5632 (through @samp{:@code{vararg}}). You can supply a default value for any
5633 macro argument by following the name with @samp{=@var{deflt}}. You
5634 cannot define two macros with the same @var{macname} unless it has been
5635 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5636 definitions. For example, these are all valid @code{.macro} statements:
5640 Begin the definition of a macro called @code{comm}, which takes no
5643 @item .macro plus1 p, p1
5644 @itemx .macro plus1 p p1
5645 Either statement begins the definition of a macro called @code{plus1},
5646 which takes two arguments; within the macro definition, write
5647 @samp{\p} or @samp{\p1} to evaluate the arguments.
5649 @item .macro reserve_str p1=0 p2
5650 Begin the definition of a macro called @code{reserve_str}, with two
5651 arguments. The first argument has a default value, but not the second.
5652 After the definition is complete, you can call the macro either as
5653 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5654 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5655 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5656 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5658 @item .macro m p1:req, p2=0, p3:vararg
5659 Begin the definition of a macro called @code{m}, with at least three
5660 arguments. The first argument must always have a value specified, but
5661 not the second, which instead has a default value. The third formal
5662 will get assigned all remaining arguments specified at invocation time.
5664 When you call a macro, you can specify the argument values either by
5665 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5666 @samp{sum to=17, from=9}.
5670 Note that since each of the @var{macargs} can be an identifier exactly
5671 as any other one permitted by the target architecture, there may be
5672 occasional problems if the target hand-crafts special meanings to certain
5673 characters when they occur in a special position. For example, if the colon
5674 (@code{:}) is generally permitted to be part of a symbol name, but the
5675 architecture specific code special-cases it when occurring as the final
5676 character of a symbol (to denote a label), then the macro parameter
5677 replacement code will have no way of knowing that and consider the whole
5678 construct (including the colon) an identifier, and check only this
5679 identifier for being the subject to parameter substitution. So for example
5680 this macro definition:
5688 might not work as expected. Invoking @samp{label foo} might not create a label
5689 called @samp{foo} but instead just insert the text @samp{\l:} into the
5690 assembler source, probably generating an error about an unrecognised
5693 Similarly problems might occur with the period character (@samp{.})
5694 which is often allowed inside opcode names (and hence identifier names). So
5695 for example constructing a macro to build an opcode from a base name and a
5696 length specifier like this:
5699 .macro opcode base length
5704 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5705 instruction but instead generate some kind of error as the assembler tries to
5706 interpret the text @samp{\base.\length}.
5708 There are several possible ways around this problem:
5711 @item Insert white space
5712 If it is possible to use white space characters then this is the simplest
5721 @item Use @samp{\()}
5722 The string @samp{\()} can be used to separate the end of a macro argument from
5723 the following text. eg:
5726 .macro opcode base length
5731 @item Use the alternate macro syntax mode
5732 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5733 used as a separator. eg:
5743 Note: this problem of correctly identifying string parameters to pseudo ops
5744 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5745 and @code{.irpc} (@pxref{Irpc}) as well.
5748 @cindex @code{endm} directive
5749 Mark the end of a macro definition.
5752 @cindex @code{exitm} directive
5753 Exit early from the current macro definition.
5755 @cindex number of macros executed
5756 @cindex macros, count executed
5758 @command{@value{AS}} maintains a counter of how many macros it has
5759 executed in this pseudo-variable; you can copy that number to your
5760 output with @samp{\@@}, but @emph{only within a macro definition}.
5762 @item LOCAL @var{name} [ , @dots{} ]
5763 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5764 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5765 @xref{Altmacro,,@code{.altmacro}}.
5769 @section @code{.mri @var{val}}
5771 @cindex @code{mri} directive
5772 @cindex MRI mode, temporarily
5773 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5774 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5775 affects code assembled until the next @code{.mri} directive, or until the end
5776 of the file. @xref{M, MRI mode, MRI mode}.
5779 @section @code{.noaltmacro}
5780 Disable alternate macro mode. @xref{Altmacro}.
5783 @section @code{.nolist}
5785 @cindex @code{nolist} directive
5786 @cindex listing control, turning off
5787 Control (in conjunction with the @code{.list} directive) whether or
5788 not assembly listings are generated. These two directives maintain an
5789 internal counter (which is zero initially). @code{.list} increments the
5790 counter, and @code{.nolist} decrements it. Assembly listings are
5791 generated whenever the counter is greater than zero.
5794 @section @code{.octa @var{bignums}}
5796 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5797 @cindex @code{octa} directive
5798 @cindex integer, 16-byte
5799 @cindex sixteen byte integer
5800 This directive expects zero or more bignums, separated by commas. For each
5801 bignum, it emits a 16-byte integer.
5803 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5804 hence @emph{octa}-word for 16 bytes.
5807 @section @code{.offset @var{loc}}
5809 @cindex @code{offset} directive
5810 Set the location counter to @var{loc} in the absolute section. @var{loc} must
5811 be an absolute expression. This directive may be useful for defining
5812 symbols with absolute values. Do not confuse it with the @code{.org}
5816 @section @code{.org @var{new-lc} , @var{fill}}
5818 @cindex @code{org} directive
5819 @cindex location counter, advancing
5820 @cindex advancing location counter
5821 @cindex current address, advancing
5822 Advance the location counter of the current section to
5823 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5824 expression with the same section as the current subsection. That is,
5825 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5826 wrong section, the @code{.org} directive is ignored. To be compatible
5827 with former assemblers, if the section of @var{new-lc} is absolute,
5828 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5829 is the same as the current subsection.
5831 @code{.org} may only increase the location counter, or leave it
5832 unchanged; you cannot use @code{.org} to move the location counter
5835 @c double negative used below "not undefined" because this is a specific
5836 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5837 @c section. doc@cygnus.com 18feb91
5838 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5839 may not be undefined. If you really detest this restriction we eagerly await
5840 a chance to share your improved assembler.
5842 Beware that the origin is relative to the start of the section, not
5843 to the start of the subsection. This is compatible with other
5844 people's assemblers.
5846 When the location counter (of the current subsection) is advanced, the
5847 intervening bytes are filled with @var{fill} which should be an
5848 absolute expression. If the comma and @var{fill} are omitted,
5849 @var{fill} defaults to zero.
5852 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5854 @cindex padding the location counter given a power of two
5855 @cindex @code{p2align} directive
5856 Pad the location counter (in the current subsection) to a particular
5857 storage boundary. The first expression (which must be absolute) is the
5858 number of low-order zero bits the location counter must have after
5859 advancement. For example @samp{.p2align 3} advances the location
5860 counter until it a multiple of 8. If the location counter is already a
5861 multiple of 8, no change is needed.
5863 The second expression (also absolute) gives the fill value to be stored in the
5864 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5865 padding bytes are normally zero. However, on some systems, if the section is
5866 marked as containing code and the fill value is omitted, the space is filled
5867 with no-op instructions.
5869 The third expression is also absolute, and is also optional. If it is present,
5870 it is the maximum number of bytes that should be skipped by this alignment
5871 directive. If doing the alignment would require skipping more bytes than the
5872 specified maximum, then the alignment is not done at all. You can omit the
5873 fill value (the second argument) entirely by simply using two commas after the
5874 required alignment; this can be useful if you want the alignment to be filled
5875 with no-op instructions when appropriate.
5877 @cindex @code{p2alignw} directive
5878 @cindex @code{p2alignl} directive
5879 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5880 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5881 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5882 fill pattern as a four byte longword value. For example, @code{.p2alignw
5883 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5884 filled in with the value 0x368d (the exact placement of the bytes depends upon
5885 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5890 @section @code{.popsection}
5892 @cindex @code{popsection} directive
5893 @cindex Section Stack
5894 This is one of the ELF section stack manipulation directives. The others are
5895 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5896 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5899 This directive replaces the current section (and subsection) with the top
5900 section (and subsection) on the section stack. This section is popped off the
5906 @section @code{.previous}
5908 @cindex @code{previous} directive
5909 @cindex Section Stack
5910 This is one of the ELF section stack manipulation directives. The others are
5911 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5912 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5913 (@pxref{PopSection}).
5915 This directive swaps the current section (and subsection) with most recently
5916 referenced section/subsection pair prior to this one. Multiple
5917 @code{.previous} directives in a row will flip between two sections (and their
5918 subsections). For example:
5930 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
5936 # Now in section A subsection 1
5940 # Now in section B subsection 0
5943 # Now in section B subsection 1
5946 # Now in section B subsection 0
5950 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
5951 section B and 0x9abc into subsection 1 of section B.
5953 In terms of the section stack, this directive swaps the current section with
5954 the top section on the section stack.
5958 @section @code{.print @var{string}}
5960 @cindex @code{print} directive
5961 @command{@value{AS}} will print @var{string} on the standard output during
5962 assembly. You must put @var{string} in double quotes.
5966 @section @code{.protected @var{names}}
5968 @cindex @code{protected} directive
5970 This is one of the ELF visibility directives. The other two are
5971 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5973 This directive overrides the named symbols default visibility (which is set by
5974 their binding: local, global or weak). The directive sets the visibility to
5975 @code{protected} which means that any references to the symbols from within the
5976 components that defines them must be resolved to the definition in that
5977 component, even if a definition in another component would normally preempt
5982 @section @code{.psize @var{lines} , @var{columns}}
5984 @cindex @code{psize} directive
5985 @cindex listing control: paper size
5986 @cindex paper size, for listings
5987 Use this directive to declare the number of lines---and, optionally, the
5988 number of columns---to use for each page, when generating listings.
5990 If you do not use @code{.psize}, listings use a default line-count
5991 of 60. You may omit the comma and @var{columns} specification; the
5992 default width is 200 columns.
5994 @command{@value{AS}} generates formfeeds whenever the specified number of
5995 lines is exceeded (or whenever you explicitly request one, using
5998 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5999 those explicitly specified with @code{.eject}.
6002 @section @code{.purgem @var{name}}
6004 @cindex @code{purgem} directive
6005 Undefine the macro @var{name}, so that later uses of the string will not be
6006 expanded. @xref{Macro}.
6010 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
6012 @cindex @code{pushsection} directive
6013 @cindex Section Stack
6014 This is one of the ELF section stack manipulation directives. The others are
6015 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6016 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6019 This directive pushes the current section (and subsection) onto the
6020 top of the section stack, and then replaces the current section and
6021 subsection with @code{name} and @code{subsection}. The optional
6022 @code{flags}, @code{type} and @code{arguments} are treated the same
6023 as in the @code{.section} (@pxref{Section}) directive.
6027 @section @code{.quad @var{bignums}}
6029 @cindex @code{quad} directive
6030 @code{.quad} expects zero or more bignums, separated by commas. For
6031 each bignum, it emits
6033 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
6034 warning message; and just takes the lowest order 8 bytes of the bignum.
6035 @cindex eight-byte integer
6036 @cindex integer, 8-byte
6038 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
6039 hence @emph{quad}-word for 8 bytes.
6042 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
6043 warning message; and just takes the lowest order 16 bytes of the bignum.
6044 @cindex sixteen-byte integer
6045 @cindex integer, 16-byte
6049 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6051 @cindex @code{reloc} directive
6052 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6053 @var{expression}. If @var{offset} is a number, the relocation is generated in
6054 the current section. If @var{offset} is an expression that resolves to a
6055 symbol plus offset, the relocation is generated in the given symbol's section.
6056 @var{expression}, if present, must resolve to a symbol plus addend or to an
6057 absolute value, but note that not all targets support an addend. e.g. ELF REL
6058 targets such as i386 store an addend in the section contents rather than in the
6059 relocation. This low level interface does not support addends stored in the
6063 @section @code{.rept @var{count}}
6065 @cindex @code{rept} directive
6066 Repeat the sequence of lines between the @code{.rept} directive and the next
6067 @code{.endr} directive @var{count} times.
6069 For example, assembling
6077 is equivalent to assembling
6086 @section @code{.sbttl "@var{subheading}"}
6088 @cindex @code{sbttl} directive
6089 @cindex subtitles for listings
6090 @cindex listing control: subtitle
6091 Use @var{subheading} as the title (third line, immediately after the
6092 title line) when generating assembly listings.
6094 This directive affects subsequent pages, as well as the current page if
6095 it appears within ten lines of the top of a page.
6099 @section @code{.scl @var{class}}
6101 @cindex @code{scl} directive
6102 @cindex symbol storage class (COFF)
6103 @cindex COFF symbol storage class
6104 Set the storage-class value for a symbol. This directive may only be
6105 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6106 whether a symbol is static or external, or it may record further
6107 symbolic debugging information.
6110 The @samp{.scl} directive is primarily associated with COFF output; when
6111 configured to generate @code{b.out} output format, @command{@value{AS}}
6112 accepts this directive but ignores it.
6118 @section @code{.section @var{name}}
6120 @cindex named section
6121 Use the @code{.section} directive to assemble the following code into a section
6124 This directive is only supported for targets that actually support arbitrarily
6125 named sections; on @code{a.out} targets, for example, it is not accepted, even
6126 with a standard @code{a.out} section name.
6130 @c only print the extra heading if both COFF and ELF are set
6131 @subheading COFF Version
6134 @cindex @code{section} directive (COFF version)
6135 For COFF targets, the @code{.section} directive is used in one of the following
6139 .section @var{name}[, "@var{flags}"]
6140 .section @var{name}[, @var{subsection}]
6143 If the optional argument is quoted, it is taken as flags to use for the
6144 section. Each flag is a single character. The following flags are recognized:
6147 bss section (uninitialized data)
6149 section is not loaded
6155 exclude section from linking
6161 shared section (meaningful for PE targets)
6163 ignored. (For compatibility with the ELF version)
6165 section is not readable (meaningful for PE targets)
6167 single-digit power-of-two section alignment (GNU extension)
6170 If no flags are specified, the default flags depend upon the section name. If
6171 the section name is not recognized, the default will be for the section to be
6172 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6173 from the section, rather than adding them, so if they are used on their own it
6174 will be as if no flags had been specified at all.
6176 If the optional argument to the @code{.section} directive is not quoted, it is
6177 taken as a subsection number (@pxref{Sub-Sections}).
6182 @c only print the extra heading if both COFF and ELF are set
6183 @subheading ELF Version
6186 @cindex Section Stack
6187 This is one of the ELF section stack manipulation directives. The others are
6188 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6189 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6190 @code{.previous} (@pxref{Previous}).
6192 @cindex @code{section} directive (ELF version)
6193 For ELF targets, the @code{.section} directive is used like this:
6196 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6199 The optional @var{flags} argument is a quoted string which may contain any
6200 combination of the following characters:
6203 section is allocatable
6205 section is excluded from executable and shared library.
6209 section is executable
6211 section is mergeable
6213 section contains zero terminated strings
6215 section is a member of a section group
6217 section is used for thread-local-storage
6219 section is a member of the previously-current section's group, if any
6222 The optional @var{type} argument may contain one of the following constants:
6225 section contains data
6227 section does not contain data (i.e., section only occupies space)
6229 section contains data which is used by things other than the program
6231 section contains an array of pointers to init functions
6233 section contains an array of pointers to finish functions
6234 @item @@preinit_array
6235 section contains an array of pointers to pre-init functions
6238 Many targets only support the first three section types.
6240 Note on targets where the @code{@@} character is the start of a comment (eg
6241 ARM) then another character is used instead. For example the ARM port uses the
6244 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6245 be specified as well as an extra argument---@var{entsize}---like this:
6248 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6251 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6252 constants, each @var{entsize} octets long. Sections with both @code{M} and
6253 @code{S} must contain zero terminated strings where each character is
6254 @var{entsize} bytes long. The linker may remove duplicates within sections with
6255 the same name, same entity size and same flags. @var{entsize} must be an
6256 absolute expression. For sections with both @code{M} and @code{S}, a string
6257 which is a suffix of a larger string is considered a duplicate. Thus
6258 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6259 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6261 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6262 be present along with an additional field like this:
6265 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6268 The @var{GroupName} field specifies the name of the section group to which this
6269 particular section belongs. The optional linkage field can contain:
6272 indicates that only one copy of this section should be retained
6277 Note: if both the @var{M} and @var{G} flags are present then the fields for
6278 the Merge flag should come first, like this:
6281 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6284 If @var{flags} contains the @code{?} symbol then it may not also contain the
6285 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6286 present. Instead, @code{?} says to consider the section that's current before
6287 this directive. If that section used @code{G}, then the new section will use
6288 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6289 If not, then the @code{?} symbol has no effect.
6291 If no flags are specified, the default flags depend upon the section name. If
6292 the section name is not recognized, the default will be for the section to have
6293 none of the above flags: it will not be allocated in memory, nor writable, nor
6294 executable. The section will contain data.
6296 For ELF targets, the assembler supports another type of @code{.section}
6297 directive for compatibility with the Solaris assembler:
6300 .section "@var{name}"[, @var{flags}...]
6303 Note that the section name is quoted. There may be a sequence of comma
6307 section is allocatable
6311 section is executable
6313 section is excluded from executable and shared library.
6315 section is used for thread local storage
6318 This directive replaces the current section and subsection. See the
6319 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6320 some examples of how this directive and the other section stack directives
6326 @section @code{.set @var{symbol}, @var{expression}}
6328 @cindex @code{set} directive
6329 @cindex symbol value, setting
6330 Set the value of @var{symbol} to @var{expression}. This
6331 changes @var{symbol}'s value and type to conform to
6332 @var{expression}. If @var{symbol} was flagged as external, it remains
6333 flagged (@pxref{Symbol Attributes}).
6335 You may @code{.set} a symbol many times in the same assembly.
6337 If you @code{.set} a global symbol, the value stored in the object
6338 file is the last value stored into it.
6341 On Z80 @code{set} is a real instruction, use
6342 @samp{@var{symbol} defl @var{expression}} instead.
6346 @section @code{.short @var{expressions}}
6348 @cindex @code{short} directive
6350 @code{.short} is normally the same as @samp{.word}.
6351 @xref{Word,,@code{.word}}.
6353 In some configurations, however, @code{.short} and @code{.word} generate
6354 numbers of different lengths. @xref{Machine Dependencies}.
6358 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6361 This expects zero or more @var{expressions}, and emits
6362 a 16 bit number for each.
6367 @section @code{.single @var{flonums}}
6369 @cindex @code{single} directive
6370 @cindex floating point numbers (single)
6371 This directive assembles zero or more flonums, separated by commas. It
6372 has the same effect as @code{.float}.
6374 The exact kind of floating point numbers emitted depends on how
6375 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6379 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6380 numbers in @sc{ieee} format.
6386 @section @code{.size}
6388 This directive is used to set the size associated with a symbol.
6392 @c only print the extra heading if both COFF and ELF are set
6393 @subheading COFF Version
6396 @cindex @code{size} directive (COFF version)
6397 For COFF targets, the @code{.size} directive is only permitted inside
6398 @code{.def}/@code{.endef} pairs. It is used like this:
6401 .size @var{expression}
6405 @samp{.size} is only meaningful when generating COFF format output; when
6406 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6413 @c only print the extra heading if both COFF and ELF are set
6414 @subheading ELF Version
6417 @cindex @code{size} directive (ELF version)
6418 For ELF targets, the @code{.size} directive is used like this:
6421 .size @var{name} , @var{expression}
6424 This directive sets the size associated with a symbol @var{name}.
6425 The size in bytes is computed from @var{expression} which can make use of label
6426 arithmetic. This directive is typically used to set the size of function
6431 @ifclear no-space-dir
6433 @section @code{.skip @var{size} , @var{fill}}
6435 @cindex @code{skip} directive
6436 @cindex filling memory
6437 This directive emits @var{size} bytes, each of value @var{fill}. Both
6438 @var{size} and @var{fill} are absolute expressions. If the comma and
6439 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6444 @section @code{.sleb128 @var{expressions}}
6446 @cindex @code{sleb128} directive
6447 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6448 compact, variable length representation of numbers used by the DWARF
6449 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6451 @ifclear no-space-dir
6453 @section @code{.space @var{size} , @var{fill}}
6455 @cindex @code{space} directive
6456 @cindex filling memory
6457 This directive emits @var{size} bytes, each of value @var{fill}. Both
6458 @var{size} and @var{fill} are absolute expressions. If the comma
6459 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6464 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6465 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6466 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6467 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6475 @section @code{.stabd, .stabn, .stabs}
6477 @cindex symbolic debuggers, information for
6478 @cindex @code{stab@var{x}} directives
6479 There are three directives that begin @samp{.stab}.
6480 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6481 The symbols are not entered in the @command{@value{AS}} hash table: they
6482 cannot be referenced elsewhere in the source file.
6483 Up to five fields are required:
6487 This is the symbol's name. It may contain any character except
6488 @samp{\000}, so is more general than ordinary symbol names. Some
6489 debuggers used to code arbitrarily complex structures into symbol names
6493 An absolute expression. The symbol's type is set to the low 8 bits of
6494 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6495 and debuggers choke on silly bit patterns.
6498 An absolute expression. The symbol's ``other'' attribute is set to the
6499 low 8 bits of this expression.
6502 An absolute expression. The symbol's descriptor is set to the low 16
6503 bits of this expression.
6506 An absolute expression which becomes the symbol's value.
6509 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6510 or @code{.stabs} statement, the symbol has probably already been created;
6511 you get a half-formed symbol in your object file. This is
6512 compatible with earlier assemblers!
6515 @cindex @code{stabd} directive
6516 @item .stabd @var{type} , @var{other} , @var{desc}
6518 The ``name'' of the symbol generated is not even an empty string.
6519 It is a null pointer, for compatibility. Older assemblers used a
6520 null pointer so they didn't waste space in object files with empty
6523 The symbol's value is set to the location counter,
6524 relocatably. When your program is linked, the value of this symbol
6525 is the address of the location counter when the @code{.stabd} was
6528 @cindex @code{stabn} directive
6529 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6530 The name of the symbol is set to the empty string @code{""}.
6532 @cindex @code{stabs} directive
6533 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6534 All five fields are specified.
6540 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6541 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6543 @cindex string, copying to object file
6544 @cindex string8, copying to object file
6545 @cindex string16, copying to object file
6546 @cindex string32, copying to object file
6547 @cindex string64, copying to object file
6548 @cindex @code{string} directive
6549 @cindex @code{string8} directive
6550 @cindex @code{string16} directive
6551 @cindex @code{string32} directive
6552 @cindex @code{string64} directive
6554 Copy the characters in @var{str} to the object file. You may specify more than
6555 one string to copy, separated by commas. Unless otherwise specified for a
6556 particular machine, the assembler marks the end of each string with a 0 byte.
6557 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6559 The variants @code{string16}, @code{string32} and @code{string64} differ from
6560 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6561 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6562 are stored in target endianness byte order.
6568 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6569 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6574 @section @code{.struct @var{expression}}
6576 @cindex @code{struct} directive
6577 Switch to the absolute section, and set the section offset to @var{expression},
6578 which must be an absolute expression. You might use this as follows:
6587 This would define the symbol @code{field1} to have the value 0, the symbol
6588 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6589 value 8. Assembly would be left in the absolute section, and you would need to
6590 use a @code{.section} directive of some sort to change to some other section
6591 before further assembly.
6595 @section @code{.subsection @var{name}}
6597 @cindex @code{subsection} directive
6598 @cindex Section Stack
6599 This is one of the ELF section stack manipulation directives. The others are
6600 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6601 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6604 This directive replaces the current subsection with @code{name}. The current
6605 section is not changed. The replaced subsection is put onto the section stack
6606 in place of the then current top of stack subsection.
6611 @section @code{.symver}
6612 @cindex @code{symver} directive
6613 @cindex symbol versioning
6614 @cindex versions of symbols
6615 Use the @code{.symver} directive to bind symbols to specific version nodes
6616 within a source file. This is only supported on ELF platforms, and is
6617 typically used when assembling files to be linked into a shared library.
6618 There are cases where it may make sense to use this in objects to be bound
6619 into an application itself so as to override a versioned symbol from a
6622 For ELF targets, the @code{.symver} directive can be used like this:
6624 .symver @var{name}, @var{name2@@nodename}
6626 If the symbol @var{name} is defined within the file
6627 being assembled, the @code{.symver} directive effectively creates a symbol
6628 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6629 just don't try and create a regular alias is that the @var{@@} character isn't
6630 permitted in symbol names. The @var{name2} part of the name is the actual name
6631 of the symbol by which it will be externally referenced. The name @var{name}
6632 itself is merely a name of convenience that is used so that it is possible to
6633 have definitions for multiple versions of a function within a single source
6634 file, and so that the compiler can unambiguously know which version of a
6635 function is being mentioned. The @var{nodename} portion of the alias should be
6636 the name of a node specified in the version script supplied to the linker when
6637 building a shared library. If you are attempting to override a versioned
6638 symbol from a shared library, then @var{nodename} should correspond to the
6639 nodename of the symbol you are trying to override.
6641 If the symbol @var{name} is not defined within the file being assembled, all
6642 references to @var{name} will be changed to @var{name2@@nodename}. If no
6643 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6646 Another usage of the @code{.symver} directive is:
6648 .symver @var{name}, @var{name2@@@@nodename}
6650 In this case, the symbol @var{name} must exist and be defined within
6651 the file being assembled. It is similar to @var{name2@@nodename}. The
6652 difference is @var{name2@@@@nodename} will also be used to resolve
6653 references to @var{name2} by the linker.
6655 The third usage of the @code{.symver} directive is:
6657 .symver @var{name}, @var{name2@@@@@@nodename}
6659 When @var{name} is not defined within the
6660 file being assembled, it is treated as @var{name2@@nodename}. When
6661 @var{name} is defined within the file being assembled, the symbol
6662 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6667 @section @code{.tag @var{structname}}
6669 @cindex COFF structure debugging
6670 @cindex structure debugging, COFF
6671 @cindex @code{tag} directive
6672 This directive is generated by compilers to include auxiliary debugging
6673 information in the symbol table. It is only permitted inside
6674 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6675 definitions in the symbol table with instances of those structures.
6678 @samp{.tag} is only used when generating COFF format output; when
6679 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6685 @section @code{.text @var{subsection}}
6687 @cindex @code{text} directive
6688 Tells @command{@value{AS}} to assemble the following statements onto the end of
6689 the text subsection numbered @var{subsection}, which is an absolute
6690 expression. If @var{subsection} is omitted, subsection number zero
6694 @section @code{.title "@var{heading}"}
6696 @cindex @code{title} directive
6697 @cindex listing control: title line
6698 Use @var{heading} as the title (second line, immediately after the
6699 source file name and pagenumber) when generating assembly listings.
6701 This directive affects subsequent pages, as well as the current page if
6702 it appears within ten lines of the top of a page.
6706 @section @code{.type}
6708 This directive is used to set the type of a symbol.
6712 @c only print the extra heading if both COFF and ELF are set
6713 @subheading COFF Version
6716 @cindex COFF symbol type
6717 @cindex symbol type, COFF
6718 @cindex @code{type} directive (COFF version)
6719 For COFF targets, this directive is permitted only within
6720 @code{.def}/@code{.endef} pairs. It is used like this:
6726 This records the integer @var{int} as the type attribute of a symbol table
6730 @samp{.type} is associated only with COFF format output; when
6731 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6732 directive but ignores it.
6738 @c only print the extra heading if both COFF and ELF are set
6739 @subheading ELF Version
6742 @cindex ELF symbol type
6743 @cindex symbol type, ELF
6744 @cindex @code{type} directive (ELF version)
6745 For ELF targets, the @code{.type} directive is used like this:
6748 .type @var{name} , @var{type description}
6751 This sets the type of symbol @var{name} to be either a
6752 function symbol or an object symbol. There are five different syntaxes
6753 supported for the @var{type description} field, in order to provide
6754 compatibility with various other assemblers.
6756 Because some of the characters used in these syntaxes (such as @samp{@@} and
6757 @samp{#}) are comment characters for some architectures, some of the syntaxes
6758 below do not work on all architectures. The first variant will be accepted by
6759 the GNU assembler on all architectures so that variant should be used for
6760 maximum portability, if you do not need to assemble your code with other
6763 The syntaxes supported are:
6766 .type <name> STT_<TYPE_IN_UPPER_CASE>
6767 .type <name>,#<type>
6768 .type <name>,@@<type>
6769 .type <name>,%<type>
6770 .type <name>,"<type>"
6773 The types supported are:
6778 Mark the symbol as being a function name.
6781 @itemx gnu_indirect_function
6782 Mark the symbol as an indirect function when evaluated during reloc
6783 processing. (This is only supported on assemblers targeting GNU systems).
6787 Mark the symbol as being a data object.
6791 Mark the symbol as being a thead-local data object.
6795 Mark the symbol as being a common data object.
6799 Does not mark the symbol in any way. It is supported just for completeness.
6801 @item gnu_unique_object
6802 Marks the symbol as being a globally unique data object. The dynamic linker
6803 will make sure that in the entire process there is just one symbol with this
6804 name and type in use. (This is only supported on assemblers targeting GNU
6809 Note: Some targets support extra types in addition to those listed above.
6815 @section @code{.uleb128 @var{expressions}}
6817 @cindex @code{uleb128} directive
6818 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6819 compact, variable length representation of numbers used by the DWARF
6820 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6824 @section @code{.val @var{addr}}
6826 @cindex @code{val} directive
6827 @cindex COFF value attribute
6828 @cindex value attribute, COFF
6829 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6830 records the address @var{addr} as the value attribute of a symbol table
6834 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6835 configured for @code{b.out}, it accepts this directive but ignores it.
6841 @section @code{.version "@var{string}"}
6843 @cindex @code{version} directive
6844 This directive creates a @code{.note} section and places into it an ELF
6845 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6850 @section @code{.vtable_entry @var{table}, @var{offset}}
6852 @cindex @code{vtable_entry} directive
6853 This directive finds or creates a symbol @code{table} and creates a
6854 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6857 @section @code{.vtable_inherit @var{child}, @var{parent}}
6859 @cindex @code{vtable_inherit} directive
6860 This directive finds the symbol @code{child} and finds or creates the symbol
6861 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6862 parent whose addend is the value of the child symbol. As a special case the
6863 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6867 @section @code{.warning "@var{string}"}
6868 @cindex warning directive
6869 Similar to the directive @code{.error}
6870 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6873 @section @code{.weak @var{names}}
6875 @cindex @code{weak} directive
6876 This directive sets the weak attribute on the comma separated list of symbol
6877 @code{names}. If the symbols do not already exist, they will be created.
6879 On COFF targets other than PE, weak symbols are a GNU extension. This
6880 directive sets the weak attribute on the comma separated list of symbol
6881 @code{names}. If the symbols do not already exist, they will be created.
6883 On the PE target, weak symbols are supported natively as weak aliases.
6884 When a weak symbol is created that is not an alias, GAS creates an
6885 alternate symbol to hold the default value.
6888 @section @code{.weakref @var{alias}, @var{target}}
6890 @cindex @code{weakref} directive
6891 This directive creates an alias to the target symbol that enables the symbol to
6892 be referenced with weak-symbol semantics, but without actually making it weak.
6893 If direct references or definitions of the symbol are present, then the symbol
6894 will not be weak, but if all references to it are through weak references, the
6895 symbol will be marked as weak in the symbol table.
6897 The effect is equivalent to moving all references to the alias to a separate
6898 assembly source file, renaming the alias to the symbol in it, declaring the
6899 symbol as weak there, and running a reloadable link to merge the object files
6900 resulting from the assembly of the new source file and the old source file that
6901 had the references to the alias removed.
6903 The alias itself never makes to the symbol table, and is entirely handled
6904 within the assembler.
6907 @section @code{.word @var{expressions}}
6909 @cindex @code{word} directive
6910 This directive expects zero or more @var{expressions}, of any section,
6911 separated by commas.
6914 For each expression, @command{@value{AS}} emits a 32-bit number.
6917 For each expression, @command{@value{AS}} emits a 16-bit number.
6922 The size of the number emitted, and its byte order,
6923 depend on what target computer the assembly is for.
6926 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
6927 @c happen---32-bit addressability, period; no long/short jumps.
6928 @ifset DIFF-TBL-KLUGE
6929 @cindex difference tables altered
6930 @cindex altered difference tables
6932 @emph{Warning: Special Treatment to support Compilers}
6936 Machines with a 32-bit address space, but that do less than 32-bit
6937 addressing, require the following special treatment. If the machine of
6938 interest to you does 32-bit addressing (or doesn't require it;
6939 @pxref{Machine Dependencies}), you can ignore this issue.
6942 In order to assemble compiler output into something that works,
6943 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6944 Directives of the form @samp{.word sym1-sym2} are often emitted by
6945 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6946 directive of the form @samp{.word sym1-sym2}, and the difference between
6947 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6948 creates a @dfn{secondary jump table}, immediately before the next label.
6949 This secondary jump table is preceded by a short-jump to the
6950 first byte after the secondary table. This short-jump prevents the flow
6951 of control from accidentally falling into the new table. Inside the
6952 table is a long-jump to @code{sym2}. The original @samp{.word}
6953 contains @code{sym1} minus the address of the long-jump to
6956 If there were several occurrences of @samp{.word sym1-sym2} before the
6957 secondary jump table, all of them are adjusted. If there was a
6958 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6959 long-jump to @code{sym4} is included in the secondary jump table,
6960 and the @code{.word} directives are adjusted to contain @code{sym3}
6961 minus the address of the long-jump to @code{sym4}; and so on, for as many
6962 entries in the original jump table as necessary.
6965 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6966 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6967 assembly language programmers.
6970 @c end DIFF-TBL-KLUGE
6973 @section Deprecated Directives
6975 @cindex deprecated directives
6976 @cindex obsolescent directives
6977 One day these directives won't work.
6978 They are included for compatibility with older assemblers.
6985 @node Object Attributes
6986 @chapter Object Attributes
6987 @cindex object attributes
6989 @command{@value{AS}} assembles source files written for a specific architecture
6990 into object files for that architecture. But not all object files are alike.
6991 Many architectures support incompatible variations. For instance, floating
6992 point arguments might be passed in floating point registers if the object file
6993 requires hardware floating point support---or floating point arguments might be
6994 passed in integer registers if the object file supports processors with no
6995 hardware floating point unit. Or, if two objects are built for different
6996 generations of the same architecture, the combination may require the
6997 newer generation at run-time.
6999 This information is useful during and after linking. At link time,
7000 @command{@value{LD}} can warn about incompatible object files. After link
7001 time, tools like @command{gdb} can use it to process the linked file
7004 Compatibility information is recorded as a series of object attributes. Each
7005 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
7006 string, and indicates who sets the meaning of the tag. The tag is an integer,
7007 and indicates what property the attribute describes. The value may be a string
7008 or an integer, and indicates how the property affects this object. Missing
7009 attributes are the same as attributes with a zero value or empty string value.
7011 Object attributes were developed as part of the ABI for the ARM Architecture.
7012 The file format is documented in @cite{ELF for the ARM Architecture}.
7015 * GNU Object Attributes:: @sc{gnu} Object Attributes
7016 * Defining New Object Attributes:: Defining New Object Attributes
7019 @node GNU Object Attributes
7020 @section @sc{gnu} Object Attributes
7022 The @code{.gnu_attribute} directive records an object attribute
7023 with vendor @samp{gnu}.
7025 Except for @samp{Tag_compatibility}, which has both an integer and a string for
7026 its value, @sc{gnu} attributes have a string value if the tag number is odd and
7027 an integer value if the tag number is even. The second bit (@code{@var{tag} &
7028 2} is set for architecture-independent attributes and clear for
7029 architecture-dependent ones.
7031 @subsection Common @sc{gnu} attributes
7033 These attributes are valid on all architectures.
7036 @item Tag_compatibility (32)
7037 The compatibility attribute takes an integer flag value and a vendor name. If
7038 the flag value is 0, the file is compatible with other toolchains. If it is 1,
7039 then the file is only compatible with the named toolchain. If it is greater
7040 than 1, the file can only be processed by other toolchains under some private
7041 arrangement indicated by the flag value and the vendor name.
7044 @subsection MIPS Attributes
7047 @item Tag_GNU_MIPS_ABI_FP (4)
7048 The floating-point ABI used by this object file. The value will be:
7052 0 for files not affected by the floating-point ABI.
7054 1 for files using the hardware floating-point ABI with a standard
7055 double-precision FPU.
7057 2 for files using the hardware floating-point ABI with a single-precision FPU.
7059 3 for files using the software floating-point ABI.
7061 4 for files using the deprecated hardware floating-point ABI which used 64-bit
7062 floating-point registers, 32-bit general-purpose registers and increased the
7063 number of callee-saved floating-point registers.
7065 5 for files using the hardware floating-point ABI with a double-precision FPU
7066 with either 32-bit or 64-bit floating-point registers and 32-bit
7067 general-purpose registers.
7069 6 for files using the hardware floating-point ABI with 64-bit floating-point
7070 registers and 32-bit general-purpose registers.
7072 7 for files using the hardware floating-point ABI with 64-bit floating-point
7073 registers, 32-bit general-purpose registers and a rule that forbids the
7074 direct use of odd-numbered single-precision floating-point registers.
7078 @subsection PowerPC Attributes
7081 @item Tag_GNU_Power_ABI_FP (4)
7082 The floating-point ABI used by this object file. The value will be:
7086 0 for files not affected by the floating-point ABI.
7088 1 for files using double-precision hardware floating-point ABI.
7090 2 for files using the software floating-point ABI.
7092 3 for files using single-precision hardware floating-point ABI.
7095 @item Tag_GNU_Power_ABI_Vector (8)
7096 The vector ABI used by this object file. The value will be:
7100 0 for files not affected by the vector ABI.
7102 1 for files using general purpose registers to pass vectors.
7104 2 for files using AltiVec registers to pass vectors.
7106 3 for files using SPE registers to pass vectors.
7110 @node Defining New Object Attributes
7111 @section Defining New Object Attributes
7113 If you want to define a new @sc{gnu} object attribute, here are the places you
7114 will need to modify. New attributes should be discussed on the @samp{binutils}
7119 This manual, which is the official register of attributes.
7121 The header for your architecture @file{include/elf}, to define the tag.
7123 The @file{bfd} support file for your architecture, to merge the attribute
7124 and issue any appropriate link warnings.
7126 Test cases in @file{ld/testsuite} for merging and link warnings.
7128 @file{binutils/readelf.c} to display your attribute.
7130 GCC, if you want the compiler to mark the attribute automatically.
7136 @node Machine Dependencies
7137 @chapter Machine Dependent Features
7139 @cindex machine dependencies
7140 The machine instruction sets are (almost by definition) different on
7141 each machine where @command{@value{AS}} runs. Floating point representations
7142 vary as well, and @command{@value{AS}} often supports a few additional
7143 directives or command-line options for compatibility with other
7144 assemblers on a particular platform. Finally, some versions of
7145 @command{@value{AS}} support special pseudo-instructions for branch
7148 This chapter discusses most of these differences, though it does not
7149 include details on any machine's instruction set. For details on that
7150 subject, see the hardware manufacturer's manual.
7154 * AArch64-Dependent:: AArch64 Dependent Features
7157 * Alpha-Dependent:: Alpha Dependent Features
7160 * ARC-Dependent:: ARC Dependent Features
7163 * ARM-Dependent:: ARM Dependent Features
7166 * AVR-Dependent:: AVR Dependent Features
7169 * Blackfin-Dependent:: Blackfin Dependent Features
7172 * CR16-Dependent:: CR16 Dependent Features
7175 * CRIS-Dependent:: CRIS Dependent Features
7178 * D10V-Dependent:: D10V Dependent Features
7181 * D30V-Dependent:: D30V Dependent Features
7184 * Epiphany-Dependent:: EPIPHANY Dependent Features
7187 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7190 * HPPA-Dependent:: HPPA Dependent Features
7193 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
7196 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7199 * i860-Dependent:: Intel 80860 Dependent Features
7202 * i960-Dependent:: Intel 80960 Dependent Features
7205 * IA-64-Dependent:: Intel IA-64 Dependent Features
7208 * IP2K-Dependent:: IP2K Dependent Features
7211 * LM32-Dependent:: LM32 Dependent Features
7214 * M32C-Dependent:: M32C Dependent Features
7217 * M32R-Dependent:: M32R Dependent Features
7220 * M68K-Dependent:: M680x0 Dependent Features
7223 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7226 * Meta-Dependent :: Meta Dependent Features
7229 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7232 * MIPS-Dependent:: MIPS Dependent Features
7235 * MMIX-Dependent:: MMIX Dependent Features
7238 * MSP430-Dependent:: MSP430 Dependent Features
7241 * NDS32-Dependent:: Andes NDS32 Dependent Features
7244 * NiosII-Dependent:: Altera Nios II Dependent Features
7247 * NS32K-Dependent:: NS32K Dependent Features
7250 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7251 * SH64-Dependent:: SuperH SH64 Dependent Features
7254 * PDP-11-Dependent:: PDP-11 Dependent Features
7257 * PJ-Dependent:: picoJava Dependent Features
7260 * PPC-Dependent:: PowerPC Dependent Features
7263 * RL78-Dependent:: RL78 Dependent Features
7266 * RX-Dependent:: RX Dependent Features
7269 * S/390-Dependent:: IBM S/390 Dependent Features
7272 * SCORE-Dependent:: SCORE Dependent Features
7275 * Sparc-Dependent:: SPARC Dependent Features
7278 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7281 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7284 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7287 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7290 * V850-Dependent:: V850 Dependent Features
7293 * Vax-Dependent:: VAX Dependent Features
7296 * Visium-Dependent:: Visium Dependent Features
7299 * XGATE-Dependent:: XGATE Features
7302 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7305 * Xtensa-Dependent:: Xtensa Dependent Features
7308 * Z80-Dependent:: Z80 Dependent Features
7311 * Z8000-Dependent:: Z8000 Dependent Features
7318 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7319 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7320 @c peculiarity: to preserve cross-references, there must be a node called
7321 @c "Machine Dependencies". Hence the conditional nodenames in each
7322 @c major node below. Node defaulting in makeinfo requires adjacency of
7323 @c node and sectioning commands; hence the repetition of @chapter BLAH
7324 @c in both conditional blocks.
7327 @include c-aarch64.texi
7331 @include c-alpha.texi
7347 @include c-bfin.texi
7351 @include c-cr16.texi
7355 @include c-cris.texi
7360 @node Machine Dependencies
7361 @chapter Machine Dependent Features
7363 The machine instruction sets are different on each Renesas chip family,
7364 and there are also some syntax differences among the families. This
7365 chapter describes the specific @command{@value{AS}} features for each
7369 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7370 * SH-Dependent:: Renesas SH Dependent Features
7377 @include c-d10v.texi
7381 @include c-d30v.texi
7385 @include c-epiphany.texi
7389 @include c-h8300.texi
7393 @include c-hppa.texi
7397 @include c-i370.texi
7401 @include c-i386.texi
7405 @include c-i860.texi
7409 @include c-i960.texi
7413 @include c-ia64.texi
7417 @include c-ip2k.texi
7421 @include c-lm32.texi
7425 @include c-m32c.texi
7429 @include c-m32r.texi
7433 @include c-m68k.texi
7437 @include c-m68hc11.texi
7441 @include c-metag.texi
7445 @include c-microblaze.texi
7449 @include c-mips.texi
7453 @include c-mmix.texi
7457 @include c-msp430.texi
7461 @include c-nds32.texi
7465 @include c-nios2.texi
7469 @include c-ns32k.texi
7473 @include c-pdp11.texi
7485 @include c-rl78.texi
7493 @include c-s390.texi
7497 @include c-score.texi
7502 @include c-sh64.texi
7506 @include c-sparc.texi
7510 @include c-tic54x.texi
7514 @include c-tic6x.texi
7518 @include c-tilegx.texi
7522 @include c-tilepro.texi
7526 @include c-v850.texi
7534 @include c-visium.texi
7538 @include c-xgate.texi
7542 @include c-xstormy16.texi
7546 @include c-xtensa.texi
7558 @c reverse effect of @down at top of generic Machine-Dep chapter
7562 @node Reporting Bugs
7563 @chapter Reporting Bugs
7564 @cindex bugs in assembler
7565 @cindex reporting bugs in assembler
7567 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7569 Reporting a bug may help you by bringing a solution to your problem, or it may
7570 not. But in any case the principal function of a bug report is to help the
7571 entire community by making the next version of @command{@value{AS}} work better.
7572 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7574 In order for a bug report to serve its purpose, you must include the
7575 information that enables us to fix the bug.
7578 * Bug Criteria:: Have you found a bug?
7579 * Bug Reporting:: How to report bugs
7583 @section Have You Found a Bug?
7584 @cindex bug criteria
7586 If you are not sure whether you have found a bug, here are some guidelines:
7589 @cindex fatal signal
7590 @cindex assembler crash
7591 @cindex crash of assembler
7593 If the assembler gets a fatal signal, for any input whatever, that is a
7594 @command{@value{AS}} bug. Reliable assemblers never crash.
7596 @cindex error on valid input
7598 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7600 @cindex invalid input
7602 If @command{@value{AS}} does not produce an error message for invalid input, that
7603 is a bug. However, you should note that your idea of ``invalid input'' might
7604 be our idea of ``an extension'' or ``support for traditional practice''.
7607 If you are an experienced user of assemblers, your suggestions for improvement
7608 of @command{@value{AS}} are welcome in any case.
7612 @section How to Report Bugs
7614 @cindex assembler bugs, reporting
7616 A number of companies and individuals offer support for @sc{gnu} products. If
7617 you obtained @command{@value{AS}} from a support organization, we recommend you
7618 contact that organization first.
7620 You can find contact information for many support companies and
7621 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7625 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7629 The fundamental principle of reporting bugs usefully is this:
7630 @strong{report all the facts}. If you are not sure whether to state a
7631 fact or leave it out, state it!
7633 Often people omit facts because they think they know what causes the problem
7634 and assume that some details do not matter. Thus, you might assume that the
7635 name of a symbol you use in an example does not matter. Well, probably it does
7636 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7637 happens to fetch from the location where that name is stored in memory;
7638 perhaps, if the name were different, the contents of that location would fool
7639 the assembler into doing the right thing despite the bug. Play it safe and
7640 give a specific, complete example. That is the easiest thing for you to do,
7641 and the most helpful.
7643 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7644 it is new to us. Therefore, always write your bug reports on the assumption
7645 that the bug has not been reported previously.
7647 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7648 bell?'' This cannot help us fix a bug, so it is basically useless. We
7649 respond by asking for enough details to enable us to investigate.
7650 You might as well expedite matters by sending them to begin with.
7652 To enable us to fix the bug, you should include all these things:
7656 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
7657 it with the @samp{--version} argument.
7659 Without this, we will not know whether there is any point in looking for
7660 the bug in the current version of @command{@value{AS}}.
7663 Any patches you may have applied to the @command{@value{AS}} source.
7666 The type of machine you are using, and the operating system name and
7670 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
7674 The command arguments you gave the assembler to assemble your example and
7675 observe the bug. To guarantee you will not omit something important, list them
7676 all. A copy of the Makefile (or the output from make) is sufficient.
7678 If we were to try to guess the arguments, we would probably guess wrong
7679 and then we might not encounter the bug.
7682 A complete input file that will reproduce the bug. If the bug is observed when
7683 the assembler is invoked via a compiler, send the assembler source, not the
7684 high level language source. Most compilers will produce the assembler source
7685 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
7686 the options @samp{-v --save-temps}; this will save the assembler source in a
7687 file with an extension of @file{.s}, and also show you exactly how
7688 @command{@value{AS}} is being run.
7691 A description of what behavior you observe that you believe is
7692 incorrect. For example, ``It gets a fatal signal.''
7694 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
7695 will certainly notice it. But if the bug is incorrect output, we might not
7696 notice unless it is glaringly wrong. You might as well not give us a chance to
7699 Even if the problem you experience is a fatal signal, you should still say so
7700 explicitly. Suppose something strange is going on, such as, your copy of
7701 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
7702 library on your system. (This has happened!) Your copy might crash and ours
7703 would not. If you told us to expect a crash, then when ours fails to crash, we
7704 would know that the bug was not happening for us. If you had not told us to
7705 expect a crash, then we would not be able to draw any conclusion from our
7709 If you wish to suggest changes to the @command{@value{AS}} source, send us context
7710 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
7711 option. Always send diffs from the old file to the new file. If you even
7712 discuss something in the @command{@value{AS}} source, refer to it by context, not
7715 The line numbers in our development sources will not match those in your
7716 sources. Your line numbers would convey no useful information to us.
7719 Here are some things that are not necessary:
7723 A description of the envelope of the bug.
7725 Often people who encounter a bug spend a lot of time investigating
7726 which changes to the input file will make the bug go away and which
7727 changes will not affect it.
7729 This is often time consuming and not very useful, because the way we
7730 will find the bug is by running a single example under the debugger
7731 with breakpoints, not by pure deduction from a series of examples.
7732 We recommend that you save your time for something else.
7734 Of course, if you can find a simpler example to report @emph{instead}
7735 of the original one, that is a convenience for us. Errors in the
7736 output will be easier to spot, running under the debugger will take
7737 less time, and so on.
7739 However, simplification is not vital; if you do not want to do this,
7740 report the bug anyway and send us the entire test case you used.
7743 A patch for the bug.
7745 A patch for the bug does help us if it is a good one. But do not omit
7746 the necessary information, such as the test case, on the assumption that
7747 a patch is all we need. We might see problems with your patch and decide
7748 to fix the problem another way, or we might not understand it at all.
7750 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
7751 construct an example that will make the program follow a certain path through
7752 the code. If you do not send us the example, we will not be able to construct
7753 one, so we will not be able to verify that the bug is fixed.
7755 And if we cannot understand what bug you are trying to fix, or why your
7756 patch should be an improvement, we will not install it. A test case will
7757 help us to understand.
7760 A guess about what the bug is or what it depends on.
7762 Such guesses are usually wrong. Even we cannot guess right about such
7763 things without first using the debugger to find the facts.
7766 @node Acknowledgements
7767 @chapter Acknowledgements
7769 If you have contributed to GAS and your name isn't listed here,
7770 it is not meant as a slight. We just don't know about it. Send mail to the
7771 maintainer, and we'll correct the situation. Currently
7773 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
7775 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
7778 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
7779 information and the 68k series machines, most of the preprocessing pass, and
7780 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
7782 K. Richard Pixley maintained GAS for a while, adding various enhancements and
7783 many bug fixes, including merging support for several processors, breaking GAS
7784 up to handle multiple object file format back ends (including heavy rewrite,
7785 testing, an integration of the coff and b.out back ends), adding configuration
7786 including heavy testing and verification of cross assemblers and file splits
7787 and renaming, converted GAS to strictly ANSI C including full prototypes, added
7788 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
7789 port (including considerable amounts of reverse engineering), a SPARC opcode
7790 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
7791 assertions and made them work, much other reorganization, cleanup, and lint.
7793 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
7794 in format-specific I/O modules.
7796 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
7797 has done much work with it since.
7799 The Intel 80386 machine description was written by Eliot Dresselhaus.
7801 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
7803 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
7804 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
7806 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
7807 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
7808 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
7809 support a.out format.
7811 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
7812 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
7813 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
7814 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
7817 John Gilmore built the AMD 29000 support, added @code{.include} support, and
7818 simplified the configuration of which versions accept which directives. He
7819 updated the 68k machine description so that Motorola's opcodes always produced
7820 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
7821 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
7822 cross-compilation support, and one bug in relaxation that took a week and
7823 required the proverbial one-bit fix.
7825 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
7826 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
7827 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
7828 PowerPC assembler, and made a few other minor patches.
7830 Steve Chamberlain made GAS able to generate listings.
7832 Hewlett-Packard contributed support for the HP9000/300.
7834 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
7835 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
7836 formats). This work was supported by both the Center for Software Science at
7837 the University of Utah and Cygnus Support.
7839 Support for ELF format files has been worked on by Mark Eichin of Cygnus
7840 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
7841 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
7842 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
7843 and some initial 64-bit support).
7845 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
7847 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
7848 support for openVMS/Alpha.
7850 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
7853 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
7854 Inc.@: added support for Xtensa processors.
7856 Several engineers at Cygnus Support have also provided many small bug fixes and
7857 configuration enhancements.
7859 Jon Beniston added support for the Lattice Mico32 architecture.
7861 Many others have contributed large or small bugfixes and enhancements. If
7862 you have contributed significant work and are not mentioned on this list, and
7863 want to be, let us know. Some of the history has been lost; we are not
7864 intentionally leaving anyone out.
7866 @node GNU Free Documentation License
7867 @appendix GNU Free Documentation License
7871 @unnumbered AS Index
projects / deliverable / binutils-gdb.git / blob