1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright (C) 1991-2014 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{} 1991-2014 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{} 1991-2014 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{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
403 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
404 [@b{-mips64}] [@b{-mips64r2}]
405 [@b{-construct-floats}] [@b{-no-construct-floats}]
406 [@b{-mnan=@var{encoding}}]
407 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
408 [@b{-mips16}] [@b{-no-mips16}]
409 [@b{-mmicromips}] [@b{-mno-micromips}]
410 [@b{-msmartmips}] [@b{-mno-smartmips}]
411 [@b{-mips3d}] [@b{-no-mips3d}]
412 [@b{-mdmx}] [@b{-no-mdmx}]
413 [@b{-mdsp}] [@b{-mno-dsp}]
414 [@b{-mdspr2}] [@b{-mno-dspr2}]
415 [@b{-mmsa}] [@b{-mno-msa}]
416 [@b{-mmt}] [@b{-mno-mt}]
417 [@b{-mmcu}] [@b{-mno-mcu}]
418 [@b{-minsn32}] [@b{-mno-insn32}]
419 [@b{-mfix7000}] [@b{-mno-fix7000}]
420 [@b{-mfix-rm7000}] [@b{-mno-fix-rm7000}]
421 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
422 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
423 [@b{-mdebug}] [@b{-no-mdebug}]
424 [@b{-mpdr}] [@b{-mno-pdr}]
428 @emph{Target MMIX options:}
429 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
430 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
431 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
432 [@b{--linker-allocated-gregs}]
436 @emph{Target Nios II options:}
437 [@b{-relax-all}] [@b{-relax-section}] [@b{-no-relax}]
442 @emph{Target NDS32 options:}
443 [@b{-EL}] [@b{-EB}] [@b{-O}] [@b{-Os}] [@b{-mcpu=@var{cpu}}]
444 [@b{-misa=@var{isa}}] [@b{-mabi=@var{abi}}] [@b{-mall-ext}]
445 [@b{-m[no-]16-bit}] [@b{-m[no-]perf-ext}] [@b{-m[no-]perf2-ext}]
446 [@b{-m[no-]string-ext}] [@b{-m[no-]dsp-ext}] [@b{-m[no-]mac}] [@b{-m[no-]div}]
447 [@b{-m[no-]audio-isa-ext}] [@b{-m[no-]fpu-sp-ext}] [@b{-m[no-]fpu-dp-ext}]
448 [@b{-m[no-]fpu-fma}] [@b{-mfpu-freg=@var{FREG}}] [@b{-mreduced-regs}]
449 [@b{-mfull-regs}] [@b{-m[no-]dx-regs}] [@b{-mpic}] [@b{-mno-relax}]
454 @emph{Target PDP11 options:}
455 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
456 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
457 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
461 @emph{Target picoJava options:}
466 @emph{Target PowerPC options:}
468 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|@b{-m403}|@b{-m405}|
469 @b{-m440}|@b{-m464}|@b{-m476}|@b{-m7400}|@b{-m7410}|@b{-m7450}|@b{-m7455}|@b{-m750cl}|@b{-mppc64}|
470 @b{-m620}|@b{-me500}|@b{-e500x2}|@b{-me500mc}|@b{-me500mc64}|@b{-me5500}|@b{-me6500}|@b{-mppc64bridge}|
471 @b{-mbooke}|@b{-mpower4}|@b{-mpwr4}|@b{-mpower5}|@b{-mpwr5}|@b{-mpwr5x}|@b{-mpower6}|@b{-mpwr6}|
472 @b{-mpower7}|@b{-mpwr7}|@b{-mpower8}|@b{-mpwr8}|@b{-ma2}|@b{-mcell}|@b{-mspe}|@b{-mtitan}|@b{-me300}|@b{-mcom}]
473 [@b{-many}] [@b{-maltivec}|@b{-mvsx}|@b{-mhtm}|@b{-mvle}]
474 [@b{-mregnames}|@b{-mno-regnames}]
475 [@b{-mrelocatable}|@b{-mrelocatable-lib}|@b{-K PIC}] [@b{-memb}]
476 [@b{-mlittle}|@b{-mlittle-endian}|@b{-le}|@b{-mbig}|@b{-mbig-endian}|@b{-be}]
477 [@b{-msolaris}|@b{-mno-solaris}]
478 [@b{-nops=@var{count}}]
482 @emph{Target RX options:}
483 [@b{-mlittle-endian}|@b{-mbig-endian}]
484 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
485 [@b{-muse-conventional-section-names}]
486 [@b{-msmall-data-limit}]
489 [@b{-mint-register=@var{number}}]
490 [@b{-mgcc-abi}|@b{-mrx-abi}]
494 @emph{Target s390 options:}
495 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
496 [@b{-mregnames}|@b{-mno-regnames}]
497 [@b{-mwarn-areg-zero}]
501 @emph{Target SCORE options:}
502 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
503 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
504 [@b{-march=score7}][@b{-march=score3}]
505 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
509 @emph{Target SPARC options:}
510 @c The order here is important. See c-sparc.texi.
511 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
512 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
513 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
518 @emph{Target TIC54X options:}
519 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
520 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
525 @emph{Target TIC6X options:}
526 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
527 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
528 [@b{-mpic}|@b{-mno-pic}]
532 @emph{Target TILE-Gx options:}
533 [@b{-m32}|@b{-m64}][@b{-EB}][@b{-EL}]
536 @c TILEPro has no machine-dependent assembler options
541 @emph{Target Xtensa options:}
542 [@b{--[no-]text-section-literals}] [@b{--[no-]absolute-literals}]
543 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
544 [@b{--[no-]transform}]
545 [@b{--rename-section} @var{oldname}=@var{newname}]
546 [@b{--[no-]trampolines}]
551 @emph{Target Z80 options:}
552 [@b{-z80}] [@b{-r800}]
553 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
554 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
555 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
556 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
557 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
558 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
562 @c Z8000 has no machine-dependent assembler options
571 @include at-file.texi
574 Turn on listings, in any of a variety of ways:
578 omit false conditionals
581 omit debugging directives
584 include general information, like @value{AS} version and options passed
587 include high-level source
593 include macro expansions
596 omit forms processing
602 set the name of the listing file
605 You may combine these options; for example, use @samp{-aln} for assembly
606 listing without forms processing. The @samp{=file} option, if used, must be
607 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
610 Begin in alternate macro mode.
612 @xref{Altmacro,,@code{.altmacro}}.
615 @item --compress-debug-sections
616 Compress DWARF debug sections using zlib. The debug sections are renamed
617 to begin with @samp{.zdebug}, and the resulting object file may not be
618 compatible with older linkers and object file utilities.
620 @item --nocompress-debug-sections
621 Do not compress DWARF debug sections. This is the default.
624 Ignored. This option is accepted for script compatibility with calls to
627 @item --debug-prefix-map @var{old}=@var{new}
628 When assembling files in directory @file{@var{old}}, record debugging
629 information describing them as in @file{@var{new}} instead.
631 @item --defsym @var{sym}=@var{value}
632 Define the symbol @var{sym} to be @var{value} before assembling the input file.
633 @var{value} must be an integer constant. As in C, a leading @samp{0x}
634 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
635 value. The value of the symbol can be overridden inside a source file via the
636 use of a @code{.set} pseudo-op.
639 ``fast''---skip whitespace and comment preprocessing (assume source is
644 Generate debugging information for each assembler source line using whichever
645 debug format is preferred by the target. This currently means either STABS,
649 Generate stabs debugging information for each assembler line. This
650 may help debugging assembler code, if the debugger can handle it.
653 Generate stabs debugging information for each assembler line, with GNU
654 extensions that probably only gdb can handle, and that could make other
655 debuggers crash or refuse to read your program. This
656 may help debugging assembler code. Currently the only GNU extension is
657 the location of the current working directory at assembling time.
660 Generate DWARF2 debugging information for each assembler line. This
661 may help debugging assembler code, if the debugger can handle it. Note---this
662 option is only supported by some targets, not all of them.
664 @item --gdwarf-sections
665 Instead of creating a .debug_line section, create a series of
666 .debug_line.@var{foo} sections where @var{foo} is the name of the
667 corresponding code section. For example a code section called @var{.text.func}
668 will have its dwarf line number information placed into a section called
669 @var{.debug_line.text.func}. If the code section is just called @var{.text}
670 then debug line section will still be called just @var{.debug_line} without any
673 @item --size-check=error
674 @itemx --size-check=warning
675 Issue an error or warning for invalid ELF .size directive.
678 Print a summary of the command line options and exit.
681 Print a summary of all target specific options and exit.
684 Add directory @var{dir} to the search list for @code{.include} directives.
687 Don't warn about signed overflow.
690 @ifclear DIFF-TBL-KLUGE
691 This option is accepted but has no effect on the @value{TARGET} family.
693 @ifset DIFF-TBL-KLUGE
694 Issue warnings when difference tables altered for long displacements.
699 Keep (in the symbol table) local symbols. These symbols start with
700 system-specific local label prefixes, typically @samp{.L} for ELF systems
701 or @samp{L} for traditional a.out systems.
706 @item --listing-lhs-width=@var{number}
707 Set the maximum width, in words, of the output data column for an assembler
708 listing to @var{number}.
710 @item --listing-lhs-width2=@var{number}
711 Set the maximum width, in words, of the output data column for continuation
712 lines in an assembler listing to @var{number}.
714 @item --listing-rhs-width=@var{number}
715 Set the maximum width of an input source line, as displayed in a listing, to
718 @item --listing-cont-lines=@var{number}
719 Set the maximum number of lines printed in a listing for a single line of input
722 @item -o @var{objfile}
723 Name the object-file output from @command{@value{AS}} @var{objfile}.
726 Fold the data section into the text section.
728 @kindex --hash-size=@var{number}
729 Set the default size of GAS's hash tables to a prime number close to
730 @var{number}. Increasing this value can reduce the length of time it takes the
731 assembler to perform its tasks, at the expense of increasing the assembler's
732 memory requirements. Similarly reducing this value can reduce the memory
733 requirements at the expense of speed.
735 @item --reduce-memory-overheads
736 This option reduces GAS's memory requirements, at the expense of making the
737 assembly processes slower. Currently this switch is a synonym for
738 @samp{--hash-size=4051}, but in the future it may have other effects as well.
741 Print the maximum space (in bytes) and total time (in seconds) used by
744 @item --strip-local-absolute
745 Remove local absolute symbols from the outgoing symbol table.
749 Print the @command{as} version.
752 Print the @command{as} version and exit.
756 Suppress warning messages.
758 @item --fatal-warnings
759 Treat warnings as errors.
762 Don't suppress warning messages or treat them as errors.
771 Generate an object file even after errors.
773 @item -- | @var{files} @dots{}
774 Standard input, or source files to assemble.
782 @xref{AArch64 Options}, for the options available when @value{AS} is configured
783 for the 64-bit mode of the ARM Architecture (AArch64).
788 The following options are available when @value{AS} is configured for the
789 64-bit mode of the ARM Architecture (AArch64).
792 @include c-aarch64.texi
793 @c ended inside the included file
801 @xref{Alpha Options}, for the options available when @value{AS} is configured
802 for an Alpha processor.
807 The following options are available when @value{AS} is configured for an Alpha
811 @include c-alpha.texi
812 @c ended inside the included file
819 The following options are available when @value{AS} is configured for
824 This option selects the core processor variant.
826 Select either big-endian (-EB) or little-endian (-EL) output.
831 The following options are available when @value{AS} is configured for the ARM
835 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
836 Specify which ARM processor variant is the target.
837 @item -march=@var{architecture}[+@var{extension}@dots{}]
838 Specify which ARM architecture variant is used by the target.
839 @item -mfpu=@var{floating-point-format}
840 Select which Floating Point architecture is the target.
841 @item -mfloat-abi=@var{abi}
842 Select which floating point ABI is in use.
844 Enable Thumb only instruction decoding.
845 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
846 Select which procedure calling convention is in use.
848 Select either big-endian (-EB) or little-endian (-EL) output.
849 @item -mthumb-interwork
850 Specify that the code has been generated with interworking between Thumb and
853 Turns on CodeComposer Studio assembly syntax compatibility mode.
855 Specify that PIC code has been generated.
863 @xref{Blackfin Options}, for the options available when @value{AS} is
864 configured for the Blackfin processor family.
869 The following options are available when @value{AS} is configured for
870 the Blackfin processor family.
874 @c ended inside the included file
881 See the info pages for documentation of the CRIS-specific options.
885 The following options are available when @value{AS} is configured for
888 @cindex D10V optimization
889 @cindex optimization, D10V
891 Optimize output by parallelizing instructions.
896 The following options are available when @value{AS} is configured for a D30V
899 @cindex D30V optimization
900 @cindex optimization, D30V
902 Optimize output by parallelizing instructions.
906 Warn when nops are generated.
908 @cindex D30V nops after 32-bit multiply
910 Warn when a nop after a 32-bit multiply instruction is generated.
916 The following options are available when @value{AS} is configured for the
917 Adapteva EPIPHANY series.
920 @xref{Epiphany Options}, for the options available when @value{AS} is
921 configured for an Epiphany processor.
926 The following options are available when @value{AS} is configured for
927 an Epiphany processor.
930 @include c-epiphany.texi
931 @c ended inside the included file
939 @xref{H8/300 Options}, for the options available when @value{AS} is configured
940 for an H8/300 processor.
945 The following options are available when @value{AS} is configured for an H8/300
949 @include c-h8300.texi
950 @c ended inside the included file
958 @xref{i386-Options}, for the options available when @value{AS} is
959 configured for an i386 processor.
964 The following options are available when @value{AS} is configured for
969 @c ended inside the included file
976 The following options are available when @value{AS} is configured for the
977 Intel 80960 processor.
980 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
981 Specify which variant of the 960 architecture is the target.
984 Add code to collect statistics about branches taken.
987 Do not alter compare-and-branch instructions for long displacements;
994 The following options are available when @value{AS} is configured for the
1000 Specifies that the extended IP2022 instructions are allowed.
1003 Restores the default behaviour, which restricts the permitted instructions to
1004 just the basic IP2022 ones.
1010 The following options are available when @value{AS} is configured for the
1011 Renesas M32C and M16C processors.
1016 Assemble M32C instructions.
1019 Assemble M16C instructions (the default).
1022 Enable support for link-time relaxations.
1025 Support H'00 style hex constants in addition to 0x00 style.
1031 The following options are available when @value{AS} is configured for the
1032 Renesas M32R (formerly Mitsubishi M32R) series.
1037 Specify which processor in the M32R family is the target. The default
1038 is normally the M32R, but this option changes it to the M32RX.
1040 @item --warn-explicit-parallel-conflicts or --Wp
1041 Produce warning messages when questionable parallel constructs are
1044 @item --no-warn-explicit-parallel-conflicts or --Wnp
1045 Do not produce warning messages when questionable parallel constructs are
1052 The following options are available when @value{AS} is configured for the
1053 Motorola 68000 series.
1058 Shorten references to undefined symbols, to one word instead of two.
1060 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1061 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1062 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1063 Specify what processor in the 68000 family is the target. The default
1064 is normally the 68020, but this can be changed at configuration time.
1066 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1067 The target machine does (or does not) have a floating-point coprocessor.
1068 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1069 the basic 68000 is not compatible with the 68881, a combination of the
1070 two can be specified, since it's possible to do emulation of the
1071 coprocessor instructions with the main processor.
1073 @item -m68851 | -mno-68851
1074 The target machine does (or does not) have a memory-management
1075 unit coprocessor. The default is to assume an MMU for 68020 and up.
1083 @xref{Nios II Options}, for the options available when @value{AS} is configured
1084 for an Altera Nios II processor.
1088 @c man begin OPTIONS
1089 The following options are available when @value{AS} is configured for an
1090 Altera Nios II processor.
1092 @c man begin INCLUDE
1093 @include c-nios2.texi
1094 @c ended inside the included file
1100 For details about the PDP-11 machine dependent features options,
1101 see @ref{PDP-11-Options}.
1104 @item -mpic | -mno-pic
1105 Generate position-independent (or position-dependent) code. The
1106 default is @option{-mpic}.
1109 @itemx -mall-extensions
1110 Enable all instruction set extensions. This is the default.
1112 @item -mno-extensions
1113 Disable all instruction set extensions.
1115 @item -m@var{extension} | -mno-@var{extension}
1116 Enable (or disable) a particular instruction set extension.
1119 Enable the instruction set extensions supported by a particular CPU, and
1120 disable all other extensions.
1122 @item -m@var{machine}
1123 Enable the instruction set extensions supported by a particular machine
1124 model, and disable all other extensions.
1130 The following options are available when @value{AS} is configured for
1131 a picoJava processor.
1135 @cindex PJ endianness
1136 @cindex endianness, PJ
1137 @cindex big endian output, PJ
1139 Generate ``big endian'' format output.
1141 @cindex little endian output, PJ
1143 Generate ``little endian'' format output.
1149 The following options are available when @value{AS} is configured for the
1150 Motorola 68HC11 or 68HC12 series.
1154 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1155 Specify what processor is the target. The default is
1156 defined by the configuration option when building the assembler.
1158 @item --xgate-ramoffset
1159 Instruct the linker to offset RAM addresses from S12X address space into
1160 XGATE address space.
1163 Specify to use the 16-bit integer ABI.
1166 Specify to use the 32-bit integer ABI.
1168 @item -mshort-double
1169 Specify to use the 32-bit double ABI.
1172 Specify to use the 64-bit double ABI.
1174 @item --force-long-branches
1175 Relative branches are turned into absolute ones. This concerns
1176 conditional branches, unconditional branches and branches to a
1179 @item -S | --short-branches
1180 Do not turn relative branches into absolute ones
1181 when the offset is out of range.
1183 @item --strict-direct-mode
1184 Do not turn the direct addressing mode into extended addressing mode
1185 when the instruction does not support direct addressing mode.
1187 @item --print-insn-syntax
1188 Print the syntax of instruction in case of error.
1190 @item --print-opcodes
1191 Print the list of instructions with syntax and then exit.
1193 @item --generate-example
1194 Print an example of instruction for each possible instruction and then exit.
1195 This option is only useful for testing @command{@value{AS}}.
1201 The following options are available when @command{@value{AS}} is configured
1202 for the SPARC architecture:
1205 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1206 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1207 Explicitly select a variant of the SPARC architecture.
1209 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1210 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1212 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1213 UltraSPARC extensions.
1215 @item -xarch=v8plus | -xarch=v8plusa
1216 For compatibility with the Solaris v9 assembler. These options are
1217 equivalent to -Av8plus and -Av8plusa, respectively.
1220 Warn when the assembler switches to another architecture.
1225 The following options are available when @value{AS} is configured for the 'c54x
1230 Enable extended addressing mode. All addresses and relocations will assume
1231 extended addressing (usually 23 bits).
1232 @item -mcpu=@var{CPU_VERSION}
1233 Sets the CPU version being compiled for.
1234 @item -merrors-to-file @var{FILENAME}
1235 Redirect error output to a file, for broken systems which don't support such
1236 behaviour in the shell.
1241 The following options are available when @value{AS} is configured for
1246 This option sets the largest size of an object that can be referenced
1247 implicitly with the @code{gp} register. It is only accepted for targets that
1248 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1250 @cindex MIPS endianness
1251 @cindex endianness, MIPS
1252 @cindex big endian output, MIPS
1254 Generate ``big endian'' format output.
1256 @cindex little endian output, MIPS
1258 Generate ``little endian'' format output.
1270 Generate code for a particular MIPS Instruction Set Architecture level.
1271 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1272 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1273 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1274 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
1275 @samp{-mips64r2} correspond to generic MIPS V, MIPS32, MIPS32 Release 2,
1276 MIPS64, and MIPS64 Release 2 ISA processors, respectively.
1278 @item -march=@var{cpu}
1279 Generate code for a particular MIPS CPU.
1281 @item -mtune=@var{cpu}
1282 Schedule and tune for a particular MIPS CPU.
1286 Cause nops to be inserted if the read of the destination register
1287 of an mfhi or mflo instruction occurs in the following two instructions.
1290 @itemx -mno-fix-rm7000
1291 Cause nops to be inserted if a dmult or dmultu instruction is
1292 followed by a load instruction.
1296 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1297 section instead of the standard ELF .stabs sections.
1301 Control generation of @code{.pdr} sections.
1305 The register sizes are normally inferred from the ISA and ABI, but these
1306 flags force a certain group of registers to be treated as 32 bits wide at
1307 all times. @samp{-mgp32} controls the size of general-purpose registers
1308 and @samp{-mfp32} controls the size of floating-point registers.
1312 Generate code for the MIPS 16 processor. This is equivalent to putting
1313 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1314 turns off this option.
1317 @itemx -mno-micromips
1318 Generate code for the microMIPS processor. This is equivalent to putting
1319 @code{.set micromips} at the start of the assembly file. @samp{-mno-micromips}
1320 turns off this option. This is equivalent to putting @code{.set nomicromips}
1321 at the start of the assembly file.
1324 @itemx -mno-smartmips
1325 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1326 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1327 @samp{-mno-smartmips} turns off this option.
1331 Generate code for the MIPS-3D Application Specific Extension.
1332 This tells the assembler to accept MIPS-3D instructions.
1333 @samp{-no-mips3d} turns off this option.
1337 Generate code for the MDMX Application Specific Extension.
1338 This tells the assembler to accept MDMX instructions.
1339 @samp{-no-mdmx} turns off this option.
1343 Generate code for the DSP Release 1 Application Specific Extension.
1344 This tells the assembler to accept DSP Release 1 instructions.
1345 @samp{-mno-dsp} turns off this option.
1349 Generate code for the DSP Release 2 Application Specific Extension.
1350 This option implies -mdsp.
1351 This tells the assembler to accept DSP Release 2 instructions.
1352 @samp{-mno-dspr2} turns off this option.
1356 Generate code for the MIPS SIMD Architecture Extension.
1357 This tells the assembler to accept MSA instructions.
1358 @samp{-mno-msa} turns off this option.
1362 Generate code for the MT Application Specific Extension.
1363 This tells the assembler to accept MT instructions.
1364 @samp{-mno-mt} turns off this option.
1368 Generate code for the MCU Application Specific Extension.
1369 This tells the assembler to accept MCU instructions.
1370 @samp{-mno-mcu} turns off this option.
1374 Only use 32-bit instruction encodings when generating code for the
1375 microMIPS processor. This option inhibits the use of any 16-bit
1376 instructions. This is equivalent to putting @code{.set insn32} at
1377 the start of the assembly file. @samp{-mno-insn32} turns off this
1378 option. This is equivalent to putting @code{.set noinsn32} at the
1379 start of the assembly file. By default @samp{-mno-insn32} is
1380 selected, allowing all instructions to be used.
1382 @item --construct-floats
1383 @itemx --no-construct-floats
1384 The @samp{--no-construct-floats} option disables the construction of
1385 double width floating point constants by loading the two halves of the
1386 value into the two single width floating point registers that make up
1387 the double width register. By default @samp{--construct-floats} is
1388 selected, allowing construction of these floating point constants.
1390 @item --relax-branch
1391 @itemx --no-relax-branch
1392 The @samp{--relax-branch} option enables the relaxation of out-of-range
1393 branches. By default @samp{--no-relax-branch} is selected, causing any
1394 out-of-range branches to produce an error.
1396 @item -mnan=@var{encoding}
1397 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1398 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1401 @item --emulation=@var{name}
1402 This option was formerly used to switch between ELF and ECOFF output
1403 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1404 removed in GAS 2.24, so the option now serves little purpose.
1405 It is retained for backwards compatibility.
1407 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1408 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1409 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1410 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1411 preferred options instead.
1414 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1421 Control how to deal with multiplication overflow and division by zero.
1422 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1423 (and only work for Instruction Set Architecture level 2 and higher);
1424 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1428 When this option is used, @command{@value{AS}} will issue a warning every
1429 time it generates a nop instruction from a macro.
1434 The following options are available when @value{AS} is configured for
1440 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1441 The command line option @samp{-nojsri2bsr} can be used to disable it.
1445 Enable or disable the silicon filter behaviour. By default this is disabled.
1446 The default can be overridden by the @samp{-sifilter} command line option.
1449 Alter jump instructions for long displacements.
1451 @item -mcpu=[210|340]
1452 Select the cpu type on the target hardware. This controls which instructions
1456 Assemble for a big endian target.
1459 Assemble for a little endian target.
1468 @xref{Meta Options}, for the options available when @value{AS} is configured
1469 for a Meta processor.
1473 @c man begin OPTIONS
1474 The following options are available when @value{AS} is configured for a
1477 @c man begin INCLUDE
1478 @include c-metag.texi
1479 @c ended inside the included file
1484 @c man begin OPTIONS
1486 See the info pages for documentation of the MMIX-specific options.
1492 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1493 for a NDS32 processor.
1495 @c ended inside the included file
1499 @c man begin OPTIONS
1500 The following options are available when @value{AS} is configured for a
1503 @c man begin INCLUDE
1504 @include c-nds32.texi
1505 @c ended inside the included file
1512 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1513 for a PowerPC processor.
1517 @c man begin OPTIONS
1518 The following options are available when @value{AS} is configured for a
1521 @c man begin INCLUDE
1523 @c ended inside the included file
1528 @c man begin OPTIONS
1530 See the info pages for documentation of the RX-specific options.
1534 The following options are available when @value{AS} is configured for the s390
1540 Select the word size, either 31/32 bits or 64 bits.
1543 Select the architecture mode, either the Enterprise System
1544 Architecture (esa) or the z/Architecture mode (zarch).
1545 @item -march=@var{processor}
1546 Specify which s390 processor variant is the target, @samp{g6}, @samp{g6},
1547 @samp{z900}, @samp{z990}, @samp{z9-109}, @samp{z9-ec}, @samp{z10},
1548 @samp{z196}, or @samp{zEC12}.
1550 @itemx -mno-regnames
1551 Allow or disallow symbolic names for registers.
1552 @item -mwarn-areg-zero
1553 Warn whenever the operand for a base or index register has been specified
1554 but evaluates to zero.
1562 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1563 for a TMS320C6000 processor.
1567 @c man begin OPTIONS
1568 The following options are available when @value{AS} is configured for a
1569 TMS320C6000 processor.
1571 @c man begin INCLUDE
1572 @include c-tic6x.texi
1573 @c ended inside the included file
1581 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1582 for a TILE-Gx processor.
1586 @c man begin OPTIONS
1587 The following options are available when @value{AS} is configured for a TILE-Gx
1590 @c man begin INCLUDE
1591 @include c-tilegx.texi
1592 @c ended inside the included file
1600 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1601 for an Xtensa processor.
1605 @c man begin OPTIONS
1606 The following options are available when @value{AS} is configured for an
1609 @c man begin INCLUDE
1610 @include c-xtensa.texi
1611 @c ended inside the included file
1616 @c man begin OPTIONS
1619 The following options are available when @value{AS} is configured for
1620 a Z80 family processor.
1623 Assemble for Z80 processor.
1625 Assemble for R800 processor.
1626 @item -ignore-undocumented-instructions
1628 Assemble undocumented Z80 instructions that also work on R800 without warning.
1629 @item -ignore-unportable-instructions
1631 Assemble all undocumented Z80 instructions without warning.
1632 @item -warn-undocumented-instructions
1634 Issue a warning for undocumented Z80 instructions that also work on R800.
1635 @item -warn-unportable-instructions
1637 Issue a warning for undocumented Z80 instructions that do not work on R800.
1638 @item -forbid-undocumented-instructions
1640 Treat all undocumented instructions as errors.
1641 @item -forbid-unportable-instructions
1643 Treat undocumented Z80 instructions that do not work on R800 as errors.
1650 * Manual:: Structure of this Manual
1651 * GNU Assembler:: The GNU Assembler
1652 * Object Formats:: Object File Formats
1653 * Command Line:: Command Line
1654 * Input Files:: Input Files
1655 * Object:: Output (Object) File
1656 * Errors:: Error and Warning Messages
1660 @section Structure of this Manual
1662 @cindex manual, structure and purpose
1663 This manual is intended to describe what you need to know to use
1664 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1665 notation for symbols, constants, and expressions; the directives that
1666 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1669 We also cover special features in the @value{TARGET}
1670 configuration of @command{@value{AS}}, including assembler directives.
1673 This manual also describes some of the machine-dependent features of
1674 various flavors of the assembler.
1677 @cindex machine instructions (not covered)
1678 On the other hand, this manual is @emph{not} intended as an introduction
1679 to programming in assembly language---let alone programming in general!
1680 In a similar vein, we make no attempt to introduce the machine
1681 architecture; we do @emph{not} describe the instruction set, standard
1682 mnemonics, registers or addressing modes that are standard to a
1683 particular architecture.
1685 You may want to consult the manufacturer's
1686 machine architecture manual for this information.
1690 For information on the H8/300 machine instruction set, see @cite{H8/300
1691 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1692 Programming Manual} (Renesas).
1695 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1696 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1697 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1698 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1701 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1705 @c I think this is premature---doc@cygnus.com, 17jan1991
1707 Throughout this manual, we assume that you are running @dfn{GNU},
1708 the portable operating system from the @dfn{Free Software
1709 Foundation, Inc.}. This restricts our attention to certain kinds of
1710 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1711 once this assumption is granted examples and definitions need less
1714 @command{@value{AS}} is part of a team of programs that turn a high-level
1715 human-readable series of instructions into a low-level
1716 computer-readable series of instructions. Different versions of
1717 @command{@value{AS}} are used for different kinds of computer.
1720 @c There used to be a section "Terminology" here, which defined
1721 @c "contents", "byte", "word", and "long". Defining "word" to any
1722 @c particular size is confusing when the .word directive may generate 16
1723 @c bits on one machine and 32 bits on another; in general, for the user
1724 @c version of this manual, none of these terms seem essential to define.
1725 @c They were used very little even in the former draft of the manual;
1726 @c this draft makes an effort to avoid them (except in names of
1730 @section The GNU Assembler
1732 @c man begin DESCRIPTION
1734 @sc{gnu} @command{as} is really a family of assemblers.
1736 This manual describes @command{@value{AS}}, a member of that family which is
1737 configured for the @value{TARGET} architectures.
1739 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1740 should find a fairly similar environment when you use it on another
1741 architecture. Each version has much in common with the others,
1742 including object file formats, most assembler directives (often called
1743 @dfn{pseudo-ops}) and assembler syntax.@refill
1745 @cindex purpose of @sc{gnu} assembler
1746 @command{@value{AS}} is primarily intended to assemble the output of the
1747 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1748 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1749 assemble correctly everything that other assemblers for the same
1750 machine would assemble.
1752 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1755 @c This remark should appear in generic version of manual; assumption
1756 @c here is that generic version sets M680x0.
1757 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1758 assembler for the same architecture; for example, we know of several
1759 incompatible versions of 680x0 assembly language syntax.
1764 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1765 program in one pass of the source file. This has a subtle impact on the
1766 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1768 @node Object Formats
1769 @section Object File Formats
1771 @cindex object file format
1772 The @sc{gnu} assembler can be configured to produce several alternative
1773 object file formats. For the most part, this does not affect how you
1774 write assembly language programs; but directives for debugging symbols
1775 are typically different in different file formats. @xref{Symbol
1776 Attributes,,Symbol Attributes}.
1779 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1780 @value{OBJ-NAME} format object files.
1782 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1784 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1785 @code{b.out} or COFF format object files.
1788 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1789 SOM or ELF format object files.
1794 @section Command Line
1796 @cindex command line conventions
1798 After the program name @command{@value{AS}}, the command line may contain
1799 options and file names. Options may appear in any order, and may be
1800 before, after, or between file names. The order of file names is
1803 @cindex standard input, as input file
1805 @file{--} (two hyphens) by itself names the standard input file
1806 explicitly, as one of the files for @command{@value{AS}} to assemble.
1808 @cindex options, command line
1809 Except for @samp{--} any command line argument that begins with a
1810 hyphen (@samp{-}) is an option. Each option changes the behavior of
1811 @command{@value{AS}}. No option changes the way another option works. An
1812 option is a @samp{-} followed by one or more letters; the case of
1813 the letter is important. All options are optional.
1815 Some options expect exactly one file name to follow them. The file
1816 name may either immediately follow the option's letter (compatible
1817 with older assemblers) or it may be the next command argument (@sc{gnu}
1818 standard). These two command lines are equivalent:
1821 @value{AS} -o my-object-file.o mumble.s
1822 @value{AS} -omy-object-file.o mumble.s
1826 @section Input Files
1829 @cindex source program
1830 @cindex files, input
1831 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1832 describe the program input to one run of @command{@value{AS}}. The program may
1833 be in one or more files; how the source is partitioned into files
1834 doesn't change the meaning of the source.
1836 @c I added "con" prefix to "catenation" just to prove I can overcome my
1837 @c APL training... doc@cygnus.com
1838 The source program is a concatenation of the text in all the files, in the
1841 @c man begin DESCRIPTION
1842 Each time you run @command{@value{AS}} it assembles exactly one source
1843 program. The source program is made up of one or more files.
1844 (The standard input is also a file.)
1846 You give @command{@value{AS}} a command line that has zero or more input file
1847 names. The input files are read (from left file name to right). A
1848 command line argument (in any position) that has no special meaning
1849 is taken to be an input file name.
1851 If you give @command{@value{AS}} no file names it attempts to read one input file
1852 from the @command{@value{AS}} standard input, which is normally your terminal. You
1853 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1856 Use @samp{--} if you need to explicitly name the standard input file
1857 in your command line.
1859 If the source is empty, @command{@value{AS}} produces a small, empty object
1864 @subheading Filenames and Line-numbers
1866 @cindex input file linenumbers
1867 @cindex line numbers, in input files
1868 There are two ways of locating a line in the input file (or files) and
1869 either may be used in reporting error messages. One way refers to a line
1870 number in a physical file; the other refers to a line number in a
1871 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1873 @dfn{Physical files} are those files named in the command line given
1874 to @command{@value{AS}}.
1876 @dfn{Logical files} are simply names declared explicitly by assembler
1877 directives; they bear no relation to physical files. Logical file names help
1878 error messages reflect the original source file, when @command{@value{AS}} source
1879 is itself synthesized from other files. @command{@value{AS}} understands the
1880 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1881 @ref{File,,@code{.file}}.
1884 @section Output (Object) File
1890 Every time you run @command{@value{AS}} it produces an output file, which is
1891 your assembly language program translated into numbers. This file
1892 is the object file. Its default name is
1900 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1902 You can give it another name by using the @option{-o} option. Conventionally,
1903 object file names end with @file{.o}. The default name is used for historical
1904 reasons: older assemblers were capable of assembling self-contained programs
1905 directly into a runnable program. (For some formats, this isn't currently
1906 possible, but it can be done for the @code{a.out} format.)
1910 The object file is meant for input to the linker @code{@value{LD}}. It contains
1911 assembled program code, information to help @code{@value{LD}} integrate
1912 the assembled program into a runnable file, and (optionally) symbolic
1913 information for the debugger.
1915 @c link above to some info file(s) like the description of a.out.
1916 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1919 @section Error and Warning Messages
1921 @c man begin DESCRIPTION
1923 @cindex error messages
1924 @cindex warning messages
1925 @cindex messages from assembler
1926 @command{@value{AS}} may write warnings and error messages to the standard error
1927 file (usually your terminal). This should not happen when a compiler
1928 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1929 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1930 grave problem that stops the assembly.
1934 @cindex format of warning messages
1935 Warning messages have the format
1938 file_name:@b{NNN}:Warning Message Text
1942 @cindex line numbers, in warnings/errors
1943 (where @b{NNN} is a line number). If a logical file name has been given
1944 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1945 the current input file is used. If a logical line number was given
1947 (@pxref{Line,,@code{.line}})
1949 then it is used to calculate the number printed,
1950 otherwise the actual line in the current source file is printed. The
1951 message text is intended to be self explanatory (in the grand Unix
1954 @cindex format of error messages
1955 Error messages have the format
1957 file_name:@b{NNN}:FATAL:Error Message Text
1959 The file name and line number are derived as for warning
1960 messages. The actual message text may be rather less explanatory
1961 because many of them aren't supposed to happen.
1964 @chapter Command-Line Options
1966 @cindex options, all versions of assembler
1967 This chapter describes command-line options available in @emph{all}
1968 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
1969 for options specific
1971 to the @value{TARGET} target.
1974 to particular machine architectures.
1977 @c man begin DESCRIPTION
1979 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1980 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1981 The assembler arguments must be separated from each other (and the @samp{-Wa})
1982 by commas. For example:
1985 gcc -c -g -O -Wa,-alh,-L file.c
1989 This passes two options to the assembler: @samp{-alh} (emit a listing to
1990 standard output with high-level and assembly source) and @samp{-L} (retain
1991 local symbols in the symbol table).
1993 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1994 command-line options are automatically passed to the assembler by the compiler.
1995 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1996 precisely what options it passes to each compilation pass, including the
2002 * a:: -a[cdghlns] enable listings
2003 * alternate:: --alternate enable alternate macro syntax
2004 * D:: -D for compatibility
2005 * f:: -f to work faster
2006 * I:: -I for .include search path
2007 @ifclear DIFF-TBL-KLUGE
2008 * K:: -K for compatibility
2010 @ifset DIFF-TBL-KLUGE
2011 * K:: -K for difference tables
2014 * L:: -L to retain local symbols
2015 * listing:: --listing-XXX to configure listing output
2016 * M:: -M or --mri to assemble in MRI compatibility mode
2017 * MD:: --MD for dependency tracking
2018 * o:: -o to name the object file
2019 * R:: -R to join data and text sections
2020 * statistics:: --statistics to see statistics about assembly
2021 * traditional-format:: --traditional-format for compatible output
2022 * v:: -v to announce version
2023 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2024 * Z:: -Z to make object file even after errors
2028 @section Enable Listings: @option{-a[cdghlns]}
2038 @cindex listings, enabling
2039 @cindex assembly listings, enabling
2041 These options enable listing output from the assembler. By itself,
2042 @samp{-a} requests high-level, assembly, and symbols listing.
2043 You can use other letters to select specific options for the list:
2044 @samp{-ah} requests a high-level language listing,
2045 @samp{-al} requests an output-program assembly listing, and
2046 @samp{-as} requests a symbol table listing.
2047 High-level listings require that a compiler debugging option like
2048 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2051 Use the @samp{-ag} option to print a first section with general assembly
2052 information, like @value{AS} version, switches passed, or time stamp.
2054 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2055 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2056 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2057 omitted from the listing.
2059 Use the @samp{-ad} option to omit debugging directives from the
2062 Once you have specified one of these options, you can further control
2063 listing output and its appearance using the directives @code{.list},
2064 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2066 The @samp{-an} option turns off all forms processing.
2067 If you do not request listing output with one of the @samp{-a} options, the
2068 listing-control directives have no effect.
2070 The letters after @samp{-a} may be combined into one option,
2071 @emph{e.g.}, @samp{-aln}.
2073 Note if the assembler source is coming from the standard input (e.g.,
2075 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
2076 is being used) then the listing will not contain any comments or preprocessor
2077 directives. This is because the listing code buffers input source lines from
2078 stdin only after they have been preprocessed by the assembler. This reduces
2079 memory usage and makes the code more efficient.
2082 @section @option{--alternate}
2085 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2088 @section @option{-D}
2091 This option has no effect whatsoever, but it is accepted to make it more
2092 likely that scripts written for other assemblers also work with
2093 @command{@value{AS}}.
2096 @section Work Faster: @option{-f}
2099 @cindex trusted compiler
2100 @cindex faster processing (@option{-f})
2101 @samp{-f} should only be used when assembling programs written by a
2102 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2103 and comment preprocessing on
2104 the input file(s) before assembling them. @xref{Preprocessing,
2108 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2109 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2114 @section @code{.include} Search Path: @option{-I} @var{path}
2116 @kindex -I @var{path}
2117 @cindex paths for @code{.include}
2118 @cindex search path for @code{.include}
2119 @cindex @code{include} directive search path
2120 Use this option to add a @var{path} to the list of directories
2121 @command{@value{AS}} searches for files specified in @code{.include}
2122 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2123 many times as necessary to include a variety of paths. The current
2124 working directory is always searched first; after that, @command{@value{AS}}
2125 searches any @samp{-I} directories in the same order as they were
2126 specified (left to right) on the command line.
2129 @section Difference Tables: @option{-K}
2132 @ifclear DIFF-TBL-KLUGE
2133 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2134 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2135 where it can be used to warn when the assembler alters the machine code
2136 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2137 family does not have the addressing limitations that sometimes lead to this
2138 alteration on other platforms.
2141 @ifset DIFF-TBL-KLUGE
2142 @cindex difference tables, warning
2143 @cindex warning for altered difference tables
2144 @command{@value{AS}} sometimes alters the code emitted for directives of the
2145 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2146 You can use the @samp{-K} option if you want a warning issued when this
2151 @section Include Local Symbols: @option{-L}
2154 @cindex local symbols, retaining in output
2155 Symbols beginning with system-specific local label prefixes, typically
2156 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2157 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2158 such symbols when debugging, because they are intended for the use of
2159 programs (like compilers) that compose assembler programs, not for your
2160 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2161 such symbols, so you do not normally debug with them.
2163 This option tells @command{@value{AS}} to retain those local symbols
2164 in the object file. Usually if you do this you also tell the linker
2165 @code{@value{LD}} to preserve those symbols.
2168 @section Configuring listing output: @option{--listing}
2170 The listing feature of the assembler can be enabled via the command line switch
2171 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2172 hex dump of the corresponding locations in the output object file, and displays
2173 them as a listing file. The format of this listing can be controlled by
2174 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2175 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2176 @code{.psize} (@pxref{Psize}), and
2177 @code{.eject} (@pxref{Eject}) and also by the following switches:
2180 @item --listing-lhs-width=@samp{number}
2181 @kindex --listing-lhs-width
2182 @cindex Width of first line disassembly output
2183 Sets the maximum width, in words, of the first line of the hex byte dump. This
2184 dump appears on the left hand side of the listing output.
2186 @item --listing-lhs-width2=@samp{number}
2187 @kindex --listing-lhs-width2
2188 @cindex Width of continuation lines of disassembly output
2189 Sets the maximum width, in words, of any further lines of the hex byte dump for
2190 a given input source line. If this value is not specified, it defaults to being
2191 the same as the value specified for @samp{--listing-lhs-width}. If neither
2192 switch is used the default is to one.
2194 @item --listing-rhs-width=@samp{number}
2195 @kindex --listing-rhs-width
2196 @cindex Width of source line output
2197 Sets the maximum width, in characters, of the source line that is displayed
2198 alongside the hex dump. The default value for this parameter is 100. The
2199 source line is displayed on the right hand side of the listing output.
2201 @item --listing-cont-lines=@samp{number}
2202 @kindex --listing-cont-lines
2203 @cindex Maximum number of continuation lines
2204 Sets the maximum number of continuation lines of hex dump that will be
2205 displayed for a given single line of source input. The default value is 4.
2209 @section Assemble in MRI Compatibility Mode: @option{-M}
2212 @cindex MRI compatibility mode
2213 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2214 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2215 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
2216 configured target) assembler from Microtec Research. The exact nature of the
2217 MRI syntax will not be documented here; see the MRI manuals for more
2218 information. Note in particular that the handling of macros and macro
2219 arguments is somewhat different. The purpose of this option is to permit
2220 assembling existing MRI assembler code using @command{@value{AS}}.
2222 The MRI compatibility is not complete. Certain operations of the MRI assembler
2223 depend upon its object file format, and can not be supported using other object
2224 file formats. Supporting these would require enhancing each object file format
2225 individually. These are:
2228 @item global symbols in common section
2230 The m68k MRI assembler supports common sections which are merged by the linker.
2231 Other object file formats do not support this. @command{@value{AS}} handles
2232 common sections by treating them as a single common symbol. It permits local
2233 symbols to be defined within a common section, but it can not support global
2234 symbols, since it has no way to describe them.
2236 @item complex relocations
2238 The MRI assemblers support relocations against a negated section address, and
2239 relocations which combine the start addresses of two or more sections. These
2240 are not support by other object file formats.
2242 @item @code{END} pseudo-op specifying start address
2244 The MRI @code{END} pseudo-op permits the specification of a start address.
2245 This is not supported by other object file formats. The start address may
2246 instead be specified using the @option{-e} option to the linker, or in a linker
2249 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2251 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2252 name to the output file. This is not supported by other object file formats.
2254 @item @code{ORG} pseudo-op
2256 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2257 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2258 which changes the location within the current section. Absolute sections are
2259 not supported by other object file formats. The address of a section may be
2260 assigned within a linker script.
2263 There are some other features of the MRI assembler which are not supported by
2264 @command{@value{AS}}, typically either because they are difficult or because they
2265 seem of little consequence. Some of these may be supported in future releases.
2269 @item EBCDIC strings
2271 EBCDIC strings are not supported.
2273 @item packed binary coded decimal
2275 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2276 and @code{DCB.P} pseudo-ops are not supported.
2278 @item @code{FEQU} pseudo-op
2280 The m68k @code{FEQU} pseudo-op is not supported.
2282 @item @code{NOOBJ} pseudo-op
2284 The m68k @code{NOOBJ} pseudo-op is not supported.
2286 @item @code{OPT} branch control options
2288 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2289 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2290 relaxes all branches, whether forward or backward, to an appropriate size, so
2291 these options serve no purpose.
2293 @item @code{OPT} list control options
2295 The following m68k @code{OPT} list control options are ignored: @code{C},
2296 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2297 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2299 @item other @code{OPT} options
2301 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2302 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2304 @item @code{OPT} @code{D} option is default
2306 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2307 @code{OPT NOD} may be used to turn it off.
2309 @item @code{XREF} pseudo-op.
2311 The m68k @code{XREF} pseudo-op is ignored.
2313 @item @code{.debug} pseudo-op
2315 The i960 @code{.debug} pseudo-op is not supported.
2317 @item @code{.extended} pseudo-op
2319 The i960 @code{.extended} pseudo-op is not supported.
2321 @item @code{.list} pseudo-op.
2323 The various options of the i960 @code{.list} pseudo-op are not supported.
2325 @item @code{.optimize} pseudo-op
2327 The i960 @code{.optimize} pseudo-op is not supported.
2329 @item @code{.output} pseudo-op
2331 The i960 @code{.output} pseudo-op is not supported.
2333 @item @code{.setreal} pseudo-op
2335 The i960 @code{.setreal} pseudo-op is not supported.
2340 @section Dependency Tracking: @option{--MD}
2343 @cindex dependency tracking
2346 @command{@value{AS}} can generate a dependency file for the file it creates. This
2347 file consists of a single rule suitable for @code{make} describing the
2348 dependencies of the main source file.
2350 The rule is written to the file named in its argument.
2352 This feature is used in the automatic updating of makefiles.
2355 @section Name the Object File: @option{-o}
2358 @cindex naming object file
2359 @cindex object file name
2360 There is always one object file output when you run @command{@value{AS}}. By
2361 default it has the name
2364 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2378 You use this option (which takes exactly one filename) to give the
2379 object file a different name.
2381 Whatever the object file is called, @command{@value{AS}} overwrites any
2382 existing file of the same name.
2385 @section Join Data and Text Sections: @option{-R}
2388 @cindex data and text sections, joining
2389 @cindex text and data sections, joining
2390 @cindex joining text and data sections
2391 @cindex merging text and data sections
2392 @option{-R} tells @command{@value{AS}} to write the object file as if all
2393 data-section data lives in the text section. This is only done at
2394 the very last moment: your binary data are the same, but data
2395 section parts are relocated differently. The data section part of
2396 your object file is zero bytes long because all its bytes are
2397 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2399 When you specify @option{-R} it would be possible to generate shorter
2400 address displacements (because we do not have to cross between text and
2401 data section). We refrain from doing this simply for compatibility with
2402 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2405 When @command{@value{AS}} is configured for COFF or ELF output,
2406 this option is only useful if you use sections named @samp{.text} and
2411 @option{-R} is not supported for any of the HPPA targets. Using
2412 @option{-R} generates a warning from @command{@value{AS}}.
2416 @section Display Assembly Statistics: @option{--statistics}
2418 @kindex --statistics
2419 @cindex statistics, about assembly
2420 @cindex time, total for assembly
2421 @cindex space used, maximum for assembly
2422 Use @samp{--statistics} to display two statistics about the resources used by
2423 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2424 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2427 @node traditional-format
2428 @section Compatible Output: @option{--traditional-format}
2430 @kindex --traditional-format
2431 For some targets, the output of @command{@value{AS}} is different in some ways
2432 from the output of some existing assembler. This switch requests
2433 @command{@value{AS}} to use the traditional format instead.
2435 For example, it disables the exception frame optimizations which
2436 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2439 @section Announce Version: @option{-v}
2443 @cindex assembler version
2444 @cindex version of assembler
2445 You can find out what version of as is running by including the
2446 option @samp{-v} (which you can also spell as @samp{-version}) on the
2450 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2452 @command{@value{AS}} should never give a warning or error message when
2453 assembling compiler output. But programs written by people often
2454 cause @command{@value{AS}} to give a warning that a particular assumption was
2455 made. All such warnings are directed to the standard error file.
2459 @cindex suppressing warnings
2460 @cindex warnings, suppressing
2461 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2462 This only affects the warning messages: it does not change any particular of
2463 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2466 @kindex --fatal-warnings
2467 @cindex errors, caused by warnings
2468 @cindex warnings, causing error
2469 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2470 files that generate warnings to be in error.
2473 @cindex warnings, switching on
2474 You can switch these options off again by specifying @option{--warn}, which
2475 causes warnings to be output as usual.
2478 @section Generate Object File in Spite of Errors: @option{-Z}
2479 @cindex object file, after errors
2480 @cindex errors, continuing after
2481 After an error message, @command{@value{AS}} normally produces no output. If for
2482 some reason you are interested in object file output even after
2483 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2484 option. If there are any errors, @command{@value{AS}} continues anyways, and
2485 writes an object file after a final warning message of the form @samp{@var{n}
2486 errors, @var{m} warnings, generating bad object file.}
2491 @cindex machine-independent syntax
2492 @cindex syntax, machine-independent
2493 This chapter describes the machine-independent syntax allowed in a
2494 source file. @command{@value{AS}} syntax is similar to what many other
2495 assemblers use; it is inspired by the BSD 4.2
2500 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2504 * Preprocessing:: Preprocessing
2505 * Whitespace:: Whitespace
2506 * Comments:: Comments
2507 * Symbol Intro:: Symbols
2508 * Statements:: Statements
2509 * Constants:: Constants
2513 @section Preprocessing
2515 @cindex preprocessing
2516 The @command{@value{AS}} internal preprocessor:
2518 @cindex whitespace, removed by preprocessor
2520 adjusts and removes extra whitespace. It leaves one space or tab before
2521 the keywords on a line, and turns any other whitespace on the line into
2524 @cindex comments, removed by preprocessor
2526 removes all comments, replacing them with a single space, or an
2527 appropriate number of newlines.
2529 @cindex constants, converted by preprocessor
2531 converts character constants into the appropriate numeric values.
2534 It does not do macro processing, include file handling, or
2535 anything else you may get from your C compiler's preprocessor. You can
2536 do include file processing with the @code{.include} directive
2537 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2538 to get other ``CPP'' style preprocessing by giving the input file a
2539 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2540 Output, gcc.info, Using GNU CC}.
2542 Excess whitespace, comments, and character constants
2543 cannot be used in the portions of the input text that are not
2546 @cindex turning preprocessing on and off
2547 @cindex preprocessing, turning on and off
2550 If the first line of an input file is @code{#NO_APP} or if you use the
2551 @samp{-f} option, whitespace and comments are not removed from the input file.
2552 Within an input file, you can ask for whitespace and comment removal in
2553 specific portions of the by putting a line that says @code{#APP} before the
2554 text that may contain whitespace or comments, and putting a line that says
2555 @code{#NO_APP} after this text. This feature is mainly intend to support
2556 @code{asm} statements in compilers whose output is otherwise free of comments
2563 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2564 Whitespace is used to separate symbols, and to make programs neater for
2565 people to read. Unless within character constants
2566 (@pxref{Characters,,Character Constants}), any whitespace means the same
2567 as exactly one space.
2573 There are two ways of rendering comments to @command{@value{AS}}. In both
2574 cases the comment is equivalent to one space.
2576 Anything from @samp{/*} through the next @samp{*/} is a comment.
2577 This means you may not nest these comments.
2581 The only way to include a newline ('\n') in a comment
2582 is to use this sort of comment.
2585 /* This sort of comment does not nest. */
2588 @cindex line comment character
2589 Anything from a @dfn{line comment} character up to the next newline is
2590 considered a comment and is ignored. The line comment character is target
2591 specific, and some targets multiple comment characters. Some targets also have
2592 line comment characters that only work if they are the first character on a
2593 line. Some targets use a sequence of two characters to introduce a line
2594 comment. Some targets can also change their line comment characters depending
2595 upon command line options that have been used. For more details see the
2596 @emph{Syntax} section in the documentation for individual targets.
2598 If the line comment character is the hash sign (@samp{#}) then it still has the
2599 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2600 to specify logical line numbers:
2603 @cindex lines starting with @code{#}
2604 @cindex logical line numbers
2605 To be compatible with past assemblers, lines that begin with @samp{#} have a
2606 special interpretation. Following the @samp{#} should be an absolute
2607 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2608 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2609 new logical file name. The rest of the line, if any, should be whitespace.
2611 If the first non-whitespace characters on the line are not numeric,
2612 the line is ignored. (Just like a comment.)
2615 # This is an ordinary comment.
2616 # 42-6 "new_file_name" # New logical file name
2617 # This is logical line # 36.
2619 This feature is deprecated, and may disappear from future versions
2620 of @command{@value{AS}}.
2625 @cindex characters used in symbols
2626 @ifclear SPECIAL-SYMS
2627 A @dfn{symbol} is one or more characters chosen from the set of all
2628 letters (both upper and lower case), digits and the three characters
2634 A @dfn{symbol} is one or more characters chosen from the set of all
2635 letters (both upper and lower case), digits and the three characters
2636 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2642 On most machines, you can also use @code{$} in symbol names; exceptions
2643 are noted in @ref{Machine Dependencies}.
2645 No symbol may begin with a digit. Case is significant.
2646 There is no length limit: all characters are significant. Multibyte characters
2647 are supported. Symbols are delimited by characters not in that set, or by the
2648 beginning of a file (since the source program must end with a newline, the end
2649 of a file is not a possible symbol delimiter). @xref{Symbols}.
2650 @cindex length of symbols
2655 @cindex statements, structure of
2656 @cindex line separator character
2657 @cindex statement separator character
2659 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2660 @dfn{line separator character}. The line separator character is target
2661 specific and described in the @emph{Syntax} section of each
2662 target's documentation. Not all targets support a line separator character.
2663 The newline or line separator character is considered to be part of the
2664 preceding statement. Newlines and separators within character constants are an
2665 exception: they do not end statements.
2667 @cindex newline, required at file end
2668 @cindex EOF, newline must precede
2669 It is an error to end any statement with end-of-file: the last
2670 character of any input file should be a newline.@refill
2672 An empty statement is allowed, and may include whitespace. It is ignored.
2674 @cindex instructions and directives
2675 @cindex directives and instructions
2676 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2677 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2679 A statement begins with zero or more labels, optionally followed by a
2680 key symbol which determines what kind of statement it is. The key
2681 symbol determines the syntax of the rest of the statement. If the
2682 symbol begins with a dot @samp{.} then the statement is an assembler
2683 directive: typically valid for any computer. If the symbol begins with
2684 a letter the statement is an assembly language @dfn{instruction}: it
2685 assembles into a machine language instruction.
2687 Different versions of @command{@value{AS}} for different computers
2688 recognize different instructions. In fact, the same symbol may
2689 represent a different instruction in a different computer's assembly
2693 @cindex @code{:} (label)
2694 @cindex label (@code{:})
2695 A label is a symbol immediately followed by a colon (@code{:}).
2696 Whitespace before a label or after a colon is permitted, but you may not
2697 have whitespace between a label's symbol and its colon. @xref{Labels}.
2700 For HPPA targets, labels need not be immediately followed by a colon, but
2701 the definition of a label must begin in column zero. This also implies that
2702 only one label may be defined on each line.
2706 label: .directive followed by something
2707 another_label: # This is an empty statement.
2708 instruction operand_1, operand_2, @dots{}
2715 A constant is a number, written so that its value is known by
2716 inspection, without knowing any context. Like this:
2719 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2720 .ascii "Ring the bell\7" # A string constant.
2721 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2722 .float 0f-314159265358979323846264338327\
2723 95028841971.693993751E-40 # - pi, a flonum.
2728 * Characters:: Character Constants
2729 * Numbers:: Number Constants
2733 @subsection Character Constants
2735 @cindex character constants
2736 @cindex constants, character
2737 There are two kinds of character constants. A @dfn{character} stands
2738 for one character in one byte and its value may be used in
2739 numeric expressions. String constants (properly called string
2740 @emph{literals}) are potentially many bytes and their values may not be
2741 used in arithmetic expressions.
2745 * Chars:: Characters
2749 @subsubsection Strings
2751 @cindex string constants
2752 @cindex constants, string
2753 A @dfn{string} is written between double-quotes. It may contain
2754 double-quotes or null characters. The way to get special characters
2755 into a string is to @dfn{escape} these characters: precede them with
2756 a backslash @samp{\} character. For example @samp{\\} represents
2757 one backslash: the first @code{\} is an escape which tells
2758 @command{@value{AS}} to interpret the second character literally as a backslash
2759 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2760 escape character). The complete list of escapes follows.
2762 @cindex escape codes, character
2763 @cindex character escape codes
2766 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2768 @cindex @code{\b} (backspace character)
2769 @cindex backspace (@code{\b})
2771 Mnemonic for backspace; for ASCII this is octal code 010.
2774 @c Mnemonic for EOText; for ASCII this is octal code 004.
2776 @cindex @code{\f} (formfeed character)
2777 @cindex formfeed (@code{\f})
2779 Mnemonic for FormFeed; for ASCII this is octal code 014.
2781 @cindex @code{\n} (newline character)
2782 @cindex newline (@code{\n})
2784 Mnemonic for newline; for ASCII this is octal code 012.
2787 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2789 @cindex @code{\r} (carriage return character)
2790 @cindex carriage return (@code{\r})
2792 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2795 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2796 @c other assemblers.
2798 @cindex @code{\t} (tab)
2799 @cindex tab (@code{\t})
2801 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2804 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2805 @c @item \x @var{digit} @var{digit} @var{digit}
2806 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2808 @cindex @code{\@var{ddd}} (octal character code)
2809 @cindex octal character code (@code{\@var{ddd}})
2810 @item \ @var{digit} @var{digit} @var{digit}
2811 An octal character code. The numeric code is 3 octal digits.
2812 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2813 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2815 @cindex @code{\@var{xd...}} (hex character code)
2816 @cindex hex character code (@code{\@var{xd...}})
2817 @item \@code{x} @var{hex-digits...}
2818 A hex character code. All trailing hex digits are combined. Either upper or
2819 lower case @code{x} works.
2821 @cindex @code{\\} (@samp{\} character)
2822 @cindex backslash (@code{\\})
2824 Represents one @samp{\} character.
2827 @c Represents one @samp{'} (accent acute) character.
2828 @c This is needed in single character literals
2829 @c (@xref{Characters,,Character Constants}.) to represent
2832 @cindex @code{\"} (doublequote character)
2833 @cindex doublequote (@code{\"})
2835 Represents one @samp{"} character. Needed in strings to represent
2836 this character, because an unescaped @samp{"} would end the string.
2838 @item \ @var{anything-else}
2839 Any other character when escaped by @kbd{\} gives a warning, but
2840 assembles as if the @samp{\} was not present. The idea is that if
2841 you used an escape sequence you clearly didn't want the literal
2842 interpretation of the following character. However @command{@value{AS}} has no
2843 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2844 code and warns you of the fact.
2847 Which characters are escapable, and what those escapes represent,
2848 varies widely among assemblers. The current set is what we think
2849 the BSD 4.2 assembler recognizes, and is a subset of what most C
2850 compilers recognize. If you are in doubt, do not use an escape
2854 @subsubsection Characters
2856 @cindex single character constant
2857 @cindex character, single
2858 @cindex constant, single character
2859 A single character may be written as a single quote immediately
2860 followed by that character. The same escapes apply to characters as
2861 to strings. So if you want to write the character backslash, you
2862 must write @kbd{'\\} where the first @code{\} escapes the second
2863 @code{\}. As you can see, the quote is an acute accent, not a
2864 grave accent. A newline
2866 @ifclear abnormal-separator
2867 (or semicolon @samp{;})
2869 @ifset abnormal-separator
2871 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2876 immediately following an acute accent is taken as a literal character
2877 and does not count as the end of a statement. The value of a character
2878 constant in a numeric expression is the machine's byte-wide code for
2879 that character. @command{@value{AS}} assumes your character code is ASCII:
2880 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2883 @subsection Number Constants
2885 @cindex constants, number
2886 @cindex number constants
2887 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2888 are stored in the target machine. @emph{Integers} are numbers that
2889 would fit into an @code{int} in the C language. @emph{Bignums} are
2890 integers, but they are stored in more than 32 bits. @emph{Flonums}
2891 are floating point numbers, described below.
2894 * Integers:: Integers
2899 * Bit Fields:: Bit Fields
2905 @subsubsection Integers
2907 @cindex constants, integer
2909 @cindex binary integers
2910 @cindex integers, binary
2911 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2912 the binary digits @samp{01}.
2914 @cindex octal integers
2915 @cindex integers, octal
2916 An octal integer is @samp{0} followed by zero or more of the octal
2917 digits (@samp{01234567}).
2919 @cindex decimal integers
2920 @cindex integers, decimal
2921 A decimal integer starts with a non-zero digit followed by zero or
2922 more digits (@samp{0123456789}).
2924 @cindex hexadecimal integers
2925 @cindex integers, hexadecimal
2926 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2927 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2929 Integers have the usual values. To denote a negative integer, use
2930 the prefix operator @samp{-} discussed under expressions
2931 (@pxref{Prefix Ops,,Prefix Operators}).
2934 @subsubsection Bignums
2937 @cindex constants, bignum
2938 A @dfn{bignum} has the same syntax and semantics as an integer
2939 except that the number (or its negative) takes more than 32 bits to
2940 represent in binary. The distinction is made because in some places
2941 integers are permitted while bignums are not.
2944 @subsubsection Flonums
2946 @cindex floating point numbers
2947 @cindex constants, floating point
2949 @cindex precision, floating point
2950 A @dfn{flonum} represents a floating point number. The translation is
2951 indirect: a decimal floating point number from the text is converted by
2952 @command{@value{AS}} to a generic binary floating point number of more than
2953 sufficient precision. This generic floating point number is converted
2954 to a particular computer's floating point format (or formats) by a
2955 portion of @command{@value{AS}} specialized to that computer.
2957 A flonum is written by writing (in order)
2962 (@samp{0} is optional on the HPPA.)
2966 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2968 @kbd{e} is recommended. Case is not important.
2970 @c FIXME: verify if flonum syntax really this vague for most cases
2971 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2972 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2975 On the H8/300, Renesas / SuperH SH,
2976 and AMD 29K architectures, the letter must be
2977 one of the letters @samp{DFPRSX} (in upper or lower case).
2979 On the ARC, the letter must be one of the letters @samp{DFRS}
2980 (in upper or lower case).
2982 On the Intel 960 architecture, the letter must be
2983 one of the letters @samp{DFT} (in upper or lower case).
2985 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2989 One of the letters @samp{DFRS} (in upper or lower case).
2992 One of the letters @samp{DFPRSX} (in upper or lower case).
2995 The letter @samp{E} (upper case only).
2998 One of the letters @samp{DFT} (in upper or lower case).
3003 An optional sign: either @samp{+} or @samp{-}.
3006 An optional @dfn{integer part}: zero or more decimal digits.
3009 An optional @dfn{fractional part}: @samp{.} followed by zero
3010 or more decimal digits.
3013 An optional exponent, consisting of:
3017 An @samp{E} or @samp{e}.
3018 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3019 @c principle this can perfectly well be different on different targets.
3021 Optional sign: either @samp{+} or @samp{-}.
3023 One or more decimal digits.
3028 At least one of the integer part or the fractional part must be
3029 present. The floating point number has the usual base-10 value.
3031 @command{@value{AS}} does all processing using integers. Flonums are computed
3032 independently of any floating point hardware in the computer running
3033 @command{@value{AS}}.
3037 @c Bit fields are written as a general facility but are also controlled
3038 @c by a conditional-compilation flag---which is as of now (21mar91)
3039 @c turned on only by the i960 config of GAS.
3041 @subsubsection Bit Fields
3044 @cindex constants, bit field
3045 You can also define numeric constants as @dfn{bit fields}.
3046 Specify two numbers separated by a colon---
3048 @var{mask}:@var{value}
3051 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
3054 The resulting number is then packed
3056 @c this conditional paren in case bit fields turned on elsewhere than 960
3057 (in host-dependent byte order)
3059 into a field whose width depends on which assembler directive has the
3060 bit-field as its argument. Overflow (a result from the bitwise and
3061 requiring more binary digits to represent) is not an error; instead,
3062 more constants are generated, of the specified width, beginning with the
3063 least significant digits.@refill
3065 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
3066 @code{.short}, and @code{.word} accept bit-field arguments.
3071 @chapter Sections and Relocation
3076 * Secs Background:: Background
3077 * Ld Sections:: Linker Sections
3078 * As Sections:: Assembler Internal Sections
3079 * Sub-Sections:: Sub-Sections
3083 @node Secs Background
3086 Roughly, a section is a range of addresses, with no gaps; all data
3087 ``in'' those addresses is treated the same for some particular purpose.
3088 For example there may be a ``read only'' section.
3090 @cindex linker, and assembler
3091 @cindex assembler, and linker
3092 The linker @code{@value{LD}} reads many object files (partial programs) and
3093 combines their contents to form a runnable program. When @command{@value{AS}}
3094 emits an object file, the partial program is assumed to start at address 0.
3095 @code{@value{LD}} assigns the final addresses for the partial program, so that
3096 different partial programs do not overlap. This is actually an
3097 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3100 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3101 addresses. These blocks slide to their run-time addresses as rigid
3102 units; their length does not change and neither does the order of bytes
3103 within them. Such a rigid unit is called a @emph{section}. Assigning
3104 run-time addresses to sections is called @dfn{relocation}. It includes
3105 the task of adjusting mentions of object-file addresses so they refer to
3106 the proper run-time addresses.
3108 For the H8/300, and for the Renesas / SuperH SH,
3109 @command{@value{AS}} pads sections if needed to
3110 ensure they end on a word (sixteen bit) boundary.
3113 @cindex standard assembler sections
3114 An object file written by @command{@value{AS}} has at least three sections, any
3115 of which may be empty. These are named @dfn{text}, @dfn{data} and
3120 When it generates COFF or ELF output,
3122 @command{@value{AS}} can also generate whatever other named sections you specify
3123 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3124 If you do not use any directives that place output in the @samp{.text}
3125 or @samp{.data} sections, these sections still exist, but are empty.
3130 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3132 @command{@value{AS}} can also generate whatever other named sections you
3133 specify using the @samp{.space} and @samp{.subspace} directives. See
3134 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3135 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3136 assembler directives.
3139 Additionally, @command{@value{AS}} uses different names for the standard
3140 text, data, and bss sections when generating SOM output. Program text
3141 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3142 BSS into @samp{$BSS$}.
3146 Within the object file, the text section starts at address @code{0}, the
3147 data section follows, and the bss section follows the data section.
3150 When generating either SOM or ELF output files on the HPPA, the text
3151 section starts at address @code{0}, the data section at address
3152 @code{0x4000000}, and the bss section follows the data section.
3155 To let @code{@value{LD}} know which data changes when the sections are
3156 relocated, and how to change that data, @command{@value{AS}} also writes to the
3157 object file details of the relocation needed. To perform relocation
3158 @code{@value{LD}} must know, each time an address in the object
3162 Where in the object file is the beginning of this reference to
3165 How long (in bytes) is this reference?
3167 Which section does the address refer to? What is the numeric value of
3169 (@var{address}) @minus{} (@var{start-address of section})?
3172 Is the reference to an address ``Program-Counter relative''?
3175 @cindex addresses, format of
3176 @cindex section-relative addressing
3177 In fact, every address @command{@value{AS}} ever uses is expressed as
3179 (@var{section}) + (@var{offset into section})
3182 Further, most expressions @command{@value{AS}} computes have this section-relative
3185 (For some object formats, such as SOM for the HPPA, some expressions are
3186 symbol-relative instead.)
3189 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3190 @var{N} into section @var{secname}.''
3192 Apart from text, data and bss sections you need to know about the
3193 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3194 addresses in the absolute section remain unchanged. For example, address
3195 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3196 @code{@value{LD}}. Although the linker never arranges two partial programs'
3197 data sections with overlapping addresses after linking, @emph{by definition}
3198 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3199 part of a program is always the same address when the program is running as
3200 address @code{@{absolute@ 239@}} in any other part of the program.
3202 The idea of sections is extended to the @dfn{undefined} section. Any
3203 address whose section is unknown at assembly time is by definition
3204 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3205 Since numbers are always defined, the only way to generate an undefined
3206 address is to mention an undefined symbol. A reference to a named
3207 common block would be such a symbol: its value is unknown at assembly
3208 time so it has section @emph{undefined}.
3210 By analogy the word @emph{section} is used to describe groups of sections in
3211 the linked program. @code{@value{LD}} puts all partial programs' text
3212 sections in contiguous addresses in the linked program. It is
3213 customary to refer to the @emph{text section} of a program, meaning all
3214 the addresses of all partial programs' text sections. Likewise for
3215 data and bss sections.
3217 Some sections are manipulated by @code{@value{LD}}; others are invented for
3218 use of @command{@value{AS}} and have no meaning except during assembly.
3221 @section Linker Sections
3222 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3227 @cindex named sections
3228 @cindex sections, named
3229 @item named sections
3232 @cindex text section
3233 @cindex data section
3237 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3238 separate but equal sections. Anything you can say of one section is
3241 When the program is running, however, it is
3242 customary for the text section to be unalterable. The
3243 text section is often shared among processes: it contains
3244 instructions, constants and the like. The data section of a running
3245 program is usually alterable: for example, C variables would be stored
3246 in the data section.
3251 This section contains zeroed bytes when your program begins running. It
3252 is used to hold uninitialized variables or common storage. The length of
3253 each partial program's bss section is important, but because it starts
3254 out containing zeroed bytes there is no need to store explicit zero
3255 bytes in the object file. The bss section was invented to eliminate
3256 those explicit zeros from object files.
3258 @cindex absolute section
3259 @item absolute section
3260 Address 0 of this section is always ``relocated'' to runtime address 0.
3261 This is useful if you want to refer to an address that @code{@value{LD}} must
3262 not change when relocating. In this sense we speak of absolute
3263 addresses being ``unrelocatable'': they do not change during relocation.
3265 @cindex undefined section
3266 @item undefined section
3267 This ``section'' is a catch-all for address references to objects not in
3268 the preceding sections.
3269 @c FIXME: ref to some other doc on obj-file formats could go here.
3272 @cindex relocation example
3273 An idealized example of three relocatable sections follows.
3275 The example uses the traditional section names @samp{.text} and @samp{.data}.
3277 Memory addresses are on the horizontal axis.
3281 @c END TEXI2ROFF-KILL
3284 partial program # 1: |ttttt|dddd|00|
3291 partial program # 2: |TTT|DDD|000|
3294 +--+---+-----+--+----+---+-----+~~
3295 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3296 +--+---+-----+--+----+---+-----+~~
3298 addresses: 0 @dots{}
3305 \line{\it Partial program \#1: \hfil}
3306 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3307 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3309 \line{\it Partial program \#2: \hfil}
3310 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3311 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3313 \line{\it linked program: \hfil}
3314 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3315 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3316 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3317 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3319 \line{\it addresses: \hfil}
3323 @c END TEXI2ROFF-KILL
3326 @section Assembler Internal Sections
3328 @cindex internal assembler sections
3329 @cindex sections in messages, internal
3330 These sections are meant only for the internal use of @command{@value{AS}}. They
3331 have no meaning at run-time. You do not really need to know about these
3332 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3333 warning messages, so it might be helpful to have an idea of their
3334 meanings to @command{@value{AS}}. These sections are used to permit the
3335 value of every expression in your assembly language program to be a
3336 section-relative address.
3339 @cindex assembler internal logic error
3340 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3341 An internal assembler logic error has been found. This means there is a
3342 bug in the assembler.
3344 @cindex expr (internal section)
3346 The assembler stores complex expression internally as combinations of
3347 symbols. When it needs to represent an expression as a symbol, it puts
3348 it in the expr section.
3350 @c FIXME item transfer[t] vector preload
3351 @c FIXME item transfer[t] vector postload
3352 @c FIXME item register
3356 @section Sub-Sections
3358 @cindex numbered subsections
3359 @cindex grouping data
3365 fall into two sections: text and data.
3367 You may have separate groups of
3369 data in named sections
3373 data in named sections
3379 that you want to end up near to each other in the object file, even though they
3380 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3381 use @dfn{subsections} for this purpose. Within each section, there can be
3382 numbered subsections with values from 0 to 8192. Objects assembled into the
3383 same subsection go into the object file together with other objects in the same
3384 subsection. For example, a compiler might want to store constants in the text
3385 section, but might not want to have them interspersed with the program being
3386 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3387 section of code being output, and a @samp{.text 1} before each group of
3388 constants being output.
3390 Subsections are optional. If you do not use subsections, everything
3391 goes in subsection number zero.
3394 Each subsection is zero-padded up to a multiple of four bytes.
3395 (Subsections may be padded a different amount on different flavors
3396 of @command{@value{AS}}.)
3400 On the H8/300 platform, each subsection is zero-padded to a word
3401 boundary (two bytes).
3402 The same is true on the Renesas SH.
3405 @c FIXME section padding (alignment)?
3406 @c Rich Pixley says padding here depends on target obj code format; that
3407 @c doesn't seem particularly useful to say without further elaboration,
3408 @c so for now I say nothing about it. If this is a generic BFD issue,
3409 @c these paragraphs might need to vanish from this manual, and be
3410 @c discussed in BFD chapter of binutils (or some such).
3414 Subsections appear in your object file in numeric order, lowest numbered
3415 to highest. (All this to be compatible with other people's assemblers.)
3416 The object file contains no representation of subsections; @code{@value{LD}} and
3417 other programs that manipulate object files see no trace of them.
3418 They just see all your text subsections as a text section, and all your
3419 data subsections as a data section.
3421 To specify which subsection you want subsequent statements assembled
3422 into, use a numeric argument to specify it, in a @samp{.text
3423 @var{expression}} or a @samp{.data @var{expression}} statement.
3426 When generating COFF output, you
3431 can also use an extra subsection
3432 argument with arbitrary named sections: @samp{.section @var{name},
3437 When generating ELF output, you
3442 can also use the @code{.subsection} directive (@pxref{SubSection})
3443 to specify a subsection: @samp{.subsection @var{expression}}.
3445 @var{Expression} should be an absolute expression
3446 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3447 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3448 begins in @code{text 0}. For instance:
3450 .text 0 # The default subsection is text 0 anyway.
3451 .ascii "This lives in the first text subsection. *"
3453 .ascii "But this lives in the second text subsection."
3455 .ascii "This lives in the data section,"
3456 .ascii "in the first data subsection."
3458 .ascii "This lives in the first text section,"
3459 .ascii "immediately following the asterisk (*)."
3462 Each section has a @dfn{location counter} incremented by one for every byte
3463 assembled into that section. Because subsections are merely a convenience
3464 restricted to @command{@value{AS}} there is no concept of a subsection location
3465 counter. There is no way to directly manipulate a location counter---but the
3466 @code{.align} directive changes it, and any label definition captures its
3467 current value. The location counter of the section where statements are being
3468 assembled is said to be the @dfn{active} location counter.
3471 @section bss Section
3474 @cindex common variable storage
3475 The bss section is used for local common variable storage.
3476 You may allocate address space in the bss section, but you may
3477 not dictate data to load into it before your program executes. When
3478 your program starts running, all the contents of the bss
3479 section are zeroed bytes.
3481 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3482 @ref{Lcomm,,@code{.lcomm}}.
3484 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3485 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3488 When assembling for a target which supports multiple sections, such as ELF or
3489 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3490 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3491 section. Typically the section will only contain symbol definitions and
3492 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3499 Symbols are a central concept: the programmer uses symbols to name
3500 things, the linker uses symbols to link, and the debugger uses symbols
3504 @cindex debuggers, and symbol order
3505 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3506 the same order they were declared. This may break some debuggers.
3511 * Setting Symbols:: Giving Symbols Other Values
3512 * Symbol Names:: Symbol Names
3513 * Dot:: The Special Dot Symbol
3514 * Symbol Attributes:: Symbol Attributes
3521 A @dfn{label} is written as a symbol immediately followed by a colon
3522 @samp{:}. The symbol then represents the current value of the
3523 active location counter, and is, for example, a suitable instruction
3524 operand. You are warned if you use the same symbol to represent two
3525 different locations: the first definition overrides any other
3529 On the HPPA, the usual form for a label need not be immediately followed by a
3530 colon, but instead must start in column zero. Only one label may be defined on
3531 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3532 provides a special directive @code{.label} for defining labels more flexibly.
3535 @node Setting Symbols
3536 @section Giving Symbols Other Values
3538 @cindex assigning values to symbols
3539 @cindex symbol values, assigning
3540 A symbol can be given an arbitrary value by writing a symbol, followed
3541 by an equals sign @samp{=}, followed by an expression
3542 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3543 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3544 equals sign @samp{=}@samp{=} here represents an equivalent of the
3545 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3548 Blackfin does not support symbol assignment with @samp{=}.
3552 @section Symbol Names
3554 @cindex symbol names
3555 @cindex names, symbol
3556 @ifclear SPECIAL-SYMS
3557 Symbol names begin with a letter or with one of @samp{._}. On most
3558 machines, you can also use @code{$} in symbol names; exceptions are
3559 noted in @ref{Machine Dependencies}. That character may be followed by any
3560 string of digits, letters, dollar signs (unless otherwise noted for a
3561 particular target machine), and underscores.
3565 Symbol names begin with a letter or with one of @samp{._}. On the
3566 Renesas SH you can also use @code{$} in symbol names. That
3567 character may be followed by any string of digits, letters, dollar signs (save
3568 on the H8/300), and underscores.
3572 Case of letters is significant: @code{foo} is a different symbol name
3575 Multibyte characters are supported. To generate a symbol name containing
3576 multibyte characters enclose it within double quotes and use escape codes. cf
3577 @xref{Strings}. Generating a multibyte symbol name from a label is not
3578 currently supported.
3580 Each symbol has exactly one name. Each name in an assembly language program
3581 refers to exactly one symbol. You may use that symbol name any number of times
3584 @subheading Local Symbol Names
3586 @cindex local symbol names
3587 @cindex symbol names, local
3588 A local symbol is any symbol beginning with certain local label prefixes.
3589 By default, the local label prefix is @samp{.L} for ELF systems or
3590 @samp{L} for traditional a.out systems, but each target may have its own
3591 set of local label prefixes.
3593 On the HPPA local symbols begin with @samp{L$}.
3596 Local symbols are defined and used within the assembler, but they are
3597 normally not saved in object files. Thus, they are not visible when debugging.
3598 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols:
3599 @option{-L}}) to retain the local symbols in the object files.
3601 @subheading Local Labels
3603 @cindex local labels
3604 @cindex temporary symbol names
3605 @cindex symbol names, temporary
3606 Local labels help compilers and programmers use names temporarily.
3607 They create symbols which are guaranteed to be unique over the entire scope of
3608 the input source code and which can be referred to by a simple notation.
3609 To define a local label, write a label of the form @samp{@b{N}:} (where @b{N}
3610 represents any positive integer). To refer to the most recent previous
3611 definition of that label write @samp{@b{N}b}, using the same number as when
3612 you defined the label. To refer to the next definition of a local label, write
3613 @samp{@b{N}f}---the @samp{b} stands for ``backwards'' and the @samp{f} stands
3616 There is no restriction on how you can use these labels, and you can reuse them
3617 too. So that it is possible to repeatedly define the same local label (using
3618 the same number @samp{@b{N}}), although you can only refer to the most recently
3619 defined local label of that number (for a backwards reference) or the next
3620 definition of a specific local label for a forward reference. It is also worth
3621 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3622 implemented in a slightly more efficient manner than the others.
3633 Which is the equivalent of:
3636 label_1: branch label_3
3637 label_2: branch label_1
3638 label_3: branch label_4
3639 label_4: branch label_3
3642 Local label names are only a notational device. They are immediately
3643 transformed into more conventional symbol names before the assembler uses them.
3644 The symbol names are stored in the symbol table, appear in error messages, and
3645 are optionally emitted to the object file. The names are constructed using
3649 @item @emph{local label prefix}
3650 All local symbols begin with the system-specific local label prefix.
3651 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3652 that start with the local label prefix. These labels are
3653 used for symbols you are never intended to see. If you use the
3654 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3655 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3656 you may use them in debugging.
3659 This is the number that was used in the local label definition. So if the
3660 label is written @samp{55:} then the number is @samp{55}.
3663 This unusual character is included so you do not accidentally invent a symbol
3664 of the same name. The character has ASCII value of @samp{\002} (control-B).
3666 @item @emph{ordinal number}
3667 This is a serial number to keep the labels distinct. The first definition of
3668 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3669 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3670 the number @samp{1} and its 15th definition gets @samp{15} as well.
3673 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3674 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3676 @subheading Dollar Local Labels
3677 @cindex dollar local symbols
3679 @code{@value{AS}} also supports an even more local form of local labels called
3680 dollar labels. These labels go out of scope (i.e., they become undefined) as
3681 soon as a non-local label is defined. Thus they remain valid for only a small
3682 region of the input source code. Normal local labels, by contrast, remain in
3683 scope for the entire file, or until they are redefined by another occurrence of
3684 the same local label.
3686 Dollar labels are defined in exactly the same way as ordinary local labels,
3687 except that they have a dollar sign suffix to their numeric value, e.g.,
3690 They can also be distinguished from ordinary local labels by their transformed
3691 names which use ASCII character @samp{\001} (control-A) as the magic character
3692 to distinguish them from ordinary labels. For example, the fifth definition of
3693 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3696 @section The Special Dot Symbol
3698 @cindex dot (symbol)
3699 @cindex @code{.} (symbol)
3700 @cindex current address
3701 @cindex location counter
3702 The special symbol @samp{.} refers to the current address that
3703 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3704 .long .} defines @code{melvin} to contain its own address.
3705 Assigning a value to @code{.} is treated the same as a @code{.org}
3707 @ifclear no-space-dir
3708 Thus, the expression @samp{.=.+4} is the same as saying
3712 @node Symbol Attributes
3713 @section Symbol Attributes
3715 @cindex symbol attributes
3716 @cindex attributes, symbol
3717 Every symbol has, as well as its name, the attributes ``Value'' and
3718 ``Type''. Depending on output format, symbols can also have auxiliary
3721 The detailed definitions are in @file{a.out.h}.
3724 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3725 all these attributes, and probably won't warn you. This makes the
3726 symbol an externally defined symbol, which is generally what you
3730 * Symbol Value:: Value
3731 * Symbol Type:: Type
3734 * a.out Symbols:: Symbol Attributes: @code{a.out}
3738 * a.out Symbols:: Symbol Attributes: @code{a.out}
3741 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3746 * COFF Symbols:: Symbol Attributes for COFF
3749 * SOM Symbols:: Symbol Attributes for SOM
3756 @cindex value of a symbol
3757 @cindex symbol value
3758 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3759 location in the text, data, bss or absolute sections the value is the
3760 number of addresses from the start of that section to the label.
3761 Naturally for text, data and bss sections the value of a symbol changes
3762 as @code{@value{LD}} changes section base addresses during linking. Absolute
3763 symbols' values do not change during linking: that is why they are
3766 The value of an undefined symbol is treated in a special way. If it is
3767 0 then the symbol is not defined in this assembler source file, and
3768 @code{@value{LD}} tries to determine its value from other files linked into the
3769 same program. You make this kind of symbol simply by mentioning a symbol
3770 name without defining it. A non-zero value represents a @code{.comm}
3771 common declaration. The value is how much common storage to reserve, in
3772 bytes (addresses). The symbol refers to the first address of the
3778 @cindex type of a symbol
3780 The type attribute of a symbol contains relocation (section)
3781 information, any flag settings indicating that a symbol is external, and
3782 (optionally), other information for linkers and debuggers. The exact
3783 format depends on the object-code output format in use.
3788 @c The following avoids a "widow" subsection title. @group would be
3789 @c better if it were available outside examples.
3792 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3794 @cindex @code{b.out} symbol attributes
3795 @cindex symbol attributes, @code{b.out}
3796 These symbol attributes appear only when @command{@value{AS}} is configured for
3797 one of the Berkeley-descended object output formats---@code{a.out} or
3803 @subsection Symbol Attributes: @code{a.out}
3805 @cindex @code{a.out} symbol attributes
3806 @cindex symbol attributes, @code{a.out}
3812 @subsection Symbol Attributes: @code{a.out}
3814 @cindex @code{a.out} symbol attributes
3815 @cindex symbol attributes, @code{a.out}
3819 * Symbol Desc:: Descriptor
3820 * Symbol Other:: Other
3824 @subsubsection Descriptor
3826 @cindex descriptor, of @code{a.out} symbol
3827 This is an arbitrary 16-bit value. You may establish a symbol's
3828 descriptor value by using a @code{.desc} statement
3829 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3830 @command{@value{AS}}.
3833 @subsubsection Other
3835 @cindex other attribute, of @code{a.out} symbol
3836 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3841 @subsection Symbol Attributes for COFF
3843 @cindex COFF symbol attributes
3844 @cindex symbol attributes, COFF
3846 The COFF format supports a multitude of auxiliary symbol attributes;
3847 like the primary symbol attributes, they are set between @code{.def} and
3848 @code{.endef} directives.
3850 @subsubsection Primary Attributes
3852 @cindex primary attributes, COFF symbols
3853 The symbol name is set with @code{.def}; the value and type,
3854 respectively, with @code{.val} and @code{.type}.
3856 @subsubsection Auxiliary Attributes
3858 @cindex auxiliary attributes, COFF symbols
3859 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3860 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3861 table information for COFF.
3866 @subsection Symbol Attributes for SOM
3868 @cindex SOM symbol attributes
3869 @cindex symbol attributes, SOM
3871 The SOM format for the HPPA supports a multitude of symbol attributes set with
3872 the @code{.EXPORT} and @code{.IMPORT} directives.
3874 The attributes are described in @cite{HP9000 Series 800 Assembly
3875 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3876 @code{EXPORT} assembler directive documentation.
3880 @chapter Expressions
3884 @cindex numeric values
3885 An @dfn{expression} specifies an address or numeric value.
3886 Whitespace may precede and/or follow an expression.
3888 The result of an expression must be an absolute number, or else an offset into
3889 a particular section. If an expression is not absolute, and there is not
3890 enough information when @command{@value{AS}} sees the expression to know its
3891 section, a second pass over the source program might be necessary to interpret
3892 the expression---but the second pass is currently not implemented.
3893 @command{@value{AS}} aborts with an error message in this situation.
3896 * Empty Exprs:: Empty Expressions
3897 * Integer Exprs:: Integer Expressions
3901 @section Empty Expressions
3903 @cindex empty expressions
3904 @cindex expressions, empty
3905 An empty expression has no value: it is just whitespace or null.
3906 Wherever an absolute expression is required, you may omit the
3907 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3908 is compatible with other assemblers.
3911 @section Integer Expressions
3913 @cindex integer expressions
3914 @cindex expressions, integer
3915 An @dfn{integer expression} is one or more @emph{arguments} delimited
3916 by @emph{operators}.
3919 * Arguments:: Arguments
3920 * Operators:: Operators
3921 * Prefix Ops:: Prefix Operators
3922 * Infix Ops:: Infix Operators
3926 @subsection Arguments
3928 @cindex expression arguments
3929 @cindex arguments in expressions
3930 @cindex operands in expressions
3931 @cindex arithmetic operands
3932 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3933 contexts arguments are sometimes called ``arithmetic operands''. In
3934 this manual, to avoid confusing them with the ``instruction operands'' of
3935 the machine language, we use the term ``argument'' to refer to parts of
3936 expressions only, reserving the word ``operand'' to refer only to machine
3937 instruction operands.
3939 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3940 @var{section} is one of text, data, bss, absolute,
3941 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3944 Numbers are usually integers.
3946 A number can be a flonum or bignum. In this case, you are warned
3947 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3948 these 32 bits are an integer. You may write integer-manipulating
3949 instructions that act on exotic constants, compatible with other
3952 @cindex subexpressions
3953 Subexpressions are a left parenthesis @samp{(} followed by an integer
3954 expression, followed by a right parenthesis @samp{)}; or a prefix
3955 operator followed by an argument.
3958 @subsection Operators
3960 @cindex operators, in expressions
3961 @cindex arithmetic functions
3962 @cindex functions, in expressions
3963 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3964 operators are followed by an argument. Infix operators appear
3965 between their arguments. Operators may be preceded and/or followed by
3969 @subsection Prefix Operator
3971 @cindex prefix operators
3972 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3973 one argument, which must be absolute.
3975 @c the tex/end tex stuff surrounding this small table is meant to make
3976 @c it align, on the printed page, with the similar table in the next
3977 @c section (which is inside an enumerate).
3979 \global\advance\leftskip by \itemindent
3984 @dfn{Negation}. Two's complement negation.
3986 @dfn{Complementation}. Bitwise not.
3990 \global\advance\leftskip by -\itemindent
3994 @subsection Infix Operators
3996 @cindex infix operators
3997 @cindex operators, permitted arguments
3998 @dfn{Infix operators} take two arguments, one on either side. Operators
3999 have precedence, but operations with equal precedence are performed left
4000 to right. Apart from @code{+} or @option{-}, both arguments must be
4001 absolute, and the result is absolute.
4004 @cindex operator precedence
4005 @cindex precedence of operators
4012 @dfn{Multiplication}.
4015 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4021 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4024 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4028 Intermediate precedence
4033 @dfn{Bitwise Inclusive Or}.
4039 @dfn{Bitwise Exclusive Or}.
4042 @dfn{Bitwise Or Not}.
4049 @cindex addition, permitted arguments
4050 @cindex plus, permitted arguments
4051 @cindex arguments for addition
4053 @dfn{Addition}. If either argument is absolute, the result has the section of
4054 the other argument. You may not add together arguments from different
4057 @cindex subtraction, permitted arguments
4058 @cindex minus, permitted arguments
4059 @cindex arguments for subtraction
4061 @dfn{Subtraction}. If the right argument is absolute, the
4062 result has the section of the left argument.
4063 If both arguments are in the same section, the result is absolute.
4064 You may not subtract arguments from different sections.
4065 @c FIXME is there still something useful to say about undefined - undefined ?
4067 @cindex comparison expressions
4068 @cindex expressions, comparison
4073 @dfn{Is Not Equal To}
4077 @dfn{Is Greater Than}
4079 @dfn{Is Greater Than Or Equal To}
4081 @dfn{Is Less Than Or Equal To}
4083 The comparison operators can be used as infix operators. A true results has a
4084 value of -1 whereas a false result has a value of 0. Note, these operators
4085 perform signed comparisons.
4088 @item Lowest Precedence
4097 These two logical operations can be used to combine the results of sub
4098 expressions. Note, unlike the comparison operators a true result returns a
4099 value of 1 but a false results does still return 0. Also note that the logical
4100 or operator has a slightly lower precedence than logical and.
4105 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4106 address; you can only have a defined section in one of the two arguments.
4109 @chapter Assembler Directives
4111 @cindex directives, machine independent
4112 @cindex pseudo-ops, machine independent
4113 @cindex machine independent directives
4114 All assembler directives have names that begin with a period (@samp{.}).
4115 The rest of the name is letters, usually in lower case.
4117 This chapter discusses directives that are available regardless of the
4118 target machine configuration for the @sc{gnu} assembler.
4120 Some machine configurations provide additional directives.
4121 @xref{Machine Dependencies}.
4124 @ifset machine-directives
4125 @xref{Machine Dependencies}, for additional directives.
4130 * Abort:: @code{.abort}
4132 * ABORT (COFF):: @code{.ABORT}
4135 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4136 * Altmacro:: @code{.altmacro}
4137 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4138 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4139 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4140 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, @code{.bundle_lock}, @code{.bundle_unlock}
4141 * Byte:: @code{.byte @var{expressions}}
4142 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4143 * Comm:: @code{.comm @var{symbol} , @var{length} }
4144 * Data:: @code{.data @var{subsection}}
4146 * Def:: @code{.def @var{name}}
4149 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4155 * Double:: @code{.double @var{flonums}}
4156 * Eject:: @code{.eject}
4157 * Else:: @code{.else}
4158 * Elseif:: @code{.elseif}
4161 * Endef:: @code{.endef}
4164 * Endfunc:: @code{.endfunc}
4165 * Endif:: @code{.endif}
4166 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4167 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4168 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4170 * Error:: @code{.error @var{string}}
4171 * Exitm:: @code{.exitm}
4172 * Extern:: @code{.extern}
4173 * Fail:: @code{.fail}
4174 * File:: @code{.file}
4175 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4176 * Float:: @code{.float @var{flonums}}
4177 * Func:: @code{.func}
4178 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4180 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4181 * Hidden:: @code{.hidden @var{names}}
4184 * hword:: @code{.hword @var{expressions}}
4185 * Ident:: @code{.ident}
4186 * If:: @code{.if @var{absolute expression}}
4187 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4188 * Include:: @code{.include "@var{file}"}
4189 * Int:: @code{.int @var{expressions}}
4191 * Internal:: @code{.internal @var{names}}
4194 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4195 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4196 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4197 * Lflags:: @code{.lflags}
4198 @ifclear no-line-dir
4199 * Line:: @code{.line @var{line-number}}
4202 * Linkonce:: @code{.linkonce [@var{type}]}
4203 * List:: @code{.list}
4204 * Ln:: @code{.ln @var{line-number}}
4205 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4206 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4208 * Local:: @code{.local @var{names}}
4211 * Long:: @code{.long @var{expressions}}
4213 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4216 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4217 * MRI:: @code{.mri @var{val}}
4218 * Noaltmacro:: @code{.noaltmacro}
4219 * Nolist:: @code{.nolist}
4220 * Octa:: @code{.octa @var{bignums}}
4221 * Offset:: @code{.offset @var{loc}}
4222 * Org:: @code{.org @var{new-lc}, @var{fill}}
4223 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4225 * PopSection:: @code{.popsection}
4226 * Previous:: @code{.previous}
4229 * Print:: @code{.print @var{string}}
4231 * Protected:: @code{.protected @var{names}}
4234 * Psize:: @code{.psize @var{lines}, @var{columns}}
4235 * Purgem:: @code{.purgem @var{name}}
4237 * PushSection:: @code{.pushsection @var{name}}
4240 * Quad:: @code{.quad @var{bignums}}
4241 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4242 * Rept:: @code{.rept @var{count}}
4243 * Sbttl:: @code{.sbttl "@var{subheading}"}
4245 * Scl:: @code{.scl @var{class}}
4248 * Section:: @code{.section @var{name}[, @var{flags}]}
4251 * Set:: @code{.set @var{symbol}, @var{expression}}
4252 * Short:: @code{.short @var{expressions}}
4253 * Single:: @code{.single @var{flonums}}
4255 * Size:: @code{.size [@var{name} , @var{expression}]}
4257 @ifclear no-space-dir
4258 * Skip:: @code{.skip @var{size} , @var{fill}}
4261 * Sleb128:: @code{.sleb128 @var{expressions}}
4262 @ifclear no-space-dir
4263 * Space:: @code{.space @var{size} , @var{fill}}
4266 * Stab:: @code{.stabd, .stabn, .stabs}
4269 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4270 * Struct:: @code{.struct @var{expression}}
4272 * SubSection:: @code{.subsection}
4273 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4277 * Tag:: @code{.tag @var{structname}}
4280 * Text:: @code{.text @var{subsection}}
4281 * Title:: @code{.title "@var{heading}"}
4283 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4286 * Uleb128:: @code{.uleb128 @var{expressions}}
4288 * Val:: @code{.val @var{addr}}
4292 * Version:: @code{.version "@var{string}"}
4293 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4294 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4297 * Warning:: @code{.warning @var{string}}
4298 * Weak:: @code{.weak @var{names}}
4299 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4300 * Word:: @code{.word @var{expressions}}
4301 * Deprecated:: Deprecated Directives
4305 @section @code{.abort}
4307 @cindex @code{abort} directive
4308 @cindex stopping the assembly
4309 This directive stops the assembly immediately. It is for
4310 compatibility with other assemblers. The original idea was that the
4311 assembly language source would be piped into the assembler. If the sender
4312 of the source quit, it could use this directive tells @command{@value{AS}} to
4313 quit also. One day @code{.abort} will not be supported.
4317 @section @code{.ABORT} (COFF)
4319 @cindex @code{ABORT} directive
4320 When producing COFF output, @command{@value{AS}} accepts this directive as a
4321 synonym for @samp{.abort}.
4324 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4330 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4332 @cindex padding the location counter
4333 @cindex @code{align} directive
4334 Pad the location counter (in the current subsection) to a particular storage
4335 boundary. The first expression (which must be absolute) is the alignment
4336 required, as described below.
4338 The second expression (also absolute) gives the fill value to be stored in the
4339 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4340 padding bytes are normally zero. However, on some systems, if the section is
4341 marked as containing code and the fill value is omitted, the space is filled
4342 with no-op instructions.
4344 The third expression is also absolute, and is also optional. If it is present,
4345 it is the maximum number of bytes that should be skipped by this alignment
4346 directive. If doing the alignment would require skipping more bytes than the
4347 specified maximum, then the alignment is not done at all. You can omit the
4348 fill value (the second argument) entirely by simply using two commas after the
4349 required alignment; this can be useful if you want the alignment to be filled
4350 with no-op instructions when appropriate.
4352 The way the required alignment is specified varies from system to system.
4353 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or32,
4354 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4355 alignment request in bytes. For example @samp{.align 8} advances
4356 the location counter until it is a multiple of 8. If the location counter
4357 is already a multiple of 8, no change is needed. For the tic54x, the
4358 first expression is the alignment request in words.
4360 For other systems, including ppc, i386 using a.out format, arm and
4361 strongarm, it is the
4362 number of low-order zero bits the location counter must have after
4363 advancement. For example @samp{.align 3} advances the location
4364 counter until it a multiple of 8. If the location counter is already a
4365 multiple of 8, no change is needed.
4367 This inconsistency is due to the different behaviors of the various
4368 native assemblers for these systems which GAS must emulate.
4369 GAS also provides @code{.balign} and @code{.p2align} directives,
4370 described later, which have a consistent behavior across all
4371 architectures (but are specific to GAS).
4374 @section @code{.altmacro}
4375 Enable alternate macro mode, enabling:
4378 @item LOCAL @var{name} [ , @dots{} ]
4379 One additional directive, @code{LOCAL}, is available. It is used to
4380 generate a string replacement for each of the @var{name} arguments, and
4381 replace any instances of @var{name} in each macro expansion. The
4382 replacement string is unique in the assembly, and different for each
4383 separate macro expansion. @code{LOCAL} allows you to write macros that
4384 define symbols, without fear of conflict between separate macro expansions.
4386 @item String delimiters
4387 You can write strings delimited in these other ways besides
4388 @code{"@var{string}"}:
4391 @item '@var{string}'
4392 You can delimit strings with single-quote characters.
4394 @item <@var{string}>
4395 You can delimit strings with matching angle brackets.
4398 @item single-character string escape
4399 To include any single character literally in a string (even if the
4400 character would otherwise have some special meaning), you can prefix the
4401 character with @samp{!} (an exclamation mark). For example, you can
4402 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4404 @item Expression results as strings
4405 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4406 and use the result as a string.
4410 @section @code{.ascii "@var{string}"}@dots{}
4412 @cindex @code{ascii} directive
4413 @cindex string literals
4414 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4415 separated by commas. It assembles each string (with no automatic
4416 trailing zero byte) into consecutive addresses.
4419 @section @code{.asciz "@var{string}"}@dots{}
4421 @cindex @code{asciz} directive
4422 @cindex zero-terminated strings
4423 @cindex null-terminated strings
4424 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4425 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4428 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4430 @cindex padding the location counter given number of bytes
4431 @cindex @code{balign} directive
4432 Pad the location counter (in the current subsection) to a particular
4433 storage boundary. The first expression (which must be absolute) is the
4434 alignment request in bytes. For example @samp{.balign 8} advances
4435 the location counter until it is a multiple of 8. If the location counter
4436 is already a multiple of 8, no change is needed.
4438 The second expression (also absolute) gives the fill value to be stored in the
4439 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4440 padding bytes are normally zero. However, on some systems, if the section is
4441 marked as containing code and the fill value is omitted, the space is filled
4442 with no-op instructions.
4444 The third expression is also absolute, and is also optional. If it is present,
4445 it is the maximum number of bytes that should be skipped by this alignment
4446 directive. If doing the alignment would require skipping more bytes than the
4447 specified maximum, then the alignment is not done at all. You can omit the
4448 fill value (the second argument) entirely by simply using two commas after the
4449 required alignment; this can be useful if you want the alignment to be filled
4450 with no-op instructions when appropriate.
4452 @cindex @code{balignw} directive
4453 @cindex @code{balignl} directive
4454 The @code{.balignw} and @code{.balignl} directives are variants of the
4455 @code{.balign} directive. The @code{.balignw} directive treats the fill
4456 pattern as a two byte word value. The @code{.balignl} directives treats the
4457 fill pattern as a four byte longword value. For example, @code{.balignw
4458 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4459 filled in with the value 0x368d (the exact placement of the bytes depends upon
4460 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4463 @node Bundle directives
4464 @section @code{.bundle_align_mode @var{abs-expr}}
4465 @cindex @code{bundle_align_mode} directive
4467 @cindex instruction bundle
4468 @cindex aligned instruction bundle
4469 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4470 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4471 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4472 disabled (which is the default state). If the argument it not zero, it
4473 gives the size of an instruction bundle as a power of two (as for the
4474 @code{.p2align} directive, @pxref{P2align}).
4476 For some targets, it's an ABI requirement that no instruction may span a
4477 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4478 instructions that starts on an aligned boundary. For example, if
4479 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4480 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4481 effect, no single instruction may span a boundary between bundles. If an
4482 instruction would start too close to the end of a bundle for the length of
4483 that particular instruction to fit within the bundle, then the space at the
4484 end of that bundle is filled with no-op instructions so the instruction
4485 starts in the next bundle. As a corollary, it's an error if any single
4486 instruction's encoding is longer than the bundle size.
4488 @section @code{.bundle_lock} and @code{.bundle_unlock}
4489 @cindex @code{bundle_lock} directive
4490 @cindex @code{bundle_unlock} directive
4491 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4492 allow explicit control over instruction bundle padding. These directives
4493 are only valid when @code{.bundle_align_mode} has been used to enable
4494 aligned instruction bundle mode. It's an error if they appear when
4495 @code{.bundle_align_mode} has not been used at all, or when the last
4496 directive was @w{@code{.bundle_align_mode 0}}.
4498 @cindex bundle-locked
4499 For some targets, it's an ABI requirement that certain instructions may
4500 appear only as part of specified permissible sequences of multiple
4501 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4502 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4503 instruction sequence. For purposes of aligned instruction bundle mode, a
4504 sequence starting with @code{.bundle_lock} and ending with
4505 @code{.bundle_unlock} is treated as a single instruction. That is, the
4506 entire sequence must fit into a single bundle and may not span a bundle
4507 boundary. If necessary, no-op instructions will be inserted before the
4508 first instruction of the sequence so that the whole sequence starts on an
4509 aligned bundle boundary. It's an error if the sequence is longer than the
4512 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4513 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4514 nested. That is, a second @code{.bundle_lock} directive before the next
4515 @code{.bundle_unlock} directive has no effect except that it must be
4516 matched by another closing @code{.bundle_unlock} so that there is the
4517 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4520 @section @code{.byte @var{expressions}}
4522 @cindex @code{byte} directive
4523 @cindex integers, one byte
4524 @code{.byte} expects zero or more expressions, separated by commas.
4525 Each expression is assembled into the next byte.
4527 @node CFI directives
4528 @section @code{.cfi_sections @var{section_list}}
4529 @cindex @code{cfi_sections} directive
4530 @code{.cfi_sections} may be used to specify whether CFI directives
4531 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4532 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4533 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4534 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4535 directive is not used is @code{.cfi_sections .eh_frame}.
4537 @section @code{.cfi_startproc [simple]}
4538 @cindex @code{cfi_startproc} directive
4539 @code{.cfi_startproc} is used at the beginning of each function that
4540 should have an entry in @code{.eh_frame}. It initializes some internal
4541 data structures. Don't forget to close the function by
4542 @code{.cfi_endproc}.
4544 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4545 it also emits some architecture dependent initial CFI instructions.
4547 @section @code{.cfi_endproc}
4548 @cindex @code{cfi_endproc} directive
4549 @code{.cfi_endproc} is used at the end of a function where it closes its
4550 unwind entry previously opened by
4551 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4553 @section @code{.cfi_personality @var{encoding} [, @var{exp}]}
4554 @code{.cfi_personality} defines personality routine and its encoding.
4555 @var{encoding} must be a constant determining how the personality
4556 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4557 argument is not present, otherwise second argument should be
4558 a constant or a symbol name. When using indirect encodings,
4559 the symbol provided should be the location where personality
4560 can be loaded from, not the personality routine itself.
4561 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4562 no personality routine.
4564 @section @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4565 @code{.cfi_lsda} defines LSDA and its encoding.
4566 @var{encoding} must be a constant determining how the LSDA
4567 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4568 argument is not present, otherwise second argument should be a constant
4569 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4572 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4573 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4574 address from @var{register} and add @var{offset} to it}.
4576 @section @code{.cfi_def_cfa_register @var{register}}
4577 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4578 now on @var{register} will be used instead of the old one. Offset
4581 @section @code{.cfi_def_cfa_offset @var{offset}}
4582 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4583 remains the same, but @var{offset} is new. Note that it is the
4584 absolute offset that will be added to a defined register to compute
4587 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4588 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4589 value that is added/substracted from the previous offset.
4591 @section @code{.cfi_offset @var{register}, @var{offset}}
4592 Previous value of @var{register} is saved at offset @var{offset} from
4595 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4596 Previous value of @var{register} is saved at offset @var{offset} from
4597 the current CFA register. This is transformed to @code{.cfi_offset}
4598 using the known displacement of the CFA register from the CFA.
4599 This is often easier to use, because the number will match the
4600 code it's annotating.
4602 @section @code{.cfi_register @var{register1}, @var{register2}}
4603 Previous value of @var{register1} is saved in register @var{register2}.
4605 @section @code{.cfi_restore @var{register}}
4606 @code{.cfi_restore} says that the rule for @var{register} is now the
4607 same as it was at the beginning of the function, after all initial
4608 instruction added by @code{.cfi_startproc} were executed.
4610 @section @code{.cfi_undefined @var{register}}
4611 From now on the previous value of @var{register} can't be restored anymore.
4613 @section @code{.cfi_same_value @var{register}}
4614 Current value of @var{register} is the same like in the previous frame,
4615 i.e. no restoration needed.
4617 @section @code{.cfi_remember_state},
4618 First save all current rules for all registers by @code{.cfi_remember_state},
4619 then totally screw them up by subsequent @code{.cfi_*} directives and when
4620 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4621 the previous saved state.
4623 @section @code{.cfi_return_column @var{register}}
4624 Change return column @var{register}, i.e. the return address is either
4625 directly in @var{register} or can be accessed by rules for @var{register}.
4627 @section @code{.cfi_signal_frame}
4628 Mark current function as signal trampoline.
4630 @section @code{.cfi_window_save}
4631 SPARC register window has been saved.
4633 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4634 Allows the user to add arbitrary bytes to the unwind info. One
4635 might use this to add OS-specific CFI opcodes, or generic CFI
4636 opcodes that GAS does not yet support.
4638 @section @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4639 The current value of @var{register} is @var{label}. The value of @var{label}
4640 will be encoded in the output file according to @var{encoding}; see the
4641 description of @code{.cfi_personality} for details on this encoding.
4643 The usefulness of equating a register to a fixed label is probably
4644 limited to the return address register. Here, it can be useful to
4645 mark a code segment that has only one return address which is reached
4646 by a direct branch and no copy of the return address exists in memory
4647 or another register.
4650 @section @code{.comm @var{symbol} , @var{length} }
4652 @cindex @code{comm} directive
4653 @cindex symbol, common
4654 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4655 common symbol in one object file may be merged with a defined or common symbol
4656 of the same name in another object file. If @code{@value{LD}} does not see a
4657 definition for the symbol--just one or more common symbols--then it will
4658 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4659 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4660 the same name, and they do not all have the same size, it will allocate space
4661 using the largest size.
4664 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4665 an optional third argument. This is the desired alignment of the symbol,
4666 specified for ELF as a byte boundary (for example, an alignment of 16 means
4667 that the least significant 4 bits of the address should be zero), and for PE
4668 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4669 boundary). The alignment must be an absolute expression, and it must be a
4670 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4671 common symbol, it will use the alignment when placing the symbol. If no
4672 alignment is specified, @command{@value{AS}} will set the alignment to the
4673 largest power of two less than or equal to the size of the symbol, up to a
4674 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4675 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4676 @samp{--section-alignment} option; image file sections in PE are aligned to
4677 multiples of 4096, which is far too large an alignment for ordinary variables.
4678 It is rather the default alignment for (non-debug) sections within object
4679 (@samp{*.o}) files, which are less strictly aligned.}.
4683 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4684 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4688 @section @code{.data @var{subsection}}
4690 @cindex @code{data} directive
4691 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4692 end of the data subsection numbered @var{subsection} (which is an
4693 absolute expression). If @var{subsection} is omitted, it defaults
4698 @section @code{.def @var{name}}
4700 @cindex @code{def} directive
4701 @cindex COFF symbols, debugging
4702 @cindex debugging COFF symbols
4703 Begin defining debugging information for a symbol @var{name}; the
4704 definition extends until the @code{.endef} directive is encountered.
4707 This directive is only observed when @command{@value{AS}} is configured for COFF
4708 format output; when producing @code{b.out}, @samp{.def} is recognized,
4715 @section @code{.desc @var{symbol}, @var{abs-expression}}
4717 @cindex @code{desc} directive
4718 @cindex COFF symbol descriptor
4719 @cindex symbol descriptor, COFF
4720 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4721 to the low 16 bits of an absolute expression.
4724 The @samp{.desc} directive is not available when @command{@value{AS}} is
4725 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4726 object format. For the sake of compatibility, @command{@value{AS}} accepts
4727 it, but produces no output, when configured for COFF.
4733 @section @code{.dim}
4735 @cindex @code{dim} directive
4736 @cindex COFF auxiliary symbol information
4737 @cindex auxiliary symbol information, COFF
4738 This directive is generated by compilers to include auxiliary debugging
4739 information in the symbol table. It is only permitted inside
4740 @code{.def}/@code{.endef} pairs.
4743 @samp{.dim} is only meaningful when generating COFF format output; when
4744 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4750 @section @code{.double @var{flonums}}
4752 @cindex @code{double} directive
4753 @cindex floating point numbers (double)
4754 @code{.double} expects zero or more flonums, separated by commas. It
4755 assembles floating point numbers.
4757 The exact kind of floating point numbers emitted depends on how
4758 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4762 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4763 in @sc{ieee} format.
4768 @section @code{.eject}
4770 @cindex @code{eject} directive
4771 @cindex new page, in listings
4772 @cindex page, in listings
4773 @cindex listing control: new page
4774 Force a page break at this point, when generating assembly listings.
4777 @section @code{.else}
4779 @cindex @code{else} directive
4780 @code{.else} is part of the @command{@value{AS}} support for conditional
4781 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4782 of code to be assembled if the condition for the preceding @code{.if}
4786 @section @code{.elseif}
4788 @cindex @code{elseif} directive
4789 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4790 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4791 @code{.if} block that would otherwise fill the entire @code{.else} section.
4794 @section @code{.end}
4796 @cindex @code{end} directive
4797 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4798 process anything in the file past the @code{.end} directive.
4802 @section @code{.endef}
4804 @cindex @code{endef} directive
4805 This directive flags the end of a symbol definition begun with
4809 @samp{.endef} is only meaningful when generating COFF format output; if
4810 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4811 directive but ignores it.
4816 @section @code{.endfunc}
4817 @cindex @code{endfunc} directive
4818 @code{.endfunc} marks the end of a function specified with @code{.func}.
4821 @section @code{.endif}
4823 @cindex @code{endif} directive
4824 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4825 it marks the end of a block of code that is only assembled
4826 conditionally. @xref{If,,@code{.if}}.
4829 @section @code{.equ @var{symbol}, @var{expression}}
4831 @cindex @code{equ} directive
4832 @cindex assigning values to symbols
4833 @cindex symbols, assigning values to
4834 This directive sets the value of @var{symbol} to @var{expression}.
4835 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4838 The syntax for @code{equ} on the HPPA is
4839 @samp{@var{symbol} .equ @var{expression}}.
4843 The syntax for @code{equ} on the Z80 is
4844 @samp{@var{symbol} equ @var{expression}}.
4845 On the Z80 it is an eror if @var{symbol} is already defined,
4846 but the symbol is not protected from later redefinition.
4847 Compare @ref{Equiv}.
4851 @section @code{.equiv @var{symbol}, @var{expression}}
4852 @cindex @code{equiv} directive
4853 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4854 the assembler will signal an error if @var{symbol} is already defined. Note a
4855 symbol which has been referenced but not actually defined is considered to be
4858 Except for the contents of the error message, this is roughly equivalent to
4865 plus it protects the symbol from later redefinition.
4868 @section @code{.eqv @var{symbol}, @var{expression}}
4869 @cindex @code{eqv} directive
4870 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
4871 evaluate the expression or any part of it immediately. Instead each time
4872 the resulting symbol is used in an expression, a snapshot of its current
4876 @section @code{.err}
4877 @cindex @code{err} directive
4878 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4879 message and, unless the @option{-Z} option was used, it will not generate an
4880 object file. This can be used to signal an error in conditionally compiled code.
4883 @section @code{.error "@var{string}"}
4884 @cindex error directive
4886 Similarly to @code{.err}, this directive emits an error, but you can specify a
4887 string that will be emitted as the error message. If you don't specify the
4888 message, it defaults to @code{".error directive invoked in source file"}.
4889 @xref{Errors, ,Error and Warning Messages}.
4892 .error "This code has not been assembled and tested."
4896 @section @code{.exitm}
4897 Exit early from the current macro definition. @xref{Macro}.
4900 @section @code{.extern}
4902 @cindex @code{extern} directive
4903 @code{.extern} is accepted in the source program---for compatibility
4904 with other assemblers---but it is ignored. @command{@value{AS}} treats
4905 all undefined symbols as external.
4908 @section @code{.fail @var{expression}}
4910 @cindex @code{fail} directive
4911 Generates an error or a warning. If the value of the @var{expression} is 500
4912 or more, @command{@value{AS}} will print a warning message. If the value is less
4913 than 500, @command{@value{AS}} will print an error message. The message will
4914 include the value of @var{expression}. This can occasionally be useful inside
4915 complex nested macros or conditional assembly.
4918 @section @code{.file}
4919 @cindex @code{file} directive
4921 @ifclear no-file-dir
4922 There are two different versions of the @code{.file} directive. Targets
4923 that support DWARF2 line number information use the DWARF2 version of
4924 @code{.file}. Other targets use the default version.
4926 @subheading Default Version
4928 @cindex logical file name
4929 @cindex file name, logical
4930 This version of the @code{.file} directive tells @command{@value{AS}} that we
4931 are about to start a new logical file. The syntax is:
4937 @var{string} is the new file name. In general, the filename is
4938 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4939 to specify an empty file name, you must give the quotes--@code{""}. This
4940 statement may go away in future: it is only recognized to be compatible with
4941 old @command{@value{AS}} programs.
4943 @subheading DWARF2 Version
4946 When emitting DWARF2 line number information, @code{.file} assigns filenames
4947 to the @code{.debug_line} file name table. The syntax is:
4950 .file @var{fileno} @var{filename}
4953 The @var{fileno} operand should be a unique positive integer to use as the
4954 index of the entry in the table. The @var{filename} operand is a C string
4957 The detail of filename indices is exposed to the user because the filename
4958 table is shared with the @code{.debug_info} section of the DWARF2 debugging
4959 information, and thus the user must know the exact indices that table
4963 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4965 @cindex @code{fill} directive
4966 @cindex writing patterns in memory
4967 @cindex patterns, writing in memory
4968 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4969 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4970 may be zero or more. @var{Size} may be zero or more, but if it is
4971 more than 8, then it is deemed to have the value 8, compatible with
4972 other people's assemblers. The contents of each @var{repeat} bytes
4973 is taken from an 8-byte number. The highest order 4 bytes are
4974 zero. The lowest order 4 bytes are @var{value} rendered in the
4975 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4976 Each @var{size} bytes in a repetition is taken from the lowest order
4977 @var{size} bytes of this number. Again, this bizarre behavior is
4978 compatible with other people's assemblers.
4980 @var{size} and @var{value} are optional.
4981 If the second comma and @var{value} are absent, @var{value} is
4982 assumed zero. If the first comma and following tokens are absent,
4983 @var{size} is assumed to be 1.
4986 @section @code{.float @var{flonums}}
4988 @cindex floating point numbers (single)
4989 @cindex @code{float} directive
4990 This directive assembles zero or more flonums, separated by commas. It
4991 has the same effect as @code{.single}.
4993 The exact kind of floating point numbers emitted depends on how
4994 @command{@value{AS}} is configured.
4995 @xref{Machine Dependencies}.
4999 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5000 in @sc{ieee} format.
5005 @section @code{.func @var{name}[,@var{label}]}
5006 @cindex @code{func} directive
5007 @code{.func} emits debugging information to denote function @var{name}, and
5008 is ignored unless the file is assembled with debugging enabled.
5009 Only @samp{--gstabs[+]} is currently supported.
5010 @var{label} is the entry point of the function and if omitted @var{name}
5011 prepended with the @samp{leading char} is used.
5012 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5013 All functions are currently defined to have @code{void} return type.
5014 The function must be terminated with @code{.endfunc}.
5017 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5019 @cindex @code{global} directive
5020 @cindex symbol, making visible to linker
5021 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5022 @var{symbol} in your partial program, its value is made available to
5023 other partial programs that are linked with it. Otherwise,
5024 @var{symbol} takes its attributes from a symbol of the same name
5025 from another file linked into the same program.
5027 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5028 compatibility with other assemblers.
5031 On the HPPA, @code{.global} is not always enough to make it accessible to other
5032 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5033 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5038 @section @code{.gnu_attribute @var{tag},@var{value}}
5039 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5042 @section @code{.hidden @var{names}}
5044 @cindex @code{hidden} directive
5046 This is one of the ELF visibility directives. The other two are
5047 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5048 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5050 This directive overrides the named symbols default visibility (which is set by
5051 their binding: local, global or weak). The directive sets the visibility to
5052 @code{hidden} which means that the symbols are not visible to other components.
5053 Such symbols are always considered to be @code{protected} as well.
5057 @section @code{.hword @var{expressions}}
5059 @cindex @code{hword} directive
5060 @cindex integers, 16-bit
5061 @cindex numbers, 16-bit
5062 @cindex sixteen bit integers
5063 This expects zero or more @var{expressions}, and emits
5064 a 16 bit number for each.
5067 This directive is a synonym for @samp{.short}; depending on the target
5068 architecture, it may also be a synonym for @samp{.word}.
5072 This directive is a synonym for @samp{.short}.
5075 This directive is a synonym for both @samp{.short} and @samp{.word}.
5080 @section @code{.ident}
5082 @cindex @code{ident} directive
5084 This directive is used by some assemblers to place tags in object files. The
5085 behavior of this directive varies depending on the target. When using the
5086 a.out object file format, @command{@value{AS}} simply accepts the directive for
5087 source-file compatibility with existing assemblers, but does not emit anything
5088 for it. When using COFF, comments are emitted to the @code{.comment} or
5089 @code{.rdata} section, depending on the target. When using ELF, comments are
5090 emitted to the @code{.comment} section.
5093 @section @code{.if @var{absolute expression}}
5095 @cindex conditional assembly
5096 @cindex @code{if} directive
5097 @code{.if} marks the beginning of a section of code which is only
5098 considered part of the source program being assembled if the argument
5099 (which must be an @var{absolute expression}) is non-zero. The end of
5100 the conditional section of code must be marked by @code{.endif}
5101 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5102 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5103 If you have several conditions to check, @code{.elseif} may be used to avoid
5104 nesting blocks if/else within each subsequent @code{.else} block.
5106 The following variants of @code{.if} are also supported:
5108 @cindex @code{ifdef} directive
5109 @item .ifdef @var{symbol}
5110 Assembles the following section of code if the specified @var{symbol}
5111 has been defined. Note a symbol which has been referenced but not yet defined
5112 is considered to be undefined.
5114 @cindex @code{ifb} directive
5115 @item .ifb @var{text}
5116 Assembles the following section of code if the operand is blank (empty).
5118 @cindex @code{ifc} directive
5119 @item .ifc @var{string1},@var{string2}
5120 Assembles the following section of code if the two strings are the same. The
5121 strings may be optionally quoted with single quotes. If they are not quoted,
5122 the first string stops at the first comma, and the second string stops at the
5123 end of the line. Strings which contain whitespace should be quoted. The
5124 string comparison is case sensitive.
5126 @cindex @code{ifeq} directive
5127 @item .ifeq @var{absolute expression}
5128 Assembles the following section of code if the argument is zero.
5130 @cindex @code{ifeqs} directive
5131 @item .ifeqs @var{string1},@var{string2}
5132 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5134 @cindex @code{ifge} directive
5135 @item .ifge @var{absolute expression}
5136 Assembles the following section of code if the argument is greater than or
5139 @cindex @code{ifgt} directive
5140 @item .ifgt @var{absolute expression}
5141 Assembles the following section of code if the argument is greater than zero.
5143 @cindex @code{ifle} directive
5144 @item .ifle @var{absolute expression}
5145 Assembles the following section of code if the argument is less than or equal
5148 @cindex @code{iflt} directive
5149 @item .iflt @var{absolute expression}
5150 Assembles the following section of code if the argument is less than zero.
5152 @cindex @code{ifnb} directive
5153 @item .ifnb @var{text}
5154 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5155 following section of code if the operand is non-blank (non-empty).
5157 @cindex @code{ifnc} directive
5158 @item .ifnc @var{string1},@var{string2}.
5159 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5160 following section of code if the two strings are not the same.
5162 @cindex @code{ifndef} directive
5163 @cindex @code{ifnotdef} directive
5164 @item .ifndef @var{symbol}
5165 @itemx .ifnotdef @var{symbol}
5166 Assembles the following section of code if the specified @var{symbol}
5167 has not been defined. Both spelling variants are equivalent. Note a symbol
5168 which has been referenced but not yet defined is considered to be undefined.
5170 @cindex @code{ifne} directive
5171 @item .ifne @var{absolute expression}
5172 Assembles the following section of code if the argument is not equal to zero
5173 (in other words, this is equivalent to @code{.if}).
5175 @cindex @code{ifnes} directive
5176 @item .ifnes @var{string1},@var{string2}
5177 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5178 following section of code if the two strings are not the same.
5182 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5184 @cindex @code{incbin} directive
5185 @cindex binary files, including
5186 The @code{incbin} directive includes @var{file} verbatim at the current
5187 location. You can control the search paths used with the @samp{-I} command-line
5188 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5191 The @var{skip} argument skips a number of bytes from the start of the
5192 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5193 read. Note that the data is not aligned in any way, so it is the user's
5194 responsibility to make sure that proper alignment is provided both before and
5195 after the @code{incbin} directive.
5198 @section @code{.include "@var{file}"}
5200 @cindex @code{include} directive
5201 @cindex supporting files, including
5202 @cindex files, including
5203 This directive provides a way to include supporting files at specified
5204 points in your source program. The code from @var{file} is assembled as
5205 if it followed the point of the @code{.include}; when the end of the
5206 included file is reached, assembly of the original file continues. You
5207 can control the search paths used with the @samp{-I} command-line option
5208 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5212 @section @code{.int @var{expressions}}
5214 @cindex @code{int} directive
5215 @cindex integers, 32-bit
5216 Expect zero or more @var{expressions}, of any section, separated by commas.
5217 For each expression, emit a number that, at run time, is the value of that
5218 expression. The byte order and bit size of the number depends on what kind
5219 of target the assembly is for.
5223 On most forms of the H8/300, @code{.int} emits 16-bit
5224 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5231 @section @code{.internal @var{names}}
5233 @cindex @code{internal} directive
5235 This is one of the ELF visibility directives. The other two are
5236 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5237 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5239 This directive overrides the named symbols default visibility (which is set by
5240 their binding: local, global or weak). The directive sets the visibility to
5241 @code{internal} which means that the symbols are considered to be @code{hidden}
5242 (i.e., not visible to other components), and that some extra, processor specific
5243 processing must also be performed upon the symbols as well.
5247 @section @code{.irp @var{symbol},@var{values}}@dots{}
5249 @cindex @code{irp} directive
5250 Evaluate a sequence of statements assigning different values to @var{symbol}.
5251 The sequence of statements starts at the @code{.irp} directive, and is
5252 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5253 set to @var{value}, and the sequence of statements is assembled. If no
5254 @var{value} is listed, the sequence of statements is assembled once, with
5255 @var{symbol} set to the null string. To refer to @var{symbol} within the
5256 sequence of statements, use @var{\symbol}.
5258 For example, assembling
5266 is equivalent to assembling
5274 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5277 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5279 @cindex @code{irpc} directive
5280 Evaluate a sequence of statements assigning different values to @var{symbol}.
5281 The sequence of statements starts at the @code{.irpc} directive, and is
5282 terminated by an @code{.endr} directive. For each character in @var{value},
5283 @var{symbol} is set to the character, and the sequence of statements is
5284 assembled. If no @var{value} is listed, the sequence of statements is
5285 assembled once, with @var{symbol} set to the null string. To refer to
5286 @var{symbol} within the sequence of statements, use @var{\symbol}.
5288 For example, assembling
5296 is equivalent to assembling
5304 For some caveats with the spelling of @var{symbol}, see also the discussion
5308 @section @code{.lcomm @var{symbol} , @var{length}}
5310 @cindex @code{lcomm} directive
5311 @cindex local common symbols
5312 @cindex symbols, local common
5313 Reserve @var{length} (an absolute expression) bytes for a local common
5314 denoted by @var{symbol}. The section and value of @var{symbol} are
5315 those of the new local common. The addresses are allocated in the bss
5316 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5317 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5318 not visible to @code{@value{LD}}.
5321 Some targets permit a third argument to be used with @code{.lcomm}. This
5322 argument specifies the desired alignment of the symbol in the bss section.
5326 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5327 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5331 @section @code{.lflags}
5333 @cindex @code{lflags} directive (ignored)
5334 @command{@value{AS}} accepts this directive, for compatibility with other
5335 assemblers, but ignores it.
5337 @ifclear no-line-dir
5339 @section @code{.line @var{line-number}}
5341 @cindex @code{line} directive
5342 @cindex logical line number
5344 Change the logical line number. @var{line-number} must be an absolute
5345 expression. The next line has that logical line number. Therefore any other
5346 statements on the current line (after a statement separator character) are
5347 reported as on logical line number @var{line-number} @minus{} 1. One day
5348 @command{@value{AS}} will no longer support this directive: it is recognized only
5349 for compatibility with existing assembler programs.
5352 Even though this is a directive associated with the @code{a.out} or
5353 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5354 when producing COFF output, and treats @samp{.line} as though it
5355 were the COFF @samp{.ln} @emph{if} it is found outside a
5356 @code{.def}/@code{.endef} pair.
5358 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5359 used by compilers to generate auxiliary symbol information for
5364 @section @code{.linkonce [@var{type}]}
5366 @cindex @code{linkonce} directive
5367 @cindex common sections
5368 Mark the current section so that the linker only includes a single copy of it.
5369 This may be used to include the same section in several different object files,
5370 but ensure that the linker will only include it once in the final output file.
5371 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5372 Duplicate sections are detected based on the section name, so it should be
5375 This directive is only supported by a few object file formats; as of this
5376 writing, the only object file format which supports it is the Portable
5377 Executable format used on Windows NT.
5379 The @var{type} argument is optional. If specified, it must be one of the
5380 following strings. For example:
5384 Not all types may be supported on all object file formats.
5388 Silently discard duplicate sections. This is the default.
5391 Warn if there are duplicate sections, but still keep only one copy.
5394 Warn if any of the duplicates have different sizes.
5397 Warn if any of the duplicates do not have exactly the same contents.
5401 @section @code{.list}
5403 @cindex @code{list} directive
5404 @cindex listing control, turning on
5405 Control (in conjunction with the @code{.nolist} directive) whether or
5406 not assembly listings are generated. These two directives maintain an
5407 internal counter (which is zero initially). @code{.list} increments the
5408 counter, and @code{.nolist} decrements it. Assembly listings are
5409 generated whenever the counter is greater than zero.
5411 By default, listings are disabled. When you enable them (with the
5412 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5413 the initial value of the listing counter is one.
5416 @section @code{.ln @var{line-number}}
5418 @cindex @code{ln} directive
5419 @ifclear no-line-dir
5420 @samp{.ln} is a synonym for @samp{.line}.
5423 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5424 must be an absolute expression. The next line has that logical
5425 line number, so any other statements on the current line (after a
5426 statement separator character @code{;}) are reported as on logical
5427 line number @var{line-number} @minus{} 1.
5430 This directive is accepted, but ignored, when @command{@value{AS}} is
5431 configured for @code{b.out}; its effect is only associated with COFF
5437 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5438 @cindex @code{loc} directive
5439 When emitting DWARF2 line number information,
5440 the @code{.loc} directive will add a row to the @code{.debug_line} line
5441 number matrix corresponding to the immediately following assembly
5442 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5443 arguments will be applied to the @code{.debug_line} state machine before
5446 The @var{options} are a sequence of the following tokens in any order:
5450 This option will set the @code{basic_block} register in the
5451 @code{.debug_line} state machine to @code{true}.
5454 This option will set the @code{prologue_end} register in the
5455 @code{.debug_line} state machine to @code{true}.
5457 @item epilogue_begin
5458 This option will set the @code{epilogue_begin} register in the
5459 @code{.debug_line} state machine to @code{true}.
5461 @item is_stmt @var{value}
5462 This option will set the @code{is_stmt} register in the
5463 @code{.debug_line} state machine to @code{value}, which must be
5466 @item isa @var{value}
5467 This directive will set the @code{isa} register in the @code{.debug_line}
5468 state machine to @var{value}, which must be an unsigned integer.
5470 @item discriminator @var{value}
5471 This directive will set the @code{discriminator} register in the @code{.debug_line}
5472 state machine to @var{value}, which must be an unsigned integer.
5476 @node Loc_mark_labels
5477 @section @code{.loc_mark_labels @var{enable}}
5478 @cindex @code{loc_mark_labels} directive
5479 When emitting DWARF2 line number information,
5480 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5481 to the @code{.debug_line} line number matrix with the @code{basic_block}
5482 register in the state machine set whenever a code label is seen.
5483 The @var{enable} argument should be either 1 or 0, to enable or disable
5484 this function respectively.
5488 @section @code{.local @var{names}}
5490 @cindex @code{local} directive
5491 This directive, which is available for ELF targets, marks each symbol in
5492 the comma-separated list of @code{names} as a local symbol so that it
5493 will not be externally visible. If the symbols do not already exist,
5494 they will be created.
5496 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5497 accept an alignment argument, which is the case for most ELF targets,
5498 the @code{.local} directive can be used in combination with @code{.comm}
5499 (@pxref{Comm}) to define aligned local common data.
5503 @section @code{.long @var{expressions}}
5505 @cindex @code{long} directive
5506 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5509 @c no one seems to know what this is for or whether this description is
5510 @c what it really ought to do
5512 @section @code{.lsym @var{symbol}, @var{expression}}
5514 @cindex @code{lsym} directive
5515 @cindex symbol, not referenced in assembly
5516 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5517 the hash table, ensuring it cannot be referenced by name during the
5518 rest of the assembly. This sets the attributes of the symbol to be
5519 the same as the expression value:
5521 @var{other} = @var{descriptor} = 0
5522 @var{type} = @r{(section of @var{expression})}
5523 @var{value} = @var{expression}
5526 The new symbol is not flagged as external.
5530 @section @code{.macro}
5533 The commands @code{.macro} and @code{.endm} allow you to define macros that
5534 generate assembly output. For example, this definition specifies a macro
5535 @code{sum} that puts a sequence of numbers into memory:
5538 .macro sum from=0, to=5
5547 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5559 @item .macro @var{macname}
5560 @itemx .macro @var{macname} @var{macargs} @dots{}
5561 @cindex @code{macro} directive
5562 Begin the definition of a macro called @var{macname}. If your macro
5563 definition requires arguments, specify their names after the macro name,
5564 separated by commas or spaces. You can qualify the macro argument to
5565 indicate whether all invocations must specify a non-blank value (through
5566 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5567 (through @samp{:@code{vararg}}). You can supply a default value for any
5568 macro argument by following the name with @samp{=@var{deflt}}. You
5569 cannot define two macros with the same @var{macname} unless it has been
5570 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5571 definitions. For example, these are all valid @code{.macro} statements:
5575 Begin the definition of a macro called @code{comm}, which takes no
5578 @item .macro plus1 p, p1
5579 @itemx .macro plus1 p p1
5580 Either statement begins the definition of a macro called @code{plus1},
5581 which takes two arguments; within the macro definition, write
5582 @samp{\p} or @samp{\p1} to evaluate the arguments.
5584 @item .macro reserve_str p1=0 p2
5585 Begin the definition of a macro called @code{reserve_str}, with two
5586 arguments. The first argument has a default value, but not the second.
5587 After the definition is complete, you can call the macro either as
5588 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5589 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5590 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5591 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5593 @item .macro m p1:req, p2=0, p3:vararg
5594 Begin the definition of a macro called @code{m}, with at least three
5595 arguments. The first argument must always have a value specified, but
5596 not the second, which instead has a default value. The third formal
5597 will get assigned all remaining arguments specified at invocation time.
5599 When you call a macro, you can specify the argument values either by
5600 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5601 @samp{sum to=17, from=9}.
5605 Note that since each of the @var{macargs} can be an identifier exactly
5606 as any other one permitted by the target architecture, there may be
5607 occasional problems if the target hand-crafts special meanings to certain
5608 characters when they occur in a special position. For example, if the colon
5609 (@code{:}) is generally permitted to be part of a symbol name, but the
5610 architecture specific code special-cases it when occurring as the final
5611 character of a symbol (to denote a label), then the macro parameter
5612 replacement code will have no way of knowing that and consider the whole
5613 construct (including the colon) an identifier, and check only this
5614 identifier for being the subject to parameter substitution. So for example
5615 this macro definition:
5623 might not work as expected. Invoking @samp{label foo} might not create a label
5624 called @samp{foo} but instead just insert the text @samp{\l:} into the
5625 assembler source, probably generating an error about an unrecognised
5628 Similarly problems might occur with the period character (@samp{.})
5629 which is often allowed inside opcode names (and hence identifier names). So
5630 for example constructing a macro to build an opcode from a base name and a
5631 length specifier like this:
5634 .macro opcode base length
5639 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5640 instruction but instead generate some kind of error as the assembler tries to
5641 interpret the text @samp{\base.\length}.
5643 There are several possible ways around this problem:
5646 @item Insert white space
5647 If it is possible to use white space characters then this is the simplest
5656 @item Use @samp{\()}
5657 The string @samp{\()} can be used to separate the end of a macro argument from
5658 the following text. eg:
5661 .macro opcode base length
5666 @item Use the alternate macro syntax mode
5667 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5668 used as a separator. eg:
5678 Note: this problem of correctly identifying string parameters to pseudo ops
5679 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5680 and @code{.irpc} (@pxref{Irpc}) as well.
5683 @cindex @code{endm} directive
5684 Mark the end of a macro definition.
5687 @cindex @code{exitm} directive
5688 Exit early from the current macro definition.
5690 @cindex number of macros executed
5691 @cindex macros, count executed
5693 @command{@value{AS}} maintains a counter of how many macros it has
5694 executed in this pseudo-variable; you can copy that number to your
5695 output with @samp{\@@}, but @emph{only within a macro definition}.
5697 @item LOCAL @var{name} [ , @dots{} ]
5698 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5699 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5700 @xref{Altmacro,,@code{.altmacro}}.
5704 @section @code{.mri @var{val}}
5706 @cindex @code{mri} directive
5707 @cindex MRI mode, temporarily
5708 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5709 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5710 affects code assembled until the next @code{.mri} directive, or until the end
5711 of the file. @xref{M, MRI mode, MRI mode}.
5714 @section @code{.noaltmacro}
5715 Disable alternate macro mode. @xref{Altmacro}.
5718 @section @code{.nolist}
5720 @cindex @code{nolist} directive
5721 @cindex listing control, turning off
5722 Control (in conjunction with the @code{.list} directive) whether or
5723 not assembly listings are generated. These two directives maintain an
5724 internal counter (which is zero initially). @code{.list} increments the
5725 counter, and @code{.nolist} decrements it. Assembly listings are
5726 generated whenever the counter is greater than zero.
5729 @section @code{.octa @var{bignums}}
5731 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5732 @cindex @code{octa} directive
5733 @cindex integer, 16-byte
5734 @cindex sixteen byte integer
5735 This directive expects zero or more bignums, separated by commas. For each
5736 bignum, it emits a 16-byte integer.
5738 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5739 hence @emph{octa}-word for 16 bytes.
5742 @section @code{.offset @var{loc}}
5744 @cindex @code{offset} directive
5745 Set the location counter to @var{loc} in the absolute section. @var{loc} must
5746 be an absolute expression. This directive may be useful for defining
5747 symbols with absolute values. Do not confuse it with the @code{.org}
5751 @section @code{.org @var{new-lc} , @var{fill}}
5753 @cindex @code{org} directive
5754 @cindex location counter, advancing
5755 @cindex advancing location counter
5756 @cindex current address, advancing
5757 Advance the location counter of the current section to
5758 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5759 expression with the same section as the current subsection. That is,
5760 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5761 wrong section, the @code{.org} directive is ignored. To be compatible
5762 with former assemblers, if the section of @var{new-lc} is absolute,
5763 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5764 is the same as the current subsection.
5766 @code{.org} may only increase the location counter, or leave it
5767 unchanged; you cannot use @code{.org} to move the location counter
5770 @c double negative used below "not undefined" because this is a specific
5771 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5772 @c section. doc@cygnus.com 18feb91
5773 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5774 may not be undefined. If you really detest this restriction we eagerly await
5775 a chance to share your improved assembler.
5777 Beware that the origin is relative to the start of the section, not
5778 to the start of the subsection. This is compatible with other
5779 people's assemblers.
5781 When the location counter (of the current subsection) is advanced, the
5782 intervening bytes are filled with @var{fill} which should be an
5783 absolute expression. If the comma and @var{fill} are omitted,
5784 @var{fill} defaults to zero.
5787 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5789 @cindex padding the location counter given a power of two
5790 @cindex @code{p2align} directive
5791 Pad the location counter (in the current subsection) to a particular
5792 storage boundary. The first expression (which must be absolute) is the
5793 number of low-order zero bits the location counter must have after
5794 advancement. For example @samp{.p2align 3} advances the location
5795 counter until it a multiple of 8. If the location counter is already a
5796 multiple of 8, no change is needed.
5798 The second expression (also absolute) gives the fill value to be stored in the
5799 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5800 padding bytes are normally zero. However, on some systems, if the section is
5801 marked as containing code and the fill value is omitted, the space is filled
5802 with no-op instructions.
5804 The third expression is also absolute, and is also optional. If it is present,
5805 it is the maximum number of bytes that should be skipped by this alignment
5806 directive. If doing the alignment would require skipping more bytes than the
5807 specified maximum, then the alignment is not done at all. You can omit the
5808 fill value (the second argument) entirely by simply using two commas after the
5809 required alignment; this can be useful if you want the alignment to be filled
5810 with no-op instructions when appropriate.
5812 @cindex @code{p2alignw} directive
5813 @cindex @code{p2alignl} directive
5814 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5815 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5816 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5817 fill pattern as a four byte longword value. For example, @code{.p2alignw
5818 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5819 filled in with the value 0x368d (the exact placement of the bytes depends upon
5820 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5825 @section @code{.popsection}
5827 @cindex @code{popsection} directive
5828 @cindex Section Stack
5829 This is one of the ELF section stack manipulation directives. The others are
5830 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5831 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5834 This directive replaces the current section (and subsection) with the top
5835 section (and subsection) on the section stack. This section is popped off the
5841 @section @code{.previous}
5843 @cindex @code{previous} directive
5844 @cindex Section Stack
5845 This is one of the ELF section stack manipulation directives. The others are
5846 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5847 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5848 (@pxref{PopSection}).
5850 This directive swaps the current section (and subsection) with most recently
5851 referenced section/subsection pair prior to this one. Multiple
5852 @code{.previous} directives in a row will flip between two sections (and their
5853 subsections). For example:
5865 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
5871 # Now in section A subsection 1
5875 # Now in section B subsection 0
5878 # Now in section B subsection 1
5881 # Now in section B subsection 0
5885 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
5886 section B and 0x9abc into subsection 1 of section B.
5888 In terms of the section stack, this directive swaps the current section with
5889 the top section on the section stack.
5893 @section @code{.print @var{string}}
5895 @cindex @code{print} directive
5896 @command{@value{AS}} will print @var{string} on the standard output during
5897 assembly. You must put @var{string} in double quotes.
5901 @section @code{.protected @var{names}}
5903 @cindex @code{protected} directive
5905 This is one of the ELF visibility directives. The other two are
5906 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5908 This directive overrides the named symbols default visibility (which is set by
5909 their binding: local, global or weak). The directive sets the visibility to
5910 @code{protected} which means that any references to the symbols from within the
5911 components that defines them must be resolved to the definition in that
5912 component, even if a definition in another component would normally preempt
5917 @section @code{.psize @var{lines} , @var{columns}}
5919 @cindex @code{psize} directive
5920 @cindex listing control: paper size
5921 @cindex paper size, for listings
5922 Use this directive to declare the number of lines---and, optionally, the
5923 number of columns---to use for each page, when generating listings.
5925 If you do not use @code{.psize}, listings use a default line-count
5926 of 60. You may omit the comma and @var{columns} specification; the
5927 default width is 200 columns.
5929 @command{@value{AS}} generates formfeeds whenever the specified number of
5930 lines is exceeded (or whenever you explicitly request one, using
5933 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5934 those explicitly specified with @code{.eject}.
5937 @section @code{.purgem @var{name}}
5939 @cindex @code{purgem} directive
5940 Undefine the macro @var{name}, so that later uses of the string will not be
5941 expanded. @xref{Macro}.
5945 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
5947 @cindex @code{pushsection} directive
5948 @cindex Section Stack
5949 This is one of the ELF section stack manipulation directives. The others are
5950 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5951 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5954 This directive pushes the current section (and subsection) onto the
5955 top of the section stack, and then replaces the current section and
5956 subsection with @code{name} and @code{subsection}. The optional
5957 @code{flags}, @code{type} and @code{arguments} are treated the same
5958 as in the @code{.section} (@pxref{Section}) directive.
5962 @section @code{.quad @var{bignums}}
5964 @cindex @code{quad} directive
5965 @code{.quad} expects zero or more bignums, separated by commas. For
5966 each bignum, it emits
5968 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5969 warning message; and just takes the lowest order 8 bytes of the bignum.
5970 @cindex eight-byte integer
5971 @cindex integer, 8-byte
5973 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5974 hence @emph{quad}-word for 8 bytes.
5977 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5978 warning message; and just takes the lowest order 16 bytes of the bignum.
5979 @cindex sixteen-byte integer
5980 @cindex integer, 16-byte
5984 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
5986 @cindex @code{reloc} directive
5987 Generate a relocation at @var{offset} of type @var{reloc_name} with value
5988 @var{expression}. If @var{offset} is a number, the relocation is generated in
5989 the current section. If @var{offset} is an expression that resolves to a
5990 symbol plus offset, the relocation is generated in the given symbol's section.
5991 @var{expression}, if present, must resolve to a symbol plus addend or to an
5992 absolute value, but note that not all targets support an addend. e.g. ELF REL
5993 targets such as i386 store an addend in the section contents rather than in the
5994 relocation. This low level interface does not support addends stored in the
5998 @section @code{.rept @var{count}}
6000 @cindex @code{rept} directive
6001 Repeat the sequence of lines between the @code{.rept} directive and the next
6002 @code{.endr} directive @var{count} times.
6004 For example, assembling
6012 is equivalent to assembling
6021 @section @code{.sbttl "@var{subheading}"}
6023 @cindex @code{sbttl} directive
6024 @cindex subtitles for listings
6025 @cindex listing control: subtitle
6026 Use @var{subheading} as the title (third line, immediately after the
6027 title line) when generating assembly listings.
6029 This directive affects subsequent pages, as well as the current page if
6030 it appears within ten lines of the top of a page.
6034 @section @code{.scl @var{class}}
6036 @cindex @code{scl} directive
6037 @cindex symbol storage class (COFF)
6038 @cindex COFF symbol storage class
6039 Set the storage-class value for a symbol. This directive may only be
6040 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6041 whether a symbol is static or external, or it may record further
6042 symbolic debugging information.
6045 The @samp{.scl} directive is primarily associated with COFF output; when
6046 configured to generate @code{b.out} output format, @command{@value{AS}}
6047 accepts this directive but ignores it.
6053 @section @code{.section @var{name}}
6055 @cindex named section
6056 Use the @code{.section} directive to assemble the following code into a section
6059 This directive is only supported for targets that actually support arbitrarily
6060 named sections; on @code{a.out} targets, for example, it is not accepted, even
6061 with a standard @code{a.out} section name.
6065 @c only print the extra heading if both COFF and ELF are set
6066 @subheading COFF Version
6069 @cindex @code{section} directive (COFF version)
6070 For COFF targets, the @code{.section} directive is used in one of the following
6074 .section @var{name}[, "@var{flags}"]
6075 .section @var{name}[, @var{subsection}]
6078 If the optional argument is quoted, it is taken as flags to use for the
6079 section. Each flag is a single character. The following flags are recognized:
6082 bss section (uninitialized data)
6084 section is not loaded
6090 exclude section from linking
6096 shared section (meaningful for PE targets)
6098 ignored. (For compatibility with the ELF version)
6100 section is not readable (meaningful for PE targets)
6102 single-digit power-of-two section alignment (GNU extension)
6105 If no flags are specified, the default flags depend upon the section name. If
6106 the section name is not recognized, the default will be for the section to be
6107 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6108 from the section, rather than adding them, so if they are used on their own it
6109 will be as if no flags had been specified at all.
6111 If the optional argument to the @code{.section} directive is not quoted, it is
6112 taken as a subsection number (@pxref{Sub-Sections}).
6117 @c only print the extra heading if both COFF and ELF are set
6118 @subheading ELF Version
6121 @cindex Section Stack
6122 This is one of the ELF section stack manipulation directives. The others are
6123 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6124 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6125 @code{.previous} (@pxref{Previous}).
6127 @cindex @code{section} directive (ELF version)
6128 For ELF targets, the @code{.section} directive is used like this:
6131 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6134 The optional @var{flags} argument is a quoted string which may contain any
6135 combination of the following characters:
6138 section is allocatable
6140 section is excluded from executable and shared library.
6144 section is executable
6146 section is mergeable
6148 section contains zero terminated strings
6150 section is a member of a section group
6152 section is used for thread-local-storage
6154 section is a member of the previously-current section's group, if any
6157 The optional @var{type} argument may contain one of the following constants:
6160 section contains data
6162 section does not contain data (i.e., section only occupies space)
6164 section contains data which is used by things other than the program
6166 section contains an array of pointers to init functions
6168 section contains an array of pointers to finish functions
6169 @item @@preinit_array
6170 section contains an array of pointers to pre-init functions
6173 Many targets only support the first three section types.
6175 Note on targets where the @code{@@} character is the start of a comment (eg
6176 ARM) then another character is used instead. For example the ARM port uses the
6179 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6180 be specified as well as an extra argument---@var{entsize}---like this:
6183 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6186 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6187 constants, each @var{entsize} octets long. Sections with both @code{M} and
6188 @code{S} must contain zero terminated strings where each character is
6189 @var{entsize} bytes long. The linker may remove duplicates within sections with
6190 the same name, same entity size and same flags. @var{entsize} must be an
6191 absolute expression. For sections with both @code{M} and @code{S}, a string
6192 which is a suffix of a larger string is considered a duplicate. Thus
6193 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6194 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6196 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6197 be present along with an additional field like this:
6200 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6203 The @var{GroupName} field specifies the name of the section group to which this
6204 particular section belongs. The optional linkage field can contain:
6207 indicates that only one copy of this section should be retained
6212 Note: if both the @var{M} and @var{G} flags are present then the fields for
6213 the Merge flag should come first, like this:
6216 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6219 If @var{flags} contains the @code{?} symbol then it may not also contain the
6220 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6221 present. Instead, @code{?} says to consider the section that's current before
6222 this directive. If that section used @code{G}, then the new section will use
6223 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6224 If not, then the @code{?} symbol has no effect.
6226 If no flags are specified, the default flags depend upon the section name. If
6227 the section name is not recognized, the default will be for the section to have
6228 none of the above flags: it will not be allocated in memory, nor writable, nor
6229 executable. The section will contain data.
6231 For ELF targets, the assembler supports another type of @code{.section}
6232 directive for compatibility with the Solaris assembler:
6235 .section "@var{name}"[, @var{flags}...]
6238 Note that the section name is quoted. There may be a sequence of comma
6242 section is allocatable
6246 section is executable
6248 section is excluded from executable and shared library.
6250 section is used for thread local storage
6253 This directive replaces the current section and subsection. See the
6254 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6255 some examples of how this directive and the other section stack directives
6261 @section @code{.set @var{symbol}, @var{expression}}
6263 @cindex @code{set} directive
6264 @cindex symbol value, setting
6265 Set the value of @var{symbol} to @var{expression}. This
6266 changes @var{symbol}'s value and type to conform to
6267 @var{expression}. If @var{symbol} was flagged as external, it remains
6268 flagged (@pxref{Symbol Attributes}).
6270 You may @code{.set} a symbol many times in the same assembly.
6272 If you @code{.set} a global symbol, the value stored in the object
6273 file is the last value stored into it.
6276 On Z80 @code{set} is a real instruction, use
6277 @samp{@var{symbol} defl @var{expression}} instead.
6281 @section @code{.short @var{expressions}}
6283 @cindex @code{short} directive
6285 @code{.short} is normally the same as @samp{.word}.
6286 @xref{Word,,@code{.word}}.
6288 In some configurations, however, @code{.short} and @code{.word} generate
6289 numbers of different lengths. @xref{Machine Dependencies}.
6293 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6296 This expects zero or more @var{expressions}, and emits
6297 a 16 bit number for each.
6302 @section @code{.single @var{flonums}}
6304 @cindex @code{single} directive
6305 @cindex floating point numbers (single)
6306 This directive assembles zero or more flonums, separated by commas. It
6307 has the same effect as @code{.float}.
6309 The exact kind of floating point numbers emitted depends on how
6310 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6314 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6315 numbers in @sc{ieee} format.
6321 @section @code{.size}
6323 This directive is used to set the size associated with a symbol.
6327 @c only print the extra heading if both COFF and ELF are set
6328 @subheading COFF Version
6331 @cindex @code{size} directive (COFF version)
6332 For COFF targets, the @code{.size} directive is only permitted inside
6333 @code{.def}/@code{.endef} pairs. It is used like this:
6336 .size @var{expression}
6340 @samp{.size} is only meaningful when generating COFF format output; when
6341 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6348 @c only print the extra heading if both COFF and ELF are set
6349 @subheading ELF Version
6352 @cindex @code{size} directive (ELF version)
6353 For ELF targets, the @code{.size} directive is used like this:
6356 .size @var{name} , @var{expression}
6359 This directive sets the size associated with a symbol @var{name}.
6360 The size in bytes is computed from @var{expression} which can make use of label
6361 arithmetic. This directive is typically used to set the size of function
6366 @ifclear no-space-dir
6368 @section @code{.skip @var{size} , @var{fill}}
6370 @cindex @code{skip} directive
6371 @cindex filling memory
6372 This directive emits @var{size} bytes, each of value @var{fill}. Both
6373 @var{size} and @var{fill} are absolute expressions. If the comma and
6374 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6379 @section @code{.sleb128 @var{expressions}}
6381 @cindex @code{sleb128} directive
6382 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6383 compact, variable length representation of numbers used by the DWARF
6384 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6386 @ifclear no-space-dir
6388 @section @code{.space @var{size} , @var{fill}}
6390 @cindex @code{space} directive
6391 @cindex filling memory
6392 This directive emits @var{size} bytes, each of value @var{fill}. Both
6393 @var{size} and @var{fill} are absolute expressions. If the comma
6394 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6399 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6400 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6401 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6402 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6410 @section @code{.stabd, .stabn, .stabs}
6412 @cindex symbolic debuggers, information for
6413 @cindex @code{stab@var{x}} directives
6414 There are three directives that begin @samp{.stab}.
6415 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6416 The symbols are not entered in the @command{@value{AS}} hash table: they
6417 cannot be referenced elsewhere in the source file.
6418 Up to five fields are required:
6422 This is the symbol's name. It may contain any character except
6423 @samp{\000}, so is more general than ordinary symbol names. Some
6424 debuggers used to code arbitrarily complex structures into symbol names
6428 An absolute expression. The symbol's type is set to the low 8 bits of
6429 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6430 and debuggers choke on silly bit patterns.
6433 An absolute expression. The symbol's ``other'' attribute is set to the
6434 low 8 bits of this expression.
6437 An absolute expression. The symbol's descriptor is set to the low 16
6438 bits of this expression.
6441 An absolute expression which becomes the symbol's value.
6444 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6445 or @code{.stabs} statement, the symbol has probably already been created;
6446 you get a half-formed symbol in your object file. This is
6447 compatible with earlier assemblers!
6450 @cindex @code{stabd} directive
6451 @item .stabd @var{type} , @var{other} , @var{desc}
6453 The ``name'' of the symbol generated is not even an empty string.
6454 It is a null pointer, for compatibility. Older assemblers used a
6455 null pointer so they didn't waste space in object files with empty
6458 The symbol's value is set to the location counter,
6459 relocatably. When your program is linked, the value of this symbol
6460 is the address of the location counter when the @code{.stabd} was
6463 @cindex @code{stabn} directive
6464 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6465 The name of the symbol is set to the empty string @code{""}.
6467 @cindex @code{stabs} directive
6468 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6469 All five fields are specified.
6475 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6476 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6478 @cindex string, copying to object file
6479 @cindex string8, copying to object file
6480 @cindex string16, copying to object file
6481 @cindex string32, copying to object file
6482 @cindex string64, copying to object file
6483 @cindex @code{string} directive
6484 @cindex @code{string8} directive
6485 @cindex @code{string16} directive
6486 @cindex @code{string32} directive
6487 @cindex @code{string64} directive
6489 Copy the characters in @var{str} to the object file. You may specify more than
6490 one string to copy, separated by commas. Unless otherwise specified for a
6491 particular machine, the assembler marks the end of each string with a 0 byte.
6492 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6494 The variants @code{string16}, @code{string32} and @code{string64} differ from
6495 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6496 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6497 are stored in target endianness byte order.
6503 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6504 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6509 @section @code{.struct @var{expression}}
6511 @cindex @code{struct} directive
6512 Switch to the absolute section, and set the section offset to @var{expression},
6513 which must be an absolute expression. You might use this as follows:
6522 This would define the symbol @code{field1} to have the value 0, the symbol
6523 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6524 value 8. Assembly would be left in the absolute section, and you would need to
6525 use a @code{.section} directive of some sort to change to some other section
6526 before further assembly.
6530 @section @code{.subsection @var{name}}
6532 @cindex @code{subsection} directive
6533 @cindex Section Stack
6534 This is one of the ELF section stack manipulation directives. The others are
6535 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6536 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6539 This directive replaces the current subsection with @code{name}. The current
6540 section is not changed. The replaced subsection is put onto the section stack
6541 in place of the then current top of stack subsection.
6546 @section @code{.symver}
6547 @cindex @code{symver} directive
6548 @cindex symbol versioning
6549 @cindex versions of symbols
6550 Use the @code{.symver} directive to bind symbols to specific version nodes
6551 within a source file. This is only supported on ELF platforms, and is
6552 typically used when assembling files to be linked into a shared library.
6553 There are cases where it may make sense to use this in objects to be bound
6554 into an application itself so as to override a versioned symbol from a
6557 For ELF targets, the @code{.symver} directive can be used like this:
6559 .symver @var{name}, @var{name2@@nodename}
6561 If the symbol @var{name} is defined within the file
6562 being assembled, the @code{.symver} directive effectively creates a symbol
6563 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6564 just don't try and create a regular alias is that the @var{@@} character isn't
6565 permitted in symbol names. The @var{name2} part of the name is the actual name
6566 of the symbol by which it will be externally referenced. The name @var{name}
6567 itself is merely a name of convenience that is used so that it is possible to
6568 have definitions for multiple versions of a function within a single source
6569 file, and so that the compiler can unambiguously know which version of a
6570 function is being mentioned. The @var{nodename} portion of the alias should be
6571 the name of a node specified in the version script supplied to the linker when
6572 building a shared library. If you are attempting to override a versioned
6573 symbol from a shared library, then @var{nodename} should correspond to the
6574 nodename of the symbol you are trying to override.
6576 If the symbol @var{name} is not defined within the file being assembled, all
6577 references to @var{name} will be changed to @var{name2@@nodename}. If no
6578 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6581 Another usage of the @code{.symver} directive is:
6583 .symver @var{name}, @var{name2@@@@nodename}
6585 In this case, the symbol @var{name} must exist and be defined within
6586 the file being assembled. It is similar to @var{name2@@nodename}. The
6587 difference is @var{name2@@@@nodename} will also be used to resolve
6588 references to @var{name2} by the linker.
6590 The third usage of the @code{.symver} directive is:
6592 .symver @var{name}, @var{name2@@@@@@nodename}
6594 When @var{name} is not defined within the
6595 file being assembled, it is treated as @var{name2@@nodename}. When
6596 @var{name} is defined within the file being assembled, the symbol
6597 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6602 @section @code{.tag @var{structname}}
6604 @cindex COFF structure debugging
6605 @cindex structure debugging, COFF
6606 @cindex @code{tag} directive
6607 This directive is generated by compilers to include auxiliary debugging
6608 information in the symbol table. It is only permitted inside
6609 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6610 definitions in the symbol table with instances of those structures.
6613 @samp{.tag} is only used when generating COFF format output; when
6614 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6620 @section @code{.text @var{subsection}}
6622 @cindex @code{text} directive
6623 Tells @command{@value{AS}} to assemble the following statements onto the end of
6624 the text subsection numbered @var{subsection}, which is an absolute
6625 expression. If @var{subsection} is omitted, subsection number zero
6629 @section @code{.title "@var{heading}"}
6631 @cindex @code{title} directive
6632 @cindex listing control: title line
6633 Use @var{heading} as the title (second line, immediately after the
6634 source file name and pagenumber) when generating assembly listings.
6636 This directive affects subsequent pages, as well as the current page if
6637 it appears within ten lines of the top of a page.
6641 @section @code{.type}
6643 This directive is used to set the type of a symbol.
6647 @c only print the extra heading if both COFF and ELF are set
6648 @subheading COFF Version
6651 @cindex COFF symbol type
6652 @cindex symbol type, COFF
6653 @cindex @code{type} directive (COFF version)
6654 For COFF targets, this directive is permitted only within
6655 @code{.def}/@code{.endef} pairs. It is used like this:
6661 This records the integer @var{int} as the type attribute of a symbol table
6665 @samp{.type} is associated only with COFF format output; when
6666 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6667 directive but ignores it.
6673 @c only print the extra heading if both COFF and ELF are set
6674 @subheading ELF Version
6677 @cindex ELF symbol type
6678 @cindex symbol type, ELF
6679 @cindex @code{type} directive (ELF version)
6680 For ELF targets, the @code{.type} directive is used like this:
6683 .type @var{name} , @var{type description}
6686 This sets the type of symbol @var{name} to be either a
6687 function symbol or an object symbol. There are five different syntaxes
6688 supported for the @var{type description} field, in order to provide
6689 compatibility with various other assemblers.
6691 Because some of the characters used in these syntaxes (such as @samp{@@} and
6692 @samp{#}) are comment characters for some architectures, some of the syntaxes
6693 below do not work on all architectures. The first variant will be accepted by
6694 the GNU assembler on all architectures so that variant should be used for
6695 maximum portability, if you do not need to assemble your code with other
6698 The syntaxes supported are:
6701 .type <name> STT_<TYPE_IN_UPPER_CASE>
6702 .type <name>,#<type>
6703 .type <name>,@@<type>
6704 .type <name>,%<type>
6705 .type <name>,"<type>"
6708 The types supported are:
6713 Mark the symbol as being a function name.
6716 @itemx gnu_indirect_function
6717 Mark the symbol as an indirect function when evaluated during reloc
6718 processing. (This is only supported on assemblers targeting GNU systems).
6722 Mark the symbol as being a data object.
6726 Mark the symbol as being a thead-local data object.
6730 Mark the symbol as being a common data object.
6734 Does not mark the symbol in any way. It is supported just for completeness.
6736 @item gnu_unique_object
6737 Marks the symbol as being a globally unique data object. The dynamic linker
6738 will make sure that in the entire process there is just one symbol with this
6739 name and type in use. (This is only supported on assemblers targeting GNU
6744 Note: Some targets support extra types in addition to those listed above.
6750 @section @code{.uleb128 @var{expressions}}
6752 @cindex @code{uleb128} directive
6753 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6754 compact, variable length representation of numbers used by the DWARF
6755 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6759 @section @code{.val @var{addr}}
6761 @cindex @code{val} directive
6762 @cindex COFF value attribute
6763 @cindex value attribute, COFF
6764 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6765 records the address @var{addr} as the value attribute of a symbol table
6769 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6770 configured for @code{b.out}, it accepts this directive but ignores it.
6776 @section @code{.version "@var{string}"}
6778 @cindex @code{version} directive
6779 This directive creates a @code{.note} section and places into it an ELF
6780 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6785 @section @code{.vtable_entry @var{table}, @var{offset}}
6787 @cindex @code{vtable_entry} directive
6788 This directive finds or creates a symbol @code{table} and creates a
6789 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6792 @section @code{.vtable_inherit @var{child}, @var{parent}}
6794 @cindex @code{vtable_inherit} directive
6795 This directive finds the symbol @code{child} and finds or creates the symbol
6796 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6797 parent whose addend is the value of the child symbol. As a special case the
6798 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6802 @section @code{.warning "@var{string}"}
6803 @cindex warning directive
6804 Similar to the directive @code{.error}
6805 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6808 @section @code{.weak @var{names}}
6810 @cindex @code{weak} directive
6811 This directive sets the weak attribute on the comma separated list of symbol
6812 @code{names}. If the symbols do not already exist, they will be created.
6814 On COFF targets other than PE, weak symbols are a GNU extension. This
6815 directive sets the weak attribute on the comma separated list of symbol
6816 @code{names}. If the symbols do not already exist, they will be created.
6818 On the PE target, weak symbols are supported natively as weak aliases.
6819 When a weak symbol is created that is not an alias, GAS creates an
6820 alternate symbol to hold the default value.
6823 @section @code{.weakref @var{alias}, @var{target}}
6825 @cindex @code{weakref} directive
6826 This directive creates an alias to the target symbol that enables the symbol to
6827 be referenced with weak-symbol semantics, but without actually making it weak.
6828 If direct references or definitions of the symbol are present, then the symbol
6829 will not be weak, but if all references to it are through weak references, the
6830 symbol will be marked as weak in the symbol table.
6832 The effect is equivalent to moving all references to the alias to a separate
6833 assembly source file, renaming the alias to the symbol in it, declaring the
6834 symbol as weak there, and running a reloadable link to merge the object files
6835 resulting from the assembly of the new source file and the old source file that
6836 had the references to the alias removed.
6838 The alias itself never makes to the symbol table, and is entirely handled
6839 within the assembler.
6842 @section @code{.word @var{expressions}}
6844 @cindex @code{word} directive
6845 This directive expects zero or more @var{expressions}, of any section,
6846 separated by commas.
6849 For each expression, @command{@value{AS}} emits a 32-bit number.
6852 For each expression, @command{@value{AS}} emits a 16-bit number.
6857 The size of the number emitted, and its byte order,
6858 depend on what target computer the assembly is for.
6861 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
6862 @c happen---32-bit addressability, period; no long/short jumps.
6863 @ifset DIFF-TBL-KLUGE
6864 @cindex difference tables altered
6865 @cindex altered difference tables
6867 @emph{Warning: Special Treatment to support Compilers}
6871 Machines with a 32-bit address space, but that do less than 32-bit
6872 addressing, require the following special treatment. If the machine of
6873 interest to you does 32-bit addressing (or doesn't require it;
6874 @pxref{Machine Dependencies}), you can ignore this issue.
6877 In order to assemble compiler output into something that works,
6878 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6879 Directives of the form @samp{.word sym1-sym2} are often emitted by
6880 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6881 directive of the form @samp{.word sym1-sym2}, and the difference between
6882 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6883 creates a @dfn{secondary jump table}, immediately before the next label.
6884 This secondary jump table is preceded by a short-jump to the
6885 first byte after the secondary table. This short-jump prevents the flow
6886 of control from accidentally falling into the new table. Inside the
6887 table is a long-jump to @code{sym2}. The original @samp{.word}
6888 contains @code{sym1} minus the address of the long-jump to
6891 If there were several occurrences of @samp{.word sym1-sym2} before the
6892 secondary jump table, all of them are adjusted. If there was a
6893 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6894 long-jump to @code{sym4} is included in the secondary jump table,
6895 and the @code{.word} directives are adjusted to contain @code{sym3}
6896 minus the address of the long-jump to @code{sym4}; and so on, for as many
6897 entries in the original jump table as necessary.
6900 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6901 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6902 assembly language programmers.
6905 @c end DIFF-TBL-KLUGE
6908 @section Deprecated Directives
6910 @cindex deprecated directives
6911 @cindex obsolescent directives
6912 One day these directives won't work.
6913 They are included for compatibility with older assemblers.
6920 @node Object Attributes
6921 @chapter Object Attributes
6922 @cindex object attributes
6924 @command{@value{AS}} assembles source files written for a specific architecture
6925 into object files for that architecture. But not all object files are alike.
6926 Many architectures support incompatible variations. For instance, floating
6927 point arguments might be passed in floating point registers if the object file
6928 requires hardware floating point support---or floating point arguments might be
6929 passed in integer registers if the object file supports processors with no
6930 hardware floating point unit. Or, if two objects are built for different
6931 generations of the same architecture, the combination may require the
6932 newer generation at run-time.
6934 This information is useful during and after linking. At link time,
6935 @command{@value{LD}} can warn about incompatible object files. After link
6936 time, tools like @command{gdb} can use it to process the linked file
6939 Compatibility information is recorded as a series of object attributes. Each
6940 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
6941 string, and indicates who sets the meaning of the tag. The tag is an integer,
6942 and indicates what property the attribute describes. The value may be a string
6943 or an integer, and indicates how the property affects this object. Missing
6944 attributes are the same as attributes with a zero value or empty string value.
6946 Object attributes were developed as part of the ABI for the ARM Architecture.
6947 The file format is documented in @cite{ELF for the ARM Architecture}.
6950 * GNU Object Attributes:: @sc{gnu} Object Attributes
6951 * Defining New Object Attributes:: Defining New Object Attributes
6954 @node GNU Object Attributes
6955 @section @sc{gnu} Object Attributes
6957 The @code{.gnu_attribute} directive records an object attribute
6958 with vendor @samp{gnu}.
6960 Except for @samp{Tag_compatibility}, which has both an integer and a string for
6961 its value, @sc{gnu} attributes have a string value if the tag number is odd and
6962 an integer value if the tag number is even. The second bit (@code{@var{tag} &
6963 2} is set for architecture-independent attributes and clear for
6964 architecture-dependent ones.
6966 @subsection Common @sc{gnu} attributes
6968 These attributes are valid on all architectures.
6971 @item Tag_compatibility (32)
6972 The compatibility attribute takes an integer flag value and a vendor name. If
6973 the flag value is 0, the file is compatible with other toolchains. If it is 1,
6974 then the file is only compatible with the named toolchain. If it is greater
6975 than 1, the file can only be processed by other toolchains under some private
6976 arrangement indicated by the flag value and the vendor name.
6979 @subsection MIPS Attributes
6982 @item Tag_GNU_MIPS_ABI_FP (4)
6983 The floating-point ABI used by this object file. The value will be:
6987 0 for files not affected by the floating-point ABI.
6989 1 for files using the hardware floating-point with a standard double-precision
6992 2 for files using the hardware floating-point ABI with a single-precision FPU.
6994 3 for files using the software floating-point ABI.
6996 4 for files using the hardware floating-point ABI with 64-bit wide
6997 double-precision floating-point registers and 32-bit wide general
7002 @subsection PowerPC Attributes
7005 @item Tag_GNU_Power_ABI_FP (4)
7006 The floating-point ABI used by this object file. The value will be:
7010 0 for files not affected by the floating-point ABI.
7012 1 for files using double-precision hardware floating-point ABI.
7014 2 for files using the software floating-point ABI.
7016 3 for files using single-precision hardware floating-point ABI.
7019 @item Tag_GNU_Power_ABI_Vector (8)
7020 The vector ABI used by this object file. The value will be:
7024 0 for files not affected by the vector ABI.
7026 1 for files using general purpose registers to pass vectors.
7028 2 for files using AltiVec registers to pass vectors.
7030 3 for files using SPE registers to pass vectors.
7034 @node Defining New Object Attributes
7035 @section Defining New Object Attributes
7037 If you want to define a new @sc{gnu} object attribute, here are the places you
7038 will need to modify. New attributes should be discussed on the @samp{binutils}
7043 This manual, which is the official register of attributes.
7045 The header for your architecture @file{include/elf}, to define the tag.
7047 The @file{bfd} support file for your architecture, to merge the attribute
7048 and issue any appropriate link warnings.
7050 Test cases in @file{ld/testsuite} for merging and link warnings.
7052 @file{binutils/readelf.c} to display your attribute.
7054 GCC, if you want the compiler to mark the attribute automatically.
7060 @node Machine Dependencies
7061 @chapter Machine Dependent Features
7063 @cindex machine dependencies
7064 The machine instruction sets are (almost by definition) different on
7065 each machine where @command{@value{AS}} runs. Floating point representations
7066 vary as well, and @command{@value{AS}} often supports a few additional
7067 directives or command-line options for compatibility with other
7068 assemblers on a particular platform. Finally, some versions of
7069 @command{@value{AS}} support special pseudo-instructions for branch
7072 This chapter discusses most of these differences, though it does not
7073 include details on any machine's instruction set. For details on that
7074 subject, see the hardware manufacturer's manual.
7078 * AArch64-Dependent:: AArch64 Dependent Features
7081 * Alpha-Dependent:: Alpha Dependent Features
7084 * ARC-Dependent:: ARC Dependent Features
7087 * ARM-Dependent:: ARM Dependent Features
7090 * AVR-Dependent:: AVR Dependent Features
7093 * Blackfin-Dependent:: Blackfin Dependent Features
7096 * CR16-Dependent:: CR16 Dependent Features
7099 * CRIS-Dependent:: CRIS Dependent Features
7102 * D10V-Dependent:: D10V Dependent Features
7105 * D30V-Dependent:: D30V Dependent Features
7108 * Epiphany-Dependent:: EPIPHANY Dependent Features
7111 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7114 * HPPA-Dependent:: HPPA Dependent Features
7117 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
7120 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7123 * i860-Dependent:: Intel 80860 Dependent Features
7126 * i960-Dependent:: Intel 80960 Dependent Features
7129 * IA-64-Dependent:: Intel IA-64 Dependent Features
7132 * IP2K-Dependent:: IP2K Dependent Features
7135 * LM32-Dependent:: LM32 Dependent Features
7138 * M32C-Dependent:: M32C Dependent Features
7141 * M32R-Dependent:: M32R Dependent Features
7144 * M68K-Dependent:: M680x0 Dependent Features
7147 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7150 * Meta-Dependent :: Meta Dependent Features
7153 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7156 * MIPS-Dependent:: MIPS Dependent Features
7159 * MMIX-Dependent:: MMIX Dependent Features
7162 * MSP430-Dependent:: MSP430 Dependent Features
7165 * NDS32-Dependent:: Andes NDS32 Dependent Features
7168 * NiosII-Dependent:: Altera Nios II Dependent Features
7171 * NS32K-Dependent:: NS32K Dependent Features
7174 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7175 * SH64-Dependent:: SuperH SH64 Dependent Features
7178 * PDP-11-Dependent:: PDP-11 Dependent Features
7181 * PJ-Dependent:: picoJava Dependent Features
7184 * PPC-Dependent:: PowerPC Dependent Features
7187 * RL78-Dependent:: RL78 Dependent Features
7190 * RX-Dependent:: RX Dependent Features
7193 * S/390-Dependent:: IBM S/390 Dependent Features
7196 * SCORE-Dependent:: SCORE Dependent Features
7199 * Sparc-Dependent:: SPARC Dependent Features
7202 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7205 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7208 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7211 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7214 * V850-Dependent:: V850 Dependent Features
7217 * XGATE-Dependent:: XGATE Features
7220 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7223 * Xtensa-Dependent:: Xtensa Dependent Features
7226 * Z80-Dependent:: Z80 Dependent Features
7229 * Z8000-Dependent:: Z8000 Dependent Features
7232 * Vax-Dependent:: VAX Dependent Features
7239 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7240 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7241 @c peculiarity: to preserve cross-references, there must be a node called
7242 @c "Machine Dependencies". Hence the conditional nodenames in each
7243 @c major node below. Node defaulting in makeinfo requires adjacency of
7244 @c node and sectioning commands; hence the repetition of @chapter BLAH
7245 @c in both conditional blocks.
7248 @include c-aarch64.texi
7252 @include c-alpha.texi
7268 @include c-bfin.texi
7272 @include c-cr16.texi
7276 @include c-cris.texi
7281 @node Machine Dependencies
7282 @chapter Machine Dependent Features
7284 The machine instruction sets are different on each Renesas chip family,
7285 and there are also some syntax differences among the families. This
7286 chapter describes the specific @command{@value{AS}} features for each
7290 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7291 * SH-Dependent:: Renesas SH Dependent Features
7298 @include c-d10v.texi
7302 @include c-d30v.texi
7306 @include c-epiphany.texi
7310 @include c-h8300.texi
7314 @include c-hppa.texi
7318 @include c-i370.texi
7322 @include c-i386.texi
7326 @include c-i860.texi
7330 @include c-i960.texi
7334 @include c-ia64.texi
7338 @include c-ip2k.texi
7342 @include c-lm32.texi
7346 @include c-m32c.texi
7350 @include c-m32r.texi
7354 @include c-m68k.texi
7358 @include c-m68hc11.texi
7362 @include c-metag.texi
7366 @include c-microblaze.texi
7370 @include c-mips.texi
7374 @include c-mmix.texi
7378 @include c-msp430.texi
7382 @include c-nds32.texi
7386 @include c-nios2.texi
7390 @include c-ns32k.texi
7394 @include c-pdp11.texi
7406 @include c-rl78.texi
7414 @include c-s390.texi
7418 @include c-score.texi
7423 @include c-sh64.texi
7427 @include c-sparc.texi
7431 @include c-tic54x.texi
7435 @include c-tic6x.texi
7439 @include c-tilegx.texi
7443 @include c-tilepro.texi
7459 @include c-v850.texi
7463 @include c-xgate.texi
7467 @include c-xstormy16.texi
7471 @include c-xtensa.texi
7475 @c reverse effect of @down at top of generic Machine-Dep chapter
7479 @node Reporting Bugs
7480 @chapter Reporting Bugs
7481 @cindex bugs in assembler
7482 @cindex reporting bugs in assembler
7484 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7486 Reporting a bug may help you by bringing a solution to your problem, or it may
7487 not. But in any case the principal function of a bug report is to help the
7488 entire community by making the next version of @command{@value{AS}} work better.
7489 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7491 In order for a bug report to serve its purpose, you must include the
7492 information that enables us to fix the bug.
7495 * Bug Criteria:: Have you found a bug?
7496 * Bug Reporting:: How to report bugs
7500 @section Have You Found a Bug?
7501 @cindex bug criteria
7503 If you are not sure whether you have found a bug, here are some guidelines:
7506 @cindex fatal signal
7507 @cindex assembler crash
7508 @cindex crash of assembler
7510 If the assembler gets a fatal signal, for any input whatever, that is a
7511 @command{@value{AS}} bug. Reliable assemblers never crash.
7513 @cindex error on valid input
7515 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7517 @cindex invalid input
7519 If @command{@value{AS}} does not produce an error message for invalid input, that
7520 is a bug. However, you should note that your idea of ``invalid input'' might
7521 be our idea of ``an extension'' or ``support for traditional practice''.
7524 If you are an experienced user of assemblers, your suggestions for improvement
7525 of @command{@value{AS}} are welcome in any case.
7529 @section How to Report Bugs
7531 @cindex assembler bugs, reporting
7533 A number of companies and individuals offer support for @sc{gnu} products. If
7534 you obtained @command{@value{AS}} from a support organization, we recommend you
7535 contact that organization first.
7537 You can find contact information for many support companies and
7538 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7542 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7546 The fundamental principle of reporting bugs usefully is this:
7547 @strong{report all the facts}. If you are not sure whether to state a
7548 fact or leave it out, state it!
7550 Often people omit facts because they think they know what causes the problem
7551 and assume that some details do not matter. Thus, you might assume that the
7552 name of a symbol you use in an example does not matter. Well, probably it does
7553 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7554 happens to fetch from the location where that name is stored in memory;
7555 perhaps, if the name were different, the contents of that location would fool
7556 the assembler into doing the right thing despite the bug. Play it safe and
7557 give a specific, complete example. That is the easiest thing for you to do,
7558 and the most helpful.
7560 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7561 it is new to us. Therefore, always write your bug reports on the assumption
7562 that the bug has not been reported previously.
7564 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7565 bell?'' This cannot help us fix a bug, so it is basically useless. We
7566 respond by asking for enough details to enable us to investigate.
7567 You might as well expedite matters by sending them to begin with.
7569 To enable us to fix the bug, you should include all these things:
7573 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
7574 it with the @samp{--version} argument.
7576 Without this, we will not know whether there is any point in looking for
7577 the bug in the current version of @command{@value{AS}}.
7580 Any patches you may have applied to the @command{@value{AS}} source.
7583 The type of machine you are using, and the operating system name and
7587 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
7591 The command arguments you gave the assembler to assemble your example and
7592 observe the bug. To guarantee you will not omit something important, list them
7593 all. A copy of the Makefile (or the output from make) is sufficient.
7595 If we were to try to guess the arguments, we would probably guess wrong
7596 and then we might not encounter the bug.
7599 A complete input file that will reproduce the bug. If the bug is observed when
7600 the assembler is invoked via a compiler, send the assembler source, not the
7601 high level language source. Most compilers will produce the assembler source
7602 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
7603 the options @samp{-v --save-temps}; this will save the assembler source in a
7604 file with an extension of @file{.s}, and also show you exactly how
7605 @command{@value{AS}} is being run.
7608 A description of what behavior you observe that you believe is
7609 incorrect. For example, ``It gets a fatal signal.''
7611 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
7612 will certainly notice it. But if the bug is incorrect output, we might not
7613 notice unless it is glaringly wrong. You might as well not give us a chance to
7616 Even if the problem you experience is a fatal signal, you should still say so
7617 explicitly. Suppose something strange is going on, such as, your copy of
7618 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
7619 library on your system. (This has happened!) Your copy might crash and ours
7620 would not. If you told us to expect a crash, then when ours fails to crash, we
7621 would know that the bug was not happening for us. If you had not told us to
7622 expect a crash, then we would not be able to draw any conclusion from our
7626 If you wish to suggest changes to the @command{@value{AS}} source, send us context
7627 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
7628 option. Always send diffs from the old file to the new file. If you even
7629 discuss something in the @command{@value{AS}} source, refer to it by context, not
7632 The line numbers in our development sources will not match those in your
7633 sources. Your line numbers would convey no useful information to us.
7636 Here are some things that are not necessary:
7640 A description of the envelope of the bug.
7642 Often people who encounter a bug spend a lot of time investigating
7643 which changes to the input file will make the bug go away and which
7644 changes will not affect it.
7646 This is often time consuming and not very useful, because the way we
7647 will find the bug is by running a single example under the debugger
7648 with breakpoints, not by pure deduction from a series of examples.
7649 We recommend that you save your time for something else.
7651 Of course, if you can find a simpler example to report @emph{instead}
7652 of the original one, that is a convenience for us. Errors in the
7653 output will be easier to spot, running under the debugger will take
7654 less time, and so on.
7656 However, simplification is not vital; if you do not want to do this,
7657 report the bug anyway and send us the entire test case you used.
7660 A patch for the bug.
7662 A patch for the bug does help us if it is a good one. But do not omit
7663 the necessary information, such as the test case, on the assumption that
7664 a patch is all we need. We might see problems with your patch and decide
7665 to fix the problem another way, or we might not understand it at all.
7667 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
7668 construct an example that will make the program follow a certain path through
7669 the code. If you do not send us the example, we will not be able to construct
7670 one, so we will not be able to verify that the bug is fixed.
7672 And if we cannot understand what bug you are trying to fix, or why your
7673 patch should be an improvement, we will not install it. A test case will
7674 help us to understand.
7677 A guess about what the bug is or what it depends on.
7679 Such guesses are usually wrong. Even we cannot guess right about such
7680 things without first using the debugger to find the facts.
7683 @node Acknowledgements
7684 @chapter Acknowledgements
7686 If you have contributed to GAS and your name isn't listed here,
7687 it is not meant as a slight. We just don't know about it. Send mail to the
7688 maintainer, and we'll correct the situation. Currently
7690 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
7692 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
7695 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
7696 information and the 68k series machines, most of the preprocessing pass, and
7697 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
7699 K. Richard Pixley maintained GAS for a while, adding various enhancements and
7700 many bug fixes, including merging support for several processors, breaking GAS
7701 up to handle multiple object file format back ends (including heavy rewrite,
7702 testing, an integration of the coff and b.out back ends), adding configuration
7703 including heavy testing and verification of cross assemblers and file splits
7704 and renaming, converted GAS to strictly ANSI C including full prototypes, added
7705 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
7706 port (including considerable amounts of reverse engineering), a SPARC opcode
7707 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
7708 assertions and made them work, much other reorganization, cleanup, and lint.
7710 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
7711 in format-specific I/O modules.
7713 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
7714 has done much work with it since.
7716 The Intel 80386 machine description was written by Eliot Dresselhaus.
7718 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
7720 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
7721 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
7723 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
7724 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
7725 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
7726 support a.out format.
7728 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
7729 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
7730 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
7731 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
7734 John Gilmore built the AMD 29000 support, added @code{.include} support, and
7735 simplified the configuration of which versions accept which directives. He
7736 updated the 68k machine description so that Motorola's opcodes always produced
7737 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
7738 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
7739 cross-compilation support, and one bug in relaxation that took a week and
7740 required the proverbial one-bit fix.
7742 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
7743 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
7744 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
7745 PowerPC assembler, and made a few other minor patches.
7747 Steve Chamberlain made GAS able to generate listings.
7749 Hewlett-Packard contributed support for the HP9000/300.
7751 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
7752 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
7753 formats). This work was supported by both the Center for Software Science at
7754 the University of Utah and Cygnus Support.
7756 Support for ELF format files has been worked on by Mark Eichin of Cygnus
7757 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
7758 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
7759 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
7760 and some initial 64-bit support).
7762 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
7764 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
7765 support for openVMS/Alpha.
7767 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
7770 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
7771 Inc.@: added support for Xtensa processors.
7773 Several engineers at Cygnus Support have also provided many small bug fixes and
7774 configuration enhancements.
7776 Jon Beniston added support for the Lattice Mico32 architecture.
7778 Many others have contributed large or small bugfixes and enhancements. If
7779 you have contributed significant work and are not mentioned on this list, and
7780 want to be, let us know. Some of the history has been lost; we are not
7781 intentionally leaving anyone out.
7783 @node GNU Free Documentation License
7784 @appendix GNU Free Documentation License
7788 @unnumbered AS Index