1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright 1991-2013 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{} 1991-2013 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{} 1991-2013 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-vr4120}] [@b{-mno-fix-vr4120}]
421 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
422 [@b{-mdebug}] [@b{-no-mdebug}]
423 [@b{-mpdr}] [@b{-mno-pdr}]
427 @emph{Target MMIX options:}
428 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
429 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
430 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
431 [@b{--linker-allocated-gregs}]
435 @emph{Target Nios II options:}
436 [@b{-relax-all}] [@b{-relax-section}] [@b{-no-relax}]
441 @emph{Target PDP11 options:}
442 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
443 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
444 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
448 @emph{Target picoJava options:}
453 @emph{Target PowerPC options:}
455 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|@b{-m403}|@b{-m405}|
456 @b{-m440}|@b{-m464}|@b{-m476}|@b{-m7400}|@b{-m7410}|@b{-m7450}|@b{-m7455}|@b{-m750cl}|@b{-mppc64}|
457 @b{-m620}|@b{-me500}|@b{-e500x2}|@b{-me500mc}|@b{-me500mc64}|@b{-me5500}|@b{-me6500}|@b{-mppc64bridge}|
458 @b{-mbooke}|@b{-mpower4}|@b{-mpwr4}|@b{-mpower5}|@b{-mpwr5}|@b{-mpwr5x}|@b{-mpower6}|@b{-mpwr6}|
459 @b{-mpower7}|@b{-mpwr7}|@b{-mpower8}|@b{-mpwr8}|@b{-ma2}|@b{-mcell}|@b{-mspe}|@b{-mtitan}|@b{-me300}|@b{-mcom}]
460 [@b{-many}] [@b{-maltivec}|@b{-mvsx}|@b{-mhtm}|@b{-mvle}]
461 [@b{-mregnames}|@b{-mno-regnames}]
462 [@b{-mrelocatable}|@b{-mrelocatable-lib}|@b{-K PIC}] [@b{-memb}]
463 [@b{-mlittle}|@b{-mlittle-endian}|@b{-le}|@b{-mbig}|@b{-mbig-endian}|@b{-be}]
464 [@b{-msolaris}|@b{-mno-solaris}]
465 [@b{-nops=@var{count}}]
469 @emph{Target RX options:}
470 [@b{-mlittle-endian}|@b{-mbig-endian}]
471 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
472 [@b{-muse-conventional-section-names}]
473 [@b{-msmall-data-limit}]
476 [@b{-mint-register=@var{number}}]
477 [@b{-mgcc-abi}|@b{-mrx-abi}]
481 @emph{Target s390 options:}
482 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
483 [@b{-mregnames}|@b{-mno-regnames}]
484 [@b{-mwarn-areg-zero}]
488 @emph{Target SCORE options:}
489 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
490 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
491 [@b{-march=score7}][@b{-march=score3}]
492 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
496 @emph{Target SPARC options:}
497 @c The order here is important. See c-sparc.texi.
498 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
499 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
500 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
505 @emph{Target TIC54X options:}
506 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
507 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
512 @emph{Target TIC6X options:}
513 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
514 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
515 [@b{-mpic}|@b{-mno-pic}]
519 @emph{Target TILE-Gx options:}
520 [@b{-m32}|@b{-m64}][@b{-EB}][@b{-EL}]
523 @c TILEPro has no machine-dependent assembler options
528 @emph{Target Xtensa options:}
529 [@b{--[no-]text-section-literals}] [@b{--[no-]absolute-literals}]
530 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
531 [@b{--[no-]transform}]
532 [@b{--rename-section} @var{oldname}=@var{newname}]
537 @emph{Target Z80 options:}
538 [@b{-z80}] [@b{-r800}]
539 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
540 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
541 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
542 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
543 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
544 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
548 @c Z8000 has no machine-dependent assembler options
557 @include at-file.texi
560 Turn on listings, in any of a variety of ways:
564 omit false conditionals
567 omit debugging directives
570 include general information, like @value{AS} version and options passed
573 include high-level source
579 include macro expansions
582 omit forms processing
588 set the name of the listing file
591 You may combine these options; for example, use @samp{-aln} for assembly
592 listing without forms processing. The @samp{=file} option, if used, must be
593 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
596 Begin in alternate macro mode.
598 @xref{Altmacro,,@code{.altmacro}}.
601 @item --compress-debug-sections
602 Compress DWARF debug sections using zlib. The debug sections are renamed
603 to begin with @samp{.zdebug}, and the resulting object file may not be
604 compatible with older linkers and object file utilities.
606 @item --nocompress-debug-sections
607 Do not compress DWARF debug sections. This is the default.
610 Ignored. This option is accepted for script compatibility with calls to
613 @item --debug-prefix-map @var{old}=@var{new}
614 When assembling files in directory @file{@var{old}}, record debugging
615 information describing them as in @file{@var{new}} instead.
617 @item --defsym @var{sym}=@var{value}
618 Define the symbol @var{sym} to be @var{value} before assembling the input file.
619 @var{value} must be an integer constant. As in C, a leading @samp{0x}
620 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
621 value. The value of the symbol can be overridden inside a source file via the
622 use of a @code{.set} pseudo-op.
625 ``fast''---skip whitespace and comment preprocessing (assume source is
630 Generate debugging information for each assembler source line using whichever
631 debug format is preferred by the target. This currently means either STABS,
635 Generate stabs debugging information for each assembler line. This
636 may help debugging assembler code, if the debugger can handle it.
639 Generate stabs debugging information for each assembler line, with GNU
640 extensions that probably only gdb can handle, and that could make other
641 debuggers crash or refuse to read your program. This
642 may help debugging assembler code. Currently the only GNU extension is
643 the location of the current working directory at assembling time.
646 Generate DWARF2 debugging information for each assembler line. This
647 may help debugging assembler code, if the debugger can handle it. Note---this
648 option is only supported by some targets, not all of them.
650 @item --gdwarf-sections
651 Instead of creating a .debug_line section, create a series of
652 .debug_line.@var{foo} sections where @var{foo} is the name of the
653 corresponding code section. For example a code section called @var{.text.func}
654 will have its dwarf line number information placed into a section called
655 @var{.debug_line.text.func}. If the code section is just called @var{.text}
656 then debug line section will still be called just @var{.debug_line} without any
659 @item --size-check=error
660 @itemx --size-check=warning
661 Issue an error or warning for invalid ELF .size directive.
664 Print a summary of the command line options and exit.
667 Print a summary of all target specific options and exit.
670 Add directory @var{dir} to the search list for @code{.include} directives.
673 Don't warn about signed overflow.
676 @ifclear DIFF-TBL-KLUGE
677 This option is accepted but has no effect on the @value{TARGET} family.
679 @ifset DIFF-TBL-KLUGE
680 Issue warnings when difference tables altered for long displacements.
685 Keep (in the symbol table) local symbols. These symbols start with
686 system-specific local label prefixes, typically @samp{.L} for ELF systems
687 or @samp{L} for traditional a.out systems.
692 @item --listing-lhs-width=@var{number}
693 Set the maximum width, in words, of the output data column for an assembler
694 listing to @var{number}.
696 @item --listing-lhs-width2=@var{number}
697 Set the maximum width, in words, of the output data column for continuation
698 lines in an assembler listing to @var{number}.
700 @item --listing-rhs-width=@var{number}
701 Set the maximum width of an input source line, as displayed in a listing, to
704 @item --listing-cont-lines=@var{number}
705 Set the maximum number of lines printed in a listing for a single line of input
708 @item -o @var{objfile}
709 Name the object-file output from @command{@value{AS}} @var{objfile}.
712 Fold the data section into the text section.
714 @kindex --hash-size=@var{number}
715 Set the default size of GAS's hash tables to a prime number close to
716 @var{number}. Increasing this value can reduce the length of time it takes the
717 assembler to perform its tasks, at the expense of increasing the assembler's
718 memory requirements. Similarly reducing this value can reduce the memory
719 requirements at the expense of speed.
721 @item --reduce-memory-overheads
722 This option reduces GAS's memory requirements, at the expense of making the
723 assembly processes slower. Currently this switch is a synonym for
724 @samp{--hash-size=4051}, but in the future it may have other effects as well.
727 Print the maximum space (in bytes) and total time (in seconds) used by
730 @item --strip-local-absolute
731 Remove local absolute symbols from the outgoing symbol table.
735 Print the @command{as} version.
738 Print the @command{as} version and exit.
742 Suppress warning messages.
744 @item --fatal-warnings
745 Treat warnings as errors.
748 Don't suppress warning messages or treat them as errors.
757 Generate an object file even after errors.
759 @item -- | @var{files} @dots{}
760 Standard input, or source files to assemble.
768 @xref{AArch64 Options}, for the options available when @value{AS} is configured
769 for the 64-bit mode of the ARM Architecture (AArch64).
774 The following options are available when @value{AS} is configured for the
775 64-bit mode of the ARM Architecture (AArch64).
778 @include c-aarch64.texi
779 @c ended inside the included file
787 @xref{Alpha Options}, for the options available when @value{AS} is configured
788 for an Alpha processor.
793 The following options are available when @value{AS} is configured for an Alpha
797 @include c-alpha.texi
798 @c ended inside the included file
805 The following options are available when @value{AS} is configured for
810 This option selects the core processor variant.
812 Select either big-endian (-EB) or little-endian (-EL) output.
817 The following options are available when @value{AS} is configured for the ARM
821 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
822 Specify which ARM processor variant is the target.
823 @item -march=@var{architecture}[+@var{extension}@dots{}]
824 Specify which ARM architecture variant is used by the target.
825 @item -mfpu=@var{floating-point-format}
826 Select which Floating Point architecture is the target.
827 @item -mfloat-abi=@var{abi}
828 Select which floating point ABI is in use.
830 Enable Thumb only instruction decoding.
831 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
832 Select which procedure calling convention is in use.
834 Select either big-endian (-EB) or little-endian (-EL) output.
835 @item -mthumb-interwork
836 Specify that the code has been generated with interworking between Thumb and
839 Specify that PIC code has been generated.
847 @xref{Blackfin Options}, for the options available when @value{AS} is
848 configured for the Blackfin processor family.
853 The following options are available when @value{AS} is configured for
854 the Blackfin processor family.
858 @c ended inside the included file
865 See the info pages for documentation of the CRIS-specific options.
869 The following options are available when @value{AS} is configured for
872 @cindex D10V optimization
873 @cindex optimization, D10V
875 Optimize output by parallelizing instructions.
880 The following options are available when @value{AS} is configured for a D30V
883 @cindex D30V optimization
884 @cindex optimization, D30V
886 Optimize output by parallelizing instructions.
890 Warn when nops are generated.
892 @cindex D30V nops after 32-bit multiply
894 Warn when a nop after a 32-bit multiply instruction is generated.
900 The following options are available when @value{AS} is configured for the
901 Adapteva EPIPHANY series.
904 @xref{Epiphany Options}, for the options available when @value{AS} is
905 configured for an Epiphany processor.
910 The following options are available when @value{AS} is configured for
911 an Epiphany processor.
914 @include c-epiphany.texi
915 @c ended inside the included file
923 @xref{H8/300 Options}, for the options available when @value{AS} is configured
924 for an H8/300 processor.
929 The following options are available when @value{AS} is configured for an H8/300
933 @include c-h8300.texi
934 @c ended inside the included file
942 @xref{i386-Options}, for the options available when @value{AS} is
943 configured for an i386 processor.
948 The following options are available when @value{AS} is configured for
953 @c ended inside the included file
960 The following options are available when @value{AS} is configured for the
961 Intel 80960 processor.
964 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
965 Specify which variant of the 960 architecture is the target.
968 Add code to collect statistics about branches taken.
971 Do not alter compare-and-branch instructions for long displacements;
978 The following options are available when @value{AS} is configured for the
984 Specifies that the extended IP2022 instructions are allowed.
987 Restores the default behaviour, which restricts the permitted instructions to
988 just the basic IP2022 ones.
994 The following options are available when @value{AS} is configured for the
995 Renesas M32C and M16C processors.
1000 Assemble M32C instructions.
1003 Assemble M16C instructions (the default).
1006 Enable support for link-time relaxations.
1009 Support H'00 style hex constants in addition to 0x00 style.
1015 The following options are available when @value{AS} is configured for the
1016 Renesas M32R (formerly Mitsubishi M32R) series.
1021 Specify which processor in the M32R family is the target. The default
1022 is normally the M32R, but this option changes it to the M32RX.
1024 @item --warn-explicit-parallel-conflicts or --Wp
1025 Produce warning messages when questionable parallel constructs are
1028 @item --no-warn-explicit-parallel-conflicts or --Wnp
1029 Do not produce warning messages when questionable parallel constructs are
1036 The following options are available when @value{AS} is configured for the
1037 Motorola 68000 series.
1042 Shorten references to undefined symbols, to one word instead of two.
1044 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1045 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1046 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1047 Specify what processor in the 68000 family is the target. The default
1048 is normally the 68020, but this can be changed at configuration time.
1050 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1051 The target machine does (or does not) have a floating-point coprocessor.
1052 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1053 the basic 68000 is not compatible with the 68881, a combination of the
1054 two can be specified, since it's possible to do emulation of the
1055 coprocessor instructions with the main processor.
1057 @item -m68851 | -mno-68851
1058 The target machine does (or does not) have a memory-management
1059 unit coprocessor. The default is to assume an MMU for 68020 and up.
1067 @xref{Nios II Options}, for the options available when @value{AS} is configured
1068 for an Altera Nios II processor.
1072 @c man begin OPTIONS
1073 The following options are available when @value{AS} is configured for an
1074 Altera Nios II processor.
1076 @c man begin INCLUDE
1077 @include c-nios2.texi
1078 @c ended inside the included file
1084 For details about the PDP-11 machine dependent features options,
1085 see @ref{PDP-11-Options}.
1088 @item -mpic | -mno-pic
1089 Generate position-independent (or position-dependent) code. The
1090 default is @option{-mpic}.
1093 @itemx -mall-extensions
1094 Enable all instruction set extensions. This is the default.
1096 @item -mno-extensions
1097 Disable all instruction set extensions.
1099 @item -m@var{extension} | -mno-@var{extension}
1100 Enable (or disable) a particular instruction set extension.
1103 Enable the instruction set extensions supported by a particular CPU, and
1104 disable all other extensions.
1106 @item -m@var{machine}
1107 Enable the instruction set extensions supported by a particular machine
1108 model, and disable all other extensions.
1114 The following options are available when @value{AS} is configured for
1115 a picoJava processor.
1119 @cindex PJ endianness
1120 @cindex endianness, PJ
1121 @cindex big endian output, PJ
1123 Generate ``big endian'' format output.
1125 @cindex little endian output, PJ
1127 Generate ``little endian'' format output.
1133 The following options are available when @value{AS} is configured for the
1134 Motorola 68HC11 or 68HC12 series.
1138 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1139 Specify what processor is the target. The default is
1140 defined by the configuration option when building the assembler.
1142 @item --xgate-ramoffset
1143 Instruct the linker to offset RAM addresses from S12X address space into
1144 XGATE address space.
1147 Specify to use the 16-bit integer ABI.
1150 Specify to use the 32-bit integer ABI.
1152 @item -mshort-double
1153 Specify to use the 32-bit double ABI.
1156 Specify to use the 64-bit double ABI.
1158 @item --force-long-branches
1159 Relative branches are turned into absolute ones. This concerns
1160 conditional branches, unconditional branches and branches to a
1163 @item -S | --short-branches
1164 Do not turn relative branches into absolute ones
1165 when the offset is out of range.
1167 @item --strict-direct-mode
1168 Do not turn the direct addressing mode into extended addressing mode
1169 when the instruction does not support direct addressing mode.
1171 @item --print-insn-syntax
1172 Print the syntax of instruction in case of error.
1174 @item --print-opcodes
1175 Print the list of instructions with syntax and then exit.
1177 @item --generate-example
1178 Print an example of instruction for each possible instruction and then exit.
1179 This option is only useful for testing @command{@value{AS}}.
1185 The following options are available when @command{@value{AS}} is configured
1186 for the SPARC architecture:
1189 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1190 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1191 Explicitly select a variant of the SPARC architecture.
1193 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1194 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1196 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1197 UltraSPARC extensions.
1199 @item -xarch=v8plus | -xarch=v8plusa
1200 For compatibility with the Solaris v9 assembler. These options are
1201 equivalent to -Av8plus and -Av8plusa, respectively.
1204 Warn when the assembler switches to another architecture.
1209 The following options are available when @value{AS} is configured for the 'c54x
1214 Enable extended addressing mode. All addresses and relocations will assume
1215 extended addressing (usually 23 bits).
1216 @item -mcpu=@var{CPU_VERSION}
1217 Sets the CPU version being compiled for.
1218 @item -merrors-to-file @var{FILENAME}
1219 Redirect error output to a file, for broken systems which don't support such
1220 behaviour in the shell.
1225 The following options are available when @value{AS} is configured for
1230 This option sets the largest size of an object that can be referenced
1231 implicitly with the @code{gp} register. It is only accepted for targets that
1232 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1234 @cindex MIPS endianness
1235 @cindex endianness, MIPS
1236 @cindex big endian output, MIPS
1238 Generate ``big endian'' format output.
1240 @cindex little endian output, MIPS
1242 Generate ``little endian'' format output.
1254 Generate code for a particular MIPS Instruction Set Architecture level.
1255 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1256 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1257 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1258 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
1259 @samp{-mips64r2} correspond to generic MIPS V, MIPS32, MIPS32 Release 2,
1260 MIPS64, and MIPS64 Release 2 ISA processors, respectively.
1262 @item -march=@var{cpu}
1263 Generate code for a particular MIPS CPU.
1265 @item -mtune=@var{cpu}
1266 Schedule and tune for a particular MIPS CPU.
1270 Cause nops to be inserted if the read of the destination register
1271 of an mfhi or mflo instruction occurs in the following two instructions.
1275 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1276 section instead of the standard ELF .stabs sections.
1280 Control generation of @code{.pdr} sections.
1284 The register sizes are normally inferred from the ISA and ABI, but these
1285 flags force a certain group of registers to be treated as 32 bits wide at
1286 all times. @samp{-mgp32} controls the size of general-purpose registers
1287 and @samp{-mfp32} controls the size of floating-point registers.
1291 Generate code for the MIPS 16 processor. This is equivalent to putting
1292 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1293 turns off this option.
1296 @itemx -mno-micromips
1297 Generate code for the microMIPS processor. This is equivalent to putting
1298 @code{.set micromips} at the start of the assembly file. @samp{-mno-micromips}
1299 turns off this option. This is equivalent to putting @code{.set nomicromips}
1300 at the start of the assembly file.
1303 @itemx -mno-smartmips
1304 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1305 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1306 @samp{-mno-smartmips} turns off this option.
1310 Generate code for the MIPS-3D Application Specific Extension.
1311 This tells the assembler to accept MIPS-3D instructions.
1312 @samp{-no-mips3d} turns off this option.
1316 Generate code for the MDMX Application Specific Extension.
1317 This tells the assembler to accept MDMX instructions.
1318 @samp{-no-mdmx} turns off this option.
1322 Generate code for the DSP Release 1 Application Specific Extension.
1323 This tells the assembler to accept DSP Release 1 instructions.
1324 @samp{-mno-dsp} turns off this option.
1328 Generate code for the DSP Release 2 Application Specific Extension.
1329 This option implies -mdsp.
1330 This tells the assembler to accept DSP Release 2 instructions.
1331 @samp{-mno-dspr2} turns off this option.
1335 Generate code for the MIPS SIMD Architecture Extension.
1336 This tells the assembler to accept MSA instructions.
1337 @samp{-mno-msa} turns off this option.
1341 Generate code for the MT Application Specific Extension.
1342 This tells the assembler to accept MT instructions.
1343 @samp{-mno-mt} turns off this option.
1347 Generate code for the MCU Application Specific Extension.
1348 This tells the assembler to accept MCU instructions.
1349 @samp{-mno-mcu} turns off this option.
1353 Only use 32-bit instruction encodings when generating code for the
1354 microMIPS processor. This option inhibits the use of any 16-bit
1355 instructions. This is equivalent to putting @code{.set insn32} at
1356 the start of the assembly file. @samp{-mno-insn32} turns off this
1357 option. This is equivalent to putting @code{.set noinsn32} at the
1358 start of the assembly file. By default @samp{-mno-insn32} is
1359 selected, allowing all instructions to be used.
1361 @item --construct-floats
1362 @itemx --no-construct-floats
1363 The @samp{--no-construct-floats} option disables the construction of
1364 double width floating point constants by loading the two halves of the
1365 value into the two single width floating point registers that make up
1366 the double width register. By default @samp{--construct-floats} is
1367 selected, allowing construction of these floating point constants.
1369 @item --relax-branch
1370 @itemx --no-relax-branch
1371 The @samp{--relax-branch} option enables the relaxation of out-of-range
1372 branches. By default @samp{--no-relax-branch} is selected, causing any
1373 out-of-range branches to produce an error.
1375 @item -mnan=@var{encoding}
1376 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1377 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1380 @item --emulation=@var{name}
1381 This option was formerly used to switch between ELF and ECOFF output
1382 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1383 removed in GAS 2.24, so the option now serves little purpose.
1384 It is retained for backwards compatibility.
1386 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1387 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1388 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1389 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1390 preferred options instead.
1393 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1400 Control how to deal with multiplication overflow and division by zero.
1401 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1402 (and only work for Instruction Set Architecture level 2 and higher);
1403 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1407 When this option is used, @command{@value{AS}} will issue a warning every
1408 time it generates a nop instruction from a macro.
1413 The following options are available when @value{AS} is configured for
1419 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1420 The command line option @samp{-nojsri2bsr} can be used to disable it.
1424 Enable or disable the silicon filter behaviour. By default this is disabled.
1425 The default can be overridden by the @samp{-sifilter} command line option.
1428 Alter jump instructions for long displacements.
1430 @item -mcpu=[210|340]
1431 Select the cpu type on the target hardware. This controls which instructions
1435 Assemble for a big endian target.
1438 Assemble for a little endian target.
1447 @xref{Meta Options}, for the options available when @value{AS} is configured
1448 for a Meta processor.
1452 @c man begin OPTIONS
1453 The following options are available when @value{AS} is configured for a
1456 @c man begin INCLUDE
1457 @include c-metag.texi
1458 @c ended inside the included file
1463 @c man begin OPTIONS
1465 See the info pages for documentation of the MMIX-specific options.
1472 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1473 for a PowerPC processor.
1477 @c man begin OPTIONS
1478 The following options are available when @value{AS} is configured for a
1481 @c man begin INCLUDE
1483 @c ended inside the included file
1488 @c man begin OPTIONS
1490 See the info pages for documentation of the RX-specific options.
1494 The following options are available when @value{AS} is configured for the s390
1500 Select the word size, either 31/32 bits or 64 bits.
1503 Select the architecture mode, either the Enterprise System
1504 Architecture (esa) or the z/Architecture mode (zarch).
1505 @item -march=@var{processor}
1506 Specify which s390 processor variant is the target, @samp{g6}, @samp{g6},
1507 @samp{z900}, @samp{z990}, @samp{z9-109}, @samp{z9-ec}, @samp{z10},
1508 @samp{z196}, or @samp{zEC12}.
1510 @itemx -mno-regnames
1511 Allow or disallow symbolic names for registers.
1512 @item -mwarn-areg-zero
1513 Warn whenever the operand for a base or index register has been specified
1514 but evaluates to zero.
1522 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1523 for a TMS320C6000 processor.
1527 @c man begin OPTIONS
1528 The following options are available when @value{AS} is configured for a
1529 TMS320C6000 processor.
1531 @c man begin INCLUDE
1532 @include c-tic6x.texi
1533 @c ended inside the included file
1541 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1542 for a TILE-Gx processor.
1546 @c man begin OPTIONS
1547 The following options are available when @value{AS} is configured for a TILE-Gx
1550 @c man begin INCLUDE
1551 @include c-tilegx.texi
1552 @c ended inside the included file
1560 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1561 for an Xtensa processor.
1565 @c man begin OPTIONS
1566 The following options are available when @value{AS} is configured for an
1569 @c man begin INCLUDE
1570 @include c-xtensa.texi
1571 @c ended inside the included file
1576 @c man begin OPTIONS
1579 The following options are available when @value{AS} is configured for
1580 a Z80 family processor.
1583 Assemble for Z80 processor.
1585 Assemble for R800 processor.
1586 @item -ignore-undocumented-instructions
1588 Assemble undocumented Z80 instructions that also work on R800 without warning.
1589 @item -ignore-unportable-instructions
1591 Assemble all undocumented Z80 instructions without warning.
1592 @item -warn-undocumented-instructions
1594 Issue a warning for undocumented Z80 instructions that also work on R800.
1595 @item -warn-unportable-instructions
1597 Issue a warning for undocumented Z80 instructions that do not work on R800.
1598 @item -forbid-undocumented-instructions
1600 Treat all undocumented instructions as errors.
1601 @item -forbid-unportable-instructions
1603 Treat undocumented Z80 instructions that do not work on R800 as errors.
1610 * Manual:: Structure of this Manual
1611 * GNU Assembler:: The GNU Assembler
1612 * Object Formats:: Object File Formats
1613 * Command Line:: Command Line
1614 * Input Files:: Input Files
1615 * Object:: Output (Object) File
1616 * Errors:: Error and Warning Messages
1620 @section Structure of this Manual
1622 @cindex manual, structure and purpose
1623 This manual is intended to describe what you need to know to use
1624 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1625 notation for symbols, constants, and expressions; the directives that
1626 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1629 We also cover special features in the @value{TARGET}
1630 configuration of @command{@value{AS}}, including assembler directives.
1633 This manual also describes some of the machine-dependent features of
1634 various flavors of the assembler.
1637 @cindex machine instructions (not covered)
1638 On the other hand, this manual is @emph{not} intended as an introduction
1639 to programming in assembly language---let alone programming in general!
1640 In a similar vein, we make no attempt to introduce the machine
1641 architecture; we do @emph{not} describe the instruction set, standard
1642 mnemonics, registers or addressing modes that are standard to a
1643 particular architecture.
1645 You may want to consult the manufacturer's
1646 machine architecture manual for this information.
1650 For information on the H8/300 machine instruction set, see @cite{H8/300
1651 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1652 Programming Manual} (Renesas).
1655 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1656 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1657 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1658 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1661 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1665 @c I think this is premature---doc@cygnus.com, 17jan1991
1667 Throughout this manual, we assume that you are running @dfn{GNU},
1668 the portable operating system from the @dfn{Free Software
1669 Foundation, Inc.}. This restricts our attention to certain kinds of
1670 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1671 once this assumption is granted examples and definitions need less
1674 @command{@value{AS}} is part of a team of programs that turn a high-level
1675 human-readable series of instructions into a low-level
1676 computer-readable series of instructions. Different versions of
1677 @command{@value{AS}} are used for different kinds of computer.
1680 @c There used to be a section "Terminology" here, which defined
1681 @c "contents", "byte", "word", and "long". Defining "word" to any
1682 @c particular size is confusing when the .word directive may generate 16
1683 @c bits on one machine and 32 bits on another; in general, for the user
1684 @c version of this manual, none of these terms seem essential to define.
1685 @c They were used very little even in the former draft of the manual;
1686 @c this draft makes an effort to avoid them (except in names of
1690 @section The GNU Assembler
1692 @c man begin DESCRIPTION
1694 @sc{gnu} @command{as} is really a family of assemblers.
1696 This manual describes @command{@value{AS}}, a member of that family which is
1697 configured for the @value{TARGET} architectures.
1699 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1700 should find a fairly similar environment when you use it on another
1701 architecture. Each version has much in common with the others,
1702 including object file formats, most assembler directives (often called
1703 @dfn{pseudo-ops}) and assembler syntax.@refill
1705 @cindex purpose of @sc{gnu} assembler
1706 @command{@value{AS}} is primarily intended to assemble the output of the
1707 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1708 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1709 assemble correctly everything that other assemblers for the same
1710 machine would assemble.
1712 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1715 @c This remark should appear in generic version of manual; assumption
1716 @c here is that generic version sets M680x0.
1717 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1718 assembler for the same architecture; for example, we know of several
1719 incompatible versions of 680x0 assembly language syntax.
1724 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1725 program in one pass of the source file. This has a subtle impact on the
1726 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1728 @node Object Formats
1729 @section Object File Formats
1731 @cindex object file format
1732 The @sc{gnu} assembler can be configured to produce several alternative
1733 object file formats. For the most part, this does not affect how you
1734 write assembly language programs; but directives for debugging symbols
1735 are typically different in different file formats. @xref{Symbol
1736 Attributes,,Symbol Attributes}.
1739 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1740 @value{OBJ-NAME} format object files.
1742 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1744 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1745 @code{b.out} or COFF format object files.
1748 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1749 SOM or ELF format object files.
1754 @section Command Line
1756 @cindex command line conventions
1758 After the program name @command{@value{AS}}, the command line may contain
1759 options and file names. Options may appear in any order, and may be
1760 before, after, or between file names. The order of file names is
1763 @cindex standard input, as input file
1765 @file{--} (two hyphens) by itself names the standard input file
1766 explicitly, as one of the files for @command{@value{AS}} to assemble.
1768 @cindex options, command line
1769 Except for @samp{--} any command line argument that begins with a
1770 hyphen (@samp{-}) is an option. Each option changes the behavior of
1771 @command{@value{AS}}. No option changes the way another option works. An
1772 option is a @samp{-} followed by one or more letters; the case of
1773 the letter is important. All options are optional.
1775 Some options expect exactly one file name to follow them. The file
1776 name may either immediately follow the option's letter (compatible
1777 with older assemblers) or it may be the next command argument (@sc{gnu}
1778 standard). These two command lines are equivalent:
1781 @value{AS} -o my-object-file.o mumble.s
1782 @value{AS} -omy-object-file.o mumble.s
1786 @section Input Files
1789 @cindex source program
1790 @cindex files, input
1791 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1792 describe the program input to one run of @command{@value{AS}}. The program may
1793 be in one or more files; how the source is partitioned into files
1794 doesn't change the meaning of the source.
1796 @c I added "con" prefix to "catenation" just to prove I can overcome my
1797 @c APL training... doc@cygnus.com
1798 The source program is a concatenation of the text in all the files, in the
1801 @c man begin DESCRIPTION
1802 Each time you run @command{@value{AS}} it assembles exactly one source
1803 program. The source program is made up of one or more files.
1804 (The standard input is also a file.)
1806 You give @command{@value{AS}} a command line that has zero or more input file
1807 names. The input files are read (from left file name to right). A
1808 command line argument (in any position) that has no special meaning
1809 is taken to be an input file name.
1811 If you give @command{@value{AS}} no file names it attempts to read one input file
1812 from the @command{@value{AS}} standard input, which is normally your terminal. You
1813 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1816 Use @samp{--} if you need to explicitly name the standard input file
1817 in your command line.
1819 If the source is empty, @command{@value{AS}} produces a small, empty object
1824 @subheading Filenames and Line-numbers
1826 @cindex input file linenumbers
1827 @cindex line numbers, in input files
1828 There are two ways of locating a line in the input file (or files) and
1829 either may be used in reporting error messages. One way refers to a line
1830 number in a physical file; the other refers to a line number in a
1831 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1833 @dfn{Physical files} are those files named in the command line given
1834 to @command{@value{AS}}.
1836 @dfn{Logical files} are simply names declared explicitly by assembler
1837 directives; they bear no relation to physical files. Logical file names help
1838 error messages reflect the original source file, when @command{@value{AS}} source
1839 is itself synthesized from other files. @command{@value{AS}} understands the
1840 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1841 @ref{File,,@code{.file}}.
1844 @section Output (Object) File
1850 Every time you run @command{@value{AS}} it produces an output file, which is
1851 your assembly language program translated into numbers. This file
1852 is the object file. Its default name is
1860 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1862 You can give it another name by using the @option{-o} option. Conventionally,
1863 object file names end with @file{.o}. The default name is used for historical
1864 reasons: older assemblers were capable of assembling self-contained programs
1865 directly into a runnable program. (For some formats, this isn't currently
1866 possible, but it can be done for the @code{a.out} format.)
1870 The object file is meant for input to the linker @code{@value{LD}}. It contains
1871 assembled program code, information to help @code{@value{LD}} integrate
1872 the assembled program into a runnable file, and (optionally) symbolic
1873 information for the debugger.
1875 @c link above to some info file(s) like the description of a.out.
1876 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1879 @section Error and Warning Messages
1881 @c man begin DESCRIPTION
1883 @cindex error messages
1884 @cindex warning messages
1885 @cindex messages from assembler
1886 @command{@value{AS}} may write warnings and error messages to the standard error
1887 file (usually your terminal). This should not happen when a compiler
1888 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1889 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1890 grave problem that stops the assembly.
1894 @cindex format of warning messages
1895 Warning messages have the format
1898 file_name:@b{NNN}:Warning Message Text
1902 @cindex line numbers, in warnings/errors
1903 (where @b{NNN} is a line number). If a logical file name has been given
1904 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1905 the current input file is used. If a logical line number was given
1907 (@pxref{Line,,@code{.line}})
1909 then it is used to calculate the number printed,
1910 otherwise the actual line in the current source file is printed. The
1911 message text is intended to be self explanatory (in the grand Unix
1914 @cindex format of error messages
1915 Error messages have the format
1917 file_name:@b{NNN}:FATAL:Error Message Text
1919 The file name and line number are derived as for warning
1920 messages. The actual message text may be rather less explanatory
1921 because many of them aren't supposed to happen.
1924 @chapter Command-Line Options
1926 @cindex options, all versions of assembler
1927 This chapter describes command-line options available in @emph{all}
1928 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
1929 for options specific
1931 to the @value{TARGET} target.
1934 to particular machine architectures.
1937 @c man begin DESCRIPTION
1939 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1940 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1941 The assembler arguments must be separated from each other (and the @samp{-Wa})
1942 by commas. For example:
1945 gcc -c -g -O -Wa,-alh,-L file.c
1949 This passes two options to the assembler: @samp{-alh} (emit a listing to
1950 standard output with high-level and assembly source) and @samp{-L} (retain
1951 local symbols in the symbol table).
1953 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1954 command-line options are automatically passed to the assembler by the compiler.
1955 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1956 precisely what options it passes to each compilation pass, including the
1962 * a:: -a[cdghlns] enable listings
1963 * alternate:: --alternate enable alternate macro syntax
1964 * D:: -D for compatibility
1965 * f:: -f to work faster
1966 * I:: -I for .include search path
1967 @ifclear DIFF-TBL-KLUGE
1968 * K:: -K for compatibility
1970 @ifset DIFF-TBL-KLUGE
1971 * K:: -K for difference tables
1974 * L:: -L to retain local symbols
1975 * listing:: --listing-XXX to configure listing output
1976 * M:: -M or --mri to assemble in MRI compatibility mode
1977 * MD:: --MD for dependency tracking
1978 * o:: -o to name the object file
1979 * R:: -R to join data and text sections
1980 * statistics:: --statistics to see statistics about assembly
1981 * traditional-format:: --traditional-format for compatible output
1982 * v:: -v to announce version
1983 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1984 * Z:: -Z to make object file even after errors
1988 @section Enable Listings: @option{-a[cdghlns]}
1998 @cindex listings, enabling
1999 @cindex assembly listings, enabling
2001 These options enable listing output from the assembler. By itself,
2002 @samp{-a} requests high-level, assembly, and symbols listing.
2003 You can use other letters to select specific options for the list:
2004 @samp{-ah} requests a high-level language listing,
2005 @samp{-al} requests an output-program assembly listing, and
2006 @samp{-as} requests a symbol table listing.
2007 High-level listings require that a compiler debugging option like
2008 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2011 Use the @samp{-ag} option to print a first section with general assembly
2012 information, like @value{AS} version, switches passed, or time stamp.
2014 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2015 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2016 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2017 omitted from the listing.
2019 Use the @samp{-ad} option to omit debugging directives from the
2022 Once you have specified one of these options, you can further control
2023 listing output and its appearance using the directives @code{.list},
2024 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2026 The @samp{-an} option turns off all forms processing.
2027 If you do not request listing output with one of the @samp{-a} options, the
2028 listing-control directives have no effect.
2030 The letters after @samp{-a} may be combined into one option,
2031 @emph{e.g.}, @samp{-aln}.
2033 Note if the assembler source is coming from the standard input (e.g.,
2035 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
2036 is being used) then the listing will not contain any comments or preprocessor
2037 directives. This is because the listing code buffers input source lines from
2038 stdin only after they have been preprocessed by the assembler. This reduces
2039 memory usage and makes the code more efficient.
2042 @section @option{--alternate}
2045 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2048 @section @option{-D}
2051 This option has no effect whatsoever, but it is accepted to make it more
2052 likely that scripts written for other assemblers also work with
2053 @command{@value{AS}}.
2056 @section Work Faster: @option{-f}
2059 @cindex trusted compiler
2060 @cindex faster processing (@option{-f})
2061 @samp{-f} should only be used when assembling programs written by a
2062 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2063 and comment preprocessing on
2064 the input file(s) before assembling them. @xref{Preprocessing,
2068 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2069 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2074 @section @code{.include} Search Path: @option{-I} @var{path}
2076 @kindex -I @var{path}
2077 @cindex paths for @code{.include}
2078 @cindex search path for @code{.include}
2079 @cindex @code{include} directive search path
2080 Use this option to add a @var{path} to the list of directories
2081 @command{@value{AS}} searches for files specified in @code{.include}
2082 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2083 many times as necessary to include a variety of paths. The current
2084 working directory is always searched first; after that, @command{@value{AS}}
2085 searches any @samp{-I} directories in the same order as they were
2086 specified (left to right) on the command line.
2089 @section Difference Tables: @option{-K}
2092 @ifclear DIFF-TBL-KLUGE
2093 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2094 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2095 where it can be used to warn when the assembler alters the machine code
2096 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2097 family does not have the addressing limitations that sometimes lead to this
2098 alteration on other platforms.
2101 @ifset DIFF-TBL-KLUGE
2102 @cindex difference tables, warning
2103 @cindex warning for altered difference tables
2104 @command{@value{AS}} sometimes alters the code emitted for directives of the
2105 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2106 You can use the @samp{-K} option if you want a warning issued when this
2111 @section Include Local Symbols: @option{-L}
2114 @cindex local symbols, retaining in output
2115 Symbols beginning with system-specific local label prefixes, typically
2116 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2117 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2118 such symbols when debugging, because they are intended for the use of
2119 programs (like compilers) that compose assembler programs, not for your
2120 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2121 such symbols, so you do not normally debug with them.
2123 This option tells @command{@value{AS}} to retain those local symbols
2124 in the object file. Usually if you do this you also tell the linker
2125 @code{@value{LD}} to preserve those symbols.
2128 @section Configuring listing output: @option{--listing}
2130 The listing feature of the assembler can be enabled via the command line switch
2131 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2132 hex dump of the corresponding locations in the output object file, and displays
2133 them as a listing file. The format of this listing can be controlled by
2134 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2135 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2136 @code{.psize} (@pxref{Psize}), and
2137 @code{.eject} (@pxref{Eject}) and also by the following switches:
2140 @item --listing-lhs-width=@samp{number}
2141 @kindex --listing-lhs-width
2142 @cindex Width of first line disassembly output
2143 Sets the maximum width, in words, of the first line of the hex byte dump. This
2144 dump appears on the left hand side of the listing output.
2146 @item --listing-lhs-width2=@samp{number}
2147 @kindex --listing-lhs-width2
2148 @cindex Width of continuation lines of disassembly output
2149 Sets the maximum width, in words, of any further lines of the hex byte dump for
2150 a given input source line. If this value is not specified, it defaults to being
2151 the same as the value specified for @samp{--listing-lhs-width}. If neither
2152 switch is used the default is to one.
2154 @item --listing-rhs-width=@samp{number}
2155 @kindex --listing-rhs-width
2156 @cindex Width of source line output
2157 Sets the maximum width, in characters, of the source line that is displayed
2158 alongside the hex dump. The default value for this parameter is 100. The
2159 source line is displayed on the right hand side of the listing output.
2161 @item --listing-cont-lines=@samp{number}
2162 @kindex --listing-cont-lines
2163 @cindex Maximum number of continuation lines
2164 Sets the maximum number of continuation lines of hex dump that will be
2165 displayed for a given single line of source input. The default value is 4.
2169 @section Assemble in MRI Compatibility Mode: @option{-M}
2172 @cindex MRI compatibility mode
2173 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2174 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2175 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
2176 configured target) assembler from Microtec Research. The exact nature of the
2177 MRI syntax will not be documented here; see the MRI manuals for more
2178 information. Note in particular that the handling of macros and macro
2179 arguments is somewhat different. The purpose of this option is to permit
2180 assembling existing MRI assembler code using @command{@value{AS}}.
2182 The MRI compatibility is not complete. Certain operations of the MRI assembler
2183 depend upon its object file format, and can not be supported using other object
2184 file formats. Supporting these would require enhancing each object file format
2185 individually. These are:
2188 @item global symbols in common section
2190 The m68k MRI assembler supports common sections which are merged by the linker.
2191 Other object file formats do not support this. @command{@value{AS}} handles
2192 common sections by treating them as a single common symbol. It permits local
2193 symbols to be defined within a common section, but it can not support global
2194 symbols, since it has no way to describe them.
2196 @item complex relocations
2198 The MRI assemblers support relocations against a negated section address, and
2199 relocations which combine the start addresses of two or more sections. These
2200 are not support by other object file formats.
2202 @item @code{END} pseudo-op specifying start address
2204 The MRI @code{END} pseudo-op permits the specification of a start address.
2205 This is not supported by other object file formats. The start address may
2206 instead be specified using the @option{-e} option to the linker, or in a linker
2209 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2211 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2212 name to the output file. This is not supported by other object file formats.
2214 @item @code{ORG} pseudo-op
2216 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2217 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2218 which changes the location within the current section. Absolute sections are
2219 not supported by other object file formats. The address of a section may be
2220 assigned within a linker script.
2223 There are some other features of the MRI assembler which are not supported by
2224 @command{@value{AS}}, typically either because they are difficult or because they
2225 seem of little consequence. Some of these may be supported in future releases.
2229 @item EBCDIC strings
2231 EBCDIC strings are not supported.
2233 @item packed binary coded decimal
2235 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2236 and @code{DCB.P} pseudo-ops are not supported.
2238 @item @code{FEQU} pseudo-op
2240 The m68k @code{FEQU} pseudo-op is not supported.
2242 @item @code{NOOBJ} pseudo-op
2244 The m68k @code{NOOBJ} pseudo-op is not supported.
2246 @item @code{OPT} branch control options
2248 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2249 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2250 relaxes all branches, whether forward or backward, to an appropriate size, so
2251 these options serve no purpose.
2253 @item @code{OPT} list control options
2255 The following m68k @code{OPT} list control options are ignored: @code{C},
2256 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2257 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2259 @item other @code{OPT} options
2261 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2262 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2264 @item @code{OPT} @code{D} option is default
2266 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2267 @code{OPT NOD} may be used to turn it off.
2269 @item @code{XREF} pseudo-op.
2271 The m68k @code{XREF} pseudo-op is ignored.
2273 @item @code{.debug} pseudo-op
2275 The i960 @code{.debug} pseudo-op is not supported.
2277 @item @code{.extended} pseudo-op
2279 The i960 @code{.extended} pseudo-op is not supported.
2281 @item @code{.list} pseudo-op.
2283 The various options of the i960 @code{.list} pseudo-op are not supported.
2285 @item @code{.optimize} pseudo-op
2287 The i960 @code{.optimize} pseudo-op is not supported.
2289 @item @code{.output} pseudo-op
2291 The i960 @code{.output} pseudo-op is not supported.
2293 @item @code{.setreal} pseudo-op
2295 The i960 @code{.setreal} pseudo-op is not supported.
2300 @section Dependency Tracking: @option{--MD}
2303 @cindex dependency tracking
2306 @command{@value{AS}} can generate a dependency file for the file it creates. This
2307 file consists of a single rule suitable for @code{make} describing the
2308 dependencies of the main source file.
2310 The rule is written to the file named in its argument.
2312 This feature is used in the automatic updating of makefiles.
2315 @section Name the Object File: @option{-o}
2318 @cindex naming object file
2319 @cindex object file name
2320 There is always one object file output when you run @command{@value{AS}}. By
2321 default it has the name
2324 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2338 You use this option (which takes exactly one filename) to give the
2339 object file a different name.
2341 Whatever the object file is called, @command{@value{AS}} overwrites any
2342 existing file of the same name.
2345 @section Join Data and Text Sections: @option{-R}
2348 @cindex data and text sections, joining
2349 @cindex text and data sections, joining
2350 @cindex joining text and data sections
2351 @cindex merging text and data sections
2352 @option{-R} tells @command{@value{AS}} to write the object file as if all
2353 data-section data lives in the text section. This is only done at
2354 the very last moment: your binary data are the same, but data
2355 section parts are relocated differently. The data section part of
2356 your object file is zero bytes long because all its bytes are
2357 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2359 When you specify @option{-R} it would be possible to generate shorter
2360 address displacements (because we do not have to cross between text and
2361 data section). We refrain from doing this simply for compatibility with
2362 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2365 When @command{@value{AS}} is configured for COFF or ELF output,
2366 this option is only useful if you use sections named @samp{.text} and
2371 @option{-R} is not supported for any of the HPPA targets. Using
2372 @option{-R} generates a warning from @command{@value{AS}}.
2376 @section Display Assembly Statistics: @option{--statistics}
2378 @kindex --statistics
2379 @cindex statistics, about assembly
2380 @cindex time, total for assembly
2381 @cindex space used, maximum for assembly
2382 Use @samp{--statistics} to display two statistics about the resources used by
2383 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2384 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2387 @node traditional-format
2388 @section Compatible Output: @option{--traditional-format}
2390 @kindex --traditional-format
2391 For some targets, the output of @command{@value{AS}} is different in some ways
2392 from the output of some existing assembler. This switch requests
2393 @command{@value{AS}} to use the traditional format instead.
2395 For example, it disables the exception frame optimizations which
2396 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2399 @section Announce Version: @option{-v}
2403 @cindex assembler version
2404 @cindex version of assembler
2405 You can find out what version of as is running by including the
2406 option @samp{-v} (which you can also spell as @samp{-version}) on the
2410 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2412 @command{@value{AS}} should never give a warning or error message when
2413 assembling compiler output. But programs written by people often
2414 cause @command{@value{AS}} to give a warning that a particular assumption was
2415 made. All such warnings are directed to the standard error file.
2419 @cindex suppressing warnings
2420 @cindex warnings, suppressing
2421 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2422 This only affects the warning messages: it does not change any particular of
2423 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2426 @kindex --fatal-warnings
2427 @cindex errors, caused by warnings
2428 @cindex warnings, causing error
2429 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2430 files that generate warnings to be in error.
2433 @cindex warnings, switching on
2434 You can switch these options off again by specifying @option{--warn}, which
2435 causes warnings to be output as usual.
2438 @section Generate Object File in Spite of Errors: @option{-Z}
2439 @cindex object file, after errors
2440 @cindex errors, continuing after
2441 After an error message, @command{@value{AS}} normally produces no output. If for
2442 some reason you are interested in object file output even after
2443 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2444 option. If there are any errors, @command{@value{AS}} continues anyways, and
2445 writes an object file after a final warning message of the form @samp{@var{n}
2446 errors, @var{m} warnings, generating bad object file.}
2451 @cindex machine-independent syntax
2452 @cindex syntax, machine-independent
2453 This chapter describes the machine-independent syntax allowed in a
2454 source file. @command{@value{AS}} syntax is similar to what many other
2455 assemblers use; it is inspired by the BSD 4.2
2460 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2464 * Preprocessing:: Preprocessing
2465 * Whitespace:: Whitespace
2466 * Comments:: Comments
2467 * Symbol Intro:: Symbols
2468 * Statements:: Statements
2469 * Constants:: Constants
2473 @section Preprocessing
2475 @cindex preprocessing
2476 The @command{@value{AS}} internal preprocessor:
2478 @cindex whitespace, removed by preprocessor
2480 adjusts and removes extra whitespace. It leaves one space or tab before
2481 the keywords on a line, and turns any other whitespace on the line into
2484 @cindex comments, removed by preprocessor
2486 removes all comments, replacing them with a single space, or an
2487 appropriate number of newlines.
2489 @cindex constants, converted by preprocessor
2491 converts character constants into the appropriate numeric values.
2494 It does not do macro processing, include file handling, or
2495 anything else you may get from your C compiler's preprocessor. You can
2496 do include file processing with the @code{.include} directive
2497 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2498 to get other ``CPP'' style preprocessing by giving the input file a
2499 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2500 Output, gcc.info, Using GNU CC}.
2502 Excess whitespace, comments, and character constants
2503 cannot be used in the portions of the input text that are not
2506 @cindex turning preprocessing on and off
2507 @cindex preprocessing, turning on and off
2510 If the first line of an input file is @code{#NO_APP} or if you use the
2511 @samp{-f} option, whitespace and comments are not removed from the input file.
2512 Within an input file, you can ask for whitespace and comment removal in
2513 specific portions of the by putting a line that says @code{#APP} before the
2514 text that may contain whitespace or comments, and putting a line that says
2515 @code{#NO_APP} after this text. This feature is mainly intend to support
2516 @code{asm} statements in compilers whose output is otherwise free of comments
2523 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2524 Whitespace is used to separate symbols, and to make programs neater for
2525 people to read. Unless within character constants
2526 (@pxref{Characters,,Character Constants}), any whitespace means the same
2527 as exactly one space.
2533 There are two ways of rendering comments to @command{@value{AS}}. In both
2534 cases the comment is equivalent to one space.
2536 Anything from @samp{/*} through the next @samp{*/} is a comment.
2537 This means you may not nest these comments.
2541 The only way to include a newline ('\n') in a comment
2542 is to use this sort of comment.
2545 /* This sort of comment does not nest. */
2548 @cindex line comment character
2549 Anything from a @dfn{line comment} character up to the next newline is
2550 considered a comment and is ignored. The line comment character is target
2551 specific, and some targets multiple comment characters. Some targets also have
2552 line comment characters that only work if they are the first character on a
2553 line. Some targets use a sequence of two characters to introduce a line
2554 comment. Some targets can also change their line comment characters depending
2555 upon command line options that have been used. For more details see the
2556 @emph{Syntax} section in the documentation for individual targets.
2558 If the line comment character is the hash sign (@samp{#}) then it still has the
2559 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2560 to specify logical line numbers:
2563 @cindex lines starting with @code{#}
2564 @cindex logical line numbers
2565 To be compatible with past assemblers, lines that begin with @samp{#} have a
2566 special interpretation. Following the @samp{#} should be an absolute
2567 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2568 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2569 new logical file name. The rest of the line, if any, should be whitespace.
2571 If the first non-whitespace characters on the line are not numeric,
2572 the line is ignored. (Just like a comment.)
2575 # This is an ordinary comment.
2576 # 42-6 "new_file_name" # New logical file name
2577 # This is logical line # 36.
2579 This feature is deprecated, and may disappear from future versions
2580 of @command{@value{AS}}.
2585 @cindex characters used in symbols
2586 @ifclear SPECIAL-SYMS
2587 A @dfn{symbol} is one or more characters chosen from the set of all
2588 letters (both upper and lower case), digits and the three characters
2594 A @dfn{symbol} is one or more characters chosen from the set of all
2595 letters (both upper and lower case), digits and the three characters
2596 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2602 On most machines, you can also use @code{$} in symbol names; exceptions
2603 are noted in @ref{Machine Dependencies}.
2605 No symbol may begin with a digit. Case is significant.
2606 There is no length limit: all characters are significant. Multibyte characters
2607 are supported. Symbols are delimited by characters not in that set, or by the
2608 beginning of a file (since the source program must end with a newline, the end
2609 of a file is not a possible symbol delimiter). @xref{Symbols}.
2610 @cindex length of symbols
2615 @cindex statements, structure of
2616 @cindex line separator character
2617 @cindex statement separator character
2619 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2620 @dfn{line separator character}. The line separator character is target
2621 specific and described in the @emph{Syntax} section of each
2622 target's documentation. Not all targets support a line separator character.
2623 The newline or line separator character is considered to be part of the
2624 preceding statement. Newlines and separators within character constants are an
2625 exception: they do not end statements.
2627 @cindex newline, required at file end
2628 @cindex EOF, newline must precede
2629 It is an error to end any statement with end-of-file: the last
2630 character of any input file should be a newline.@refill
2632 An empty statement is allowed, and may include whitespace. It is ignored.
2634 @cindex instructions and directives
2635 @cindex directives and instructions
2636 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2637 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2639 A statement begins with zero or more labels, optionally followed by a
2640 key symbol which determines what kind of statement it is. The key
2641 symbol determines the syntax of the rest of the statement. If the
2642 symbol begins with a dot @samp{.} then the statement is an assembler
2643 directive: typically valid for any computer. If the symbol begins with
2644 a letter the statement is an assembly language @dfn{instruction}: it
2645 assembles into a machine language instruction.
2647 Different versions of @command{@value{AS}} for different computers
2648 recognize different instructions. In fact, the same symbol may
2649 represent a different instruction in a different computer's assembly
2653 @cindex @code{:} (label)
2654 @cindex label (@code{:})
2655 A label is a symbol immediately followed by a colon (@code{:}).
2656 Whitespace before a label or after a colon is permitted, but you may not
2657 have whitespace between a label's symbol and its colon. @xref{Labels}.
2660 For HPPA targets, labels need not be immediately followed by a colon, but
2661 the definition of a label must begin in column zero. This also implies that
2662 only one label may be defined on each line.
2666 label: .directive followed by something
2667 another_label: # This is an empty statement.
2668 instruction operand_1, operand_2, @dots{}
2675 A constant is a number, written so that its value is known by
2676 inspection, without knowing any context. Like this:
2679 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2680 .ascii "Ring the bell\7" # A string constant.
2681 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2682 .float 0f-314159265358979323846264338327\
2683 95028841971.693993751E-40 # - pi, a flonum.
2688 * Characters:: Character Constants
2689 * Numbers:: Number Constants
2693 @subsection Character Constants
2695 @cindex character constants
2696 @cindex constants, character
2697 There are two kinds of character constants. A @dfn{character} stands
2698 for one character in one byte and its value may be used in
2699 numeric expressions. String constants (properly called string
2700 @emph{literals}) are potentially many bytes and their values may not be
2701 used in arithmetic expressions.
2705 * Chars:: Characters
2709 @subsubsection Strings
2711 @cindex string constants
2712 @cindex constants, string
2713 A @dfn{string} is written between double-quotes. It may contain
2714 double-quotes or null characters. The way to get special characters
2715 into a string is to @dfn{escape} these characters: precede them with
2716 a backslash @samp{\} character. For example @samp{\\} represents
2717 one backslash: the first @code{\} is an escape which tells
2718 @command{@value{AS}} to interpret the second character literally as a backslash
2719 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2720 escape character). The complete list of escapes follows.
2722 @cindex escape codes, character
2723 @cindex character escape codes
2726 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2728 @cindex @code{\b} (backspace character)
2729 @cindex backspace (@code{\b})
2731 Mnemonic for backspace; for ASCII this is octal code 010.
2734 @c Mnemonic for EOText; for ASCII this is octal code 004.
2736 @cindex @code{\f} (formfeed character)
2737 @cindex formfeed (@code{\f})
2739 Mnemonic for FormFeed; for ASCII this is octal code 014.
2741 @cindex @code{\n} (newline character)
2742 @cindex newline (@code{\n})
2744 Mnemonic for newline; for ASCII this is octal code 012.
2747 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2749 @cindex @code{\r} (carriage return character)
2750 @cindex carriage return (@code{\r})
2752 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2755 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2756 @c other assemblers.
2758 @cindex @code{\t} (tab)
2759 @cindex tab (@code{\t})
2761 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2764 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2765 @c @item \x @var{digit} @var{digit} @var{digit}
2766 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2768 @cindex @code{\@var{ddd}} (octal character code)
2769 @cindex octal character code (@code{\@var{ddd}})
2770 @item \ @var{digit} @var{digit} @var{digit}
2771 An octal character code. The numeric code is 3 octal digits.
2772 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2773 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2775 @cindex @code{\@var{xd...}} (hex character code)
2776 @cindex hex character code (@code{\@var{xd...}})
2777 @item \@code{x} @var{hex-digits...}
2778 A hex character code. All trailing hex digits are combined. Either upper or
2779 lower case @code{x} works.
2781 @cindex @code{\\} (@samp{\} character)
2782 @cindex backslash (@code{\\})
2784 Represents one @samp{\} character.
2787 @c Represents one @samp{'} (accent acute) character.
2788 @c This is needed in single character literals
2789 @c (@xref{Characters,,Character Constants}.) to represent
2792 @cindex @code{\"} (doublequote character)
2793 @cindex doublequote (@code{\"})
2795 Represents one @samp{"} character. Needed in strings to represent
2796 this character, because an unescaped @samp{"} would end the string.
2798 @item \ @var{anything-else}
2799 Any other character when escaped by @kbd{\} gives a warning, but
2800 assembles as if the @samp{\} was not present. The idea is that if
2801 you used an escape sequence you clearly didn't want the literal
2802 interpretation of the following character. However @command{@value{AS}} has no
2803 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2804 code and warns you of the fact.
2807 Which characters are escapable, and what those escapes represent,
2808 varies widely among assemblers. The current set is what we think
2809 the BSD 4.2 assembler recognizes, and is a subset of what most C
2810 compilers recognize. If you are in doubt, do not use an escape
2814 @subsubsection Characters
2816 @cindex single character constant
2817 @cindex character, single
2818 @cindex constant, single character
2819 A single character may be written as a single quote immediately
2820 followed by that character. The same escapes apply to characters as
2821 to strings. So if you want to write the character backslash, you
2822 must write @kbd{'\\} where the first @code{\} escapes the second
2823 @code{\}. As you can see, the quote is an acute accent, not a
2824 grave accent. A newline
2826 @ifclear abnormal-separator
2827 (or semicolon @samp{;})
2829 @ifset abnormal-separator
2831 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2836 immediately following an acute accent is taken as a literal character
2837 and does not count as the end of a statement. The value of a character
2838 constant in a numeric expression is the machine's byte-wide code for
2839 that character. @command{@value{AS}} assumes your character code is ASCII:
2840 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2843 @subsection Number Constants
2845 @cindex constants, number
2846 @cindex number constants
2847 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2848 are stored in the target machine. @emph{Integers} are numbers that
2849 would fit into an @code{int} in the C language. @emph{Bignums} are
2850 integers, but they are stored in more than 32 bits. @emph{Flonums}
2851 are floating point numbers, described below.
2854 * Integers:: Integers
2859 * Bit Fields:: Bit Fields
2865 @subsubsection Integers
2867 @cindex constants, integer
2869 @cindex binary integers
2870 @cindex integers, binary
2871 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2872 the binary digits @samp{01}.
2874 @cindex octal integers
2875 @cindex integers, octal
2876 An octal integer is @samp{0} followed by zero or more of the octal
2877 digits (@samp{01234567}).
2879 @cindex decimal integers
2880 @cindex integers, decimal
2881 A decimal integer starts with a non-zero digit followed by zero or
2882 more digits (@samp{0123456789}).
2884 @cindex hexadecimal integers
2885 @cindex integers, hexadecimal
2886 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2887 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2889 Integers have the usual values. To denote a negative integer, use
2890 the prefix operator @samp{-} discussed under expressions
2891 (@pxref{Prefix Ops,,Prefix Operators}).
2894 @subsubsection Bignums
2897 @cindex constants, bignum
2898 A @dfn{bignum} has the same syntax and semantics as an integer
2899 except that the number (or its negative) takes more than 32 bits to
2900 represent in binary. The distinction is made because in some places
2901 integers are permitted while bignums are not.
2904 @subsubsection Flonums
2906 @cindex floating point numbers
2907 @cindex constants, floating point
2909 @cindex precision, floating point
2910 A @dfn{flonum} represents a floating point number. The translation is
2911 indirect: a decimal floating point number from the text is converted by
2912 @command{@value{AS}} to a generic binary floating point number of more than
2913 sufficient precision. This generic floating point number is converted
2914 to a particular computer's floating point format (or formats) by a
2915 portion of @command{@value{AS}} specialized to that computer.
2917 A flonum is written by writing (in order)
2922 (@samp{0} is optional on the HPPA.)
2926 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2928 @kbd{e} is recommended. Case is not important.
2930 @c FIXME: verify if flonum syntax really this vague for most cases
2931 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2932 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2935 On the H8/300, Renesas / SuperH SH,
2936 and AMD 29K architectures, the letter must be
2937 one of the letters @samp{DFPRSX} (in upper or lower case).
2939 On the ARC, the letter must be one of the letters @samp{DFRS}
2940 (in upper or lower case).
2942 On the Intel 960 architecture, the letter must be
2943 one of the letters @samp{DFT} (in upper or lower case).
2945 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2949 One of the letters @samp{DFRS} (in upper or lower case).
2952 One of the letters @samp{DFPRSX} (in upper or lower case).
2955 The letter @samp{E} (upper case only).
2958 One of the letters @samp{DFT} (in upper or lower case).
2963 An optional sign: either @samp{+} or @samp{-}.
2966 An optional @dfn{integer part}: zero or more decimal digits.
2969 An optional @dfn{fractional part}: @samp{.} followed by zero
2970 or more decimal digits.
2973 An optional exponent, consisting of:
2977 An @samp{E} or @samp{e}.
2978 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2979 @c principle this can perfectly well be different on different targets.
2981 Optional sign: either @samp{+} or @samp{-}.
2983 One or more decimal digits.
2988 At least one of the integer part or the fractional part must be
2989 present. The floating point number has the usual base-10 value.
2991 @command{@value{AS}} does all processing using integers. Flonums are computed
2992 independently of any floating point hardware in the computer running
2993 @command{@value{AS}}.
2997 @c Bit fields are written as a general facility but are also controlled
2998 @c by a conditional-compilation flag---which is as of now (21mar91)
2999 @c turned on only by the i960 config of GAS.
3001 @subsubsection Bit Fields
3004 @cindex constants, bit field
3005 You can also define numeric constants as @dfn{bit fields}.
3006 Specify two numbers separated by a colon---
3008 @var{mask}:@var{value}
3011 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
3014 The resulting number is then packed
3016 @c this conditional paren in case bit fields turned on elsewhere than 960
3017 (in host-dependent byte order)
3019 into a field whose width depends on which assembler directive has the
3020 bit-field as its argument. Overflow (a result from the bitwise and
3021 requiring more binary digits to represent) is not an error; instead,
3022 more constants are generated, of the specified width, beginning with the
3023 least significant digits.@refill
3025 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
3026 @code{.short}, and @code{.word} accept bit-field arguments.
3031 @chapter Sections and Relocation
3036 * Secs Background:: Background
3037 * Ld Sections:: Linker Sections
3038 * As Sections:: Assembler Internal Sections
3039 * Sub-Sections:: Sub-Sections
3043 @node Secs Background
3046 Roughly, a section is a range of addresses, with no gaps; all data
3047 ``in'' those addresses is treated the same for some particular purpose.
3048 For example there may be a ``read only'' section.
3050 @cindex linker, and assembler
3051 @cindex assembler, and linker
3052 The linker @code{@value{LD}} reads many object files (partial programs) and
3053 combines their contents to form a runnable program. When @command{@value{AS}}
3054 emits an object file, the partial program is assumed to start at address 0.
3055 @code{@value{LD}} assigns the final addresses for the partial program, so that
3056 different partial programs do not overlap. This is actually an
3057 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3060 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3061 addresses. These blocks slide to their run-time addresses as rigid
3062 units; their length does not change and neither does the order of bytes
3063 within them. Such a rigid unit is called a @emph{section}. Assigning
3064 run-time addresses to sections is called @dfn{relocation}. It includes
3065 the task of adjusting mentions of object-file addresses so they refer to
3066 the proper run-time addresses.
3068 For the H8/300, and for the Renesas / SuperH SH,
3069 @command{@value{AS}} pads sections if needed to
3070 ensure they end on a word (sixteen bit) boundary.
3073 @cindex standard assembler sections
3074 An object file written by @command{@value{AS}} has at least three sections, any
3075 of which may be empty. These are named @dfn{text}, @dfn{data} and
3080 When it generates COFF or ELF output,
3082 @command{@value{AS}} can also generate whatever other named sections you specify
3083 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3084 If you do not use any directives that place output in the @samp{.text}
3085 or @samp{.data} sections, these sections still exist, but are empty.
3090 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3092 @command{@value{AS}} can also generate whatever other named sections you
3093 specify using the @samp{.space} and @samp{.subspace} directives. See
3094 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3095 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3096 assembler directives.
3099 Additionally, @command{@value{AS}} uses different names for the standard
3100 text, data, and bss sections when generating SOM output. Program text
3101 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3102 BSS into @samp{$BSS$}.
3106 Within the object file, the text section starts at address @code{0}, the
3107 data section follows, and the bss section follows the data section.
3110 When generating either SOM or ELF output files on the HPPA, the text
3111 section starts at address @code{0}, the data section at address
3112 @code{0x4000000}, and the bss section follows the data section.
3115 To let @code{@value{LD}} know which data changes when the sections are
3116 relocated, and how to change that data, @command{@value{AS}} also writes to the
3117 object file details of the relocation needed. To perform relocation
3118 @code{@value{LD}} must know, each time an address in the object
3122 Where in the object file is the beginning of this reference to
3125 How long (in bytes) is this reference?
3127 Which section does the address refer to? What is the numeric value of
3129 (@var{address}) @minus{} (@var{start-address of section})?
3132 Is the reference to an address ``Program-Counter relative''?
3135 @cindex addresses, format of
3136 @cindex section-relative addressing
3137 In fact, every address @command{@value{AS}} ever uses is expressed as
3139 (@var{section}) + (@var{offset into section})
3142 Further, most expressions @command{@value{AS}} computes have this section-relative
3145 (For some object formats, such as SOM for the HPPA, some expressions are
3146 symbol-relative instead.)
3149 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3150 @var{N} into section @var{secname}.''
3152 Apart from text, data and bss sections you need to know about the
3153 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3154 addresses in the absolute section remain unchanged. For example, address
3155 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3156 @code{@value{LD}}. Although the linker never arranges two partial programs'
3157 data sections with overlapping addresses after linking, @emph{by definition}
3158 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3159 part of a program is always the same address when the program is running as
3160 address @code{@{absolute@ 239@}} in any other part of the program.
3162 The idea of sections is extended to the @dfn{undefined} section. Any
3163 address whose section is unknown at assembly time is by definition
3164 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3165 Since numbers are always defined, the only way to generate an undefined
3166 address is to mention an undefined symbol. A reference to a named
3167 common block would be such a symbol: its value is unknown at assembly
3168 time so it has section @emph{undefined}.
3170 By analogy the word @emph{section} is used to describe groups of sections in
3171 the linked program. @code{@value{LD}} puts all partial programs' text
3172 sections in contiguous addresses in the linked program. It is
3173 customary to refer to the @emph{text section} of a program, meaning all
3174 the addresses of all partial programs' text sections. Likewise for
3175 data and bss sections.
3177 Some sections are manipulated by @code{@value{LD}}; others are invented for
3178 use of @command{@value{AS}} and have no meaning except during assembly.
3181 @section Linker Sections
3182 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3187 @cindex named sections
3188 @cindex sections, named
3189 @item named sections
3192 @cindex text section
3193 @cindex data section
3197 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3198 separate but equal sections. Anything you can say of one section is
3201 When the program is running, however, it is
3202 customary for the text section to be unalterable. The
3203 text section is often shared among processes: it contains
3204 instructions, constants and the like. The data section of a running
3205 program is usually alterable: for example, C variables would be stored
3206 in the data section.
3211 This section contains zeroed bytes when your program begins running. It
3212 is used to hold uninitialized variables or common storage. The length of
3213 each partial program's bss section is important, but because it starts
3214 out containing zeroed bytes there is no need to store explicit zero
3215 bytes in the object file. The bss section was invented to eliminate
3216 those explicit zeros from object files.
3218 @cindex absolute section
3219 @item absolute section
3220 Address 0 of this section is always ``relocated'' to runtime address 0.
3221 This is useful if you want to refer to an address that @code{@value{LD}} must
3222 not change when relocating. In this sense we speak of absolute
3223 addresses being ``unrelocatable'': they do not change during relocation.
3225 @cindex undefined section
3226 @item undefined section
3227 This ``section'' is a catch-all for address references to objects not in
3228 the preceding sections.
3229 @c FIXME: ref to some other doc on obj-file formats could go here.
3232 @cindex relocation example
3233 An idealized example of three relocatable sections follows.
3235 The example uses the traditional section names @samp{.text} and @samp{.data}.
3237 Memory addresses are on the horizontal axis.
3241 @c END TEXI2ROFF-KILL
3244 partial program # 1: |ttttt|dddd|00|
3251 partial program # 2: |TTT|DDD|000|
3254 +--+---+-----+--+----+---+-----+~~
3255 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3256 +--+---+-----+--+----+---+-----+~~
3258 addresses: 0 @dots{}
3265 \line{\it Partial program \#1: \hfil}
3266 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3267 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3269 \line{\it Partial program \#2: \hfil}
3270 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3271 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3273 \line{\it linked program: \hfil}
3274 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3275 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3276 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3277 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3279 \line{\it addresses: \hfil}
3283 @c END TEXI2ROFF-KILL
3286 @section Assembler Internal Sections
3288 @cindex internal assembler sections
3289 @cindex sections in messages, internal
3290 These sections are meant only for the internal use of @command{@value{AS}}. They
3291 have no meaning at run-time. You do not really need to know about these
3292 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3293 warning messages, so it might be helpful to have an idea of their
3294 meanings to @command{@value{AS}}. These sections are used to permit the
3295 value of every expression in your assembly language program to be a
3296 section-relative address.
3299 @cindex assembler internal logic error
3300 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3301 An internal assembler logic error has been found. This means there is a
3302 bug in the assembler.
3304 @cindex expr (internal section)
3306 The assembler stores complex expression internally as combinations of
3307 symbols. When it needs to represent an expression as a symbol, it puts
3308 it in the expr section.
3310 @c FIXME item transfer[t] vector preload
3311 @c FIXME item transfer[t] vector postload
3312 @c FIXME item register
3316 @section Sub-Sections
3318 @cindex numbered subsections
3319 @cindex grouping data
3325 fall into two sections: text and data.
3327 You may have separate groups of
3329 data in named sections
3333 data in named sections
3339 that you want to end up near to each other in the object file, even though they
3340 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3341 use @dfn{subsections} for this purpose. Within each section, there can be
3342 numbered subsections with values from 0 to 8192. Objects assembled into the
3343 same subsection go into the object file together with other objects in the same
3344 subsection. For example, a compiler might want to store constants in the text
3345 section, but might not want to have them interspersed with the program being
3346 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3347 section of code being output, and a @samp{.text 1} before each group of
3348 constants being output.
3350 Subsections are optional. If you do not use subsections, everything
3351 goes in subsection number zero.
3354 Each subsection is zero-padded up to a multiple of four bytes.
3355 (Subsections may be padded a different amount on different flavors
3356 of @command{@value{AS}}.)
3360 On the H8/300 platform, each subsection is zero-padded to a word
3361 boundary (two bytes).
3362 The same is true on the Renesas SH.
3365 @c FIXME section padding (alignment)?
3366 @c Rich Pixley says padding here depends on target obj code format; that
3367 @c doesn't seem particularly useful to say without further elaboration,
3368 @c so for now I say nothing about it. If this is a generic BFD issue,
3369 @c these paragraphs might need to vanish from this manual, and be
3370 @c discussed in BFD chapter of binutils (or some such).
3374 Subsections appear in your object file in numeric order, lowest numbered
3375 to highest. (All this to be compatible with other people's assemblers.)
3376 The object file contains no representation of subsections; @code{@value{LD}} and
3377 other programs that manipulate object files see no trace of them.
3378 They just see all your text subsections as a text section, and all your
3379 data subsections as a data section.
3381 To specify which subsection you want subsequent statements assembled
3382 into, use a numeric argument to specify it, in a @samp{.text
3383 @var{expression}} or a @samp{.data @var{expression}} statement.
3386 When generating COFF output, you
3391 can also use an extra subsection
3392 argument with arbitrary named sections: @samp{.section @var{name},
3397 When generating ELF output, you
3402 can also use the @code{.subsection} directive (@pxref{SubSection})
3403 to specify a subsection: @samp{.subsection @var{expression}}.
3405 @var{Expression} should be an absolute expression
3406 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3407 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3408 begins in @code{text 0}. For instance:
3410 .text 0 # The default subsection is text 0 anyway.
3411 .ascii "This lives in the first text subsection. *"
3413 .ascii "But this lives in the second text subsection."
3415 .ascii "This lives in the data section,"
3416 .ascii "in the first data subsection."
3418 .ascii "This lives in the first text section,"
3419 .ascii "immediately following the asterisk (*)."
3422 Each section has a @dfn{location counter} incremented by one for every byte
3423 assembled into that section. Because subsections are merely a convenience
3424 restricted to @command{@value{AS}} there is no concept of a subsection location
3425 counter. There is no way to directly manipulate a location counter---but the
3426 @code{.align} directive changes it, and any label definition captures its
3427 current value. The location counter of the section where statements are being
3428 assembled is said to be the @dfn{active} location counter.
3431 @section bss Section
3434 @cindex common variable storage
3435 The bss section is used for local common variable storage.
3436 You may allocate address space in the bss section, but you may
3437 not dictate data to load into it before your program executes. When
3438 your program starts running, all the contents of the bss
3439 section are zeroed bytes.
3441 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3442 @ref{Lcomm,,@code{.lcomm}}.
3444 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3445 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3448 When assembling for a target which supports multiple sections, such as ELF or
3449 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3450 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3451 section. Typically the section will only contain symbol definitions and
3452 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3459 Symbols are a central concept: the programmer uses symbols to name
3460 things, the linker uses symbols to link, and the debugger uses symbols
3464 @cindex debuggers, and symbol order
3465 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3466 the same order they were declared. This may break some debuggers.
3471 * Setting Symbols:: Giving Symbols Other Values
3472 * Symbol Names:: Symbol Names
3473 * Dot:: The Special Dot Symbol
3474 * Symbol Attributes:: Symbol Attributes
3481 A @dfn{label} is written as a symbol immediately followed by a colon
3482 @samp{:}. The symbol then represents the current value of the
3483 active location counter, and is, for example, a suitable instruction
3484 operand. You are warned if you use the same symbol to represent two
3485 different locations: the first definition overrides any other
3489 On the HPPA, the usual form for a label need not be immediately followed by a
3490 colon, but instead must start in column zero. Only one label may be defined on
3491 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3492 provides a special directive @code{.label} for defining labels more flexibly.
3495 @node Setting Symbols
3496 @section Giving Symbols Other Values
3498 @cindex assigning values to symbols
3499 @cindex symbol values, assigning
3500 A symbol can be given an arbitrary value by writing a symbol, followed
3501 by an equals sign @samp{=}, followed by an expression
3502 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3503 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3504 equals sign @samp{=}@samp{=} here represents an equivalent of the
3505 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3508 Blackfin does not support symbol assignment with @samp{=}.
3512 @section Symbol Names
3514 @cindex symbol names
3515 @cindex names, symbol
3516 @ifclear SPECIAL-SYMS
3517 Symbol names begin with a letter or with one of @samp{._}. On most
3518 machines, you can also use @code{$} in symbol names; exceptions are
3519 noted in @ref{Machine Dependencies}. That character may be followed by any
3520 string of digits, letters, dollar signs (unless otherwise noted for a
3521 particular target machine), and underscores.
3525 Symbol names begin with a letter or with one of @samp{._}. On the
3526 Renesas SH you can also use @code{$} in symbol names. That
3527 character may be followed by any string of digits, letters, dollar signs (save
3528 on the H8/300), and underscores.
3532 Case of letters is significant: @code{foo} is a different symbol name
3535 Multibyte characters are supported. To generate a symbol name containing
3536 multibyte characters enclose it within double quotes and use escape codes. cf
3537 @xref{Strings}. Generating a multibyte symbol name from a label is not
3538 currently supported.
3540 Each symbol has exactly one name. Each name in an assembly language program
3541 refers to exactly one symbol. You may use that symbol name any number of times
3544 @subheading Local Symbol Names
3546 @cindex local symbol names
3547 @cindex symbol names, local
3548 A local symbol is any symbol beginning with certain local label prefixes.
3549 By default, the local label prefix is @samp{.L} for ELF systems or
3550 @samp{L} for traditional a.out systems, but each target may have its own
3551 set of local label prefixes.
3553 On the HPPA local symbols begin with @samp{L$}.
3556 Local symbols are defined and used within the assembler, but they are
3557 normally not saved in object files. Thus, they are not visible when debugging.
3558 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols:
3559 @option{-L}}) to retain the local symbols in the object files.
3561 @subheading Local Labels
3563 @cindex local labels
3564 @cindex temporary symbol names
3565 @cindex symbol names, temporary
3566 Local labels help compilers and programmers use names temporarily.
3567 They create symbols which are guaranteed to be unique over the entire scope of
3568 the input source code and which can be referred to by a simple notation.
3569 To define a local label, write a label of the form @samp{@b{N}:} (where @b{N}
3570 represents any positive integer). To refer to the most recent previous
3571 definition of that label write @samp{@b{N}b}, using the same number as when
3572 you defined the label. To refer to the next definition of a local label, write
3573 @samp{@b{N}f}---the @samp{b} stands for ``backwards'' and the @samp{f} stands
3576 There is no restriction on how you can use these labels, and you can reuse them
3577 too. So that it is possible to repeatedly define the same local label (using
3578 the same number @samp{@b{N}}), although you can only refer to the most recently
3579 defined local label of that number (for a backwards reference) or the next
3580 definition of a specific local label for a forward reference. It is also worth
3581 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3582 implemented in a slightly more efficient manner than the others.
3593 Which is the equivalent of:
3596 label_1: branch label_3
3597 label_2: branch label_1
3598 label_3: branch label_4
3599 label_4: branch label_3
3602 Local label names are only a notational device. They are immediately
3603 transformed into more conventional symbol names before the assembler uses them.
3604 The symbol names are stored in the symbol table, appear in error messages, and
3605 are optionally emitted to the object file. The names are constructed using
3609 @item @emph{local label prefix}
3610 All local symbols begin with the system-specific local label prefix.
3611 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3612 that start with the local label prefix. These labels are
3613 used for symbols you are never intended to see. If you use the
3614 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3615 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3616 you may use them in debugging.
3619 This is the number that was used in the local label definition. So if the
3620 label is written @samp{55:} then the number is @samp{55}.
3623 This unusual character is included so you do not accidentally invent a symbol
3624 of the same name. The character has ASCII value of @samp{\002} (control-B).
3626 @item @emph{ordinal number}
3627 This is a serial number to keep the labels distinct. The first definition of
3628 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3629 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3630 the number @samp{1} and its 15th definition gets @samp{15} as well.
3633 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3634 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3636 @subheading Dollar Local Labels
3637 @cindex dollar local symbols
3639 @code{@value{AS}} also supports an even more local form of local labels called
3640 dollar labels. These labels go out of scope (i.e., they become undefined) as
3641 soon as a non-local label is defined. Thus they remain valid for only a small
3642 region of the input source code. Normal local labels, by contrast, remain in
3643 scope for the entire file, or until they are redefined by another occurrence of
3644 the same local label.
3646 Dollar labels are defined in exactly the same way as ordinary local labels,
3647 except that they have a dollar sign suffix to their numeric value, e.g.,
3650 They can also be distinguished from ordinary local labels by their transformed
3651 names which use ASCII character @samp{\001} (control-A) as the magic character
3652 to distinguish them from ordinary labels. For example, the fifth definition of
3653 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3656 @section The Special Dot Symbol
3658 @cindex dot (symbol)
3659 @cindex @code{.} (symbol)
3660 @cindex current address
3661 @cindex location counter
3662 The special symbol @samp{.} refers to the current address that
3663 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3664 .long .} defines @code{melvin} to contain its own address.
3665 Assigning a value to @code{.} is treated the same as a @code{.org}
3667 @ifclear no-space-dir
3668 Thus, the expression @samp{.=.+4} is the same as saying
3672 @node Symbol Attributes
3673 @section Symbol Attributes
3675 @cindex symbol attributes
3676 @cindex attributes, symbol
3677 Every symbol has, as well as its name, the attributes ``Value'' and
3678 ``Type''. Depending on output format, symbols can also have auxiliary
3681 The detailed definitions are in @file{a.out.h}.
3684 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3685 all these attributes, and probably won't warn you. This makes the
3686 symbol an externally defined symbol, which is generally what you
3690 * Symbol Value:: Value
3691 * Symbol Type:: Type
3694 * a.out Symbols:: Symbol Attributes: @code{a.out}
3698 * a.out Symbols:: Symbol Attributes: @code{a.out}
3701 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3706 * COFF Symbols:: Symbol Attributes for COFF
3709 * SOM Symbols:: Symbol Attributes for SOM
3716 @cindex value of a symbol
3717 @cindex symbol value
3718 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3719 location in the text, data, bss or absolute sections the value is the
3720 number of addresses from the start of that section to the label.
3721 Naturally for text, data and bss sections the value of a symbol changes
3722 as @code{@value{LD}} changes section base addresses during linking. Absolute
3723 symbols' values do not change during linking: that is why they are
3726 The value of an undefined symbol is treated in a special way. If it is
3727 0 then the symbol is not defined in this assembler source file, and
3728 @code{@value{LD}} tries to determine its value from other files linked into the
3729 same program. You make this kind of symbol simply by mentioning a symbol
3730 name without defining it. A non-zero value represents a @code{.comm}
3731 common declaration. The value is how much common storage to reserve, in
3732 bytes (addresses). The symbol refers to the first address of the
3738 @cindex type of a symbol
3740 The type attribute of a symbol contains relocation (section)
3741 information, any flag settings indicating that a symbol is external, and
3742 (optionally), other information for linkers and debuggers. The exact
3743 format depends on the object-code output format in use.
3748 @c The following avoids a "widow" subsection title. @group would be
3749 @c better if it were available outside examples.
3752 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3754 @cindex @code{b.out} symbol attributes
3755 @cindex symbol attributes, @code{b.out}
3756 These symbol attributes appear only when @command{@value{AS}} is configured for
3757 one of the Berkeley-descended object output formats---@code{a.out} or
3763 @subsection Symbol Attributes: @code{a.out}
3765 @cindex @code{a.out} symbol attributes
3766 @cindex symbol attributes, @code{a.out}
3772 @subsection Symbol Attributes: @code{a.out}
3774 @cindex @code{a.out} symbol attributes
3775 @cindex symbol attributes, @code{a.out}
3779 * Symbol Desc:: Descriptor
3780 * Symbol Other:: Other
3784 @subsubsection Descriptor
3786 @cindex descriptor, of @code{a.out} symbol
3787 This is an arbitrary 16-bit value. You may establish a symbol's
3788 descriptor value by using a @code{.desc} statement
3789 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3790 @command{@value{AS}}.
3793 @subsubsection Other
3795 @cindex other attribute, of @code{a.out} symbol
3796 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3801 @subsection Symbol Attributes for COFF
3803 @cindex COFF symbol attributes
3804 @cindex symbol attributes, COFF
3806 The COFF format supports a multitude of auxiliary symbol attributes;
3807 like the primary symbol attributes, they are set between @code{.def} and
3808 @code{.endef} directives.
3810 @subsubsection Primary Attributes
3812 @cindex primary attributes, COFF symbols
3813 The symbol name is set with @code{.def}; the value and type,
3814 respectively, with @code{.val} and @code{.type}.
3816 @subsubsection Auxiliary Attributes
3818 @cindex auxiliary attributes, COFF symbols
3819 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3820 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3821 table information for COFF.
3826 @subsection Symbol Attributes for SOM
3828 @cindex SOM symbol attributes
3829 @cindex symbol attributes, SOM
3831 The SOM format for the HPPA supports a multitude of symbol attributes set with
3832 the @code{.EXPORT} and @code{.IMPORT} directives.
3834 The attributes are described in @cite{HP9000 Series 800 Assembly
3835 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3836 @code{EXPORT} assembler directive documentation.
3840 @chapter Expressions
3844 @cindex numeric values
3845 An @dfn{expression} specifies an address or numeric value.
3846 Whitespace may precede and/or follow an expression.
3848 The result of an expression must be an absolute number, or else an offset into
3849 a particular section. If an expression is not absolute, and there is not
3850 enough information when @command{@value{AS}} sees the expression to know its
3851 section, a second pass over the source program might be necessary to interpret
3852 the expression---but the second pass is currently not implemented.
3853 @command{@value{AS}} aborts with an error message in this situation.
3856 * Empty Exprs:: Empty Expressions
3857 * Integer Exprs:: Integer Expressions
3861 @section Empty Expressions
3863 @cindex empty expressions
3864 @cindex expressions, empty
3865 An empty expression has no value: it is just whitespace or null.
3866 Wherever an absolute expression is required, you may omit the
3867 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3868 is compatible with other assemblers.
3871 @section Integer Expressions
3873 @cindex integer expressions
3874 @cindex expressions, integer
3875 An @dfn{integer expression} is one or more @emph{arguments} delimited
3876 by @emph{operators}.
3879 * Arguments:: Arguments
3880 * Operators:: Operators
3881 * Prefix Ops:: Prefix Operators
3882 * Infix Ops:: Infix Operators
3886 @subsection Arguments
3888 @cindex expression arguments
3889 @cindex arguments in expressions
3890 @cindex operands in expressions
3891 @cindex arithmetic operands
3892 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3893 contexts arguments are sometimes called ``arithmetic operands''. In
3894 this manual, to avoid confusing them with the ``instruction operands'' of
3895 the machine language, we use the term ``argument'' to refer to parts of
3896 expressions only, reserving the word ``operand'' to refer only to machine
3897 instruction operands.
3899 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3900 @var{section} is one of text, data, bss, absolute,
3901 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3904 Numbers are usually integers.
3906 A number can be a flonum or bignum. In this case, you are warned
3907 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3908 these 32 bits are an integer. You may write integer-manipulating
3909 instructions that act on exotic constants, compatible with other
3912 @cindex subexpressions
3913 Subexpressions are a left parenthesis @samp{(} followed by an integer
3914 expression, followed by a right parenthesis @samp{)}; or a prefix
3915 operator followed by an argument.
3918 @subsection Operators
3920 @cindex operators, in expressions
3921 @cindex arithmetic functions
3922 @cindex functions, in expressions
3923 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3924 operators are followed by an argument. Infix operators appear
3925 between their arguments. Operators may be preceded and/or followed by
3929 @subsection Prefix Operator
3931 @cindex prefix operators
3932 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3933 one argument, which must be absolute.
3935 @c the tex/end tex stuff surrounding this small table is meant to make
3936 @c it align, on the printed page, with the similar table in the next
3937 @c section (which is inside an enumerate).
3939 \global\advance\leftskip by \itemindent
3944 @dfn{Negation}. Two's complement negation.
3946 @dfn{Complementation}. Bitwise not.
3950 \global\advance\leftskip by -\itemindent
3954 @subsection Infix Operators
3956 @cindex infix operators
3957 @cindex operators, permitted arguments
3958 @dfn{Infix operators} take two arguments, one on either side. Operators
3959 have precedence, but operations with equal precedence are performed left
3960 to right. Apart from @code{+} or @option{-}, both arguments must be
3961 absolute, and the result is absolute.
3964 @cindex operator precedence
3965 @cindex precedence of operators
3972 @dfn{Multiplication}.
3975 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3981 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3984 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3988 Intermediate precedence
3993 @dfn{Bitwise Inclusive Or}.
3999 @dfn{Bitwise Exclusive Or}.
4002 @dfn{Bitwise Or Not}.
4009 @cindex addition, permitted arguments
4010 @cindex plus, permitted arguments
4011 @cindex arguments for addition
4013 @dfn{Addition}. If either argument is absolute, the result has the section of
4014 the other argument. You may not add together arguments from different
4017 @cindex subtraction, permitted arguments
4018 @cindex minus, permitted arguments
4019 @cindex arguments for subtraction
4021 @dfn{Subtraction}. If the right argument is absolute, the
4022 result has the section of the left argument.
4023 If both arguments are in the same section, the result is absolute.
4024 You may not subtract arguments from different sections.
4025 @c FIXME is there still something useful to say about undefined - undefined ?
4027 @cindex comparison expressions
4028 @cindex expressions, comparison
4033 @dfn{Is Not Equal To}
4037 @dfn{Is Greater Than}
4039 @dfn{Is Greater Than Or Equal To}
4041 @dfn{Is Less Than Or Equal To}
4043 The comparison operators can be used as infix operators. A true results has a
4044 value of -1 whereas a false result has a value of 0. Note, these operators
4045 perform signed comparisons.
4048 @item Lowest Precedence
4057 These two logical operations can be used to combine the results of sub
4058 expressions. Note, unlike the comparison operators a true result returns a
4059 value of 1 but a false results does still return 0. Also note that the logical
4060 or operator has a slightly lower precedence than logical and.
4065 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4066 address; you can only have a defined section in one of the two arguments.
4069 @chapter Assembler Directives
4071 @cindex directives, machine independent
4072 @cindex pseudo-ops, machine independent
4073 @cindex machine independent directives
4074 All assembler directives have names that begin with a period (@samp{.}).
4075 The rest of the name is letters, usually in lower case.
4077 This chapter discusses directives that are available regardless of the
4078 target machine configuration for the @sc{gnu} assembler.
4080 Some machine configurations provide additional directives.
4081 @xref{Machine Dependencies}.
4084 @ifset machine-directives
4085 @xref{Machine Dependencies}, for additional directives.
4090 * Abort:: @code{.abort}
4092 * ABORT (COFF):: @code{.ABORT}
4095 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4096 * Altmacro:: @code{.altmacro}
4097 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4098 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4099 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4100 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, @code{.bundle_lock}, @code{.bundle_unlock}
4101 * Byte:: @code{.byte @var{expressions}}
4102 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4103 * Comm:: @code{.comm @var{symbol} , @var{length} }
4104 * Data:: @code{.data @var{subsection}}
4106 * Def:: @code{.def @var{name}}
4109 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4115 * Double:: @code{.double @var{flonums}}
4116 * Eject:: @code{.eject}
4117 * Else:: @code{.else}
4118 * Elseif:: @code{.elseif}
4121 * Endef:: @code{.endef}
4124 * Endfunc:: @code{.endfunc}
4125 * Endif:: @code{.endif}
4126 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4127 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4128 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4130 * Error:: @code{.error @var{string}}
4131 * Exitm:: @code{.exitm}
4132 * Extern:: @code{.extern}
4133 * Fail:: @code{.fail}
4134 * File:: @code{.file}
4135 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4136 * Float:: @code{.float @var{flonums}}
4137 * Func:: @code{.func}
4138 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4140 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4141 * Hidden:: @code{.hidden @var{names}}
4144 * hword:: @code{.hword @var{expressions}}
4145 * Ident:: @code{.ident}
4146 * If:: @code{.if @var{absolute expression}}
4147 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4148 * Include:: @code{.include "@var{file}"}
4149 * Int:: @code{.int @var{expressions}}
4151 * Internal:: @code{.internal @var{names}}
4154 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4155 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4156 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4157 * Lflags:: @code{.lflags}
4158 @ifclear no-line-dir
4159 * Line:: @code{.line @var{line-number}}
4162 * Linkonce:: @code{.linkonce [@var{type}]}
4163 * List:: @code{.list}
4164 * Ln:: @code{.ln @var{line-number}}
4165 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4166 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4168 * Local:: @code{.local @var{names}}
4171 * Long:: @code{.long @var{expressions}}
4173 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4176 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4177 * MRI:: @code{.mri @var{val}}
4178 * Noaltmacro:: @code{.noaltmacro}
4179 * Nolist:: @code{.nolist}
4180 * Octa:: @code{.octa @var{bignums}}
4181 * Offset:: @code{.offset @var{loc}}
4182 * Org:: @code{.org @var{new-lc}, @var{fill}}
4183 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4185 * PopSection:: @code{.popsection}
4186 * Previous:: @code{.previous}
4189 * Print:: @code{.print @var{string}}
4191 * Protected:: @code{.protected @var{names}}
4194 * Psize:: @code{.psize @var{lines}, @var{columns}}
4195 * Purgem:: @code{.purgem @var{name}}
4197 * PushSection:: @code{.pushsection @var{name}}
4200 * Quad:: @code{.quad @var{bignums}}
4201 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4202 * Rept:: @code{.rept @var{count}}
4203 * Sbttl:: @code{.sbttl "@var{subheading}"}
4205 * Scl:: @code{.scl @var{class}}
4208 * Section:: @code{.section @var{name}[, @var{flags}]}
4211 * Set:: @code{.set @var{symbol}, @var{expression}}
4212 * Short:: @code{.short @var{expressions}}
4213 * Single:: @code{.single @var{flonums}}
4215 * Size:: @code{.size [@var{name} , @var{expression}]}
4217 @ifclear no-space-dir
4218 * Skip:: @code{.skip @var{size} , @var{fill}}
4221 * Sleb128:: @code{.sleb128 @var{expressions}}
4222 @ifclear no-space-dir
4223 * Space:: @code{.space @var{size} , @var{fill}}
4226 * Stab:: @code{.stabd, .stabn, .stabs}
4229 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4230 * Struct:: @code{.struct @var{expression}}
4232 * SubSection:: @code{.subsection}
4233 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4237 * Tag:: @code{.tag @var{structname}}
4240 * Text:: @code{.text @var{subsection}}
4241 * Title:: @code{.title "@var{heading}"}
4243 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4246 * Uleb128:: @code{.uleb128 @var{expressions}}
4248 * Val:: @code{.val @var{addr}}
4252 * Version:: @code{.version "@var{string}"}
4253 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4254 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4257 * Warning:: @code{.warning @var{string}}
4258 * Weak:: @code{.weak @var{names}}
4259 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4260 * Word:: @code{.word @var{expressions}}
4261 * Deprecated:: Deprecated Directives
4265 @section @code{.abort}
4267 @cindex @code{abort} directive
4268 @cindex stopping the assembly
4269 This directive stops the assembly immediately. It is for
4270 compatibility with other assemblers. The original idea was that the
4271 assembly language source would be piped into the assembler. If the sender
4272 of the source quit, it could use this directive tells @command{@value{AS}} to
4273 quit also. One day @code{.abort} will not be supported.
4277 @section @code{.ABORT} (COFF)
4279 @cindex @code{ABORT} directive
4280 When producing COFF output, @command{@value{AS}} accepts this directive as a
4281 synonym for @samp{.abort}.
4284 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4290 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4292 @cindex padding the location counter
4293 @cindex @code{align} directive
4294 Pad the location counter (in the current subsection) to a particular storage
4295 boundary. The first expression (which must be absolute) is the alignment
4296 required, as described below.
4298 The second expression (also absolute) gives the fill value to be stored in the
4299 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4300 padding bytes are normally zero. However, on some systems, if the section is
4301 marked as containing code and the fill value is omitted, the space is filled
4302 with no-op instructions.
4304 The third expression is also absolute, and is also optional. If it is present,
4305 it is the maximum number of bytes that should be skipped by this alignment
4306 directive. If doing the alignment would require skipping more bytes than the
4307 specified maximum, then the alignment is not done at all. You can omit the
4308 fill value (the second argument) entirely by simply using two commas after the
4309 required alignment; this can be useful if you want the alignment to be filled
4310 with no-op instructions when appropriate.
4312 The way the required alignment is specified varies from system to system.
4313 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or32,
4314 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4315 alignment request in bytes. For example @samp{.align 8} advances
4316 the location counter until it is a multiple of 8. If the location counter
4317 is already a multiple of 8, no change is needed. For the tic54x, the
4318 first expression is the alignment request in words.
4320 For other systems, including ppc, i386 using a.out format, arm and
4321 strongarm, it is the
4322 number of low-order zero bits the location counter must have after
4323 advancement. For example @samp{.align 3} advances the location
4324 counter until it a multiple of 8. If the location counter is already a
4325 multiple of 8, no change is needed.
4327 This inconsistency is due to the different behaviors of the various
4328 native assemblers for these systems which GAS must emulate.
4329 GAS also provides @code{.balign} and @code{.p2align} directives,
4330 described later, which have a consistent behavior across all
4331 architectures (but are specific to GAS).
4334 @section @code{.altmacro}
4335 Enable alternate macro mode, enabling:
4338 @item LOCAL @var{name} [ , @dots{} ]
4339 One additional directive, @code{LOCAL}, is available. It is used to
4340 generate a string replacement for each of the @var{name} arguments, and
4341 replace any instances of @var{name} in each macro expansion. The
4342 replacement string is unique in the assembly, and different for each
4343 separate macro expansion. @code{LOCAL} allows you to write macros that
4344 define symbols, without fear of conflict between separate macro expansions.
4346 @item String delimiters
4347 You can write strings delimited in these other ways besides
4348 @code{"@var{string}"}:
4351 @item '@var{string}'
4352 You can delimit strings with single-quote characters.
4354 @item <@var{string}>
4355 You can delimit strings with matching angle brackets.
4358 @item single-character string escape
4359 To include any single character literally in a string (even if the
4360 character would otherwise have some special meaning), you can prefix the
4361 character with @samp{!} (an exclamation mark). For example, you can
4362 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4364 @item Expression results as strings
4365 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4366 and use the result as a string.
4370 @section @code{.ascii "@var{string}"}@dots{}
4372 @cindex @code{ascii} directive
4373 @cindex string literals
4374 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4375 separated by commas. It assembles each string (with no automatic
4376 trailing zero byte) into consecutive addresses.
4379 @section @code{.asciz "@var{string}"}@dots{}
4381 @cindex @code{asciz} directive
4382 @cindex zero-terminated strings
4383 @cindex null-terminated strings
4384 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4385 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4388 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4390 @cindex padding the location counter given number of bytes
4391 @cindex @code{balign} directive
4392 Pad the location counter (in the current subsection) to a particular
4393 storage boundary. The first expression (which must be absolute) is the
4394 alignment request in bytes. For example @samp{.balign 8} advances
4395 the location counter until it is a multiple of 8. If the location counter
4396 is already a multiple of 8, no change is needed.
4398 The second expression (also absolute) gives the fill value to be stored in the
4399 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4400 padding bytes are normally zero. However, on some systems, if the section is
4401 marked as containing code and the fill value is omitted, the space is filled
4402 with no-op instructions.
4404 The third expression is also absolute, and is also optional. If it is present,
4405 it is the maximum number of bytes that should be skipped by this alignment
4406 directive. If doing the alignment would require skipping more bytes than the
4407 specified maximum, then the alignment is not done at all. You can omit the
4408 fill value (the second argument) entirely by simply using two commas after the
4409 required alignment; this can be useful if you want the alignment to be filled
4410 with no-op instructions when appropriate.
4412 @cindex @code{balignw} directive
4413 @cindex @code{balignl} directive
4414 The @code{.balignw} and @code{.balignl} directives are variants of the
4415 @code{.balign} directive. The @code{.balignw} directive treats the fill
4416 pattern as a two byte word value. The @code{.balignl} directives treats the
4417 fill pattern as a four byte longword value. For example, @code{.balignw
4418 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4419 filled in with the value 0x368d (the exact placement of the bytes depends upon
4420 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4423 @node Bundle directives
4424 @section @code{.bundle_align_mode @var{abs-expr}}
4425 @cindex @code{bundle_align_mode} directive
4427 @cindex instruction bundle
4428 @cindex aligned instruction bundle
4429 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4430 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4431 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4432 disabled (which is the default state). If the argument it not zero, it
4433 gives the size of an instruction bundle as a power of two (as for the
4434 @code{.p2align} directive, @pxref{P2align}).
4436 For some targets, it's an ABI requirement that no instruction may span a
4437 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4438 instructions that starts on an aligned boundary. For example, if
4439 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4440 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4441 effect, no single instruction may span a boundary between bundles. If an
4442 instruction would start too close to the end of a bundle for the length of
4443 that particular instruction to fit within the bundle, then the space at the
4444 end of that bundle is filled with no-op instructions so the instruction
4445 starts in the next bundle. As a corollary, it's an error if any single
4446 instruction's encoding is longer than the bundle size.
4448 @section @code{.bundle_lock} and @code{.bundle_unlock}
4449 @cindex @code{bundle_lock} directive
4450 @cindex @code{bundle_unlock} directive
4451 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4452 allow explicit control over instruction bundle padding. These directives
4453 are only valid when @code{.bundle_align_mode} has been used to enable
4454 aligned instruction bundle mode. It's an error if they appear when
4455 @code{.bundle_align_mode} has not been used at all, or when the last
4456 directive was @w{@code{.bundle_align_mode 0}}.
4458 @cindex bundle-locked
4459 For some targets, it's an ABI requirement that certain instructions may
4460 appear only as part of specified permissible sequences of multiple
4461 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4462 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4463 instruction sequence. For purposes of aligned instruction bundle mode, a
4464 sequence starting with @code{.bundle_lock} and ending with
4465 @code{.bundle_unlock} is treated as a single instruction. That is, the
4466 entire sequence must fit into a single bundle and may not span a bundle
4467 boundary. If necessary, no-op instructions will be inserted before the
4468 first instruction of the sequence so that the whole sequence starts on an
4469 aligned bundle boundary. It's an error if the sequence is longer than the
4472 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4473 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4474 nested. That is, a second @code{.bundle_lock} directive before the next
4475 @code{.bundle_unlock} directive has no effect except that it must be
4476 matched by another closing @code{.bundle_unlock} so that there is the
4477 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4480 @section @code{.byte @var{expressions}}
4482 @cindex @code{byte} directive
4483 @cindex integers, one byte
4484 @code{.byte} expects zero or more expressions, separated by commas.
4485 Each expression is assembled into the next byte.
4487 @node CFI directives
4488 @section @code{.cfi_sections @var{section_list}}
4489 @cindex @code{cfi_sections} directive
4490 @code{.cfi_sections} may be used to specify whether CFI directives
4491 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4492 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4493 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4494 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4495 directive is not used is @code{.cfi_sections .eh_frame}.
4497 @section @code{.cfi_startproc [simple]}
4498 @cindex @code{cfi_startproc} directive
4499 @code{.cfi_startproc} is used at the beginning of each function that
4500 should have an entry in @code{.eh_frame}. It initializes some internal
4501 data structures. Don't forget to close the function by
4502 @code{.cfi_endproc}.
4504 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4505 it also emits some architecture dependent initial CFI instructions.
4507 @section @code{.cfi_endproc}
4508 @cindex @code{cfi_endproc} directive
4509 @code{.cfi_endproc} is used at the end of a function where it closes its
4510 unwind entry previously opened by
4511 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4513 @section @code{.cfi_personality @var{encoding} [, @var{exp}]}
4514 @code{.cfi_personality} defines personality routine and its encoding.
4515 @var{encoding} must be a constant determining how the personality
4516 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4517 argument is not present, otherwise second argument should be
4518 a constant or a symbol name. When using indirect encodings,
4519 the symbol provided should be the location where personality
4520 can be loaded from, not the personality routine itself.
4521 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4522 no personality routine.
4524 @section @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4525 @code{.cfi_lsda} defines LSDA and its encoding.
4526 @var{encoding} must be a constant determining how the LSDA
4527 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4528 argument is not present, otherwise second argument should be a constant
4529 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4532 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4533 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4534 address from @var{register} and add @var{offset} to it}.
4536 @section @code{.cfi_def_cfa_register @var{register}}
4537 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4538 now on @var{register} will be used instead of the old one. Offset
4541 @section @code{.cfi_def_cfa_offset @var{offset}}
4542 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4543 remains the same, but @var{offset} is new. Note that it is the
4544 absolute offset that will be added to a defined register to compute
4547 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4548 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4549 value that is added/substracted from the previous offset.
4551 @section @code{.cfi_offset @var{register}, @var{offset}}
4552 Previous value of @var{register} is saved at offset @var{offset} from
4555 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4556 Previous value of @var{register} is saved at offset @var{offset} from
4557 the current CFA register. This is transformed to @code{.cfi_offset}
4558 using the known displacement of the CFA register from the CFA.
4559 This is often easier to use, because the number will match the
4560 code it's annotating.
4562 @section @code{.cfi_register @var{register1}, @var{register2}}
4563 Previous value of @var{register1} is saved in register @var{register2}.
4565 @section @code{.cfi_restore @var{register}}
4566 @code{.cfi_restore} says that the rule for @var{register} is now the
4567 same as it was at the beginning of the function, after all initial
4568 instruction added by @code{.cfi_startproc} were executed.
4570 @section @code{.cfi_undefined @var{register}}
4571 From now on the previous value of @var{register} can't be restored anymore.
4573 @section @code{.cfi_same_value @var{register}}
4574 Current value of @var{register} is the same like in the previous frame,
4575 i.e. no restoration needed.
4577 @section @code{.cfi_remember_state},
4578 First save all current rules for all registers by @code{.cfi_remember_state},
4579 then totally screw them up by subsequent @code{.cfi_*} directives and when
4580 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4581 the previous saved state.
4583 @section @code{.cfi_return_column @var{register}}
4584 Change return column @var{register}, i.e. the return address is either
4585 directly in @var{register} or can be accessed by rules for @var{register}.
4587 @section @code{.cfi_signal_frame}
4588 Mark current function as signal trampoline.
4590 @section @code{.cfi_window_save}
4591 SPARC register window has been saved.
4593 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4594 Allows the user to add arbitrary bytes to the unwind info. One
4595 might use this to add OS-specific CFI opcodes, or generic CFI
4596 opcodes that GAS does not yet support.
4598 @section @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4599 The current value of @var{register} is @var{label}. The value of @var{label}
4600 will be encoded in the output file according to @var{encoding}; see the
4601 description of @code{.cfi_personality} for details on this encoding.
4603 The usefulness of equating a register to a fixed label is probably
4604 limited to the return address register. Here, it can be useful to
4605 mark a code segment that has only one return address which is reached
4606 by a direct branch and no copy of the return address exists in memory
4607 or another register.
4610 @section @code{.comm @var{symbol} , @var{length} }
4612 @cindex @code{comm} directive
4613 @cindex symbol, common
4614 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4615 common symbol in one object file may be merged with a defined or common symbol
4616 of the same name in another object file. If @code{@value{LD}} does not see a
4617 definition for the symbol--just one or more common symbols--then it will
4618 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4619 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4620 the same name, and they do not all have the same size, it will allocate space
4621 using the largest size.
4624 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4625 an optional third argument. This is the desired alignment of the symbol,
4626 specified for ELF as a byte boundary (for example, an alignment of 16 means
4627 that the least significant 4 bits of the address should be zero), and for PE
4628 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4629 boundary). The alignment must be an absolute expression, and it must be a
4630 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4631 common symbol, it will use the alignment when placing the symbol. If no
4632 alignment is specified, @command{@value{AS}} will set the alignment to the
4633 largest power of two less than or equal to the size of the symbol, up to a
4634 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4635 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4636 @samp{--section-alignment} option; image file sections in PE are aligned to
4637 multiples of 4096, which is far too large an alignment for ordinary variables.
4638 It is rather the default alignment for (non-debug) sections within object
4639 (@samp{*.o}) files, which are less strictly aligned.}.
4643 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4644 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4648 @section @code{.data @var{subsection}}
4650 @cindex @code{data} directive
4651 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4652 end of the data subsection numbered @var{subsection} (which is an
4653 absolute expression). If @var{subsection} is omitted, it defaults
4658 @section @code{.def @var{name}}
4660 @cindex @code{def} directive
4661 @cindex COFF symbols, debugging
4662 @cindex debugging COFF symbols
4663 Begin defining debugging information for a symbol @var{name}; the
4664 definition extends until the @code{.endef} directive is encountered.
4667 This directive is only observed when @command{@value{AS}} is configured for COFF
4668 format output; when producing @code{b.out}, @samp{.def} is recognized,
4675 @section @code{.desc @var{symbol}, @var{abs-expression}}
4677 @cindex @code{desc} directive
4678 @cindex COFF symbol descriptor
4679 @cindex symbol descriptor, COFF
4680 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4681 to the low 16 bits of an absolute expression.
4684 The @samp{.desc} directive is not available when @command{@value{AS}} is
4685 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4686 object format. For the sake of compatibility, @command{@value{AS}} accepts
4687 it, but produces no output, when configured for COFF.
4693 @section @code{.dim}
4695 @cindex @code{dim} directive
4696 @cindex COFF auxiliary symbol information
4697 @cindex auxiliary symbol information, COFF
4698 This directive is generated by compilers to include auxiliary debugging
4699 information in the symbol table. It is only permitted inside
4700 @code{.def}/@code{.endef} pairs.
4703 @samp{.dim} is only meaningful when generating COFF format output; when
4704 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4710 @section @code{.double @var{flonums}}
4712 @cindex @code{double} directive
4713 @cindex floating point numbers (double)
4714 @code{.double} expects zero or more flonums, separated by commas. It
4715 assembles floating point numbers.
4717 The exact kind of floating point numbers emitted depends on how
4718 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4722 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4723 in @sc{ieee} format.
4728 @section @code{.eject}
4730 @cindex @code{eject} directive
4731 @cindex new page, in listings
4732 @cindex page, in listings
4733 @cindex listing control: new page
4734 Force a page break at this point, when generating assembly listings.
4737 @section @code{.else}
4739 @cindex @code{else} directive
4740 @code{.else} is part of the @command{@value{AS}} support for conditional
4741 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4742 of code to be assembled if the condition for the preceding @code{.if}
4746 @section @code{.elseif}
4748 @cindex @code{elseif} directive
4749 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4750 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4751 @code{.if} block that would otherwise fill the entire @code{.else} section.
4754 @section @code{.end}
4756 @cindex @code{end} directive
4757 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4758 process anything in the file past the @code{.end} directive.
4762 @section @code{.endef}
4764 @cindex @code{endef} directive
4765 This directive flags the end of a symbol definition begun with
4769 @samp{.endef} is only meaningful when generating COFF format output; if
4770 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4771 directive but ignores it.
4776 @section @code{.endfunc}
4777 @cindex @code{endfunc} directive
4778 @code{.endfunc} marks the end of a function specified with @code{.func}.
4781 @section @code{.endif}
4783 @cindex @code{endif} directive
4784 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4785 it marks the end of a block of code that is only assembled
4786 conditionally. @xref{If,,@code{.if}}.
4789 @section @code{.equ @var{symbol}, @var{expression}}
4791 @cindex @code{equ} directive
4792 @cindex assigning values to symbols
4793 @cindex symbols, assigning values to
4794 This directive sets the value of @var{symbol} to @var{expression}.
4795 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4798 The syntax for @code{equ} on the HPPA is
4799 @samp{@var{symbol} .equ @var{expression}}.
4803 The syntax for @code{equ} on the Z80 is
4804 @samp{@var{symbol} equ @var{expression}}.
4805 On the Z80 it is an eror if @var{symbol} is already defined,
4806 but the symbol is not protected from later redefinition.
4807 Compare @ref{Equiv}.
4811 @section @code{.equiv @var{symbol}, @var{expression}}
4812 @cindex @code{equiv} directive
4813 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4814 the assembler will signal an error if @var{symbol} is already defined. Note a
4815 symbol which has been referenced but not actually defined is considered to be
4818 Except for the contents of the error message, this is roughly equivalent to
4825 plus it protects the symbol from later redefinition.
4828 @section @code{.eqv @var{symbol}, @var{expression}}
4829 @cindex @code{eqv} directive
4830 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
4831 evaluate the expression or any part of it immediately. Instead each time
4832 the resulting symbol is used in an expression, a snapshot of its current
4836 @section @code{.err}
4837 @cindex @code{err} directive
4838 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4839 message and, unless the @option{-Z} option was used, it will not generate an
4840 object file. This can be used to signal an error in conditionally compiled code.
4843 @section @code{.error "@var{string}"}
4844 @cindex error directive
4846 Similarly to @code{.err}, this directive emits an error, but you can specify a
4847 string that will be emitted as the error message. If you don't specify the
4848 message, it defaults to @code{".error directive invoked in source file"}.
4849 @xref{Errors, ,Error and Warning Messages}.
4852 .error "This code has not been assembled and tested."
4856 @section @code{.exitm}
4857 Exit early from the current macro definition. @xref{Macro}.
4860 @section @code{.extern}
4862 @cindex @code{extern} directive
4863 @code{.extern} is accepted in the source program---for compatibility
4864 with other assemblers---but it is ignored. @command{@value{AS}} treats
4865 all undefined symbols as external.
4868 @section @code{.fail @var{expression}}
4870 @cindex @code{fail} directive
4871 Generates an error or a warning. If the value of the @var{expression} is 500
4872 or more, @command{@value{AS}} will print a warning message. If the value is less
4873 than 500, @command{@value{AS}} will print an error message. The message will
4874 include the value of @var{expression}. This can occasionally be useful inside
4875 complex nested macros or conditional assembly.
4878 @section @code{.file}
4879 @cindex @code{file} directive
4881 @ifclear no-file-dir
4882 There are two different versions of the @code{.file} directive. Targets
4883 that support DWARF2 line number information use the DWARF2 version of
4884 @code{.file}. Other targets use the default version.
4886 @subheading Default Version
4888 @cindex logical file name
4889 @cindex file name, logical
4890 This version of the @code{.file} directive tells @command{@value{AS}} that we
4891 are about to start a new logical file. The syntax is:
4897 @var{string} is the new file name. In general, the filename is
4898 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4899 to specify an empty file name, you must give the quotes--@code{""}. This
4900 statement may go away in future: it is only recognized to be compatible with
4901 old @command{@value{AS}} programs.
4903 @subheading DWARF2 Version
4906 When emitting DWARF2 line number information, @code{.file} assigns filenames
4907 to the @code{.debug_line} file name table. The syntax is:
4910 .file @var{fileno} @var{filename}
4913 The @var{fileno} operand should be a unique positive integer to use as the
4914 index of the entry in the table. The @var{filename} operand is a C string
4917 The detail of filename indices is exposed to the user because the filename
4918 table is shared with the @code{.debug_info} section of the DWARF2 debugging
4919 information, and thus the user must know the exact indices that table
4923 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4925 @cindex @code{fill} directive
4926 @cindex writing patterns in memory
4927 @cindex patterns, writing in memory
4928 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4929 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4930 may be zero or more. @var{Size} may be zero or more, but if it is
4931 more than 8, then it is deemed to have the value 8, compatible with
4932 other people's assemblers. The contents of each @var{repeat} bytes
4933 is taken from an 8-byte number. The highest order 4 bytes are
4934 zero. The lowest order 4 bytes are @var{value} rendered in the
4935 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4936 Each @var{size} bytes in a repetition is taken from the lowest order
4937 @var{size} bytes of this number. Again, this bizarre behavior is
4938 compatible with other people's assemblers.
4940 @var{size} and @var{value} are optional.
4941 If the second comma and @var{value} are absent, @var{value} is
4942 assumed zero. If the first comma and following tokens are absent,
4943 @var{size} is assumed to be 1.
4946 @section @code{.float @var{flonums}}
4948 @cindex floating point numbers (single)
4949 @cindex @code{float} directive
4950 This directive assembles zero or more flonums, separated by commas. It
4951 has the same effect as @code{.single}.
4953 The exact kind of floating point numbers emitted depends on how
4954 @command{@value{AS}} is configured.
4955 @xref{Machine Dependencies}.
4959 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4960 in @sc{ieee} format.
4965 @section @code{.func @var{name}[,@var{label}]}
4966 @cindex @code{func} directive
4967 @code{.func} emits debugging information to denote function @var{name}, and
4968 is ignored unless the file is assembled with debugging enabled.
4969 Only @samp{--gstabs[+]} is currently supported.
4970 @var{label} is the entry point of the function and if omitted @var{name}
4971 prepended with the @samp{leading char} is used.
4972 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4973 All functions are currently defined to have @code{void} return type.
4974 The function must be terminated with @code{.endfunc}.
4977 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4979 @cindex @code{global} directive
4980 @cindex symbol, making visible to linker
4981 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4982 @var{symbol} in your partial program, its value is made available to
4983 other partial programs that are linked with it. Otherwise,
4984 @var{symbol} takes its attributes from a symbol of the same name
4985 from another file linked into the same program.
4987 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4988 compatibility with other assemblers.
4991 On the HPPA, @code{.global} is not always enough to make it accessible to other
4992 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4993 @xref{HPPA Directives, ,HPPA Assembler Directives}.
4998 @section @code{.gnu_attribute @var{tag},@var{value}}
4999 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5002 @section @code{.hidden @var{names}}
5004 @cindex @code{hidden} directive
5006 This is one of the ELF visibility directives. The other two are
5007 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5008 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5010 This directive overrides the named symbols default visibility (which is set by
5011 their binding: local, global or weak). The directive sets the visibility to
5012 @code{hidden} which means that the symbols are not visible to other components.
5013 Such symbols are always considered to be @code{protected} as well.
5017 @section @code{.hword @var{expressions}}
5019 @cindex @code{hword} directive
5020 @cindex integers, 16-bit
5021 @cindex numbers, 16-bit
5022 @cindex sixteen bit integers
5023 This expects zero or more @var{expressions}, and emits
5024 a 16 bit number for each.
5027 This directive is a synonym for @samp{.short}; depending on the target
5028 architecture, it may also be a synonym for @samp{.word}.
5032 This directive is a synonym for @samp{.short}.
5035 This directive is a synonym for both @samp{.short} and @samp{.word}.
5040 @section @code{.ident}
5042 @cindex @code{ident} directive
5044 This directive is used by some assemblers to place tags in object files. The
5045 behavior of this directive varies depending on the target. When using the
5046 a.out object file format, @command{@value{AS}} simply accepts the directive for
5047 source-file compatibility with existing assemblers, but does not emit anything
5048 for it. When using COFF, comments are emitted to the @code{.comment} or
5049 @code{.rdata} section, depending on the target. When using ELF, comments are
5050 emitted to the @code{.comment} section.
5053 @section @code{.if @var{absolute expression}}
5055 @cindex conditional assembly
5056 @cindex @code{if} directive
5057 @code{.if} marks the beginning of a section of code which is only
5058 considered part of the source program being assembled if the argument
5059 (which must be an @var{absolute expression}) is non-zero. The end of
5060 the conditional section of code must be marked by @code{.endif}
5061 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5062 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5063 If you have several conditions to check, @code{.elseif} may be used to avoid
5064 nesting blocks if/else within each subsequent @code{.else} block.
5066 The following variants of @code{.if} are also supported:
5068 @cindex @code{ifdef} directive
5069 @item .ifdef @var{symbol}
5070 Assembles the following section of code if the specified @var{symbol}
5071 has been defined. Note a symbol which has been referenced but not yet defined
5072 is considered to be undefined.
5074 @cindex @code{ifb} directive
5075 @item .ifb @var{text}
5076 Assembles the following section of code if the operand is blank (empty).
5078 @cindex @code{ifc} directive
5079 @item .ifc @var{string1},@var{string2}
5080 Assembles the following section of code if the two strings are the same. The
5081 strings may be optionally quoted with single quotes. If they are not quoted,
5082 the first string stops at the first comma, and the second string stops at the
5083 end of the line. Strings which contain whitespace should be quoted. The
5084 string comparison is case sensitive.
5086 @cindex @code{ifeq} directive
5087 @item .ifeq @var{absolute expression}
5088 Assembles the following section of code if the argument is zero.
5090 @cindex @code{ifeqs} directive
5091 @item .ifeqs @var{string1},@var{string2}
5092 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5094 @cindex @code{ifge} directive
5095 @item .ifge @var{absolute expression}
5096 Assembles the following section of code if the argument is greater than or
5099 @cindex @code{ifgt} directive
5100 @item .ifgt @var{absolute expression}
5101 Assembles the following section of code if the argument is greater than zero.
5103 @cindex @code{ifle} directive
5104 @item .ifle @var{absolute expression}
5105 Assembles the following section of code if the argument is less than or equal
5108 @cindex @code{iflt} directive
5109 @item .iflt @var{absolute expression}
5110 Assembles the following section of code if the argument is less than zero.
5112 @cindex @code{ifnb} directive
5113 @item .ifnb @var{text}
5114 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5115 following section of code if the operand is non-blank (non-empty).
5117 @cindex @code{ifnc} directive
5118 @item .ifnc @var{string1},@var{string2}.
5119 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5120 following section of code if the two strings are not the same.
5122 @cindex @code{ifndef} directive
5123 @cindex @code{ifnotdef} directive
5124 @item .ifndef @var{symbol}
5125 @itemx .ifnotdef @var{symbol}
5126 Assembles the following section of code if the specified @var{symbol}
5127 has not been defined. Both spelling variants are equivalent. Note a symbol
5128 which has been referenced but not yet defined is considered to be undefined.
5130 @cindex @code{ifne} directive
5131 @item .ifne @var{absolute expression}
5132 Assembles the following section of code if the argument is not equal to zero
5133 (in other words, this is equivalent to @code{.if}).
5135 @cindex @code{ifnes} directive
5136 @item .ifnes @var{string1},@var{string2}
5137 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5138 following section of code if the two strings are not the same.
5142 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5144 @cindex @code{incbin} directive
5145 @cindex binary files, including
5146 The @code{incbin} directive includes @var{file} verbatim at the current
5147 location. You can control the search paths used with the @samp{-I} command-line
5148 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5151 The @var{skip} argument skips a number of bytes from the start of the
5152 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5153 read. Note that the data is not aligned in any way, so it is the user's
5154 responsibility to make sure that proper alignment is provided both before and
5155 after the @code{incbin} directive.
5158 @section @code{.include "@var{file}"}
5160 @cindex @code{include} directive
5161 @cindex supporting files, including
5162 @cindex files, including
5163 This directive provides a way to include supporting files at specified
5164 points in your source program. The code from @var{file} is assembled as
5165 if it followed the point of the @code{.include}; when the end of the
5166 included file is reached, assembly of the original file continues. You
5167 can control the search paths used with the @samp{-I} command-line option
5168 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5172 @section @code{.int @var{expressions}}
5174 @cindex @code{int} directive
5175 @cindex integers, 32-bit
5176 Expect zero or more @var{expressions}, of any section, separated by commas.
5177 For each expression, emit a number that, at run time, is the value of that
5178 expression. The byte order and bit size of the number depends on what kind
5179 of target the assembly is for.
5183 On most forms of the H8/300, @code{.int} emits 16-bit
5184 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5191 @section @code{.internal @var{names}}
5193 @cindex @code{internal} directive
5195 This is one of the ELF visibility directives. The other two are
5196 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5197 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5199 This directive overrides the named symbols default visibility (which is set by
5200 their binding: local, global or weak). The directive sets the visibility to
5201 @code{internal} which means that the symbols are considered to be @code{hidden}
5202 (i.e., not visible to other components), and that some extra, processor specific
5203 processing must also be performed upon the symbols as well.
5207 @section @code{.irp @var{symbol},@var{values}}@dots{}
5209 @cindex @code{irp} directive
5210 Evaluate a sequence of statements assigning different values to @var{symbol}.
5211 The sequence of statements starts at the @code{.irp} directive, and is
5212 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5213 set to @var{value}, and the sequence of statements is assembled. If no
5214 @var{value} is listed, the sequence of statements is assembled once, with
5215 @var{symbol} set to the null string. To refer to @var{symbol} within the
5216 sequence of statements, use @var{\symbol}.
5218 For example, assembling
5226 is equivalent to assembling
5234 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5237 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5239 @cindex @code{irpc} directive
5240 Evaluate a sequence of statements assigning different values to @var{symbol}.
5241 The sequence of statements starts at the @code{.irpc} directive, and is
5242 terminated by an @code{.endr} directive. For each character in @var{value},
5243 @var{symbol} is set to the character, and the sequence of statements is
5244 assembled. If no @var{value} is listed, the sequence of statements is
5245 assembled once, with @var{symbol} set to the null string. To refer to
5246 @var{symbol} within the sequence of statements, use @var{\symbol}.
5248 For example, assembling
5256 is equivalent to assembling
5264 For some caveats with the spelling of @var{symbol}, see also the discussion
5268 @section @code{.lcomm @var{symbol} , @var{length}}
5270 @cindex @code{lcomm} directive
5271 @cindex local common symbols
5272 @cindex symbols, local common
5273 Reserve @var{length} (an absolute expression) bytes for a local common
5274 denoted by @var{symbol}. The section and value of @var{symbol} are
5275 those of the new local common. The addresses are allocated in the bss
5276 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5277 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5278 not visible to @code{@value{LD}}.
5281 Some targets permit a third argument to be used with @code{.lcomm}. This
5282 argument specifies the desired alignment of the symbol in the bss section.
5286 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5287 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5291 @section @code{.lflags}
5293 @cindex @code{lflags} directive (ignored)
5294 @command{@value{AS}} accepts this directive, for compatibility with other
5295 assemblers, but ignores it.
5297 @ifclear no-line-dir
5299 @section @code{.line @var{line-number}}
5301 @cindex @code{line} directive
5302 @cindex logical line number
5304 Change the logical line number. @var{line-number} must be an absolute
5305 expression. The next line has that logical line number. Therefore any other
5306 statements on the current line (after a statement separator character) are
5307 reported as on logical line number @var{line-number} @minus{} 1. One day
5308 @command{@value{AS}} will no longer support this directive: it is recognized only
5309 for compatibility with existing assembler programs.
5312 Even though this is a directive associated with the @code{a.out} or
5313 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5314 when producing COFF output, and treats @samp{.line} as though it
5315 were the COFF @samp{.ln} @emph{if} it is found outside a
5316 @code{.def}/@code{.endef} pair.
5318 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5319 used by compilers to generate auxiliary symbol information for
5324 @section @code{.linkonce [@var{type}]}
5326 @cindex @code{linkonce} directive
5327 @cindex common sections
5328 Mark the current section so that the linker only includes a single copy of it.
5329 This may be used to include the same section in several different object files,
5330 but ensure that the linker will only include it once in the final output file.
5331 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5332 Duplicate sections are detected based on the section name, so it should be
5335 This directive is only supported by a few object file formats; as of this
5336 writing, the only object file format which supports it is the Portable
5337 Executable format used on Windows NT.
5339 The @var{type} argument is optional. If specified, it must be one of the
5340 following strings. For example:
5344 Not all types may be supported on all object file formats.
5348 Silently discard duplicate sections. This is the default.
5351 Warn if there are duplicate sections, but still keep only one copy.
5354 Warn if any of the duplicates have different sizes.
5357 Warn if any of the duplicates do not have exactly the same contents.
5361 @section @code{.list}
5363 @cindex @code{list} directive
5364 @cindex listing control, turning on
5365 Control (in conjunction with the @code{.nolist} directive) whether or
5366 not assembly listings are generated. These two directives maintain an
5367 internal counter (which is zero initially). @code{.list} increments the
5368 counter, and @code{.nolist} decrements it. Assembly listings are
5369 generated whenever the counter is greater than zero.
5371 By default, listings are disabled. When you enable them (with the
5372 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5373 the initial value of the listing counter is one.
5376 @section @code{.ln @var{line-number}}
5378 @cindex @code{ln} directive
5379 @ifclear no-line-dir
5380 @samp{.ln} is a synonym for @samp{.line}.
5383 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5384 must be an absolute expression. The next line has that logical
5385 line number, so any other statements on the current line (after a
5386 statement separator character @code{;}) are reported as on logical
5387 line number @var{line-number} @minus{} 1.
5390 This directive is accepted, but ignored, when @command{@value{AS}} is
5391 configured for @code{b.out}; its effect is only associated with COFF
5397 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5398 @cindex @code{loc} directive
5399 When emitting DWARF2 line number information,
5400 the @code{.loc} directive will add a row to the @code{.debug_line} line
5401 number matrix corresponding to the immediately following assembly
5402 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5403 arguments will be applied to the @code{.debug_line} state machine before
5406 The @var{options} are a sequence of the following tokens in any order:
5410 This option will set the @code{basic_block} register in the
5411 @code{.debug_line} state machine to @code{true}.
5414 This option will set the @code{prologue_end} register in the
5415 @code{.debug_line} state machine to @code{true}.
5417 @item epilogue_begin
5418 This option will set the @code{epilogue_begin} register in the
5419 @code{.debug_line} state machine to @code{true}.
5421 @item is_stmt @var{value}
5422 This option will set the @code{is_stmt} register in the
5423 @code{.debug_line} state machine to @code{value}, which must be
5426 @item isa @var{value}
5427 This directive will set the @code{isa} register in the @code{.debug_line}
5428 state machine to @var{value}, which must be an unsigned integer.
5430 @item discriminator @var{value}
5431 This directive will set the @code{discriminator} register in the @code{.debug_line}
5432 state machine to @var{value}, which must be an unsigned integer.
5436 @node Loc_mark_labels
5437 @section @code{.loc_mark_labels @var{enable}}
5438 @cindex @code{loc_mark_labels} directive
5439 When emitting DWARF2 line number information,
5440 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5441 to the @code{.debug_line} line number matrix with the @code{basic_block}
5442 register in the state machine set whenever a code label is seen.
5443 The @var{enable} argument should be either 1 or 0, to enable or disable
5444 this function respectively.
5448 @section @code{.local @var{names}}
5450 @cindex @code{local} directive
5451 This directive, which is available for ELF targets, marks each symbol in
5452 the comma-separated list of @code{names} as a local symbol so that it
5453 will not be externally visible. If the symbols do not already exist,
5454 they will be created.
5456 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5457 accept an alignment argument, which is the case for most ELF targets,
5458 the @code{.local} directive can be used in combination with @code{.comm}
5459 (@pxref{Comm}) to define aligned local common data.
5463 @section @code{.long @var{expressions}}
5465 @cindex @code{long} directive
5466 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5469 @c no one seems to know what this is for or whether this description is
5470 @c what it really ought to do
5472 @section @code{.lsym @var{symbol}, @var{expression}}
5474 @cindex @code{lsym} directive
5475 @cindex symbol, not referenced in assembly
5476 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5477 the hash table, ensuring it cannot be referenced by name during the
5478 rest of the assembly. This sets the attributes of the symbol to be
5479 the same as the expression value:
5481 @var{other} = @var{descriptor} = 0
5482 @var{type} = @r{(section of @var{expression})}
5483 @var{value} = @var{expression}
5486 The new symbol is not flagged as external.
5490 @section @code{.macro}
5493 The commands @code{.macro} and @code{.endm} allow you to define macros that
5494 generate assembly output. For example, this definition specifies a macro
5495 @code{sum} that puts a sequence of numbers into memory:
5498 .macro sum from=0, to=5
5507 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5519 @item .macro @var{macname}
5520 @itemx .macro @var{macname} @var{macargs} @dots{}
5521 @cindex @code{macro} directive
5522 Begin the definition of a macro called @var{macname}. If your macro
5523 definition requires arguments, specify their names after the macro name,
5524 separated by commas or spaces. You can qualify the macro argument to
5525 indicate whether all invocations must specify a non-blank value (through
5526 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5527 (through @samp{:@code{vararg}}). You can supply a default value for any
5528 macro argument by following the name with @samp{=@var{deflt}}. You
5529 cannot define two macros with the same @var{macname} unless it has been
5530 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5531 definitions. For example, these are all valid @code{.macro} statements:
5535 Begin the definition of a macro called @code{comm}, which takes no
5538 @item .macro plus1 p, p1
5539 @itemx .macro plus1 p p1
5540 Either statement begins the definition of a macro called @code{plus1},
5541 which takes two arguments; within the macro definition, write
5542 @samp{\p} or @samp{\p1} to evaluate the arguments.
5544 @item .macro reserve_str p1=0 p2
5545 Begin the definition of a macro called @code{reserve_str}, with two
5546 arguments. The first argument has a default value, but not the second.
5547 After the definition is complete, you can call the macro either as
5548 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5549 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5550 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5551 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5553 @item .macro m p1:req, p2=0, p3:vararg
5554 Begin the definition of a macro called @code{m}, with at least three
5555 arguments. The first argument must always have a value specified, but
5556 not the second, which instead has a default value. The third formal
5557 will get assigned all remaining arguments specified at invocation time.
5559 When you call a macro, you can specify the argument values either by
5560 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5561 @samp{sum to=17, from=9}.
5565 Note that since each of the @var{macargs} can be an identifier exactly
5566 as any other one permitted by the target architecture, there may be
5567 occasional problems if the target hand-crafts special meanings to certain
5568 characters when they occur in a special position. For example, if the colon
5569 (@code{:}) is generally permitted to be part of a symbol name, but the
5570 architecture specific code special-cases it when occurring as the final
5571 character of a symbol (to denote a label), then the macro parameter
5572 replacement code will have no way of knowing that and consider the whole
5573 construct (including the colon) an identifier, and check only this
5574 identifier for being the subject to parameter substitution. So for example
5575 this macro definition:
5583 might not work as expected. Invoking @samp{label foo} might not create a label
5584 called @samp{foo} but instead just insert the text @samp{\l:} into the
5585 assembler source, probably generating an error about an unrecognised
5588 Similarly problems might occur with the period character (@samp{.})
5589 which is often allowed inside opcode names (and hence identifier names). So
5590 for example constructing a macro to build an opcode from a base name and a
5591 length specifier like this:
5594 .macro opcode base length
5599 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5600 instruction but instead generate some kind of error as the assembler tries to
5601 interpret the text @samp{\base.\length}.
5603 There are several possible ways around this problem:
5606 @item Insert white space
5607 If it is possible to use white space characters then this is the simplest
5616 @item Use @samp{\()}
5617 The string @samp{\()} can be used to separate the end of a macro argument from
5618 the following text. eg:
5621 .macro opcode base length
5626 @item Use the alternate macro syntax mode
5627 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5628 used as a separator. eg:
5638 Note: this problem of correctly identifying string parameters to pseudo ops
5639 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5640 and @code{.irpc} (@pxref{Irpc}) as well.
5643 @cindex @code{endm} directive
5644 Mark the end of a macro definition.
5647 @cindex @code{exitm} directive
5648 Exit early from the current macro definition.
5650 @cindex number of macros executed
5651 @cindex macros, count executed
5653 @command{@value{AS}} maintains a counter of how many macros it has
5654 executed in this pseudo-variable; you can copy that number to your
5655 output with @samp{\@@}, but @emph{only within a macro definition}.
5657 @item LOCAL @var{name} [ , @dots{} ]
5658 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5659 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5660 @xref{Altmacro,,@code{.altmacro}}.
5664 @section @code{.mri @var{val}}
5666 @cindex @code{mri} directive
5667 @cindex MRI mode, temporarily
5668 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5669 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5670 affects code assembled until the next @code{.mri} directive, or until the end
5671 of the file. @xref{M, MRI mode, MRI mode}.
5674 @section @code{.noaltmacro}
5675 Disable alternate macro mode. @xref{Altmacro}.
5678 @section @code{.nolist}
5680 @cindex @code{nolist} directive
5681 @cindex listing control, turning off
5682 Control (in conjunction with the @code{.list} directive) whether or
5683 not assembly listings are generated. These two directives maintain an
5684 internal counter (which is zero initially). @code{.list} increments the
5685 counter, and @code{.nolist} decrements it. Assembly listings are
5686 generated whenever the counter is greater than zero.
5689 @section @code{.octa @var{bignums}}
5691 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5692 @cindex @code{octa} directive
5693 @cindex integer, 16-byte
5694 @cindex sixteen byte integer
5695 This directive expects zero or more bignums, separated by commas. For each
5696 bignum, it emits a 16-byte integer.
5698 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5699 hence @emph{octa}-word for 16 bytes.
5702 @section @code{.offset @var{loc}}
5704 @cindex @code{offset} directive
5705 Set the location counter to @var{loc} in the absolute section. @var{loc} must
5706 be an absolute expression. This directive may be useful for defining
5707 symbols with absolute values. Do not confuse it with the @code{.org}
5711 @section @code{.org @var{new-lc} , @var{fill}}
5713 @cindex @code{org} directive
5714 @cindex location counter, advancing
5715 @cindex advancing location counter
5716 @cindex current address, advancing
5717 Advance the location counter of the current section to
5718 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5719 expression with the same section as the current subsection. That is,
5720 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5721 wrong section, the @code{.org} directive is ignored. To be compatible
5722 with former assemblers, if the section of @var{new-lc} is absolute,
5723 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5724 is the same as the current subsection.
5726 @code{.org} may only increase the location counter, or leave it
5727 unchanged; you cannot use @code{.org} to move the location counter
5730 @c double negative used below "not undefined" because this is a specific
5731 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5732 @c section. doc@cygnus.com 18feb91
5733 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5734 may not be undefined. If you really detest this restriction we eagerly await
5735 a chance to share your improved assembler.
5737 Beware that the origin is relative to the start of the section, not
5738 to the start of the subsection. This is compatible with other
5739 people's assemblers.
5741 When the location counter (of the current subsection) is advanced, the
5742 intervening bytes are filled with @var{fill} which should be an
5743 absolute expression. If the comma and @var{fill} are omitted,
5744 @var{fill} defaults to zero.
5747 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5749 @cindex padding the location counter given a power of two
5750 @cindex @code{p2align} directive
5751 Pad the location counter (in the current subsection) to a particular
5752 storage boundary. The first expression (which must be absolute) is the
5753 number of low-order zero bits the location counter must have after
5754 advancement. For example @samp{.p2align 3} advances the location
5755 counter until it a multiple of 8. If the location counter is already a
5756 multiple of 8, no change is needed.
5758 The second expression (also absolute) gives the fill value to be stored in the
5759 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5760 padding bytes are normally zero. However, on some systems, if the section is
5761 marked as containing code and the fill value is omitted, the space is filled
5762 with no-op instructions.
5764 The third expression is also absolute, and is also optional. If it is present,
5765 it is the maximum number of bytes that should be skipped by this alignment
5766 directive. If doing the alignment would require skipping more bytes than the
5767 specified maximum, then the alignment is not done at all. You can omit the
5768 fill value (the second argument) entirely by simply using two commas after the
5769 required alignment; this can be useful if you want the alignment to be filled
5770 with no-op instructions when appropriate.
5772 @cindex @code{p2alignw} directive
5773 @cindex @code{p2alignl} directive
5774 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5775 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5776 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5777 fill pattern as a four byte longword value. For example, @code{.p2alignw
5778 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5779 filled in with the value 0x368d (the exact placement of the bytes depends upon
5780 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5785 @section @code{.popsection}
5787 @cindex @code{popsection} directive
5788 @cindex Section Stack
5789 This is one of the ELF section stack manipulation directives. The others are
5790 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5791 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5794 This directive replaces the current section (and subsection) with the top
5795 section (and subsection) on the section stack. This section is popped off the
5801 @section @code{.previous}
5803 @cindex @code{previous} directive
5804 @cindex Section Stack
5805 This is one of the ELF section stack manipulation directives. The others are
5806 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5807 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5808 (@pxref{PopSection}).
5810 This directive swaps the current section (and subsection) with most recently
5811 referenced section/subsection pair prior to this one. Multiple
5812 @code{.previous} directives in a row will flip between two sections (and their
5813 subsections). For example:
5825 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
5831 # Now in section A subsection 1
5835 # Now in section B subsection 0
5838 # Now in section B subsection 1
5841 # Now in section B subsection 0
5845 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
5846 section B and 0x9abc into subsection 1 of section B.
5848 In terms of the section stack, this directive swaps the current section with
5849 the top section on the section stack.
5853 @section @code{.print @var{string}}
5855 @cindex @code{print} directive
5856 @command{@value{AS}} will print @var{string} on the standard output during
5857 assembly. You must put @var{string} in double quotes.
5861 @section @code{.protected @var{names}}
5863 @cindex @code{protected} directive
5865 This is one of the ELF visibility directives. The other two are
5866 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5868 This directive overrides the named symbols default visibility (which is set by
5869 their binding: local, global or weak). The directive sets the visibility to
5870 @code{protected} which means that any references to the symbols from within the
5871 components that defines them must be resolved to the definition in that
5872 component, even if a definition in another component would normally preempt
5877 @section @code{.psize @var{lines} , @var{columns}}
5879 @cindex @code{psize} directive
5880 @cindex listing control: paper size
5881 @cindex paper size, for listings
5882 Use this directive to declare the number of lines---and, optionally, the
5883 number of columns---to use for each page, when generating listings.
5885 If you do not use @code{.psize}, listings use a default line-count
5886 of 60. You may omit the comma and @var{columns} specification; the
5887 default width is 200 columns.
5889 @command{@value{AS}} generates formfeeds whenever the specified number of
5890 lines is exceeded (or whenever you explicitly request one, using
5893 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5894 those explicitly specified with @code{.eject}.
5897 @section @code{.purgem @var{name}}
5899 @cindex @code{purgem} directive
5900 Undefine the macro @var{name}, so that later uses of the string will not be
5901 expanded. @xref{Macro}.
5905 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
5907 @cindex @code{pushsection} directive
5908 @cindex Section Stack
5909 This is one of the ELF section stack manipulation directives. The others are
5910 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5911 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5914 This directive pushes the current section (and subsection) onto the
5915 top of the section stack, and then replaces the current section and
5916 subsection with @code{name} and @code{subsection}. The optional
5917 @code{flags}, @code{type} and @code{arguments} are treated the same
5918 as in the @code{.section} (@pxref{Section}) directive.
5922 @section @code{.quad @var{bignums}}
5924 @cindex @code{quad} directive
5925 @code{.quad} expects zero or more bignums, separated by commas. For
5926 each bignum, it emits
5928 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5929 warning message; and just takes the lowest order 8 bytes of the bignum.
5930 @cindex eight-byte integer
5931 @cindex integer, 8-byte
5933 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5934 hence @emph{quad}-word for 8 bytes.
5937 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5938 warning message; and just takes the lowest order 16 bytes of the bignum.
5939 @cindex sixteen-byte integer
5940 @cindex integer, 16-byte
5944 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
5946 @cindex @code{reloc} directive
5947 Generate a relocation at @var{offset} of type @var{reloc_name} with value
5948 @var{expression}. If @var{offset} is a number, the relocation is generated in
5949 the current section. If @var{offset} is an expression that resolves to a
5950 symbol plus offset, the relocation is generated in the given symbol's section.
5951 @var{expression}, if present, must resolve to a symbol plus addend or to an
5952 absolute value, but note that not all targets support an addend. e.g. ELF REL
5953 targets such as i386 store an addend in the section contents rather than in the
5954 relocation. This low level interface does not support addends stored in the
5958 @section @code{.rept @var{count}}
5960 @cindex @code{rept} directive
5961 Repeat the sequence of lines between the @code{.rept} directive and the next
5962 @code{.endr} directive @var{count} times.
5964 For example, assembling
5972 is equivalent to assembling
5981 @section @code{.sbttl "@var{subheading}"}
5983 @cindex @code{sbttl} directive
5984 @cindex subtitles for listings
5985 @cindex listing control: subtitle
5986 Use @var{subheading} as the title (third line, immediately after the
5987 title line) when generating assembly listings.
5989 This directive affects subsequent pages, as well as the current page if
5990 it appears within ten lines of the top of a page.
5994 @section @code{.scl @var{class}}
5996 @cindex @code{scl} directive
5997 @cindex symbol storage class (COFF)
5998 @cindex COFF symbol storage class
5999 Set the storage-class value for a symbol. This directive may only be
6000 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6001 whether a symbol is static or external, or it may record further
6002 symbolic debugging information.
6005 The @samp{.scl} directive is primarily associated with COFF output; when
6006 configured to generate @code{b.out} output format, @command{@value{AS}}
6007 accepts this directive but ignores it.
6013 @section @code{.section @var{name}}
6015 @cindex named section
6016 Use the @code{.section} directive to assemble the following code into a section
6019 This directive is only supported for targets that actually support arbitrarily
6020 named sections; on @code{a.out} targets, for example, it is not accepted, even
6021 with a standard @code{a.out} section name.
6025 @c only print the extra heading if both COFF and ELF are set
6026 @subheading COFF Version
6029 @cindex @code{section} directive (COFF version)
6030 For COFF targets, the @code{.section} directive is used in one of the following
6034 .section @var{name}[, "@var{flags}"]
6035 .section @var{name}[, @var{subsection}]
6038 If the optional argument is quoted, it is taken as flags to use for the
6039 section. Each flag is a single character. The following flags are recognized:
6042 bss section (uninitialized data)
6044 section is not loaded
6050 exclude section from linking
6056 shared section (meaningful for PE targets)
6058 ignored. (For compatibility with the ELF version)
6060 section is not readable (meaningful for PE targets)
6062 single-digit power-of-two section alignment (GNU extension)
6065 If no flags are specified, the default flags depend upon the section name. If
6066 the section name is not recognized, the default will be for the section to be
6067 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6068 from the section, rather than adding them, so if they are used on their own it
6069 will be as if no flags had been specified at all.
6071 If the optional argument to the @code{.section} directive is not quoted, it is
6072 taken as a subsection number (@pxref{Sub-Sections}).
6077 @c only print the extra heading if both COFF and ELF are set
6078 @subheading ELF Version
6081 @cindex Section Stack
6082 This is one of the ELF section stack manipulation directives. The others are
6083 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6084 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6085 @code{.previous} (@pxref{Previous}).
6087 @cindex @code{section} directive (ELF version)
6088 For ELF targets, the @code{.section} directive is used like this:
6091 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6094 The optional @var{flags} argument is a quoted string which may contain any
6095 combination of the following characters:
6098 section is allocatable
6100 section is excluded from executable and shared library.
6104 section is executable
6106 section is mergeable
6108 section contains zero terminated strings
6110 section is a member of a section group
6112 section is used for thread-local-storage
6114 section is a member of the previously-current section's group, if any
6117 The optional @var{type} argument may contain one of the following constants:
6120 section contains data
6122 section does not contain data (i.e., section only occupies space)
6124 section contains data which is used by things other than the program
6126 section contains an array of pointers to init functions
6128 section contains an array of pointers to finish functions
6129 @item @@preinit_array
6130 section contains an array of pointers to pre-init functions
6133 Many targets only support the first three section types.
6135 Note on targets where the @code{@@} character is the start of a comment (eg
6136 ARM) then another character is used instead. For example the ARM port uses the
6139 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6140 be specified as well as an extra argument---@var{entsize}---like this:
6143 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6146 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6147 constants, each @var{entsize} octets long. Sections with both @code{M} and
6148 @code{S} must contain zero terminated strings where each character is
6149 @var{entsize} bytes long. The linker may remove duplicates within sections with
6150 the same name, same entity size and same flags. @var{entsize} must be an
6151 absolute expression. For sections with both @code{M} and @code{S}, a string
6152 which is a suffix of a larger string is considered a duplicate. Thus
6153 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6154 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6156 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6157 be present along with an additional field like this:
6160 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6163 The @var{GroupName} field specifies the name of the section group to which this
6164 particular section belongs. The optional linkage field can contain:
6167 indicates that only one copy of this section should be retained
6172 Note: if both the @var{M} and @var{G} flags are present then the fields for
6173 the Merge flag should come first, like this:
6176 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6179 If @var{flags} contains the @code{?} symbol then it may not also contain the
6180 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6181 present. Instead, @code{?} says to consider the section that's current before
6182 this directive. If that section used @code{G}, then the new section will use
6183 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6184 If not, then the @code{?} symbol has no effect.
6186 If no flags are specified, the default flags depend upon the section name. If
6187 the section name is not recognized, the default will be for the section to have
6188 none of the above flags: it will not be allocated in memory, nor writable, nor
6189 executable. The section will contain data.
6191 For ELF targets, the assembler supports another type of @code{.section}
6192 directive for compatibility with the Solaris assembler:
6195 .section "@var{name}"[, @var{flags}...]
6198 Note that the section name is quoted. There may be a sequence of comma
6202 section is allocatable
6206 section is executable
6208 section is excluded from executable and shared library.
6210 section is used for thread local storage
6213 This directive replaces the current section and subsection. See the
6214 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6215 some examples of how this directive and the other section stack directives
6221 @section @code{.set @var{symbol}, @var{expression}}
6223 @cindex @code{set} directive
6224 @cindex symbol value, setting
6225 Set the value of @var{symbol} to @var{expression}. This
6226 changes @var{symbol}'s value and type to conform to
6227 @var{expression}. If @var{symbol} was flagged as external, it remains
6228 flagged (@pxref{Symbol Attributes}).
6230 You may @code{.set} a symbol many times in the same assembly.
6232 If you @code{.set} a global symbol, the value stored in the object
6233 file is the last value stored into it.
6236 On Z80 @code{set} is a real instruction, use
6237 @samp{@var{symbol} defl @var{expression}} instead.
6241 @section @code{.short @var{expressions}}
6243 @cindex @code{short} directive
6245 @code{.short} is normally the same as @samp{.word}.
6246 @xref{Word,,@code{.word}}.
6248 In some configurations, however, @code{.short} and @code{.word} generate
6249 numbers of different lengths. @xref{Machine Dependencies}.
6253 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6256 This expects zero or more @var{expressions}, and emits
6257 a 16 bit number for each.
6262 @section @code{.single @var{flonums}}
6264 @cindex @code{single} directive
6265 @cindex floating point numbers (single)
6266 This directive assembles zero or more flonums, separated by commas. It
6267 has the same effect as @code{.float}.
6269 The exact kind of floating point numbers emitted depends on how
6270 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6274 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6275 numbers in @sc{ieee} format.
6281 @section @code{.size}
6283 This directive is used to set the size associated with a symbol.
6287 @c only print the extra heading if both COFF and ELF are set
6288 @subheading COFF Version
6291 @cindex @code{size} directive (COFF version)
6292 For COFF targets, the @code{.size} directive is only permitted inside
6293 @code{.def}/@code{.endef} pairs. It is used like this:
6296 .size @var{expression}
6300 @samp{.size} is only meaningful when generating COFF format output; when
6301 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6308 @c only print the extra heading if both COFF and ELF are set
6309 @subheading ELF Version
6312 @cindex @code{size} directive (ELF version)
6313 For ELF targets, the @code{.size} directive is used like this:
6316 .size @var{name} , @var{expression}
6319 This directive sets the size associated with a symbol @var{name}.
6320 The size in bytes is computed from @var{expression} which can make use of label
6321 arithmetic. This directive is typically used to set the size of function
6326 @ifclear no-space-dir
6328 @section @code{.skip @var{size} , @var{fill}}
6330 @cindex @code{skip} directive
6331 @cindex filling memory
6332 This directive emits @var{size} bytes, each of value @var{fill}. Both
6333 @var{size} and @var{fill} are absolute expressions. If the comma and
6334 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6339 @section @code{.sleb128 @var{expressions}}
6341 @cindex @code{sleb128} directive
6342 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6343 compact, variable length representation of numbers used by the DWARF
6344 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6346 @ifclear no-space-dir
6348 @section @code{.space @var{size} , @var{fill}}
6350 @cindex @code{space} directive
6351 @cindex filling memory
6352 This directive emits @var{size} bytes, each of value @var{fill}. Both
6353 @var{size} and @var{fill} are absolute expressions. If the comma
6354 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6359 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6360 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6361 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6362 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6370 @section @code{.stabd, .stabn, .stabs}
6372 @cindex symbolic debuggers, information for
6373 @cindex @code{stab@var{x}} directives
6374 There are three directives that begin @samp{.stab}.
6375 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6376 The symbols are not entered in the @command{@value{AS}} hash table: they
6377 cannot be referenced elsewhere in the source file.
6378 Up to five fields are required:
6382 This is the symbol's name. It may contain any character except
6383 @samp{\000}, so is more general than ordinary symbol names. Some
6384 debuggers used to code arbitrarily complex structures into symbol names
6388 An absolute expression. The symbol's type is set to the low 8 bits of
6389 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6390 and debuggers choke on silly bit patterns.
6393 An absolute expression. The symbol's ``other'' attribute is set to the
6394 low 8 bits of this expression.
6397 An absolute expression. The symbol's descriptor is set to the low 16
6398 bits of this expression.
6401 An absolute expression which becomes the symbol's value.
6404 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6405 or @code{.stabs} statement, the symbol has probably already been created;
6406 you get a half-formed symbol in your object file. This is
6407 compatible with earlier assemblers!
6410 @cindex @code{stabd} directive
6411 @item .stabd @var{type} , @var{other} , @var{desc}
6413 The ``name'' of the symbol generated is not even an empty string.
6414 It is a null pointer, for compatibility. Older assemblers used a
6415 null pointer so they didn't waste space in object files with empty
6418 The symbol's value is set to the location counter,
6419 relocatably. When your program is linked, the value of this symbol
6420 is the address of the location counter when the @code{.stabd} was
6423 @cindex @code{stabn} directive
6424 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6425 The name of the symbol is set to the empty string @code{""}.
6427 @cindex @code{stabs} directive
6428 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6429 All five fields are specified.
6435 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6436 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6438 @cindex string, copying to object file
6439 @cindex string8, copying to object file
6440 @cindex string16, copying to object file
6441 @cindex string32, copying to object file
6442 @cindex string64, copying to object file
6443 @cindex @code{string} directive
6444 @cindex @code{string8} directive
6445 @cindex @code{string16} directive
6446 @cindex @code{string32} directive
6447 @cindex @code{string64} directive
6449 Copy the characters in @var{str} to the object file. You may specify more than
6450 one string to copy, separated by commas. Unless otherwise specified for a
6451 particular machine, the assembler marks the end of each string with a 0 byte.
6452 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6454 The variants @code{string16}, @code{string32} and @code{string64} differ from
6455 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6456 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6457 are stored in target endianness byte order.
6463 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6464 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6469 @section @code{.struct @var{expression}}
6471 @cindex @code{struct} directive
6472 Switch to the absolute section, and set the section offset to @var{expression},
6473 which must be an absolute expression. You might use this as follows:
6482 This would define the symbol @code{field1} to have the value 0, the symbol
6483 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6484 value 8. Assembly would be left in the absolute section, and you would need to
6485 use a @code{.section} directive of some sort to change to some other section
6486 before further assembly.
6490 @section @code{.subsection @var{name}}
6492 @cindex @code{subsection} directive
6493 @cindex Section Stack
6494 This is one of the ELF section stack manipulation directives. The others are
6495 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6496 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6499 This directive replaces the current subsection with @code{name}. The current
6500 section is not changed. The replaced subsection is put onto the section stack
6501 in place of the then current top of stack subsection.
6506 @section @code{.symver}
6507 @cindex @code{symver} directive
6508 @cindex symbol versioning
6509 @cindex versions of symbols
6510 Use the @code{.symver} directive to bind symbols to specific version nodes
6511 within a source file. This is only supported on ELF platforms, and is
6512 typically used when assembling files to be linked into a shared library.
6513 There are cases where it may make sense to use this in objects to be bound
6514 into an application itself so as to override a versioned symbol from a
6517 For ELF targets, the @code{.symver} directive can be used like this:
6519 .symver @var{name}, @var{name2@@nodename}
6521 If the symbol @var{name} is defined within the file
6522 being assembled, the @code{.symver} directive effectively creates a symbol
6523 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6524 just don't try and create a regular alias is that the @var{@@} character isn't
6525 permitted in symbol names. The @var{name2} part of the name is the actual name
6526 of the symbol by which it will be externally referenced. The name @var{name}
6527 itself is merely a name of convenience that is used so that it is possible to
6528 have definitions for multiple versions of a function within a single source
6529 file, and so that the compiler can unambiguously know which version of a
6530 function is being mentioned. The @var{nodename} portion of the alias should be
6531 the name of a node specified in the version script supplied to the linker when
6532 building a shared library. If you are attempting to override a versioned
6533 symbol from a shared library, then @var{nodename} should correspond to the
6534 nodename of the symbol you are trying to override.
6536 If the symbol @var{name} is not defined within the file being assembled, all
6537 references to @var{name} will be changed to @var{name2@@nodename}. If no
6538 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6541 Another usage of the @code{.symver} directive is:
6543 .symver @var{name}, @var{name2@@@@nodename}
6545 In this case, the symbol @var{name} must exist and be defined within
6546 the file being assembled. It is similar to @var{name2@@nodename}. The
6547 difference is @var{name2@@@@nodename} will also be used to resolve
6548 references to @var{name2} by the linker.
6550 The third usage of the @code{.symver} directive is:
6552 .symver @var{name}, @var{name2@@@@@@nodename}
6554 When @var{name} is not defined within the
6555 file being assembled, it is treated as @var{name2@@nodename}. When
6556 @var{name} is defined within the file being assembled, the symbol
6557 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6562 @section @code{.tag @var{structname}}
6564 @cindex COFF structure debugging
6565 @cindex structure debugging, COFF
6566 @cindex @code{tag} directive
6567 This directive is generated by compilers to include auxiliary debugging
6568 information in the symbol table. It is only permitted inside
6569 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6570 definitions in the symbol table with instances of those structures.
6573 @samp{.tag} is only used when generating COFF format output; when
6574 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6580 @section @code{.text @var{subsection}}
6582 @cindex @code{text} directive
6583 Tells @command{@value{AS}} to assemble the following statements onto the end of
6584 the text subsection numbered @var{subsection}, which is an absolute
6585 expression. If @var{subsection} is omitted, subsection number zero
6589 @section @code{.title "@var{heading}"}
6591 @cindex @code{title} directive
6592 @cindex listing control: title line
6593 Use @var{heading} as the title (second line, immediately after the
6594 source file name and pagenumber) when generating assembly listings.
6596 This directive affects subsequent pages, as well as the current page if
6597 it appears within ten lines of the top of a page.
6601 @section @code{.type}
6603 This directive is used to set the type of a symbol.
6607 @c only print the extra heading if both COFF and ELF are set
6608 @subheading COFF Version
6611 @cindex COFF symbol type
6612 @cindex symbol type, COFF
6613 @cindex @code{type} directive (COFF version)
6614 For COFF targets, this directive is permitted only within
6615 @code{.def}/@code{.endef} pairs. It is used like this:
6621 This records the integer @var{int} as the type attribute of a symbol table
6625 @samp{.type} is associated only with COFF format output; when
6626 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6627 directive but ignores it.
6633 @c only print the extra heading if both COFF and ELF are set
6634 @subheading ELF Version
6637 @cindex ELF symbol type
6638 @cindex symbol type, ELF
6639 @cindex @code{type} directive (ELF version)
6640 For ELF targets, the @code{.type} directive is used like this:
6643 .type @var{name} , @var{type description}
6646 This sets the type of symbol @var{name} to be either a
6647 function symbol or an object symbol. There are five different syntaxes
6648 supported for the @var{type description} field, in order to provide
6649 compatibility with various other assemblers.
6651 Because some of the characters used in these syntaxes (such as @samp{@@} and
6652 @samp{#}) are comment characters for some architectures, some of the syntaxes
6653 below do not work on all architectures. The first variant will be accepted by
6654 the GNU assembler on all architectures so that variant should be used for
6655 maximum portability, if you do not need to assemble your code with other
6658 The syntaxes supported are:
6661 .type <name> STT_<TYPE_IN_UPPER_CASE>
6662 .type <name>,#<type>
6663 .type <name>,@@<type>
6664 .type <name>,%<type>
6665 .type <name>,"<type>"
6668 The types supported are:
6673 Mark the symbol as being a function name.
6676 @itemx gnu_indirect_function
6677 Mark the symbol as an indirect function when evaluated during reloc
6678 processing. (This is only supported on assemblers targeting GNU systems).
6682 Mark the symbol as being a data object.
6686 Mark the symbol as being a thead-local data object.
6690 Mark the symbol as being a common data object.
6694 Does not mark the symbol in any way. It is supported just for completeness.
6696 @item gnu_unique_object
6697 Marks the symbol as being a globally unique data object. The dynamic linker
6698 will make sure that in the entire process there is just one symbol with this
6699 name and type in use. (This is only supported on assemblers targeting GNU
6704 Note: Some targets support extra types in addition to those listed above.
6710 @section @code{.uleb128 @var{expressions}}
6712 @cindex @code{uleb128} directive
6713 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6714 compact, variable length representation of numbers used by the DWARF
6715 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6719 @section @code{.val @var{addr}}
6721 @cindex @code{val} directive
6722 @cindex COFF value attribute
6723 @cindex value attribute, COFF
6724 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6725 records the address @var{addr} as the value attribute of a symbol table
6729 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6730 configured for @code{b.out}, it accepts this directive but ignores it.
6736 @section @code{.version "@var{string}"}
6738 @cindex @code{version} directive
6739 This directive creates a @code{.note} section and places into it an ELF
6740 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6745 @section @code{.vtable_entry @var{table}, @var{offset}}
6747 @cindex @code{vtable_entry} directive
6748 This directive finds or creates a symbol @code{table} and creates a
6749 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6752 @section @code{.vtable_inherit @var{child}, @var{parent}}
6754 @cindex @code{vtable_inherit} directive
6755 This directive finds the symbol @code{child} and finds or creates the symbol
6756 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6757 parent whose addend is the value of the child symbol. As a special case the
6758 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6762 @section @code{.warning "@var{string}"}
6763 @cindex warning directive
6764 Similar to the directive @code{.error}
6765 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6768 @section @code{.weak @var{names}}
6770 @cindex @code{weak} directive
6771 This directive sets the weak attribute on the comma separated list of symbol
6772 @code{names}. If the symbols do not already exist, they will be created.
6774 On COFF targets other than PE, weak symbols are a GNU extension. This
6775 directive sets the weak attribute on the comma separated list of symbol
6776 @code{names}. If the symbols do not already exist, they will be created.
6778 On the PE target, weak symbols are supported natively as weak aliases.
6779 When a weak symbol is created that is not an alias, GAS creates an
6780 alternate symbol to hold the default value.
6783 @section @code{.weakref @var{alias}, @var{target}}
6785 @cindex @code{weakref} directive
6786 This directive creates an alias to the target symbol that enables the symbol to
6787 be referenced with weak-symbol semantics, but without actually making it weak.
6788 If direct references or definitions of the symbol are present, then the symbol
6789 will not be weak, but if all references to it are through weak references, the
6790 symbol will be marked as weak in the symbol table.
6792 The effect is equivalent to moving all references to the alias to a separate
6793 assembly source file, renaming the alias to the symbol in it, declaring the
6794 symbol as weak there, and running a reloadable link to merge the object files
6795 resulting from the assembly of the new source file and the old source file that
6796 had the references to the alias removed.
6798 The alias itself never makes to the symbol table, and is entirely handled
6799 within the assembler.
6802 @section @code{.word @var{expressions}}
6804 @cindex @code{word} directive
6805 This directive expects zero or more @var{expressions}, of any section,
6806 separated by commas.
6809 For each expression, @command{@value{AS}} emits a 32-bit number.
6812 For each expression, @command{@value{AS}} emits a 16-bit number.
6817 The size of the number emitted, and its byte order,
6818 depend on what target computer the assembly is for.
6821 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
6822 @c happen---32-bit addressability, period; no long/short jumps.
6823 @ifset DIFF-TBL-KLUGE
6824 @cindex difference tables altered
6825 @cindex altered difference tables
6827 @emph{Warning: Special Treatment to support Compilers}
6831 Machines with a 32-bit address space, but that do less than 32-bit
6832 addressing, require the following special treatment. If the machine of
6833 interest to you does 32-bit addressing (or doesn't require it;
6834 @pxref{Machine Dependencies}), you can ignore this issue.
6837 In order to assemble compiler output into something that works,
6838 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6839 Directives of the form @samp{.word sym1-sym2} are often emitted by
6840 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6841 directive of the form @samp{.word sym1-sym2}, and the difference between
6842 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6843 creates a @dfn{secondary jump table}, immediately before the next label.
6844 This secondary jump table is preceded by a short-jump to the
6845 first byte after the secondary table. This short-jump prevents the flow
6846 of control from accidentally falling into the new table. Inside the
6847 table is a long-jump to @code{sym2}. The original @samp{.word}
6848 contains @code{sym1} minus the address of the long-jump to
6851 If there were several occurrences of @samp{.word sym1-sym2} before the
6852 secondary jump table, all of them are adjusted. If there was a
6853 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6854 long-jump to @code{sym4} is included in the secondary jump table,
6855 and the @code{.word} directives are adjusted to contain @code{sym3}
6856 minus the address of the long-jump to @code{sym4}; and so on, for as many
6857 entries in the original jump table as necessary.
6860 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6861 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6862 assembly language programmers.
6865 @c end DIFF-TBL-KLUGE
6868 @section Deprecated Directives
6870 @cindex deprecated directives
6871 @cindex obsolescent directives
6872 One day these directives won't work.
6873 They are included for compatibility with older assemblers.
6880 @node Object Attributes
6881 @chapter Object Attributes
6882 @cindex object attributes
6884 @command{@value{AS}} assembles source files written for a specific architecture
6885 into object files for that architecture. But not all object files are alike.
6886 Many architectures support incompatible variations. For instance, floating
6887 point arguments might be passed in floating point registers if the object file
6888 requires hardware floating point support---or floating point arguments might be
6889 passed in integer registers if the object file supports processors with no
6890 hardware floating point unit. Or, if two objects are built for different
6891 generations of the same architecture, the combination may require the
6892 newer generation at run-time.
6894 This information is useful during and after linking. At link time,
6895 @command{@value{LD}} can warn about incompatible object files. After link
6896 time, tools like @command{gdb} can use it to process the linked file
6899 Compatibility information is recorded as a series of object attributes. Each
6900 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
6901 string, and indicates who sets the meaning of the tag. The tag is an integer,
6902 and indicates what property the attribute describes. The value may be a string
6903 or an integer, and indicates how the property affects this object. Missing
6904 attributes are the same as attributes with a zero value or empty string value.
6906 Object attributes were developed as part of the ABI for the ARM Architecture.
6907 The file format is documented in @cite{ELF for the ARM Architecture}.
6910 * GNU Object Attributes:: @sc{gnu} Object Attributes
6911 * Defining New Object Attributes:: Defining New Object Attributes
6914 @node GNU Object Attributes
6915 @section @sc{gnu} Object Attributes
6917 The @code{.gnu_attribute} directive records an object attribute
6918 with vendor @samp{gnu}.
6920 Except for @samp{Tag_compatibility}, which has both an integer and a string for
6921 its value, @sc{gnu} attributes have a string value if the tag number is odd and
6922 an integer value if the tag number is even. The second bit (@code{@var{tag} &
6923 2} is set for architecture-independent attributes and clear for
6924 architecture-dependent ones.
6926 @subsection Common @sc{gnu} attributes
6928 These attributes are valid on all architectures.
6931 @item Tag_compatibility (32)
6932 The compatibility attribute takes an integer flag value and a vendor name. If
6933 the flag value is 0, the file is compatible with other toolchains. If it is 1,
6934 then the file is only compatible with the named toolchain. If it is greater
6935 than 1, the file can only be processed by other toolchains under some private
6936 arrangement indicated by the flag value and the vendor name.
6939 @subsection MIPS Attributes
6942 @item Tag_GNU_MIPS_ABI_FP (4)
6943 The floating-point ABI used by this object file. The value will be:
6947 0 for files not affected by the floating-point ABI.
6949 1 for files using the hardware floating-point with a standard double-precision
6952 2 for files using the hardware floating-point ABI with a single-precision FPU.
6954 3 for files using the software floating-point ABI.
6956 4 for files using the hardware floating-point ABI with 64-bit wide
6957 double-precision floating-point registers and 32-bit wide general
6962 @subsection PowerPC Attributes
6965 @item Tag_GNU_Power_ABI_FP (4)
6966 The floating-point ABI used by this object file. The value will be:
6970 0 for files not affected by the floating-point ABI.
6972 1 for files using double-precision hardware floating-point ABI.
6974 2 for files using the software floating-point ABI.
6976 3 for files using single-precision hardware floating-point ABI.
6979 @item Tag_GNU_Power_ABI_Vector (8)
6980 The vector ABI used by this object file. The value will be:
6984 0 for files not affected by the vector ABI.
6986 1 for files using general purpose registers to pass vectors.
6988 2 for files using AltiVec registers to pass vectors.
6990 3 for files using SPE registers to pass vectors.
6994 @node Defining New Object Attributes
6995 @section Defining New Object Attributes
6997 If you want to define a new @sc{gnu} object attribute, here are the places you
6998 will need to modify. New attributes should be discussed on the @samp{binutils}
7003 This manual, which is the official register of attributes.
7005 The header for your architecture @file{include/elf}, to define the tag.
7007 The @file{bfd} support file for your architecture, to merge the attribute
7008 and issue any appropriate link warnings.
7010 Test cases in @file{ld/testsuite} for merging and link warnings.
7012 @file{binutils/readelf.c} to display your attribute.
7014 GCC, if you want the compiler to mark the attribute automatically.
7020 @node Machine Dependencies
7021 @chapter Machine Dependent Features
7023 @cindex machine dependencies
7024 The machine instruction sets are (almost by definition) different on
7025 each machine where @command{@value{AS}} runs. Floating point representations
7026 vary as well, and @command{@value{AS}} often supports a few additional
7027 directives or command-line options for compatibility with other
7028 assemblers on a particular platform. Finally, some versions of
7029 @command{@value{AS}} support special pseudo-instructions for branch
7032 This chapter discusses most of these differences, though it does not
7033 include details on any machine's instruction set. For details on that
7034 subject, see the hardware manufacturer's manual.
7038 * AArch64-Dependent:: AArch64 Dependent Features
7041 * Alpha-Dependent:: Alpha Dependent Features
7044 * ARC-Dependent:: ARC Dependent Features
7047 * ARM-Dependent:: ARM Dependent Features
7050 * AVR-Dependent:: AVR Dependent Features
7053 * Blackfin-Dependent:: Blackfin Dependent Features
7056 * CR16-Dependent:: CR16 Dependent Features
7059 * CRIS-Dependent:: CRIS Dependent Features
7062 * D10V-Dependent:: D10V Dependent Features
7065 * D30V-Dependent:: D30V Dependent Features
7068 * Epiphany-Dependent:: EPIPHANY Dependent Features
7071 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7074 * HPPA-Dependent:: HPPA Dependent Features
7077 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
7080 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7083 * i860-Dependent:: Intel 80860 Dependent Features
7086 * i960-Dependent:: Intel 80960 Dependent Features
7089 * IA-64-Dependent:: Intel IA-64 Dependent Features
7092 * IP2K-Dependent:: IP2K Dependent Features
7095 * LM32-Dependent:: LM32 Dependent Features
7098 * M32C-Dependent:: M32C Dependent Features
7101 * M32R-Dependent:: M32R Dependent Features
7104 * M68K-Dependent:: M680x0 Dependent Features
7107 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7110 * Meta-Dependent :: Meta Dependent Features
7113 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7116 * MIPS-Dependent:: MIPS Dependent Features
7119 * MMIX-Dependent:: MMIX Dependent Features
7122 * MSP430-Dependent:: MSP430 Dependent Features
7125 * NiosII-Dependent:: Altera Nios II Dependent Features
7128 * NS32K-Dependent:: NS32K Dependent Features
7131 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7132 * SH64-Dependent:: SuperH SH64 Dependent Features
7135 * PDP-11-Dependent:: PDP-11 Dependent Features
7138 * PJ-Dependent:: picoJava Dependent Features
7141 * PPC-Dependent:: PowerPC Dependent Features
7144 * RL78-Dependent:: RL78 Dependent Features
7147 * RX-Dependent:: RX Dependent Features
7150 * S/390-Dependent:: IBM S/390 Dependent Features
7153 * SCORE-Dependent:: SCORE Dependent Features
7156 * Sparc-Dependent:: SPARC Dependent Features
7159 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7162 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7165 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7168 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7171 * V850-Dependent:: V850 Dependent Features
7174 * XGATE-Dependent:: XGATE Features
7177 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7180 * Xtensa-Dependent:: Xtensa Dependent Features
7183 * Z80-Dependent:: Z80 Dependent Features
7186 * Z8000-Dependent:: Z8000 Dependent Features
7189 * Vax-Dependent:: VAX Dependent Features
7196 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7197 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7198 @c peculiarity: to preserve cross-references, there must be a node called
7199 @c "Machine Dependencies". Hence the conditional nodenames in each
7200 @c major node below. Node defaulting in makeinfo requires adjacency of
7201 @c node and sectioning commands; hence the repetition of @chapter BLAH
7202 @c in both conditional blocks.
7205 @include c-aarch64.texi
7209 @include c-alpha.texi
7225 @include c-bfin.texi
7229 @include c-cr16.texi
7233 @include c-cris.texi
7238 @node Machine Dependencies
7239 @chapter Machine Dependent Features
7241 The machine instruction sets are different on each Renesas chip family,
7242 and there are also some syntax differences among the families. This
7243 chapter describes the specific @command{@value{AS}} features for each
7247 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7248 * SH-Dependent:: Renesas SH Dependent Features
7255 @include c-d10v.texi
7259 @include c-d30v.texi
7263 @include c-epiphany.texi
7267 @include c-h8300.texi
7271 @include c-hppa.texi
7275 @include c-i370.texi
7279 @include c-i386.texi
7283 @include c-i860.texi
7287 @include c-i960.texi
7291 @include c-ia64.texi
7295 @include c-ip2k.texi
7299 @include c-lm32.texi
7303 @include c-m32c.texi
7307 @include c-m32r.texi
7311 @include c-m68k.texi
7315 @include c-m68hc11.texi
7319 @include c-metag.texi
7323 @include c-microblaze.texi
7327 @include c-mips.texi
7331 @include c-mmix.texi
7335 @include c-msp430.texi
7339 @include c-nios2.texi
7343 @include c-ns32k.texi
7347 @include c-pdp11.texi
7359 @include c-rl78.texi
7367 @include c-s390.texi
7371 @include c-score.texi
7376 @include c-sh64.texi
7380 @include c-sparc.texi
7384 @include c-tic54x.texi
7388 @include c-tic6x.texi
7392 @include c-tilegx.texi
7396 @include c-tilepro.texi
7412 @include c-v850.texi
7416 @include c-xgate.texi
7420 @include c-xstormy16.texi
7424 @include c-xtensa.texi
7428 @c reverse effect of @down at top of generic Machine-Dep chapter
7432 @node Reporting Bugs
7433 @chapter Reporting Bugs
7434 @cindex bugs in assembler
7435 @cindex reporting bugs in assembler
7437 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7439 Reporting a bug may help you by bringing a solution to your problem, or it may
7440 not. But in any case the principal function of a bug report is to help the
7441 entire community by making the next version of @command{@value{AS}} work better.
7442 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7444 In order for a bug report to serve its purpose, you must include the
7445 information that enables us to fix the bug.
7448 * Bug Criteria:: Have you found a bug?
7449 * Bug Reporting:: How to report bugs
7453 @section Have You Found a Bug?
7454 @cindex bug criteria
7456 If you are not sure whether you have found a bug, here are some guidelines:
7459 @cindex fatal signal
7460 @cindex assembler crash
7461 @cindex crash of assembler
7463 If the assembler gets a fatal signal, for any input whatever, that is a
7464 @command{@value{AS}} bug. Reliable assemblers never crash.
7466 @cindex error on valid input
7468 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7470 @cindex invalid input
7472 If @command{@value{AS}} does not produce an error message for invalid input, that
7473 is a bug. However, you should note that your idea of ``invalid input'' might
7474 be our idea of ``an extension'' or ``support for traditional practice''.
7477 If you are an experienced user of assemblers, your suggestions for improvement
7478 of @command{@value{AS}} are welcome in any case.
7482 @section How to Report Bugs
7484 @cindex assembler bugs, reporting
7486 A number of companies and individuals offer support for @sc{gnu} products. If
7487 you obtained @command{@value{AS}} from a support organization, we recommend you
7488 contact that organization first.
7490 You can find contact information for many support companies and
7491 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7495 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7499 The fundamental principle of reporting bugs usefully is this:
7500 @strong{report all the facts}. If you are not sure whether to state a
7501 fact or leave it out, state it!
7503 Often people omit facts because they think they know what causes the problem
7504 and assume that some details do not matter. Thus, you might assume that the
7505 name of a symbol you use in an example does not matter. Well, probably it does
7506 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7507 happens to fetch from the location where that name is stored in memory;
7508 perhaps, if the name were different, the contents of that location would fool
7509 the assembler into doing the right thing despite the bug. Play it safe and
7510 give a specific, complete example. That is the easiest thing for you to do,
7511 and the most helpful.
7513 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7514 it is new to us. Therefore, always write your bug reports on the assumption
7515 that the bug has not been reported previously.
7517 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7518 bell?'' This cannot help us fix a bug, so it is basically useless. We
7519 respond by asking for enough details to enable us to investigate.
7520 You might as well expedite matters by sending them to begin with.
7522 To enable us to fix the bug, you should include all these things:
7526 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
7527 it with the @samp{--version} argument.
7529 Without this, we will not know whether there is any point in looking for
7530 the bug in the current version of @command{@value{AS}}.
7533 Any patches you may have applied to the @command{@value{AS}} source.
7536 The type of machine you are using, and the operating system name and
7540 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
7544 The command arguments you gave the assembler to assemble your example and
7545 observe the bug. To guarantee you will not omit something important, list them
7546 all. A copy of the Makefile (or the output from make) is sufficient.
7548 If we were to try to guess the arguments, we would probably guess wrong
7549 and then we might not encounter the bug.
7552 A complete input file that will reproduce the bug. If the bug is observed when
7553 the assembler is invoked via a compiler, send the assembler source, not the
7554 high level language source. Most compilers will produce the assembler source
7555 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
7556 the options @samp{-v --save-temps}; this will save the assembler source in a
7557 file with an extension of @file{.s}, and also show you exactly how
7558 @command{@value{AS}} is being run.
7561 A description of what behavior you observe that you believe is
7562 incorrect. For example, ``It gets a fatal signal.''
7564 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
7565 will certainly notice it. But if the bug is incorrect output, we might not
7566 notice unless it is glaringly wrong. You might as well not give us a chance to
7569 Even if the problem you experience is a fatal signal, you should still say so
7570 explicitly. Suppose something strange is going on, such as, your copy of
7571 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
7572 library on your system. (This has happened!) Your copy might crash and ours
7573 would not. If you told us to expect a crash, then when ours fails to crash, we
7574 would know that the bug was not happening for us. If you had not told us to
7575 expect a crash, then we would not be able to draw any conclusion from our
7579 If you wish to suggest changes to the @command{@value{AS}} source, send us context
7580 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
7581 option. Always send diffs from the old file to the new file. If you even
7582 discuss something in the @command{@value{AS}} source, refer to it by context, not
7585 The line numbers in our development sources will not match those in your
7586 sources. Your line numbers would convey no useful information to us.
7589 Here are some things that are not necessary:
7593 A description of the envelope of the bug.
7595 Often people who encounter a bug spend a lot of time investigating
7596 which changes to the input file will make the bug go away and which
7597 changes will not affect it.
7599 This is often time consuming and not very useful, because the way we
7600 will find the bug is by running a single example under the debugger
7601 with breakpoints, not by pure deduction from a series of examples.
7602 We recommend that you save your time for something else.
7604 Of course, if you can find a simpler example to report @emph{instead}
7605 of the original one, that is a convenience for us. Errors in the
7606 output will be easier to spot, running under the debugger will take
7607 less time, and so on.
7609 However, simplification is not vital; if you do not want to do this,
7610 report the bug anyway and send us the entire test case you used.
7613 A patch for the bug.
7615 A patch for the bug does help us if it is a good one. But do not omit
7616 the necessary information, such as the test case, on the assumption that
7617 a patch is all we need. We might see problems with your patch and decide
7618 to fix the problem another way, or we might not understand it at all.
7620 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
7621 construct an example that will make the program follow a certain path through
7622 the code. If you do not send us the example, we will not be able to construct
7623 one, so we will not be able to verify that the bug is fixed.
7625 And if we cannot understand what bug you are trying to fix, or why your
7626 patch should be an improvement, we will not install it. A test case will
7627 help us to understand.
7630 A guess about what the bug is or what it depends on.
7632 Such guesses are usually wrong. Even we cannot guess right about such
7633 things without first using the debugger to find the facts.
7636 @node Acknowledgements
7637 @chapter Acknowledgements
7639 If you have contributed to GAS and your name isn't listed here,
7640 it is not meant as a slight. We just don't know about it. Send mail to the
7641 maintainer, and we'll correct the situation. Currently
7643 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
7645 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
7648 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
7649 information and the 68k series machines, most of the preprocessing pass, and
7650 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
7652 K. Richard Pixley maintained GAS for a while, adding various enhancements and
7653 many bug fixes, including merging support for several processors, breaking GAS
7654 up to handle multiple object file format back ends (including heavy rewrite,
7655 testing, an integration of the coff and b.out back ends), adding configuration
7656 including heavy testing and verification of cross assemblers and file splits
7657 and renaming, converted GAS to strictly ANSI C including full prototypes, added
7658 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
7659 port (including considerable amounts of reverse engineering), a SPARC opcode
7660 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
7661 assertions and made them work, much other reorganization, cleanup, and lint.
7663 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
7664 in format-specific I/O modules.
7666 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
7667 has done much work with it since.
7669 The Intel 80386 machine description was written by Eliot Dresselhaus.
7671 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
7673 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
7674 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
7676 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
7677 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
7678 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
7679 support a.out format.
7681 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
7682 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
7683 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
7684 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
7687 John Gilmore built the AMD 29000 support, added @code{.include} support, and
7688 simplified the configuration of which versions accept which directives. He
7689 updated the 68k machine description so that Motorola's opcodes always produced
7690 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
7691 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
7692 cross-compilation support, and one bug in relaxation that took a week and
7693 required the proverbial one-bit fix.
7695 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
7696 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
7697 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
7698 PowerPC assembler, and made a few other minor patches.
7700 Steve Chamberlain made GAS able to generate listings.
7702 Hewlett-Packard contributed support for the HP9000/300.
7704 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
7705 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
7706 formats). This work was supported by both the Center for Software Science at
7707 the University of Utah and Cygnus Support.
7709 Support for ELF format files has been worked on by Mark Eichin of Cygnus
7710 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
7711 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
7712 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
7713 and some initial 64-bit support).
7715 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
7717 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
7718 support for openVMS/Alpha.
7720 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
7723 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
7724 Inc.@: added support for Xtensa processors.
7726 Several engineers at Cygnus Support have also provided many small bug fixes and
7727 configuration enhancements.
7729 Jon Beniston added support for the Lattice Mico32 architecture.
7731 Many others have contributed large or small bugfixes and enhancements. If
7732 you have contributed significant work and are not mentioned on this list, and
7733 want to be, let us know. Some of the history has been lost; we are not
7734 intentionally leaving anyone out.
7736 @node GNU Free Documentation License
7737 @appendix GNU Free Documentation License
7741 @unnumbered AS Index
projects / deliverable / binutils-gdb.git / blob