1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright (C) 1991-2014 Free Software Foundation, Inc.
3 @c UPDATE!! On future updates--
4 @c (1) check for new machine-dep cmdline options in
5 @c md_parse_option definitions in config/tc-*.c
6 @c (2) for platform-specific directives, examine md_pseudo_op
8 @c (3) for object-format specific directives, examine obj_pseudo_op
10 @c (4) portable directives in potable[] in read.c
14 @macro gcctabopt{body}
17 @c defaults, config file may override:
22 @include asconfig.texi
27 @c common OR combinations of conditions
53 @set abnormal-separator
57 @settitle Using @value{AS}
60 @settitle Using @value{AS} (@value{TARGET})
62 @setchapternewpage odd
67 @c WARE! Some of the machine-dependent sections contain tables of machine
68 @c instructions. Except in multi-column format, these tables look silly.
69 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
70 @c the multi-col format is faked within @example sections.
72 @c Again unfortunately, the natural size that fits on a page, for these tables,
73 @c is different depending on whether or not smallbook is turned on.
74 @c This matters, because of order: text flow switches columns at each page
77 @c The format faked in this source works reasonably well for smallbook,
78 @c not well for the default large-page format. This manual expects that if you
79 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
80 @c tables in question. You can turn on one without the other at your
81 @c discretion, of course.
84 @c the insn tables look just as silly in info files regardless of smallbook,
85 @c might as well show 'em anyways.
89 @dircategory Software development
91 * As: (as). The GNU assembler.
92 * Gas: (as). The GNU assembler.
100 This file documents the GNU Assembler "@value{AS}".
102 @c man begin COPYRIGHT
103 Copyright @copyright{} 1991-2014 Free Software Foundation, Inc.
105 Permission is granted to copy, distribute and/or modify this document
106 under the terms of the GNU Free Documentation License, Version 1.3
107 or any later version published by the Free Software Foundation;
108 with no Invariant Sections, with no Front-Cover Texts, and with no
109 Back-Cover Texts. A copy of the license is included in the
110 section entitled ``GNU Free Documentation License''.
116 @title Using @value{AS}
117 @subtitle The @sc{gnu} Assembler
119 @subtitle for the @value{TARGET} family
121 @ifset VERSION_PACKAGE
123 @subtitle @value{VERSION_PACKAGE}
126 @subtitle Version @value{VERSION}
129 The Free Software Foundation Inc.@: thanks The Nice Computer
130 Company of Australia for loaning Dean Elsner to write the
131 first (Vax) version of @command{as} for Project @sc{gnu}.
132 The proprietors, management and staff of TNCCA thank FSF for
133 distracting the boss while they got some work
136 @author Dean Elsner, Jay Fenlason & friends
140 \hfill {\it Using {\tt @value{AS}}}\par
141 \hfill Edited by Cygnus Support\par
143 %"boxit" macro for figures:
144 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
145 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
146 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
147 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
148 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
151 @vskip 0pt plus 1filll
152 Copyright @copyright{} 1991-2014 Free Software Foundation, Inc.
154 Permission is granted to copy, distribute and/or modify this document
155 under the terms of the GNU Free Documentation License, Version 1.3
156 or any later version published by the Free Software Foundation;
157 with no Invariant Sections, with no Front-Cover Texts, and with no
158 Back-Cover Texts. A copy of the license is included in the
159 section entitled ``GNU Free Documentation License''.
166 @top Using @value{AS}
168 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}}
169 @ifset VERSION_PACKAGE
170 @value{VERSION_PACKAGE}
172 version @value{VERSION}.
174 This version of the file describes @command{@value{AS}} configured to generate
175 code for @value{TARGET} architectures.
178 This document is distributed under the terms of the GNU Free
179 Documentation License. A copy of the license is included in the
180 section entitled ``GNU Free Documentation License''.
183 * Overview:: Overview
184 * Invoking:: Command-Line Options
186 * Sections:: Sections and Relocation
188 * Expressions:: Expressions
189 * Pseudo Ops:: Assembler Directives
191 * Object Attributes:: Object Attributes
193 * Machine Dependencies:: Machine Dependent Features
194 * Reporting Bugs:: Reporting Bugs
195 * Acknowledgements:: Who Did What
196 * GNU Free Documentation License:: GNU Free Documentation License
197 * AS Index:: AS Index
204 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
206 This version of the manual describes @command{@value{AS}} configured to generate
207 code for @value{TARGET} architectures.
211 @cindex invocation summary
212 @cindex option summary
213 @cindex summary of options
214 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
215 see @ref{Invoking,,Command-Line Options}.
217 @c man title AS the portable GNU assembler.
221 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
225 @c We don't use deffn and friends for the following because they seem
226 @c to be limited to one line for the header.
228 @c man begin SYNOPSIS
229 @value{AS} [@b{-a}[@b{cdghlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
230 [@b{--compress-debug-sections}] [@b{--nocompress-debug-sections}]
231 [@b{--debug-prefix-map} @var{old}=@var{new}]
232 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
233 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--gdwarf-sections}]
234 [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
235 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
236 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
237 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}] [@b{-o}
238 @var{objfile}] [@b{-R}] [@b{--reduce-memory-overheads}] [@b{--statistics}]
239 [@b{-v}] [@b{-version}] [@b{--version}] [@b{-W}] [@b{--warn}]
240 [@b{--fatal-warnings}] [@b{-w}] [@b{-x}] [@b{-Z}] [@b{@@@var{FILE}}]
241 [@b{--size-check=[error|warning]}]
242 [@b{--target-help}] [@var{target-options}]
243 [@b{--}|@var{files} @dots{}]
245 @c Target dependent options are listed below. Keep the list sorted.
246 @c Add an empty line for separation.
249 @emph{Target AArch64 options:}
251 [@b{-mabi}=@var{ABI}]
255 @emph{Target Alpha options:}
257 [@b{-mdebug} | @b{-no-mdebug}]
258 [@b{-replace} | @b{-noreplace}]
259 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
260 [@b{-F}] [@b{-32addr}]
264 @emph{Target ARC options:}
270 @emph{Target ARM options:}
271 @c Don't document the deprecated options
272 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
273 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
274 [@b{-mfpu}=@var{floating-point-format}]
275 [@b{-mfloat-abi}=@var{abi}]
276 [@b{-meabi}=@var{ver}]
279 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
280 @b{-mapcs-reentrant}]
281 [@b{-mthumb-interwork}] [@b{-k}]
285 @emph{Target Blackfin options:}
286 [@b{-mcpu}=@var{processor}[-@var{sirevision}]]
293 @emph{Target CRIS options:}
294 [@b{--underscore} | @b{--no-underscore}]
296 [@b{--emulation=criself} | @b{--emulation=crisaout}]
297 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
298 @c Deprecated -- deliberately not documented.
303 @emph{Target D10V options:}
308 @emph{Target D30V options:}
309 [@b{-O}|@b{-n}|@b{-N}]
313 @emph{Target EPIPHANY options:}
314 [@b{-mepiphany}|@b{-mepiphany16}]
318 @emph{Target H8/300 options:}
322 @c HPPA has no machine-dependent assembler options (yet).
326 @emph{Target i386 options:}
327 [@b{--32}|@b{--x32}|@b{--64}] [@b{-n}]
328 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
332 @emph{Target i960 options:}
333 @c see md_parse_option in tc-i960.c
334 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
336 [@b{-b}] [@b{-no-relax}]
340 @emph{Target IA-64 options:}
341 [@b{-mconstant-gp}|@b{-mauto-pic}]
342 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
344 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
345 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
346 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
347 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
351 @emph{Target IP2K options:}
352 [@b{-mip2022}|@b{-mip2022ext}]
356 @emph{Target M32C options:}
357 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
361 @emph{Target M32R options:}
362 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
367 @emph{Target M680X0 options:}
368 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
372 @emph{Target M68HC11 options:}
373 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}|@b{-mm9s12x}|@b{-mm9s12xg}]
374 [@b{-mshort}|@b{-mlong}]
375 [@b{-mshort-double}|@b{-mlong-double}]
376 [@b{--force-long-branches}] [@b{--short-branches}]
377 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
378 [@b{--print-opcodes}] [@b{--generate-example}]
382 @emph{Target MCORE options:}
383 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
384 [@b{-mcpu=[210|340]}]
388 @emph{Target Meta options:}
389 [@b{-mcpu=@var{cpu}}] [@b{-mfpu=@var{cpu}}] [@b{-mdsp=@var{cpu}}]
392 @emph{Target MICROBLAZE options:}
393 @c MicroBlaze has no machine-dependent assembler options.
397 @emph{Target MIPS options:}
398 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
399 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
400 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
401 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
402 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
403 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
404 [@b{-mips32r3}] [@b{-mips32r5}] [@b{-mips64}] [@b{-mips64r2}]
405 [@b{-mips64r3}] [@b{-mips64r5}]
406 [@b{-construct-floats}] [@b{-no-construct-floats}]
407 [@b{-mnan=@var{encoding}}]
408 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
409 [@b{-mips16}] [@b{-no-mips16}]
410 [@b{-mmicromips}] [@b{-mno-micromips}]
411 [@b{-msmartmips}] [@b{-mno-smartmips}]
412 [@b{-mips3d}] [@b{-no-mips3d}]
413 [@b{-mdmx}] [@b{-no-mdmx}]
414 [@b{-mdsp}] [@b{-mno-dsp}]
415 [@b{-mdspr2}] [@b{-mno-dspr2}]
416 [@b{-mmsa}] [@b{-mno-msa}]
417 [@b{-mxpa}] [@b{-mno-xpa}]
418 [@b{-mmt}] [@b{-mno-mt}]
419 [@b{-mmcu}] [@b{-mno-mcu}]
420 [@b{-minsn32}] [@b{-mno-insn32}]
421 [@b{-mfix7000}] [@b{-mno-fix7000}]
422 [@b{-mfix-rm7000}] [@b{-mno-fix-rm7000}]
423 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
424 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
425 [@b{-mdebug}] [@b{-no-mdebug}]
426 [@b{-mpdr}] [@b{-mno-pdr}]
430 @emph{Target MMIX options:}
431 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
432 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
433 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
434 [@b{--linker-allocated-gregs}]
438 @emph{Target Nios II options:}
439 [@b{-relax-all}] [@b{-relax-section}] [@b{-no-relax}]
444 @emph{Target NDS32 options:}
445 [@b{-EL}] [@b{-EB}] [@b{-O}] [@b{-Os}] [@b{-mcpu=@var{cpu}}]
446 [@b{-misa=@var{isa}}] [@b{-mabi=@var{abi}}] [@b{-mall-ext}]
447 [@b{-m[no-]16-bit}] [@b{-m[no-]perf-ext}] [@b{-m[no-]perf2-ext}]
448 [@b{-m[no-]string-ext}] [@b{-m[no-]dsp-ext}] [@b{-m[no-]mac}] [@b{-m[no-]div}]
449 [@b{-m[no-]audio-isa-ext}] [@b{-m[no-]fpu-sp-ext}] [@b{-m[no-]fpu-dp-ext}]
450 [@b{-m[no-]fpu-fma}] [@b{-mfpu-freg=@var{FREG}}] [@b{-mreduced-regs}]
451 [@b{-mfull-regs}] [@b{-m[no-]dx-regs}] [@b{-mpic}] [@b{-mno-relax}]
456 @emph{Target PDP11 options:}
457 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
458 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
459 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
463 @emph{Target picoJava options:}
468 @emph{Target PowerPC options:}
470 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|@b{-m403}|@b{-m405}|
471 @b{-m440}|@b{-m464}|@b{-m476}|@b{-m7400}|@b{-m7410}|@b{-m7450}|@b{-m7455}|@b{-m750cl}|@b{-mppc64}|
472 @b{-m620}|@b{-me500}|@b{-e500x2}|@b{-me500mc}|@b{-me500mc64}|@b{-me5500}|@b{-me6500}|@b{-mppc64bridge}|
473 @b{-mbooke}|@b{-mpower4}|@b{-mpwr4}|@b{-mpower5}|@b{-mpwr5}|@b{-mpwr5x}|@b{-mpower6}|@b{-mpwr6}|
474 @b{-mpower7}|@b{-mpwr7}|@b{-mpower8}|@b{-mpwr8}|@b{-ma2}|@b{-mcell}|@b{-mspe}|@b{-mtitan}|@b{-me300}|@b{-mcom}]
475 [@b{-many}] [@b{-maltivec}|@b{-mvsx}|@b{-mhtm}|@b{-mvle}]
476 [@b{-mregnames}|@b{-mno-regnames}]
477 [@b{-mrelocatable}|@b{-mrelocatable-lib}|@b{-K PIC}] [@b{-memb}]
478 [@b{-mlittle}|@b{-mlittle-endian}|@b{-le}|@b{-mbig}|@b{-mbig-endian}|@b{-be}]
479 [@b{-msolaris}|@b{-mno-solaris}]
480 [@b{-nops=@var{count}}]
484 @emph{Target RL78 options:}
486 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
490 @emph{Target RX options:}
491 [@b{-mlittle-endian}|@b{-mbig-endian}]
492 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
493 [@b{-muse-conventional-section-names}]
494 [@b{-msmall-data-limit}]
497 [@b{-mint-register=@var{number}}]
498 [@b{-mgcc-abi}|@b{-mrx-abi}]
502 @emph{Target s390 options:}
503 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
504 [@b{-mregnames}|@b{-mno-regnames}]
505 [@b{-mwarn-areg-zero}]
509 @emph{Target SCORE options:}
510 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
511 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
512 [@b{-march=score7}][@b{-march=score3}]
513 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
517 @emph{Target SPARC options:}
518 @c The order here is important. See c-sparc.texi.
519 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
520 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
521 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
526 @emph{Target TIC54X options:}
527 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
528 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
533 @emph{Target TIC6X options:}
534 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
535 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
536 [@b{-mpic}|@b{-mno-pic}]
540 @emph{Target TILE-Gx options:}
541 [@b{-m32}|@b{-m64}][@b{-EB}][@b{-EL}]
544 @c TILEPro has no machine-dependent assembler options
549 @emph{Target Xtensa options:}
550 [@b{--[no-]text-section-literals}] [@b{--[no-]absolute-literals}]
551 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
552 [@b{--[no-]transform}]
553 [@b{--rename-section} @var{oldname}=@var{newname}]
554 [@b{--[no-]trampolines}]
559 @emph{Target Z80 options:}
560 [@b{-z80}] [@b{-r800}]
561 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
562 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
563 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
564 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
565 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
566 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
570 @c Z8000 has no machine-dependent assembler options
579 @include at-file.texi
582 Turn on listings, in any of a variety of ways:
586 omit false conditionals
589 omit debugging directives
592 include general information, like @value{AS} version and options passed
595 include high-level source
601 include macro expansions
604 omit forms processing
610 set the name of the listing file
613 You may combine these options; for example, use @samp{-aln} for assembly
614 listing without forms processing. The @samp{=file} option, if used, must be
615 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
618 Begin in alternate macro mode.
620 @xref{Altmacro,,@code{.altmacro}}.
623 @item --compress-debug-sections
624 Compress DWARF debug sections using zlib. The debug sections are renamed
625 to begin with @samp{.zdebug}, and the resulting object file may not be
626 compatible with older linkers and object file utilities.
628 @item --nocompress-debug-sections
629 Do not compress DWARF debug sections. This is the default.
632 Ignored. This option is accepted for script compatibility with calls to
635 @item --debug-prefix-map @var{old}=@var{new}
636 When assembling files in directory @file{@var{old}}, record debugging
637 information describing them as in @file{@var{new}} instead.
639 @item --defsym @var{sym}=@var{value}
640 Define the symbol @var{sym} to be @var{value} before assembling the input file.
641 @var{value} must be an integer constant. As in C, a leading @samp{0x}
642 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
643 value. The value of the symbol can be overridden inside a source file via the
644 use of a @code{.set} pseudo-op.
647 ``fast''---skip whitespace and comment preprocessing (assume source is
652 Generate debugging information for each assembler source line using whichever
653 debug format is preferred by the target. This currently means either STABS,
657 Generate stabs debugging information for each assembler line. This
658 may help debugging assembler code, if the debugger can handle it.
661 Generate stabs debugging information for each assembler line, with GNU
662 extensions that probably only gdb can handle, and that could make other
663 debuggers crash or refuse to read your program. This
664 may help debugging assembler code. Currently the only GNU extension is
665 the location of the current working directory at assembling time.
668 Generate DWARF2 debugging information for each assembler line. This
669 may help debugging assembler code, if the debugger can handle it. Note---this
670 option is only supported by some targets, not all of them.
672 @item --gdwarf-sections
673 Instead of creating a .debug_line section, create a series of
674 .debug_line.@var{foo} sections where @var{foo} is the name of the
675 corresponding code section. For example a code section called @var{.text.func}
676 will have its dwarf line number information placed into a section called
677 @var{.debug_line.text.func}. If the code section is just called @var{.text}
678 then debug line section will still be called just @var{.debug_line} without any
681 @item --size-check=error
682 @itemx --size-check=warning
683 Issue an error or warning for invalid ELF .size directive.
686 Print a summary of the command line options and exit.
689 Print a summary of all target specific options and exit.
692 Add directory @var{dir} to the search list for @code{.include} directives.
695 Don't warn about signed overflow.
698 @ifclear DIFF-TBL-KLUGE
699 This option is accepted but has no effect on the @value{TARGET} family.
701 @ifset DIFF-TBL-KLUGE
702 Issue warnings when difference tables altered for long displacements.
707 Keep (in the symbol table) local symbols. These symbols start with
708 system-specific local label prefixes, typically @samp{.L} for ELF systems
709 or @samp{L} for traditional a.out systems.
714 @item --listing-lhs-width=@var{number}
715 Set the maximum width, in words, of the output data column for an assembler
716 listing to @var{number}.
718 @item --listing-lhs-width2=@var{number}
719 Set the maximum width, in words, of the output data column for continuation
720 lines in an assembler listing to @var{number}.
722 @item --listing-rhs-width=@var{number}
723 Set the maximum width of an input source line, as displayed in a listing, to
726 @item --listing-cont-lines=@var{number}
727 Set the maximum number of lines printed in a listing for a single line of input
730 @item -o @var{objfile}
731 Name the object-file output from @command{@value{AS}} @var{objfile}.
734 Fold the data section into the text section.
736 @kindex --hash-size=@var{number}
737 Set the default size of GAS's hash tables to a prime number close to
738 @var{number}. Increasing this value can reduce the length of time it takes the
739 assembler to perform its tasks, at the expense of increasing the assembler's
740 memory requirements. Similarly reducing this value can reduce the memory
741 requirements at the expense of speed.
743 @item --reduce-memory-overheads
744 This option reduces GAS's memory requirements, at the expense of making the
745 assembly processes slower. Currently this switch is a synonym for
746 @samp{--hash-size=4051}, but in the future it may have other effects as well.
749 Print the maximum space (in bytes) and total time (in seconds) used by
752 @item --strip-local-absolute
753 Remove local absolute symbols from the outgoing symbol table.
757 Print the @command{as} version.
760 Print the @command{as} version and exit.
764 Suppress warning messages.
766 @item --fatal-warnings
767 Treat warnings as errors.
770 Don't suppress warning messages or treat them as errors.
779 Generate an object file even after errors.
781 @item -- | @var{files} @dots{}
782 Standard input, or source files to assemble.
790 @xref{AArch64 Options}, for the options available when @value{AS} is configured
791 for the 64-bit mode of the ARM Architecture (AArch64).
796 The following options are available when @value{AS} is configured for the
797 64-bit mode of the ARM Architecture (AArch64).
800 @include c-aarch64.texi
801 @c ended inside the included file
809 @xref{Alpha Options}, for the options available when @value{AS} is configured
810 for an Alpha processor.
815 The following options are available when @value{AS} is configured for an Alpha
819 @include c-alpha.texi
820 @c ended inside the included file
827 The following options are available when @value{AS} is configured for
832 This option selects the core processor variant.
834 Select either big-endian (-EB) or little-endian (-EL) output.
839 The following options are available when @value{AS} is configured for the ARM
843 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
844 Specify which ARM processor variant is the target.
845 @item -march=@var{architecture}[+@var{extension}@dots{}]
846 Specify which ARM architecture variant is used by the target.
847 @item -mfpu=@var{floating-point-format}
848 Select which Floating Point architecture is the target.
849 @item -mfloat-abi=@var{abi}
850 Select which floating point ABI is in use.
852 Enable Thumb only instruction decoding.
853 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
854 Select which procedure calling convention is in use.
856 Select either big-endian (-EB) or little-endian (-EL) output.
857 @item -mthumb-interwork
858 Specify that the code has been generated with interworking between Thumb and
861 Turns on CodeComposer Studio assembly syntax compatibility mode.
863 Specify that PIC code has been generated.
871 @xref{Blackfin Options}, for the options available when @value{AS} is
872 configured for the Blackfin processor family.
877 The following options are available when @value{AS} is configured for
878 the Blackfin processor family.
882 @c ended inside the included file
889 See the info pages for documentation of the CRIS-specific options.
893 The following options are available when @value{AS} is configured for
896 @cindex D10V optimization
897 @cindex optimization, D10V
899 Optimize output by parallelizing instructions.
904 The following options are available when @value{AS} is configured for a D30V
907 @cindex D30V optimization
908 @cindex optimization, D30V
910 Optimize output by parallelizing instructions.
914 Warn when nops are generated.
916 @cindex D30V nops after 32-bit multiply
918 Warn when a nop after a 32-bit multiply instruction is generated.
924 The following options are available when @value{AS} is configured for the
925 Adapteva EPIPHANY series.
928 @xref{Epiphany Options}, for the options available when @value{AS} is
929 configured for an Epiphany processor.
934 The following options are available when @value{AS} is configured for
935 an Epiphany processor.
938 @include c-epiphany.texi
939 @c ended inside the included file
947 @xref{H8/300 Options}, for the options available when @value{AS} is configured
948 for an H8/300 processor.
953 The following options are available when @value{AS} is configured for an H8/300
957 @include c-h8300.texi
958 @c ended inside the included file
966 @xref{i386-Options}, for the options available when @value{AS} is
967 configured for an i386 processor.
972 The following options are available when @value{AS} is configured for
977 @c ended inside the included file
984 The following options are available when @value{AS} is configured for the
985 Intel 80960 processor.
988 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
989 Specify which variant of the 960 architecture is the target.
992 Add code to collect statistics about branches taken.
995 Do not alter compare-and-branch instructions for long displacements;
1002 The following options are available when @value{AS} is configured for the
1008 Specifies that the extended IP2022 instructions are allowed.
1011 Restores the default behaviour, which restricts the permitted instructions to
1012 just the basic IP2022 ones.
1018 The following options are available when @value{AS} is configured for the
1019 Renesas M32C and M16C processors.
1024 Assemble M32C instructions.
1027 Assemble M16C instructions (the default).
1030 Enable support for link-time relaxations.
1033 Support H'00 style hex constants in addition to 0x00 style.
1039 The following options are available when @value{AS} is configured for the
1040 Renesas M32R (formerly Mitsubishi M32R) series.
1045 Specify which processor in the M32R family is the target. The default
1046 is normally the M32R, but this option changes it to the M32RX.
1048 @item --warn-explicit-parallel-conflicts or --Wp
1049 Produce warning messages when questionable parallel constructs are
1052 @item --no-warn-explicit-parallel-conflicts or --Wnp
1053 Do not produce warning messages when questionable parallel constructs are
1060 The following options are available when @value{AS} is configured for the
1061 Motorola 68000 series.
1066 Shorten references to undefined symbols, to one word instead of two.
1068 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1069 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1070 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1071 Specify what processor in the 68000 family is the target. The default
1072 is normally the 68020, but this can be changed at configuration time.
1074 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1075 The target machine does (or does not) have a floating-point coprocessor.
1076 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1077 the basic 68000 is not compatible with the 68881, a combination of the
1078 two can be specified, since it's possible to do emulation of the
1079 coprocessor instructions with the main processor.
1081 @item -m68851 | -mno-68851
1082 The target machine does (or does not) have a memory-management
1083 unit coprocessor. The default is to assume an MMU for 68020 and up.
1091 @xref{Nios II Options}, for the options available when @value{AS} is configured
1092 for an Altera Nios II processor.
1096 @c man begin OPTIONS
1097 The following options are available when @value{AS} is configured for an
1098 Altera Nios II processor.
1100 @c man begin INCLUDE
1101 @include c-nios2.texi
1102 @c ended inside the included file
1108 For details about the PDP-11 machine dependent features options,
1109 see @ref{PDP-11-Options}.
1112 @item -mpic | -mno-pic
1113 Generate position-independent (or position-dependent) code. The
1114 default is @option{-mpic}.
1117 @itemx -mall-extensions
1118 Enable all instruction set extensions. This is the default.
1120 @item -mno-extensions
1121 Disable all instruction set extensions.
1123 @item -m@var{extension} | -mno-@var{extension}
1124 Enable (or disable) a particular instruction set extension.
1127 Enable the instruction set extensions supported by a particular CPU, and
1128 disable all other extensions.
1130 @item -m@var{machine}
1131 Enable the instruction set extensions supported by a particular machine
1132 model, and disable all other extensions.
1138 The following options are available when @value{AS} is configured for
1139 a picoJava processor.
1143 @cindex PJ endianness
1144 @cindex endianness, PJ
1145 @cindex big endian output, PJ
1147 Generate ``big endian'' format output.
1149 @cindex little endian output, PJ
1151 Generate ``little endian'' format output.
1157 The following options are available when @value{AS} is configured for the
1158 Motorola 68HC11 or 68HC12 series.
1162 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1163 Specify what processor is the target. The default is
1164 defined by the configuration option when building the assembler.
1166 @item --xgate-ramoffset
1167 Instruct the linker to offset RAM addresses from S12X address space into
1168 XGATE address space.
1171 Specify to use the 16-bit integer ABI.
1174 Specify to use the 32-bit integer ABI.
1176 @item -mshort-double
1177 Specify to use the 32-bit double ABI.
1180 Specify to use the 64-bit double ABI.
1182 @item --force-long-branches
1183 Relative branches are turned into absolute ones. This concerns
1184 conditional branches, unconditional branches and branches to a
1187 @item -S | --short-branches
1188 Do not turn relative branches into absolute ones
1189 when the offset is out of range.
1191 @item --strict-direct-mode
1192 Do not turn the direct addressing mode into extended addressing mode
1193 when the instruction does not support direct addressing mode.
1195 @item --print-insn-syntax
1196 Print the syntax of instruction in case of error.
1198 @item --print-opcodes
1199 Print the list of instructions with syntax and then exit.
1201 @item --generate-example
1202 Print an example of instruction for each possible instruction and then exit.
1203 This option is only useful for testing @command{@value{AS}}.
1209 The following options are available when @command{@value{AS}} is configured
1210 for the SPARC architecture:
1213 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1214 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1215 Explicitly select a variant of the SPARC architecture.
1217 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1218 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1220 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1221 UltraSPARC extensions.
1223 @item -xarch=v8plus | -xarch=v8plusa
1224 For compatibility with the Solaris v9 assembler. These options are
1225 equivalent to -Av8plus and -Av8plusa, respectively.
1228 Warn when the assembler switches to another architecture.
1233 The following options are available when @value{AS} is configured for the 'c54x
1238 Enable extended addressing mode. All addresses and relocations will assume
1239 extended addressing (usually 23 bits).
1240 @item -mcpu=@var{CPU_VERSION}
1241 Sets the CPU version being compiled for.
1242 @item -merrors-to-file @var{FILENAME}
1243 Redirect error output to a file, for broken systems which don't support such
1244 behaviour in the shell.
1249 The following options are available when @value{AS} is configured for
1254 This option sets the largest size of an object that can be referenced
1255 implicitly with the @code{gp} register. It is only accepted for targets that
1256 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1258 @cindex MIPS endianness
1259 @cindex endianness, MIPS
1260 @cindex big endian output, MIPS
1262 Generate ``big endian'' format output.
1264 @cindex little endian output, MIPS
1266 Generate ``little endian'' format output.
1282 Generate code for a particular MIPS Instruction Set Architecture level.
1283 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1284 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1285 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1286 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1287 @samp{-mips32r5}, @samp{-mips64}, @samp{-mips64r2}, @samp{-mips64r3}, and
1288 @samp{-mips64r5} correspond to generic MIPS V, MIPS32, MIPS32 Release 2,
1289 MIPS32 Release 3, MIPS32 Release 5, MIPS64, MIPS64 Release 2,
1290 MIPS64 Release 3, and MIPS64 Release 5 ISA processors, respectively.
1292 @item -march=@var{cpu}
1293 Generate code for a particular MIPS CPU.
1295 @item -mtune=@var{cpu}
1296 Schedule and tune for a particular MIPS CPU.
1300 Cause nops to be inserted if the read of the destination register
1301 of an mfhi or mflo instruction occurs in the following two instructions.
1304 @itemx -mno-fix-rm7000
1305 Cause nops to be inserted if a dmult or dmultu instruction is
1306 followed by a load instruction.
1310 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1311 section instead of the standard ELF .stabs sections.
1315 Control generation of @code{.pdr} sections.
1319 The register sizes are normally inferred from the ISA and ABI, but these
1320 flags force a certain group of registers to be treated as 32 bits wide at
1321 all times. @samp{-mgp32} controls the size of general-purpose registers
1322 and @samp{-mfp32} controls the size of floating-point registers.
1326 Generate code for the MIPS 16 processor. This is equivalent to putting
1327 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1328 turns off this option.
1331 @itemx -mno-micromips
1332 Generate code for the microMIPS processor. This is equivalent to putting
1333 @code{.set micromips} at the start of the assembly file. @samp{-mno-micromips}
1334 turns off this option. This is equivalent to putting @code{.set nomicromips}
1335 at the start of the assembly file.
1338 @itemx -mno-smartmips
1339 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1340 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1341 @samp{-mno-smartmips} turns off this option.
1345 Generate code for the MIPS-3D Application Specific Extension.
1346 This tells the assembler to accept MIPS-3D instructions.
1347 @samp{-no-mips3d} turns off this option.
1351 Generate code for the MDMX Application Specific Extension.
1352 This tells the assembler to accept MDMX instructions.
1353 @samp{-no-mdmx} turns off this option.
1357 Generate code for the DSP Release 1 Application Specific Extension.
1358 This tells the assembler to accept DSP Release 1 instructions.
1359 @samp{-mno-dsp} turns off this option.
1363 Generate code for the DSP Release 2 Application Specific Extension.
1364 This option implies -mdsp.
1365 This tells the assembler to accept DSP Release 2 instructions.
1366 @samp{-mno-dspr2} turns off this option.
1370 Generate code for the MIPS SIMD Architecture Extension.
1371 This tells the assembler to accept MSA instructions.
1372 @samp{-mno-msa} turns off this option.
1376 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1377 This tells the assembler to accept XPA instructions.
1378 @samp{-mno-xpa} turns off this option.
1382 Generate code for the MT Application Specific Extension.
1383 This tells the assembler to accept MT instructions.
1384 @samp{-mno-mt} turns off this option.
1388 Generate code for the MCU Application Specific Extension.
1389 This tells the assembler to accept MCU instructions.
1390 @samp{-mno-mcu} turns off this option.
1394 Only use 32-bit instruction encodings when generating code for the
1395 microMIPS processor. This option inhibits the use of any 16-bit
1396 instructions. This is equivalent to putting @code{.set insn32} at
1397 the start of the assembly file. @samp{-mno-insn32} turns off this
1398 option. This is equivalent to putting @code{.set noinsn32} at the
1399 start of the assembly file. By default @samp{-mno-insn32} is
1400 selected, allowing all instructions to be used.
1402 @item --construct-floats
1403 @itemx --no-construct-floats
1404 The @samp{--no-construct-floats} option disables the construction of
1405 double width floating point constants by loading the two halves of the
1406 value into the two single width floating point registers that make up
1407 the double width register. By default @samp{--construct-floats} is
1408 selected, allowing construction of these floating point constants.
1410 @item --relax-branch
1411 @itemx --no-relax-branch
1412 The @samp{--relax-branch} option enables the relaxation of out-of-range
1413 branches. By default @samp{--no-relax-branch} is selected, causing any
1414 out-of-range branches to produce an error.
1416 @item -mnan=@var{encoding}
1417 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1418 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1421 @item --emulation=@var{name}
1422 This option was formerly used to switch between ELF and ECOFF output
1423 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1424 removed in GAS 2.24, so the option now serves little purpose.
1425 It is retained for backwards compatibility.
1427 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1428 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1429 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1430 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1431 preferred options instead.
1434 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1441 Control how to deal with multiplication overflow and division by zero.
1442 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1443 (and only work for Instruction Set Architecture level 2 and higher);
1444 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1448 When this option is used, @command{@value{AS}} will issue a warning every
1449 time it generates a nop instruction from a macro.
1454 The following options are available when @value{AS} is configured for
1460 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1461 The command line option @samp{-nojsri2bsr} can be used to disable it.
1465 Enable or disable the silicon filter behaviour. By default this is disabled.
1466 The default can be overridden by the @samp{-sifilter} command line option.
1469 Alter jump instructions for long displacements.
1471 @item -mcpu=[210|340]
1472 Select the cpu type on the target hardware. This controls which instructions
1476 Assemble for a big endian target.
1479 Assemble for a little endian target.
1488 @xref{Meta Options}, for the options available when @value{AS} is configured
1489 for a Meta processor.
1493 @c man begin OPTIONS
1494 The following options are available when @value{AS} is configured for a
1497 @c man begin INCLUDE
1498 @include c-metag.texi
1499 @c ended inside the included file
1504 @c man begin OPTIONS
1506 See the info pages for documentation of the MMIX-specific options.
1512 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1513 for a NDS32 processor.
1515 @c ended inside the included file
1519 @c man begin OPTIONS
1520 The following options are available when @value{AS} is configured for a
1523 @c man begin INCLUDE
1524 @include c-nds32.texi
1525 @c ended inside the included file
1532 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1533 for a PowerPC processor.
1537 @c man begin OPTIONS
1538 The following options are available when @value{AS} is configured for a
1541 @c man begin INCLUDE
1543 @c ended inside the included file
1548 @c man begin OPTIONS
1550 See the info pages for documentation of the RX-specific options.
1554 The following options are available when @value{AS} is configured for the s390
1560 Select the word size, either 31/32 bits or 64 bits.
1563 Select the architecture mode, either the Enterprise System
1564 Architecture (esa) or the z/Architecture mode (zarch).
1565 @item -march=@var{processor}
1566 Specify which s390 processor variant is the target, @samp{g6}, @samp{g6},
1567 @samp{z900}, @samp{z990}, @samp{z9-109}, @samp{z9-ec}, @samp{z10},
1568 @samp{z196}, or @samp{zEC12}.
1570 @itemx -mno-regnames
1571 Allow or disallow symbolic names for registers.
1572 @item -mwarn-areg-zero
1573 Warn whenever the operand for a base or index register has been specified
1574 but evaluates to zero.
1582 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1583 for a TMS320C6000 processor.
1587 @c man begin OPTIONS
1588 The following options are available when @value{AS} is configured for a
1589 TMS320C6000 processor.
1591 @c man begin INCLUDE
1592 @include c-tic6x.texi
1593 @c ended inside the included file
1601 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1602 for a TILE-Gx processor.
1606 @c man begin OPTIONS
1607 The following options are available when @value{AS} is configured for a TILE-Gx
1610 @c man begin INCLUDE
1611 @include c-tilegx.texi
1612 @c ended inside the included file
1620 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1621 for an Xtensa processor.
1625 @c man begin OPTIONS
1626 The following options are available when @value{AS} is configured for an
1629 @c man begin INCLUDE
1630 @include c-xtensa.texi
1631 @c ended inside the included file
1636 @c man begin OPTIONS
1639 The following options are available when @value{AS} is configured for
1640 a Z80 family processor.
1643 Assemble for Z80 processor.
1645 Assemble for R800 processor.
1646 @item -ignore-undocumented-instructions
1648 Assemble undocumented Z80 instructions that also work on R800 without warning.
1649 @item -ignore-unportable-instructions
1651 Assemble all undocumented Z80 instructions without warning.
1652 @item -warn-undocumented-instructions
1654 Issue a warning for undocumented Z80 instructions that also work on R800.
1655 @item -warn-unportable-instructions
1657 Issue a warning for undocumented Z80 instructions that do not work on R800.
1658 @item -forbid-undocumented-instructions
1660 Treat all undocumented instructions as errors.
1661 @item -forbid-unportable-instructions
1663 Treat undocumented Z80 instructions that do not work on R800 as errors.
1670 * Manual:: Structure of this Manual
1671 * GNU Assembler:: The GNU Assembler
1672 * Object Formats:: Object File Formats
1673 * Command Line:: Command Line
1674 * Input Files:: Input Files
1675 * Object:: Output (Object) File
1676 * Errors:: Error and Warning Messages
1680 @section Structure of this Manual
1682 @cindex manual, structure and purpose
1683 This manual is intended to describe what you need to know to use
1684 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1685 notation for symbols, constants, and expressions; the directives that
1686 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1689 We also cover special features in the @value{TARGET}
1690 configuration of @command{@value{AS}}, including assembler directives.
1693 This manual also describes some of the machine-dependent features of
1694 various flavors of the assembler.
1697 @cindex machine instructions (not covered)
1698 On the other hand, this manual is @emph{not} intended as an introduction
1699 to programming in assembly language---let alone programming in general!
1700 In a similar vein, we make no attempt to introduce the machine
1701 architecture; we do @emph{not} describe the instruction set, standard
1702 mnemonics, registers or addressing modes that are standard to a
1703 particular architecture.
1705 You may want to consult the manufacturer's
1706 machine architecture manual for this information.
1710 For information on the H8/300 machine instruction set, see @cite{H8/300
1711 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1712 Programming Manual} (Renesas).
1715 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1716 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1717 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1718 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1721 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1725 @c I think this is premature---doc@cygnus.com, 17jan1991
1727 Throughout this manual, we assume that you are running @dfn{GNU},
1728 the portable operating system from the @dfn{Free Software
1729 Foundation, Inc.}. This restricts our attention to certain kinds of
1730 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1731 once this assumption is granted examples and definitions need less
1734 @command{@value{AS}} is part of a team of programs that turn a high-level
1735 human-readable series of instructions into a low-level
1736 computer-readable series of instructions. Different versions of
1737 @command{@value{AS}} are used for different kinds of computer.
1740 @c There used to be a section "Terminology" here, which defined
1741 @c "contents", "byte", "word", and "long". Defining "word" to any
1742 @c particular size is confusing when the .word directive may generate 16
1743 @c bits on one machine and 32 bits on another; in general, for the user
1744 @c version of this manual, none of these terms seem essential to define.
1745 @c They were used very little even in the former draft of the manual;
1746 @c this draft makes an effort to avoid them (except in names of
1750 @section The GNU Assembler
1752 @c man begin DESCRIPTION
1754 @sc{gnu} @command{as} is really a family of assemblers.
1756 This manual describes @command{@value{AS}}, a member of that family which is
1757 configured for the @value{TARGET} architectures.
1759 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1760 should find a fairly similar environment when you use it on another
1761 architecture. Each version has much in common with the others,
1762 including object file formats, most assembler directives (often called
1763 @dfn{pseudo-ops}) and assembler syntax.@refill
1765 @cindex purpose of @sc{gnu} assembler
1766 @command{@value{AS}} is primarily intended to assemble the output of the
1767 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1768 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1769 assemble correctly everything that other assemblers for the same
1770 machine would assemble.
1772 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1775 @c This remark should appear in generic version of manual; assumption
1776 @c here is that generic version sets M680x0.
1777 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1778 assembler for the same architecture; for example, we know of several
1779 incompatible versions of 680x0 assembly language syntax.
1784 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1785 program in one pass of the source file. This has a subtle impact on the
1786 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1788 @node Object Formats
1789 @section Object File Formats
1791 @cindex object file format
1792 The @sc{gnu} assembler can be configured to produce several alternative
1793 object file formats. For the most part, this does not affect how you
1794 write assembly language programs; but directives for debugging symbols
1795 are typically different in different file formats. @xref{Symbol
1796 Attributes,,Symbol Attributes}.
1799 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1800 @value{OBJ-NAME} format object files.
1802 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1804 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1805 @code{b.out} or COFF format object files.
1808 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1809 SOM or ELF format object files.
1814 @section Command Line
1816 @cindex command line conventions
1818 After the program name @command{@value{AS}}, the command line may contain
1819 options and file names. Options may appear in any order, and may be
1820 before, after, or between file names. The order of file names is
1823 @cindex standard input, as input file
1825 @file{--} (two hyphens) by itself names the standard input file
1826 explicitly, as one of the files for @command{@value{AS}} to assemble.
1828 @cindex options, command line
1829 Except for @samp{--} any command line argument that begins with a
1830 hyphen (@samp{-}) is an option. Each option changes the behavior of
1831 @command{@value{AS}}. No option changes the way another option works. An
1832 option is a @samp{-} followed by one or more letters; the case of
1833 the letter is important. All options are optional.
1835 Some options expect exactly one file name to follow them. The file
1836 name may either immediately follow the option's letter (compatible
1837 with older assemblers) or it may be the next command argument (@sc{gnu}
1838 standard). These two command lines are equivalent:
1841 @value{AS} -o my-object-file.o mumble.s
1842 @value{AS} -omy-object-file.o mumble.s
1846 @section Input Files
1849 @cindex source program
1850 @cindex files, input
1851 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1852 describe the program input to one run of @command{@value{AS}}. The program may
1853 be in one or more files; how the source is partitioned into files
1854 doesn't change the meaning of the source.
1856 @c I added "con" prefix to "catenation" just to prove I can overcome my
1857 @c APL training... doc@cygnus.com
1858 The source program is a concatenation of the text in all the files, in the
1861 @c man begin DESCRIPTION
1862 Each time you run @command{@value{AS}} it assembles exactly one source
1863 program. The source program is made up of one or more files.
1864 (The standard input is also a file.)
1866 You give @command{@value{AS}} a command line that has zero or more input file
1867 names. The input files are read (from left file name to right). A
1868 command line argument (in any position) that has no special meaning
1869 is taken to be an input file name.
1871 If you give @command{@value{AS}} no file names it attempts to read one input file
1872 from the @command{@value{AS}} standard input, which is normally your terminal. You
1873 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1876 Use @samp{--} if you need to explicitly name the standard input file
1877 in your command line.
1879 If the source is empty, @command{@value{AS}} produces a small, empty object
1884 @subheading Filenames and Line-numbers
1886 @cindex input file linenumbers
1887 @cindex line numbers, in input files
1888 There are two ways of locating a line in the input file (or files) and
1889 either may be used in reporting error messages. One way refers to a line
1890 number in a physical file; the other refers to a line number in a
1891 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1893 @dfn{Physical files} are those files named in the command line given
1894 to @command{@value{AS}}.
1896 @dfn{Logical files} are simply names declared explicitly by assembler
1897 directives; they bear no relation to physical files. Logical file names help
1898 error messages reflect the original source file, when @command{@value{AS}} source
1899 is itself synthesized from other files. @command{@value{AS}} understands the
1900 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1901 @ref{File,,@code{.file}}.
1904 @section Output (Object) File
1910 Every time you run @command{@value{AS}} it produces an output file, which is
1911 your assembly language program translated into numbers. This file
1912 is the object file. Its default name is
1920 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1922 You can give it another name by using the @option{-o} option. Conventionally,
1923 object file names end with @file{.o}. The default name is used for historical
1924 reasons: older assemblers were capable of assembling self-contained programs
1925 directly into a runnable program. (For some formats, this isn't currently
1926 possible, but it can be done for the @code{a.out} format.)
1930 The object file is meant for input to the linker @code{@value{LD}}. It contains
1931 assembled program code, information to help @code{@value{LD}} integrate
1932 the assembled program into a runnable file, and (optionally) symbolic
1933 information for the debugger.
1935 @c link above to some info file(s) like the description of a.out.
1936 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1939 @section Error and Warning Messages
1941 @c man begin DESCRIPTION
1943 @cindex error messages
1944 @cindex warning messages
1945 @cindex messages from assembler
1946 @command{@value{AS}} may write warnings and error messages to the standard error
1947 file (usually your terminal). This should not happen when a compiler
1948 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1949 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1950 grave problem that stops the assembly.
1954 @cindex format of warning messages
1955 Warning messages have the format
1958 file_name:@b{NNN}:Warning Message Text
1962 @cindex line numbers, in warnings/errors
1963 (where @b{NNN} is a line number). If a logical file name has been given
1964 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1965 the current input file is used. If a logical line number was given
1967 (@pxref{Line,,@code{.line}})
1969 then it is used to calculate the number printed,
1970 otherwise the actual line in the current source file is printed. The
1971 message text is intended to be self explanatory (in the grand Unix
1974 @cindex format of error messages
1975 Error messages have the format
1977 file_name:@b{NNN}:FATAL:Error Message Text
1979 The file name and line number are derived as for warning
1980 messages. The actual message text may be rather less explanatory
1981 because many of them aren't supposed to happen.
1984 @chapter Command-Line Options
1986 @cindex options, all versions of assembler
1987 This chapter describes command-line options available in @emph{all}
1988 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
1989 for options specific
1991 to the @value{TARGET} target.
1994 to particular machine architectures.
1997 @c man begin DESCRIPTION
1999 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2000 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2001 The assembler arguments must be separated from each other (and the @samp{-Wa})
2002 by commas. For example:
2005 gcc -c -g -O -Wa,-alh,-L file.c
2009 This passes two options to the assembler: @samp{-alh} (emit a listing to
2010 standard output with high-level and assembly source) and @samp{-L} (retain
2011 local symbols in the symbol table).
2013 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2014 command-line options are automatically passed to the assembler by the compiler.
2015 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2016 precisely what options it passes to each compilation pass, including the
2022 * a:: -a[cdghlns] enable listings
2023 * alternate:: --alternate enable alternate macro syntax
2024 * D:: -D for compatibility
2025 * f:: -f to work faster
2026 * I:: -I for .include search path
2027 @ifclear DIFF-TBL-KLUGE
2028 * K:: -K for compatibility
2030 @ifset DIFF-TBL-KLUGE
2031 * K:: -K for difference tables
2034 * L:: -L to retain local symbols
2035 * listing:: --listing-XXX to configure listing output
2036 * M:: -M or --mri to assemble in MRI compatibility mode
2037 * MD:: --MD for dependency tracking
2038 * o:: -o to name the object file
2039 * R:: -R to join data and text sections
2040 * statistics:: --statistics to see statistics about assembly
2041 * traditional-format:: --traditional-format for compatible output
2042 * v:: -v to announce version
2043 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2044 * Z:: -Z to make object file even after errors
2048 @section Enable Listings: @option{-a[cdghlns]}
2058 @cindex listings, enabling
2059 @cindex assembly listings, enabling
2061 These options enable listing output from the assembler. By itself,
2062 @samp{-a} requests high-level, assembly, and symbols listing.
2063 You can use other letters to select specific options for the list:
2064 @samp{-ah} requests a high-level language listing,
2065 @samp{-al} requests an output-program assembly listing, and
2066 @samp{-as} requests a symbol table listing.
2067 High-level listings require that a compiler debugging option like
2068 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2071 Use the @samp{-ag} option to print a first section with general assembly
2072 information, like @value{AS} version, switches passed, or time stamp.
2074 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2075 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2076 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2077 omitted from the listing.
2079 Use the @samp{-ad} option to omit debugging directives from the
2082 Once you have specified one of these options, you can further control
2083 listing output and its appearance using the directives @code{.list},
2084 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2086 The @samp{-an} option turns off all forms processing.
2087 If you do not request listing output with one of the @samp{-a} options, the
2088 listing-control directives have no effect.
2090 The letters after @samp{-a} may be combined into one option,
2091 @emph{e.g.}, @samp{-aln}.
2093 Note if the assembler source is coming from the standard input (e.g.,
2095 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
2096 is being used) then the listing will not contain any comments or preprocessor
2097 directives. This is because the listing code buffers input source lines from
2098 stdin only after they have been preprocessed by the assembler. This reduces
2099 memory usage and makes the code more efficient.
2102 @section @option{--alternate}
2105 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2108 @section @option{-D}
2111 This option has no effect whatsoever, but it is accepted to make it more
2112 likely that scripts written for other assemblers also work with
2113 @command{@value{AS}}.
2116 @section Work Faster: @option{-f}
2119 @cindex trusted compiler
2120 @cindex faster processing (@option{-f})
2121 @samp{-f} should only be used when assembling programs written by a
2122 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2123 and comment preprocessing on
2124 the input file(s) before assembling them. @xref{Preprocessing,
2128 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2129 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2134 @section @code{.include} Search Path: @option{-I} @var{path}
2136 @kindex -I @var{path}
2137 @cindex paths for @code{.include}
2138 @cindex search path for @code{.include}
2139 @cindex @code{include} directive search path
2140 Use this option to add a @var{path} to the list of directories
2141 @command{@value{AS}} searches for files specified in @code{.include}
2142 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2143 many times as necessary to include a variety of paths. The current
2144 working directory is always searched first; after that, @command{@value{AS}}
2145 searches any @samp{-I} directories in the same order as they were
2146 specified (left to right) on the command line.
2149 @section Difference Tables: @option{-K}
2152 @ifclear DIFF-TBL-KLUGE
2153 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2154 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2155 where it can be used to warn when the assembler alters the machine code
2156 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2157 family does not have the addressing limitations that sometimes lead to this
2158 alteration on other platforms.
2161 @ifset DIFF-TBL-KLUGE
2162 @cindex difference tables, warning
2163 @cindex warning for altered difference tables
2164 @command{@value{AS}} sometimes alters the code emitted for directives of the
2165 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2166 You can use the @samp{-K} option if you want a warning issued when this
2171 @section Include Local Symbols: @option{-L}
2174 @cindex local symbols, retaining in output
2175 Symbols beginning with system-specific local label prefixes, typically
2176 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2177 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2178 such symbols when debugging, because they are intended for the use of
2179 programs (like compilers) that compose assembler programs, not for your
2180 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2181 such symbols, so you do not normally debug with them.
2183 This option tells @command{@value{AS}} to retain those local symbols
2184 in the object file. Usually if you do this you also tell the linker
2185 @code{@value{LD}} to preserve those symbols.
2188 @section Configuring listing output: @option{--listing}
2190 The listing feature of the assembler can be enabled via the command line switch
2191 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2192 hex dump of the corresponding locations in the output object file, and displays
2193 them as a listing file. The format of this listing can be controlled by
2194 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2195 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2196 @code{.psize} (@pxref{Psize}), and
2197 @code{.eject} (@pxref{Eject}) and also by the following switches:
2200 @item --listing-lhs-width=@samp{number}
2201 @kindex --listing-lhs-width
2202 @cindex Width of first line disassembly output
2203 Sets the maximum width, in words, of the first line of the hex byte dump. This
2204 dump appears on the left hand side of the listing output.
2206 @item --listing-lhs-width2=@samp{number}
2207 @kindex --listing-lhs-width2
2208 @cindex Width of continuation lines of disassembly output
2209 Sets the maximum width, in words, of any further lines of the hex byte dump for
2210 a given input source line. If this value is not specified, it defaults to being
2211 the same as the value specified for @samp{--listing-lhs-width}. If neither
2212 switch is used the default is to one.
2214 @item --listing-rhs-width=@samp{number}
2215 @kindex --listing-rhs-width
2216 @cindex Width of source line output
2217 Sets the maximum width, in characters, of the source line that is displayed
2218 alongside the hex dump. The default value for this parameter is 100. The
2219 source line is displayed on the right hand side of the listing output.
2221 @item --listing-cont-lines=@samp{number}
2222 @kindex --listing-cont-lines
2223 @cindex Maximum number of continuation lines
2224 Sets the maximum number of continuation lines of hex dump that will be
2225 displayed for a given single line of source input. The default value is 4.
2229 @section Assemble in MRI Compatibility Mode: @option{-M}
2232 @cindex MRI compatibility mode
2233 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2234 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2235 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
2236 configured target) assembler from Microtec Research. The exact nature of the
2237 MRI syntax will not be documented here; see the MRI manuals for more
2238 information. Note in particular that the handling of macros and macro
2239 arguments is somewhat different. The purpose of this option is to permit
2240 assembling existing MRI assembler code using @command{@value{AS}}.
2242 The MRI compatibility is not complete. Certain operations of the MRI assembler
2243 depend upon its object file format, and can not be supported using other object
2244 file formats. Supporting these would require enhancing each object file format
2245 individually. These are:
2248 @item global symbols in common section
2250 The m68k MRI assembler supports common sections which are merged by the linker.
2251 Other object file formats do not support this. @command{@value{AS}} handles
2252 common sections by treating them as a single common symbol. It permits local
2253 symbols to be defined within a common section, but it can not support global
2254 symbols, since it has no way to describe them.
2256 @item complex relocations
2258 The MRI assemblers support relocations against a negated section address, and
2259 relocations which combine the start addresses of two or more sections. These
2260 are not support by other object file formats.
2262 @item @code{END} pseudo-op specifying start address
2264 The MRI @code{END} pseudo-op permits the specification of a start address.
2265 This is not supported by other object file formats. The start address may
2266 instead be specified using the @option{-e} option to the linker, or in a linker
2269 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2271 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2272 name to the output file. This is not supported by other object file formats.
2274 @item @code{ORG} pseudo-op
2276 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2277 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2278 which changes the location within the current section. Absolute sections are
2279 not supported by other object file formats. The address of a section may be
2280 assigned within a linker script.
2283 There are some other features of the MRI assembler which are not supported by
2284 @command{@value{AS}}, typically either because they are difficult or because they
2285 seem of little consequence. Some of these may be supported in future releases.
2289 @item EBCDIC strings
2291 EBCDIC strings are not supported.
2293 @item packed binary coded decimal
2295 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2296 and @code{DCB.P} pseudo-ops are not supported.
2298 @item @code{FEQU} pseudo-op
2300 The m68k @code{FEQU} pseudo-op is not supported.
2302 @item @code{NOOBJ} pseudo-op
2304 The m68k @code{NOOBJ} pseudo-op is not supported.
2306 @item @code{OPT} branch control options
2308 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2309 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2310 relaxes all branches, whether forward or backward, to an appropriate size, so
2311 these options serve no purpose.
2313 @item @code{OPT} list control options
2315 The following m68k @code{OPT} list control options are ignored: @code{C},
2316 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2317 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2319 @item other @code{OPT} options
2321 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2322 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2324 @item @code{OPT} @code{D} option is default
2326 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2327 @code{OPT NOD} may be used to turn it off.
2329 @item @code{XREF} pseudo-op.
2331 The m68k @code{XREF} pseudo-op is ignored.
2333 @item @code{.debug} pseudo-op
2335 The i960 @code{.debug} pseudo-op is not supported.
2337 @item @code{.extended} pseudo-op
2339 The i960 @code{.extended} pseudo-op is not supported.
2341 @item @code{.list} pseudo-op.
2343 The various options of the i960 @code{.list} pseudo-op are not supported.
2345 @item @code{.optimize} pseudo-op
2347 The i960 @code{.optimize} pseudo-op is not supported.
2349 @item @code{.output} pseudo-op
2351 The i960 @code{.output} pseudo-op is not supported.
2353 @item @code{.setreal} pseudo-op
2355 The i960 @code{.setreal} pseudo-op is not supported.
2360 @section Dependency Tracking: @option{--MD}
2363 @cindex dependency tracking
2366 @command{@value{AS}} can generate a dependency file for the file it creates. This
2367 file consists of a single rule suitable for @code{make} describing the
2368 dependencies of the main source file.
2370 The rule is written to the file named in its argument.
2372 This feature is used in the automatic updating of makefiles.
2375 @section Name the Object File: @option{-o}
2378 @cindex naming object file
2379 @cindex object file name
2380 There is always one object file output when you run @command{@value{AS}}. By
2381 default it has the name
2384 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2398 You use this option (which takes exactly one filename) to give the
2399 object file a different name.
2401 Whatever the object file is called, @command{@value{AS}} overwrites any
2402 existing file of the same name.
2405 @section Join Data and Text Sections: @option{-R}
2408 @cindex data and text sections, joining
2409 @cindex text and data sections, joining
2410 @cindex joining text and data sections
2411 @cindex merging text and data sections
2412 @option{-R} tells @command{@value{AS}} to write the object file as if all
2413 data-section data lives in the text section. This is only done at
2414 the very last moment: your binary data are the same, but data
2415 section parts are relocated differently. The data section part of
2416 your object file is zero bytes long because all its bytes are
2417 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2419 When you specify @option{-R} it would be possible to generate shorter
2420 address displacements (because we do not have to cross between text and
2421 data section). We refrain from doing this simply for compatibility with
2422 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2425 When @command{@value{AS}} is configured for COFF or ELF output,
2426 this option is only useful if you use sections named @samp{.text} and
2431 @option{-R} is not supported for any of the HPPA targets. Using
2432 @option{-R} generates a warning from @command{@value{AS}}.
2436 @section Display Assembly Statistics: @option{--statistics}
2438 @kindex --statistics
2439 @cindex statistics, about assembly
2440 @cindex time, total for assembly
2441 @cindex space used, maximum for assembly
2442 Use @samp{--statistics} to display two statistics about the resources used by
2443 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2444 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2447 @node traditional-format
2448 @section Compatible Output: @option{--traditional-format}
2450 @kindex --traditional-format
2451 For some targets, the output of @command{@value{AS}} is different in some ways
2452 from the output of some existing assembler. This switch requests
2453 @command{@value{AS}} to use the traditional format instead.
2455 For example, it disables the exception frame optimizations which
2456 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2459 @section Announce Version: @option{-v}
2463 @cindex assembler version
2464 @cindex version of assembler
2465 You can find out what version of as is running by including the
2466 option @samp{-v} (which you can also spell as @samp{-version}) on the
2470 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2472 @command{@value{AS}} should never give a warning or error message when
2473 assembling compiler output. But programs written by people often
2474 cause @command{@value{AS}} to give a warning that a particular assumption was
2475 made. All such warnings are directed to the standard error file.
2479 @cindex suppressing warnings
2480 @cindex warnings, suppressing
2481 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2482 This only affects the warning messages: it does not change any particular of
2483 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2486 @kindex --fatal-warnings
2487 @cindex errors, caused by warnings
2488 @cindex warnings, causing error
2489 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2490 files that generate warnings to be in error.
2493 @cindex warnings, switching on
2494 You can switch these options off again by specifying @option{--warn}, which
2495 causes warnings to be output as usual.
2498 @section Generate Object File in Spite of Errors: @option{-Z}
2499 @cindex object file, after errors
2500 @cindex errors, continuing after
2501 After an error message, @command{@value{AS}} normally produces no output. If for
2502 some reason you are interested in object file output even after
2503 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2504 option. If there are any errors, @command{@value{AS}} continues anyways, and
2505 writes an object file after a final warning message of the form @samp{@var{n}
2506 errors, @var{m} warnings, generating bad object file.}
2511 @cindex machine-independent syntax
2512 @cindex syntax, machine-independent
2513 This chapter describes the machine-independent syntax allowed in a
2514 source file. @command{@value{AS}} syntax is similar to what many other
2515 assemblers use; it is inspired by the BSD 4.2
2520 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2524 * Preprocessing:: Preprocessing
2525 * Whitespace:: Whitespace
2526 * Comments:: Comments
2527 * Symbol Intro:: Symbols
2528 * Statements:: Statements
2529 * Constants:: Constants
2533 @section Preprocessing
2535 @cindex preprocessing
2536 The @command{@value{AS}} internal preprocessor:
2538 @cindex whitespace, removed by preprocessor
2540 adjusts and removes extra whitespace. It leaves one space or tab before
2541 the keywords on a line, and turns any other whitespace on the line into
2544 @cindex comments, removed by preprocessor
2546 removes all comments, replacing them with a single space, or an
2547 appropriate number of newlines.
2549 @cindex constants, converted by preprocessor
2551 converts character constants into the appropriate numeric values.
2554 It does not do macro processing, include file handling, or
2555 anything else you may get from your C compiler's preprocessor. You can
2556 do include file processing with the @code{.include} directive
2557 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2558 to get other ``CPP'' style preprocessing by giving the input file a
2559 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2560 Output, gcc.info, Using GNU CC}.
2562 Excess whitespace, comments, and character constants
2563 cannot be used in the portions of the input text that are not
2566 @cindex turning preprocessing on and off
2567 @cindex preprocessing, turning on and off
2570 If the first line of an input file is @code{#NO_APP} or if you use the
2571 @samp{-f} option, whitespace and comments are not removed from the input file.
2572 Within an input file, you can ask for whitespace and comment removal in
2573 specific portions of the by putting a line that says @code{#APP} before the
2574 text that may contain whitespace or comments, and putting a line that says
2575 @code{#NO_APP} after this text. This feature is mainly intend to support
2576 @code{asm} statements in compilers whose output is otherwise free of comments
2583 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2584 Whitespace is used to separate symbols, and to make programs neater for
2585 people to read. Unless within character constants
2586 (@pxref{Characters,,Character Constants}), any whitespace means the same
2587 as exactly one space.
2593 There are two ways of rendering comments to @command{@value{AS}}. In both
2594 cases the comment is equivalent to one space.
2596 Anything from @samp{/*} through the next @samp{*/} is a comment.
2597 This means you may not nest these comments.
2601 The only way to include a newline ('\n') in a comment
2602 is to use this sort of comment.
2605 /* This sort of comment does not nest. */
2608 @cindex line comment character
2609 Anything from a @dfn{line comment} character up to the next newline is
2610 considered a comment and is ignored. The line comment character is target
2611 specific, and some targets multiple comment characters. Some targets also have
2612 line comment characters that only work if they are the first character on a
2613 line. Some targets use a sequence of two characters to introduce a line
2614 comment. Some targets can also change their line comment characters depending
2615 upon command line options that have been used. For more details see the
2616 @emph{Syntax} section in the documentation for individual targets.
2618 If the line comment character is the hash sign (@samp{#}) then it still has the
2619 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2620 to specify logical line numbers:
2623 @cindex lines starting with @code{#}
2624 @cindex logical line numbers
2625 To be compatible with past assemblers, lines that begin with @samp{#} have a
2626 special interpretation. Following the @samp{#} should be an absolute
2627 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2628 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2629 new logical file name. The rest of the line, if any, should be whitespace.
2631 If the first non-whitespace characters on the line are not numeric,
2632 the line is ignored. (Just like a comment.)
2635 # This is an ordinary comment.
2636 # 42-6 "new_file_name" # New logical file name
2637 # This is logical line # 36.
2639 This feature is deprecated, and may disappear from future versions
2640 of @command{@value{AS}}.
2645 @cindex characters used in symbols
2646 @ifclear SPECIAL-SYMS
2647 A @dfn{symbol} is one or more characters chosen from the set of all
2648 letters (both upper and lower case), digits and the three characters
2654 A @dfn{symbol} is one or more characters chosen from the set of all
2655 letters (both upper and lower case), digits and the three characters
2656 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2662 On most machines, you can also use @code{$} in symbol names; exceptions
2663 are noted in @ref{Machine Dependencies}.
2665 No symbol may begin with a digit. Case is significant.
2666 There is no length limit: all characters are significant. Multibyte characters
2667 are supported. Symbols are delimited by characters not in that set, or by the
2668 beginning of a file (since the source program must end with a newline, the end
2669 of a file is not a possible symbol delimiter). @xref{Symbols}.
2670 @cindex length of symbols
2675 @cindex statements, structure of
2676 @cindex line separator character
2677 @cindex statement separator character
2679 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2680 @dfn{line separator character}. The line separator character is target
2681 specific and described in the @emph{Syntax} section of each
2682 target's documentation. Not all targets support a line separator character.
2683 The newline or line separator character is considered to be part of the
2684 preceding statement. Newlines and separators within character constants are an
2685 exception: they do not end statements.
2687 @cindex newline, required at file end
2688 @cindex EOF, newline must precede
2689 It is an error to end any statement with end-of-file: the last
2690 character of any input file should be a newline.@refill
2692 An empty statement is allowed, and may include whitespace. It is ignored.
2694 @cindex instructions and directives
2695 @cindex directives and instructions
2696 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2697 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2699 A statement begins with zero or more labels, optionally followed by a
2700 key symbol which determines what kind of statement it is. The key
2701 symbol determines the syntax of the rest of the statement. If the
2702 symbol begins with a dot @samp{.} then the statement is an assembler
2703 directive: typically valid for any computer. If the symbol begins with
2704 a letter the statement is an assembly language @dfn{instruction}: it
2705 assembles into a machine language instruction.
2707 Different versions of @command{@value{AS}} for different computers
2708 recognize different instructions. In fact, the same symbol may
2709 represent a different instruction in a different computer's assembly
2713 @cindex @code{:} (label)
2714 @cindex label (@code{:})
2715 A label is a symbol immediately followed by a colon (@code{:}).
2716 Whitespace before a label or after a colon is permitted, but you may not
2717 have whitespace between a label's symbol and its colon. @xref{Labels}.
2720 For HPPA targets, labels need not be immediately followed by a colon, but
2721 the definition of a label must begin in column zero. This also implies that
2722 only one label may be defined on each line.
2726 label: .directive followed by something
2727 another_label: # This is an empty statement.
2728 instruction operand_1, operand_2, @dots{}
2735 A constant is a number, written so that its value is known by
2736 inspection, without knowing any context. Like this:
2739 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2740 .ascii "Ring the bell\7" # A string constant.
2741 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2742 .float 0f-314159265358979323846264338327\
2743 95028841971.693993751E-40 # - pi, a flonum.
2748 * Characters:: Character Constants
2749 * Numbers:: Number Constants
2753 @subsection Character Constants
2755 @cindex character constants
2756 @cindex constants, character
2757 There are two kinds of character constants. A @dfn{character} stands
2758 for one character in one byte and its value may be used in
2759 numeric expressions. String constants (properly called string
2760 @emph{literals}) are potentially many bytes and their values may not be
2761 used in arithmetic expressions.
2765 * Chars:: Characters
2769 @subsubsection Strings
2771 @cindex string constants
2772 @cindex constants, string
2773 A @dfn{string} is written between double-quotes. It may contain
2774 double-quotes or null characters. The way to get special characters
2775 into a string is to @dfn{escape} these characters: precede them with
2776 a backslash @samp{\} character. For example @samp{\\} represents
2777 one backslash: the first @code{\} is an escape which tells
2778 @command{@value{AS}} to interpret the second character literally as a backslash
2779 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2780 escape character). The complete list of escapes follows.
2782 @cindex escape codes, character
2783 @cindex character escape codes
2786 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2788 @cindex @code{\b} (backspace character)
2789 @cindex backspace (@code{\b})
2791 Mnemonic for backspace; for ASCII this is octal code 010.
2794 @c Mnemonic for EOText; for ASCII this is octal code 004.
2796 @cindex @code{\f} (formfeed character)
2797 @cindex formfeed (@code{\f})
2799 Mnemonic for FormFeed; for ASCII this is octal code 014.
2801 @cindex @code{\n} (newline character)
2802 @cindex newline (@code{\n})
2804 Mnemonic for newline; for ASCII this is octal code 012.
2807 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2809 @cindex @code{\r} (carriage return character)
2810 @cindex carriage return (@code{\r})
2812 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2815 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2816 @c other assemblers.
2818 @cindex @code{\t} (tab)
2819 @cindex tab (@code{\t})
2821 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2824 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2825 @c @item \x @var{digit} @var{digit} @var{digit}
2826 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2828 @cindex @code{\@var{ddd}} (octal character code)
2829 @cindex octal character code (@code{\@var{ddd}})
2830 @item \ @var{digit} @var{digit} @var{digit}
2831 An octal character code. The numeric code is 3 octal digits.
2832 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2833 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2835 @cindex @code{\@var{xd...}} (hex character code)
2836 @cindex hex character code (@code{\@var{xd...}})
2837 @item \@code{x} @var{hex-digits...}
2838 A hex character code. All trailing hex digits are combined. Either upper or
2839 lower case @code{x} works.
2841 @cindex @code{\\} (@samp{\} character)
2842 @cindex backslash (@code{\\})
2844 Represents one @samp{\} character.
2847 @c Represents one @samp{'} (accent acute) character.
2848 @c This is needed in single character literals
2849 @c (@xref{Characters,,Character Constants}.) to represent
2852 @cindex @code{\"} (doublequote character)
2853 @cindex doublequote (@code{\"})
2855 Represents one @samp{"} character. Needed in strings to represent
2856 this character, because an unescaped @samp{"} would end the string.
2858 @item \ @var{anything-else}
2859 Any other character when escaped by @kbd{\} gives a warning, but
2860 assembles as if the @samp{\} was not present. The idea is that if
2861 you used an escape sequence you clearly didn't want the literal
2862 interpretation of the following character. However @command{@value{AS}} has no
2863 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2864 code and warns you of the fact.
2867 Which characters are escapable, and what those escapes represent,
2868 varies widely among assemblers. The current set is what we think
2869 the BSD 4.2 assembler recognizes, and is a subset of what most C
2870 compilers recognize. If you are in doubt, do not use an escape
2874 @subsubsection Characters
2876 @cindex single character constant
2877 @cindex character, single
2878 @cindex constant, single character
2879 A single character may be written as a single quote immediately
2880 followed by that character. The same escapes apply to characters as
2881 to strings. So if you want to write the character backslash, you
2882 must write @kbd{'\\} where the first @code{\} escapes the second
2883 @code{\}. As you can see, the quote is an acute accent, not a
2884 grave accent. A newline
2886 @ifclear abnormal-separator
2887 (or semicolon @samp{;})
2889 @ifset abnormal-separator
2891 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2896 immediately following an acute accent is taken as a literal character
2897 and does not count as the end of a statement. The value of a character
2898 constant in a numeric expression is the machine's byte-wide code for
2899 that character. @command{@value{AS}} assumes your character code is ASCII:
2900 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2903 @subsection Number Constants
2905 @cindex constants, number
2906 @cindex number constants
2907 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2908 are stored in the target machine. @emph{Integers} are numbers that
2909 would fit into an @code{int} in the C language. @emph{Bignums} are
2910 integers, but they are stored in more than 32 bits. @emph{Flonums}
2911 are floating point numbers, described below.
2914 * Integers:: Integers
2919 * Bit Fields:: Bit Fields
2925 @subsubsection Integers
2927 @cindex constants, integer
2929 @cindex binary integers
2930 @cindex integers, binary
2931 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2932 the binary digits @samp{01}.
2934 @cindex octal integers
2935 @cindex integers, octal
2936 An octal integer is @samp{0} followed by zero or more of the octal
2937 digits (@samp{01234567}).
2939 @cindex decimal integers
2940 @cindex integers, decimal
2941 A decimal integer starts with a non-zero digit followed by zero or
2942 more digits (@samp{0123456789}).
2944 @cindex hexadecimal integers
2945 @cindex integers, hexadecimal
2946 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2947 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2949 Integers have the usual values. To denote a negative integer, use
2950 the prefix operator @samp{-} discussed under expressions
2951 (@pxref{Prefix Ops,,Prefix Operators}).
2954 @subsubsection Bignums
2957 @cindex constants, bignum
2958 A @dfn{bignum} has the same syntax and semantics as an integer
2959 except that the number (or its negative) takes more than 32 bits to
2960 represent in binary. The distinction is made because in some places
2961 integers are permitted while bignums are not.
2964 @subsubsection Flonums
2966 @cindex floating point numbers
2967 @cindex constants, floating point
2969 @cindex precision, floating point
2970 A @dfn{flonum} represents a floating point number. The translation is
2971 indirect: a decimal floating point number from the text is converted by
2972 @command{@value{AS}} to a generic binary floating point number of more than
2973 sufficient precision. This generic floating point number is converted
2974 to a particular computer's floating point format (or formats) by a
2975 portion of @command{@value{AS}} specialized to that computer.
2977 A flonum is written by writing (in order)
2982 (@samp{0} is optional on the HPPA.)
2986 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2988 @kbd{e} is recommended. Case is not important.
2990 @c FIXME: verify if flonum syntax really this vague for most cases
2991 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2992 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2995 On the H8/300, Renesas / SuperH SH,
2996 and AMD 29K architectures, the letter must be
2997 one of the letters @samp{DFPRSX} (in upper or lower case).
2999 On the ARC, the letter must be one of the letters @samp{DFRS}
3000 (in upper or lower case).
3002 On the Intel 960 architecture, the letter must be
3003 one of the letters @samp{DFT} (in upper or lower case).
3005 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3009 One of the letters @samp{DFRS} (in upper or lower case).
3012 One of the letters @samp{DFPRSX} (in upper or lower case).
3015 The letter @samp{E} (upper case only).
3018 One of the letters @samp{DFT} (in upper or lower case).
3023 An optional sign: either @samp{+} or @samp{-}.
3026 An optional @dfn{integer part}: zero or more decimal digits.
3029 An optional @dfn{fractional part}: @samp{.} followed by zero
3030 or more decimal digits.
3033 An optional exponent, consisting of:
3037 An @samp{E} or @samp{e}.
3038 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3039 @c principle this can perfectly well be different on different targets.
3041 Optional sign: either @samp{+} or @samp{-}.
3043 One or more decimal digits.
3048 At least one of the integer part or the fractional part must be
3049 present. The floating point number has the usual base-10 value.
3051 @command{@value{AS}} does all processing using integers. Flonums are computed
3052 independently of any floating point hardware in the computer running
3053 @command{@value{AS}}.
3057 @c Bit fields are written as a general facility but are also controlled
3058 @c by a conditional-compilation flag---which is as of now (21mar91)
3059 @c turned on only by the i960 config of GAS.
3061 @subsubsection Bit Fields
3064 @cindex constants, bit field
3065 You can also define numeric constants as @dfn{bit fields}.
3066 Specify two numbers separated by a colon---
3068 @var{mask}:@var{value}
3071 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
3074 The resulting number is then packed
3076 @c this conditional paren in case bit fields turned on elsewhere than 960
3077 (in host-dependent byte order)
3079 into a field whose width depends on which assembler directive has the
3080 bit-field as its argument. Overflow (a result from the bitwise and
3081 requiring more binary digits to represent) is not an error; instead,
3082 more constants are generated, of the specified width, beginning with the
3083 least significant digits.@refill
3085 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
3086 @code{.short}, and @code{.word} accept bit-field arguments.
3091 @chapter Sections and Relocation
3096 * Secs Background:: Background
3097 * Ld Sections:: Linker Sections
3098 * As Sections:: Assembler Internal Sections
3099 * Sub-Sections:: Sub-Sections
3103 @node Secs Background
3106 Roughly, a section is a range of addresses, with no gaps; all data
3107 ``in'' those addresses is treated the same for some particular purpose.
3108 For example there may be a ``read only'' section.
3110 @cindex linker, and assembler
3111 @cindex assembler, and linker
3112 The linker @code{@value{LD}} reads many object files (partial programs) and
3113 combines their contents to form a runnable program. When @command{@value{AS}}
3114 emits an object file, the partial program is assumed to start at address 0.
3115 @code{@value{LD}} assigns the final addresses for the partial program, so that
3116 different partial programs do not overlap. This is actually an
3117 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3120 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3121 addresses. These blocks slide to their run-time addresses as rigid
3122 units; their length does not change and neither does the order of bytes
3123 within them. Such a rigid unit is called a @emph{section}. Assigning
3124 run-time addresses to sections is called @dfn{relocation}. It includes
3125 the task of adjusting mentions of object-file addresses so they refer to
3126 the proper run-time addresses.
3128 For the H8/300, and for the Renesas / SuperH SH,
3129 @command{@value{AS}} pads sections if needed to
3130 ensure they end on a word (sixteen bit) boundary.
3133 @cindex standard assembler sections
3134 An object file written by @command{@value{AS}} has at least three sections, any
3135 of which may be empty. These are named @dfn{text}, @dfn{data} and
3140 When it generates COFF or ELF output,
3142 @command{@value{AS}} can also generate whatever other named sections you specify
3143 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3144 If you do not use any directives that place output in the @samp{.text}
3145 or @samp{.data} sections, these sections still exist, but are empty.
3150 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3152 @command{@value{AS}} can also generate whatever other named sections you
3153 specify using the @samp{.space} and @samp{.subspace} directives. See
3154 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3155 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3156 assembler directives.
3159 Additionally, @command{@value{AS}} uses different names for the standard
3160 text, data, and bss sections when generating SOM output. Program text
3161 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3162 BSS into @samp{$BSS$}.
3166 Within the object file, the text section starts at address @code{0}, the
3167 data section follows, and the bss section follows the data section.
3170 When generating either SOM or ELF output files on the HPPA, the text
3171 section starts at address @code{0}, the data section at address
3172 @code{0x4000000}, and the bss section follows the data section.
3175 To let @code{@value{LD}} know which data changes when the sections are
3176 relocated, and how to change that data, @command{@value{AS}} also writes to the
3177 object file details of the relocation needed. To perform relocation
3178 @code{@value{LD}} must know, each time an address in the object
3182 Where in the object file is the beginning of this reference to
3185 How long (in bytes) is this reference?
3187 Which section does the address refer to? What is the numeric value of
3189 (@var{address}) @minus{} (@var{start-address of section})?
3192 Is the reference to an address ``Program-Counter relative''?
3195 @cindex addresses, format of
3196 @cindex section-relative addressing
3197 In fact, every address @command{@value{AS}} ever uses is expressed as
3199 (@var{section}) + (@var{offset into section})
3202 Further, most expressions @command{@value{AS}} computes have this section-relative
3205 (For some object formats, such as SOM for the HPPA, some expressions are
3206 symbol-relative instead.)
3209 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3210 @var{N} into section @var{secname}.''
3212 Apart from text, data and bss sections you need to know about the
3213 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3214 addresses in the absolute section remain unchanged. For example, address
3215 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3216 @code{@value{LD}}. Although the linker never arranges two partial programs'
3217 data sections with overlapping addresses after linking, @emph{by definition}
3218 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3219 part of a program is always the same address when the program is running as
3220 address @code{@{absolute@ 239@}} in any other part of the program.
3222 The idea of sections is extended to the @dfn{undefined} section. Any
3223 address whose section is unknown at assembly time is by definition
3224 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3225 Since numbers are always defined, the only way to generate an undefined
3226 address is to mention an undefined symbol. A reference to a named
3227 common block would be such a symbol: its value is unknown at assembly
3228 time so it has section @emph{undefined}.
3230 By analogy the word @emph{section} is used to describe groups of sections in
3231 the linked program. @code{@value{LD}} puts all partial programs' text
3232 sections in contiguous addresses in the linked program. It is
3233 customary to refer to the @emph{text section} of a program, meaning all
3234 the addresses of all partial programs' text sections. Likewise for
3235 data and bss sections.
3237 Some sections are manipulated by @code{@value{LD}}; others are invented for
3238 use of @command{@value{AS}} and have no meaning except during assembly.
3241 @section Linker Sections
3242 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3247 @cindex named sections
3248 @cindex sections, named
3249 @item named sections
3252 @cindex text section
3253 @cindex data section
3257 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3258 separate but equal sections. Anything you can say of one section is
3261 When the program is running, however, it is
3262 customary for the text section to be unalterable. The
3263 text section is often shared among processes: it contains
3264 instructions, constants and the like. The data section of a running
3265 program is usually alterable: for example, C variables would be stored
3266 in the data section.
3271 This section contains zeroed bytes when your program begins running. It
3272 is used to hold uninitialized variables or common storage. The length of
3273 each partial program's bss section is important, but because it starts
3274 out containing zeroed bytes there is no need to store explicit zero
3275 bytes in the object file. The bss section was invented to eliminate
3276 those explicit zeros from object files.
3278 @cindex absolute section
3279 @item absolute section
3280 Address 0 of this section is always ``relocated'' to runtime address 0.
3281 This is useful if you want to refer to an address that @code{@value{LD}} must
3282 not change when relocating. In this sense we speak of absolute
3283 addresses being ``unrelocatable'': they do not change during relocation.
3285 @cindex undefined section
3286 @item undefined section
3287 This ``section'' is a catch-all for address references to objects not in
3288 the preceding sections.
3289 @c FIXME: ref to some other doc on obj-file formats could go here.
3292 @cindex relocation example
3293 An idealized example of three relocatable sections follows.
3295 The example uses the traditional section names @samp{.text} and @samp{.data}.
3297 Memory addresses are on the horizontal axis.
3301 @c END TEXI2ROFF-KILL
3304 partial program # 1: |ttttt|dddd|00|
3311 partial program # 2: |TTT|DDD|000|
3314 +--+---+-----+--+----+---+-----+~~
3315 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3316 +--+---+-----+--+----+---+-----+~~
3318 addresses: 0 @dots{}
3325 \line{\it Partial program \#1: \hfil}
3326 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3327 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3329 \line{\it Partial program \#2: \hfil}
3330 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3331 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3333 \line{\it linked program: \hfil}
3334 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3335 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3336 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3337 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3339 \line{\it addresses: \hfil}
3343 @c END TEXI2ROFF-KILL
3346 @section Assembler Internal Sections
3348 @cindex internal assembler sections
3349 @cindex sections in messages, internal
3350 These sections are meant only for the internal use of @command{@value{AS}}. They
3351 have no meaning at run-time. You do not really need to know about these
3352 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3353 warning messages, so it might be helpful to have an idea of their
3354 meanings to @command{@value{AS}}. These sections are used to permit the
3355 value of every expression in your assembly language program to be a
3356 section-relative address.
3359 @cindex assembler internal logic error
3360 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3361 An internal assembler logic error has been found. This means there is a
3362 bug in the assembler.
3364 @cindex expr (internal section)
3366 The assembler stores complex expression internally as combinations of
3367 symbols. When it needs to represent an expression as a symbol, it puts
3368 it in the expr section.
3370 @c FIXME item transfer[t] vector preload
3371 @c FIXME item transfer[t] vector postload
3372 @c FIXME item register
3376 @section Sub-Sections
3378 @cindex numbered subsections
3379 @cindex grouping data
3385 fall into two sections: text and data.
3387 You may have separate groups of
3389 data in named sections
3393 data in named sections
3399 that you want to end up near to each other in the object file, even though they
3400 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3401 use @dfn{subsections} for this purpose. Within each section, there can be
3402 numbered subsections with values from 0 to 8192. Objects assembled into the
3403 same subsection go into the object file together with other objects in the same
3404 subsection. For example, a compiler might want to store constants in the text
3405 section, but might not want to have them interspersed with the program being
3406 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3407 section of code being output, and a @samp{.text 1} before each group of
3408 constants being output.
3410 Subsections are optional. If you do not use subsections, everything
3411 goes in subsection number zero.
3414 Each subsection is zero-padded up to a multiple of four bytes.
3415 (Subsections may be padded a different amount on different flavors
3416 of @command{@value{AS}}.)
3420 On the H8/300 platform, each subsection is zero-padded to a word
3421 boundary (two bytes).
3422 The same is true on the Renesas SH.
3425 @c FIXME section padding (alignment)?
3426 @c Rich Pixley says padding here depends on target obj code format; that
3427 @c doesn't seem particularly useful to say without further elaboration,
3428 @c so for now I say nothing about it. If this is a generic BFD issue,
3429 @c these paragraphs might need to vanish from this manual, and be
3430 @c discussed in BFD chapter of binutils (or some such).
3434 Subsections appear in your object file in numeric order, lowest numbered
3435 to highest. (All this to be compatible with other people's assemblers.)
3436 The object file contains no representation of subsections; @code{@value{LD}} and
3437 other programs that manipulate object files see no trace of them.
3438 They just see all your text subsections as a text section, and all your
3439 data subsections as a data section.
3441 To specify which subsection you want subsequent statements assembled
3442 into, use a numeric argument to specify it, in a @samp{.text
3443 @var{expression}} or a @samp{.data @var{expression}} statement.
3446 When generating COFF output, you
3451 can also use an extra subsection
3452 argument with arbitrary named sections: @samp{.section @var{name},
3457 When generating ELF output, you
3462 can also use the @code{.subsection} directive (@pxref{SubSection})
3463 to specify a subsection: @samp{.subsection @var{expression}}.
3465 @var{Expression} should be an absolute expression
3466 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3467 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3468 begins in @code{text 0}. For instance:
3470 .text 0 # The default subsection is text 0 anyway.
3471 .ascii "This lives in the first text subsection. *"
3473 .ascii "But this lives in the second text subsection."
3475 .ascii "This lives in the data section,"
3476 .ascii "in the first data subsection."
3478 .ascii "This lives in the first text section,"
3479 .ascii "immediately following the asterisk (*)."
3482 Each section has a @dfn{location counter} incremented by one for every byte
3483 assembled into that section. Because subsections are merely a convenience
3484 restricted to @command{@value{AS}} there is no concept of a subsection location
3485 counter. There is no way to directly manipulate a location counter---but the
3486 @code{.align} directive changes it, and any label definition captures its
3487 current value. The location counter of the section where statements are being
3488 assembled is said to be the @dfn{active} location counter.
3491 @section bss Section
3494 @cindex common variable storage
3495 The bss section is used for local common variable storage.
3496 You may allocate address space in the bss section, but you may
3497 not dictate data to load into it before your program executes. When
3498 your program starts running, all the contents of the bss
3499 section are zeroed bytes.
3501 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3502 @ref{Lcomm,,@code{.lcomm}}.
3504 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3505 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3508 When assembling for a target which supports multiple sections, such as ELF or
3509 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3510 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3511 section. Typically the section will only contain symbol definitions and
3512 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3519 Symbols are a central concept: the programmer uses symbols to name
3520 things, the linker uses symbols to link, and the debugger uses symbols
3524 @cindex debuggers, and symbol order
3525 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3526 the same order they were declared. This may break some debuggers.
3531 * Setting Symbols:: Giving Symbols Other Values
3532 * Symbol Names:: Symbol Names
3533 * Dot:: The Special Dot Symbol
3534 * Symbol Attributes:: Symbol Attributes
3541 A @dfn{label} is written as a symbol immediately followed by a colon
3542 @samp{:}. The symbol then represents the current value of the
3543 active location counter, and is, for example, a suitable instruction
3544 operand. You are warned if you use the same symbol to represent two
3545 different locations: the first definition overrides any other
3549 On the HPPA, the usual form for a label need not be immediately followed by a
3550 colon, but instead must start in column zero. Only one label may be defined on
3551 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3552 provides a special directive @code{.label} for defining labels more flexibly.
3555 @node Setting Symbols
3556 @section Giving Symbols Other Values
3558 @cindex assigning values to symbols
3559 @cindex symbol values, assigning
3560 A symbol can be given an arbitrary value by writing a symbol, followed
3561 by an equals sign @samp{=}, followed by an expression
3562 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3563 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3564 equals sign @samp{=}@samp{=} here represents an equivalent of the
3565 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3568 Blackfin does not support symbol assignment with @samp{=}.
3572 @section Symbol Names
3574 @cindex symbol names
3575 @cindex names, symbol
3576 @ifclear SPECIAL-SYMS
3577 Symbol names begin with a letter or with one of @samp{._}. On most
3578 machines, you can also use @code{$} in symbol names; exceptions are
3579 noted in @ref{Machine Dependencies}. That character may be followed by any
3580 string of digits, letters, dollar signs (unless otherwise noted for a
3581 particular target machine), and underscores.
3585 Symbol names begin with a letter or with one of @samp{._}. On the
3586 Renesas SH you can also use @code{$} in symbol names. That
3587 character may be followed by any string of digits, letters, dollar signs (save
3588 on the H8/300), and underscores.
3592 Case of letters is significant: @code{foo} is a different symbol name
3595 Multibyte characters are supported. To generate a symbol name containing
3596 multibyte characters enclose it within double quotes and use escape codes. cf
3597 @xref{Strings}. Generating a multibyte symbol name from a label is not
3598 currently supported.
3600 Each symbol has exactly one name. Each name in an assembly language program
3601 refers to exactly one symbol. You may use that symbol name any number of times
3604 @subheading Local Symbol Names
3606 @cindex local symbol names
3607 @cindex symbol names, local
3608 A local symbol is any symbol beginning with certain local label prefixes.
3609 By default, the local label prefix is @samp{.L} for ELF systems or
3610 @samp{L} for traditional a.out systems, but each target may have its own
3611 set of local label prefixes.
3613 On the HPPA local symbols begin with @samp{L$}.
3616 Local symbols are defined and used within the assembler, but they are
3617 normally not saved in object files. Thus, they are not visible when debugging.
3618 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols:
3619 @option{-L}}) to retain the local symbols in the object files.
3621 @subheading Local Labels
3623 @cindex local labels
3624 @cindex temporary symbol names
3625 @cindex symbol names, temporary
3626 Local labels help compilers and programmers use names temporarily.
3627 They create symbols which are guaranteed to be unique over the entire scope of
3628 the input source code and which can be referred to by a simple notation.
3629 To define a local label, write a label of the form @samp{@b{N}:} (where @b{N}
3630 represents any positive integer). To refer to the most recent previous
3631 definition of that label write @samp{@b{N}b}, using the same number as when
3632 you defined the label. To refer to the next definition of a local label, write
3633 @samp{@b{N}f}---the @samp{b} stands for ``backwards'' and the @samp{f} stands
3636 There is no restriction on how you can use these labels, and you can reuse them
3637 too. So that it is possible to repeatedly define the same local label (using
3638 the same number @samp{@b{N}}), although you can only refer to the most recently
3639 defined local label of that number (for a backwards reference) or the next
3640 definition of a specific local label for a forward reference. It is also worth
3641 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3642 implemented in a slightly more efficient manner than the others.
3653 Which is the equivalent of:
3656 label_1: branch label_3
3657 label_2: branch label_1
3658 label_3: branch label_4
3659 label_4: branch label_3
3662 Local label names are only a notational device. They are immediately
3663 transformed into more conventional symbol names before the assembler uses them.
3664 The symbol names are stored in the symbol table, appear in error messages, and
3665 are optionally emitted to the object file. The names are constructed using
3669 @item @emph{local label prefix}
3670 All local symbols begin with the system-specific local label prefix.
3671 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3672 that start with the local label prefix. These labels are
3673 used for symbols you are never intended to see. If you use the
3674 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3675 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3676 you may use them in debugging.
3679 This is the number that was used in the local label definition. So if the
3680 label is written @samp{55:} then the number is @samp{55}.
3683 This unusual character is included so you do not accidentally invent a symbol
3684 of the same name. The character has ASCII value of @samp{\002} (control-B).
3686 @item @emph{ordinal number}
3687 This is a serial number to keep the labels distinct. The first definition of
3688 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3689 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3690 the number @samp{1} and its 15th definition gets @samp{15} as well.
3693 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3694 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3696 @subheading Dollar Local Labels
3697 @cindex dollar local symbols
3699 @code{@value{AS}} also supports an even more local form of local labels called
3700 dollar labels. These labels go out of scope (i.e., they become undefined) as
3701 soon as a non-local label is defined. Thus they remain valid for only a small
3702 region of the input source code. Normal local labels, by contrast, remain in
3703 scope for the entire file, or until they are redefined by another occurrence of
3704 the same local label.
3706 Dollar labels are defined in exactly the same way as ordinary local labels,
3707 except that they have a dollar sign suffix to their numeric value, e.g.,
3710 They can also be distinguished from ordinary local labels by their transformed
3711 names which use ASCII character @samp{\001} (control-A) as the magic character
3712 to distinguish them from ordinary labels. For example, the fifth definition of
3713 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3716 @section The Special Dot Symbol
3718 @cindex dot (symbol)
3719 @cindex @code{.} (symbol)
3720 @cindex current address
3721 @cindex location counter
3722 The special symbol @samp{.} refers to the current address that
3723 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3724 .long .} defines @code{melvin} to contain its own address.
3725 Assigning a value to @code{.} is treated the same as a @code{.org}
3727 @ifclear no-space-dir
3728 Thus, the expression @samp{.=.+4} is the same as saying
3732 @node Symbol Attributes
3733 @section Symbol Attributes
3735 @cindex symbol attributes
3736 @cindex attributes, symbol
3737 Every symbol has, as well as its name, the attributes ``Value'' and
3738 ``Type''. Depending on output format, symbols can also have auxiliary
3741 The detailed definitions are in @file{a.out.h}.
3744 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3745 all these attributes, and probably won't warn you. This makes the
3746 symbol an externally defined symbol, which is generally what you
3750 * Symbol Value:: Value
3751 * Symbol Type:: Type
3754 * a.out Symbols:: Symbol Attributes: @code{a.out}
3758 * a.out Symbols:: Symbol Attributes: @code{a.out}
3761 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3766 * COFF Symbols:: Symbol Attributes for COFF
3769 * SOM Symbols:: Symbol Attributes for SOM
3776 @cindex value of a symbol
3777 @cindex symbol value
3778 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3779 location in the text, data, bss or absolute sections the value is the
3780 number of addresses from the start of that section to the label.
3781 Naturally for text, data and bss sections the value of a symbol changes
3782 as @code{@value{LD}} changes section base addresses during linking. Absolute
3783 symbols' values do not change during linking: that is why they are
3786 The value of an undefined symbol is treated in a special way. If it is
3787 0 then the symbol is not defined in this assembler source file, and
3788 @code{@value{LD}} tries to determine its value from other files linked into the
3789 same program. You make this kind of symbol simply by mentioning a symbol
3790 name without defining it. A non-zero value represents a @code{.comm}
3791 common declaration. The value is how much common storage to reserve, in
3792 bytes (addresses). The symbol refers to the first address of the
3798 @cindex type of a symbol
3800 The type attribute of a symbol contains relocation (section)
3801 information, any flag settings indicating that a symbol is external, and
3802 (optionally), other information for linkers and debuggers. The exact
3803 format depends on the object-code output format in use.
3808 @c The following avoids a "widow" subsection title. @group would be
3809 @c better if it were available outside examples.
3812 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3814 @cindex @code{b.out} symbol attributes
3815 @cindex symbol attributes, @code{b.out}
3816 These symbol attributes appear only when @command{@value{AS}} is configured for
3817 one of the Berkeley-descended object output formats---@code{a.out} or
3823 @subsection Symbol Attributes: @code{a.out}
3825 @cindex @code{a.out} symbol attributes
3826 @cindex symbol attributes, @code{a.out}
3832 @subsection Symbol Attributes: @code{a.out}
3834 @cindex @code{a.out} symbol attributes
3835 @cindex symbol attributes, @code{a.out}
3839 * Symbol Desc:: Descriptor
3840 * Symbol Other:: Other
3844 @subsubsection Descriptor
3846 @cindex descriptor, of @code{a.out} symbol
3847 This is an arbitrary 16-bit value. You may establish a symbol's
3848 descriptor value by using a @code{.desc} statement
3849 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3850 @command{@value{AS}}.
3853 @subsubsection Other
3855 @cindex other attribute, of @code{a.out} symbol
3856 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3861 @subsection Symbol Attributes for COFF
3863 @cindex COFF symbol attributes
3864 @cindex symbol attributes, COFF
3866 The COFF format supports a multitude of auxiliary symbol attributes;
3867 like the primary symbol attributes, they are set between @code{.def} and
3868 @code{.endef} directives.
3870 @subsubsection Primary Attributes
3872 @cindex primary attributes, COFF symbols
3873 The symbol name is set with @code{.def}; the value and type,
3874 respectively, with @code{.val} and @code{.type}.
3876 @subsubsection Auxiliary Attributes
3878 @cindex auxiliary attributes, COFF symbols
3879 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3880 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3881 table information for COFF.
3886 @subsection Symbol Attributes for SOM
3888 @cindex SOM symbol attributes
3889 @cindex symbol attributes, SOM
3891 The SOM format for the HPPA supports a multitude of symbol attributes set with
3892 the @code{.EXPORT} and @code{.IMPORT} directives.
3894 The attributes are described in @cite{HP9000 Series 800 Assembly
3895 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3896 @code{EXPORT} assembler directive documentation.
3900 @chapter Expressions
3904 @cindex numeric values
3905 An @dfn{expression} specifies an address or numeric value.
3906 Whitespace may precede and/or follow an expression.
3908 The result of an expression must be an absolute number, or else an offset into
3909 a particular section. If an expression is not absolute, and there is not
3910 enough information when @command{@value{AS}} sees the expression to know its
3911 section, a second pass over the source program might be necessary to interpret
3912 the expression---but the second pass is currently not implemented.
3913 @command{@value{AS}} aborts with an error message in this situation.
3916 * Empty Exprs:: Empty Expressions
3917 * Integer Exprs:: Integer Expressions
3921 @section Empty Expressions
3923 @cindex empty expressions
3924 @cindex expressions, empty
3925 An empty expression has no value: it is just whitespace or null.
3926 Wherever an absolute expression is required, you may omit the
3927 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3928 is compatible with other assemblers.
3931 @section Integer Expressions
3933 @cindex integer expressions
3934 @cindex expressions, integer
3935 An @dfn{integer expression} is one or more @emph{arguments} delimited
3936 by @emph{operators}.
3939 * Arguments:: Arguments
3940 * Operators:: Operators
3941 * Prefix Ops:: Prefix Operators
3942 * Infix Ops:: Infix Operators
3946 @subsection Arguments
3948 @cindex expression arguments
3949 @cindex arguments in expressions
3950 @cindex operands in expressions
3951 @cindex arithmetic operands
3952 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3953 contexts arguments are sometimes called ``arithmetic operands''. In
3954 this manual, to avoid confusing them with the ``instruction operands'' of
3955 the machine language, we use the term ``argument'' to refer to parts of
3956 expressions only, reserving the word ``operand'' to refer only to machine
3957 instruction operands.
3959 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3960 @var{section} is one of text, data, bss, absolute,
3961 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3964 Numbers are usually integers.
3966 A number can be a flonum or bignum. In this case, you are warned
3967 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3968 these 32 bits are an integer. You may write integer-manipulating
3969 instructions that act on exotic constants, compatible with other
3972 @cindex subexpressions
3973 Subexpressions are a left parenthesis @samp{(} followed by an integer
3974 expression, followed by a right parenthesis @samp{)}; or a prefix
3975 operator followed by an argument.
3978 @subsection Operators
3980 @cindex operators, in expressions
3981 @cindex arithmetic functions
3982 @cindex functions, in expressions
3983 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3984 operators are followed by an argument. Infix operators appear
3985 between their arguments. Operators may be preceded and/or followed by
3989 @subsection Prefix Operator
3991 @cindex prefix operators
3992 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3993 one argument, which must be absolute.
3995 @c the tex/end tex stuff surrounding this small table is meant to make
3996 @c it align, on the printed page, with the similar table in the next
3997 @c section (which is inside an enumerate).
3999 \global\advance\leftskip by \itemindent
4004 @dfn{Negation}. Two's complement negation.
4006 @dfn{Complementation}. Bitwise not.
4010 \global\advance\leftskip by -\itemindent
4014 @subsection Infix Operators
4016 @cindex infix operators
4017 @cindex operators, permitted arguments
4018 @dfn{Infix operators} take two arguments, one on either side. Operators
4019 have precedence, but operations with equal precedence are performed left
4020 to right. Apart from @code{+} or @option{-}, both arguments must be
4021 absolute, and the result is absolute.
4024 @cindex operator precedence
4025 @cindex precedence of operators
4032 @dfn{Multiplication}.
4035 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4041 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4044 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4048 Intermediate precedence
4053 @dfn{Bitwise Inclusive Or}.
4059 @dfn{Bitwise Exclusive Or}.
4062 @dfn{Bitwise Or Not}.
4069 @cindex addition, permitted arguments
4070 @cindex plus, permitted arguments
4071 @cindex arguments for addition
4073 @dfn{Addition}. If either argument is absolute, the result has the section of
4074 the other argument. You may not add together arguments from different
4077 @cindex subtraction, permitted arguments
4078 @cindex minus, permitted arguments
4079 @cindex arguments for subtraction
4081 @dfn{Subtraction}. If the right argument is absolute, the
4082 result has the section of the left argument.
4083 If both arguments are in the same section, the result is absolute.
4084 You may not subtract arguments from different sections.
4085 @c FIXME is there still something useful to say about undefined - undefined ?
4087 @cindex comparison expressions
4088 @cindex expressions, comparison
4093 @dfn{Is Not Equal To}
4097 @dfn{Is Greater Than}
4099 @dfn{Is Greater Than Or Equal To}
4101 @dfn{Is Less Than Or Equal To}
4103 The comparison operators can be used as infix operators. A true results has a
4104 value of -1 whereas a false result has a value of 0. Note, these operators
4105 perform signed comparisons.
4108 @item Lowest Precedence
4117 These two logical operations can be used to combine the results of sub
4118 expressions. Note, unlike the comparison operators a true result returns a
4119 value of 1 but a false results does still return 0. Also note that the logical
4120 or operator has a slightly lower precedence than logical and.
4125 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4126 address; you can only have a defined section in one of the two arguments.
4129 @chapter Assembler Directives
4131 @cindex directives, machine independent
4132 @cindex pseudo-ops, machine independent
4133 @cindex machine independent directives
4134 All assembler directives have names that begin with a period (@samp{.}).
4135 The rest of the name is letters, usually in lower case.
4137 This chapter discusses directives that are available regardless of the
4138 target machine configuration for the @sc{gnu} assembler.
4140 Some machine configurations provide additional directives.
4141 @xref{Machine Dependencies}.
4144 @ifset machine-directives
4145 @xref{Machine Dependencies}, for additional directives.
4150 * Abort:: @code{.abort}
4152 * ABORT (COFF):: @code{.ABORT}
4155 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4156 * Altmacro:: @code{.altmacro}
4157 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4158 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4159 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4160 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, @code{.bundle_lock}, @code{.bundle_unlock}
4161 * Byte:: @code{.byte @var{expressions}}
4162 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4163 * Comm:: @code{.comm @var{symbol} , @var{length} }
4164 * Data:: @code{.data @var{subsection}}
4166 * Def:: @code{.def @var{name}}
4169 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4175 * Double:: @code{.double @var{flonums}}
4176 * Eject:: @code{.eject}
4177 * Else:: @code{.else}
4178 * Elseif:: @code{.elseif}
4181 * Endef:: @code{.endef}
4184 * Endfunc:: @code{.endfunc}
4185 * Endif:: @code{.endif}
4186 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4187 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4188 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4190 * Error:: @code{.error @var{string}}
4191 * Exitm:: @code{.exitm}
4192 * Extern:: @code{.extern}
4193 * Fail:: @code{.fail}
4194 * File:: @code{.file}
4195 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4196 * Float:: @code{.float @var{flonums}}
4197 * Func:: @code{.func}
4198 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4200 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4201 * Hidden:: @code{.hidden @var{names}}
4204 * hword:: @code{.hword @var{expressions}}
4205 * Ident:: @code{.ident}
4206 * If:: @code{.if @var{absolute expression}}
4207 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4208 * Include:: @code{.include "@var{file}"}
4209 * Int:: @code{.int @var{expressions}}
4211 * Internal:: @code{.internal @var{names}}
4214 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4215 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4216 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4217 * Lflags:: @code{.lflags}
4218 @ifclear no-line-dir
4219 * Line:: @code{.line @var{line-number}}
4222 * Linkonce:: @code{.linkonce [@var{type}]}
4223 * List:: @code{.list}
4224 * Ln:: @code{.ln @var{line-number}}
4225 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4226 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4228 * Local:: @code{.local @var{names}}
4231 * Long:: @code{.long @var{expressions}}
4233 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4236 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4237 * MRI:: @code{.mri @var{val}}
4238 * Noaltmacro:: @code{.noaltmacro}
4239 * Nolist:: @code{.nolist}
4240 * Octa:: @code{.octa @var{bignums}}
4241 * Offset:: @code{.offset @var{loc}}
4242 * Org:: @code{.org @var{new-lc}, @var{fill}}
4243 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4245 * PopSection:: @code{.popsection}
4246 * Previous:: @code{.previous}
4249 * Print:: @code{.print @var{string}}
4251 * Protected:: @code{.protected @var{names}}
4254 * Psize:: @code{.psize @var{lines}, @var{columns}}
4255 * Purgem:: @code{.purgem @var{name}}
4257 * PushSection:: @code{.pushsection @var{name}}
4260 * Quad:: @code{.quad @var{bignums}}
4261 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4262 * Rept:: @code{.rept @var{count}}
4263 * Sbttl:: @code{.sbttl "@var{subheading}"}
4265 * Scl:: @code{.scl @var{class}}
4268 * Section:: @code{.section @var{name}[, @var{flags}]}
4271 * Set:: @code{.set @var{symbol}, @var{expression}}
4272 * Short:: @code{.short @var{expressions}}
4273 * Single:: @code{.single @var{flonums}}
4275 * Size:: @code{.size [@var{name} , @var{expression}]}
4277 @ifclear no-space-dir
4278 * Skip:: @code{.skip @var{size} , @var{fill}}
4281 * Sleb128:: @code{.sleb128 @var{expressions}}
4282 @ifclear no-space-dir
4283 * Space:: @code{.space @var{size} , @var{fill}}
4286 * Stab:: @code{.stabd, .stabn, .stabs}
4289 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4290 * Struct:: @code{.struct @var{expression}}
4292 * SubSection:: @code{.subsection}
4293 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4297 * Tag:: @code{.tag @var{structname}}
4300 * Text:: @code{.text @var{subsection}}
4301 * Title:: @code{.title "@var{heading}"}
4303 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4306 * Uleb128:: @code{.uleb128 @var{expressions}}
4308 * Val:: @code{.val @var{addr}}
4312 * Version:: @code{.version "@var{string}"}
4313 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4314 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4317 * Warning:: @code{.warning @var{string}}
4318 * Weak:: @code{.weak @var{names}}
4319 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4320 * Word:: @code{.word @var{expressions}}
4321 * Deprecated:: Deprecated Directives
4325 @section @code{.abort}
4327 @cindex @code{abort} directive
4328 @cindex stopping the assembly
4329 This directive stops the assembly immediately. It is for
4330 compatibility with other assemblers. The original idea was that the
4331 assembly language source would be piped into the assembler. If the sender
4332 of the source quit, it could use this directive tells @command{@value{AS}} to
4333 quit also. One day @code{.abort} will not be supported.
4337 @section @code{.ABORT} (COFF)
4339 @cindex @code{ABORT} directive
4340 When producing COFF output, @command{@value{AS}} accepts this directive as a
4341 synonym for @samp{.abort}.
4344 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4350 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4352 @cindex padding the location counter
4353 @cindex @code{align} directive
4354 Pad the location counter (in the current subsection) to a particular storage
4355 boundary. The first expression (which must be absolute) is the alignment
4356 required, as described below.
4358 The second expression (also absolute) gives the fill value to be stored in the
4359 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4360 padding bytes are normally zero. However, on some systems, if the section is
4361 marked as containing code and the fill value is omitted, the space is filled
4362 with no-op instructions.
4364 The third expression is also absolute, and is also optional. If it is present,
4365 it is the maximum number of bytes that should be skipped by this alignment
4366 directive. If doing the alignment would require skipping more bytes than the
4367 specified maximum, then the alignment is not done at all. You can omit the
4368 fill value (the second argument) entirely by simply using two commas after the
4369 required alignment; this can be useful if you want the alignment to be filled
4370 with no-op instructions when appropriate.
4372 The way the required alignment is specified varies from system to system.
4373 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or1k,
4374 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4375 alignment request in bytes. For example @samp{.align 8} advances
4376 the location counter until it is a multiple of 8. If the location counter
4377 is already a multiple of 8, no change is needed. For the tic54x, the
4378 first expression is the alignment request in words.
4380 For other systems, including ppc, i386 using a.out format, arm and
4381 strongarm, it is the
4382 number of low-order zero bits the location counter must have after
4383 advancement. For example @samp{.align 3} advances the location
4384 counter until it a multiple of 8. If the location counter is already a
4385 multiple of 8, no change is needed.
4387 This inconsistency is due to the different behaviors of the various
4388 native assemblers for these systems which GAS must emulate.
4389 GAS also provides @code{.balign} and @code{.p2align} directives,
4390 described later, which have a consistent behavior across all
4391 architectures (but are specific to GAS).
4394 @section @code{.altmacro}
4395 Enable alternate macro mode, enabling:
4398 @item LOCAL @var{name} [ , @dots{} ]
4399 One additional directive, @code{LOCAL}, is available. It is used to
4400 generate a string replacement for each of the @var{name} arguments, and
4401 replace any instances of @var{name} in each macro expansion. The
4402 replacement string is unique in the assembly, and different for each
4403 separate macro expansion. @code{LOCAL} allows you to write macros that
4404 define symbols, without fear of conflict between separate macro expansions.
4406 @item String delimiters
4407 You can write strings delimited in these other ways besides
4408 @code{"@var{string}"}:
4411 @item '@var{string}'
4412 You can delimit strings with single-quote characters.
4414 @item <@var{string}>
4415 You can delimit strings with matching angle brackets.
4418 @item single-character string escape
4419 To include any single character literally in a string (even if the
4420 character would otherwise have some special meaning), you can prefix the
4421 character with @samp{!} (an exclamation mark). For example, you can
4422 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4424 @item Expression results as strings
4425 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4426 and use the result as a string.
4430 @section @code{.ascii "@var{string}"}@dots{}
4432 @cindex @code{ascii} directive
4433 @cindex string literals
4434 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4435 separated by commas. It assembles each string (with no automatic
4436 trailing zero byte) into consecutive addresses.
4439 @section @code{.asciz "@var{string}"}@dots{}
4441 @cindex @code{asciz} directive
4442 @cindex zero-terminated strings
4443 @cindex null-terminated strings
4444 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4445 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4448 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4450 @cindex padding the location counter given number of bytes
4451 @cindex @code{balign} directive
4452 Pad the location counter (in the current subsection) to a particular
4453 storage boundary. The first expression (which must be absolute) is the
4454 alignment request in bytes. For example @samp{.balign 8} advances
4455 the location counter until it is a multiple of 8. If the location counter
4456 is already a multiple of 8, no change is needed.
4458 The second expression (also absolute) gives the fill value to be stored in the
4459 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4460 padding bytes are normally zero. However, on some systems, if the section is
4461 marked as containing code and the fill value is omitted, the space is filled
4462 with no-op instructions.
4464 The third expression is also absolute, and is also optional. If it is present,
4465 it is the maximum number of bytes that should be skipped by this alignment
4466 directive. If doing the alignment would require skipping more bytes than the
4467 specified maximum, then the alignment is not done at all. You can omit the
4468 fill value (the second argument) entirely by simply using two commas after the
4469 required alignment; this can be useful if you want the alignment to be filled
4470 with no-op instructions when appropriate.
4472 @cindex @code{balignw} directive
4473 @cindex @code{balignl} directive
4474 The @code{.balignw} and @code{.balignl} directives are variants of the
4475 @code{.balign} directive. The @code{.balignw} directive treats the fill
4476 pattern as a two byte word value. The @code{.balignl} directives treats the
4477 fill pattern as a four byte longword value. For example, @code{.balignw
4478 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4479 filled in with the value 0x368d (the exact placement of the bytes depends upon
4480 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4483 @node Bundle directives
4484 @section @code{.bundle_align_mode @var{abs-expr}}
4485 @cindex @code{bundle_align_mode} directive
4487 @cindex instruction bundle
4488 @cindex aligned instruction bundle
4489 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4490 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4491 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4492 disabled (which is the default state). If the argument it not zero, it
4493 gives the size of an instruction bundle as a power of two (as for the
4494 @code{.p2align} directive, @pxref{P2align}).
4496 For some targets, it's an ABI requirement that no instruction may span a
4497 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4498 instructions that starts on an aligned boundary. For example, if
4499 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4500 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4501 effect, no single instruction may span a boundary between bundles. If an
4502 instruction would start too close to the end of a bundle for the length of
4503 that particular instruction to fit within the bundle, then the space at the
4504 end of that bundle is filled with no-op instructions so the instruction
4505 starts in the next bundle. As a corollary, it's an error if any single
4506 instruction's encoding is longer than the bundle size.
4508 @section @code{.bundle_lock} and @code{.bundle_unlock}
4509 @cindex @code{bundle_lock} directive
4510 @cindex @code{bundle_unlock} directive
4511 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4512 allow explicit control over instruction bundle padding. These directives
4513 are only valid when @code{.bundle_align_mode} has been used to enable
4514 aligned instruction bundle mode. It's an error if they appear when
4515 @code{.bundle_align_mode} has not been used at all, or when the last
4516 directive was @w{@code{.bundle_align_mode 0}}.
4518 @cindex bundle-locked
4519 For some targets, it's an ABI requirement that certain instructions may
4520 appear only as part of specified permissible sequences of multiple
4521 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4522 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4523 instruction sequence. For purposes of aligned instruction bundle mode, a
4524 sequence starting with @code{.bundle_lock} and ending with
4525 @code{.bundle_unlock} is treated as a single instruction. That is, the
4526 entire sequence must fit into a single bundle and may not span a bundle
4527 boundary. If necessary, no-op instructions will be inserted before the
4528 first instruction of the sequence so that the whole sequence starts on an
4529 aligned bundle boundary. It's an error if the sequence is longer than the
4532 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4533 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4534 nested. That is, a second @code{.bundle_lock} directive before the next
4535 @code{.bundle_unlock} directive has no effect except that it must be
4536 matched by another closing @code{.bundle_unlock} so that there is the
4537 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4540 @section @code{.byte @var{expressions}}
4542 @cindex @code{byte} directive
4543 @cindex integers, one byte
4544 @code{.byte} expects zero or more expressions, separated by commas.
4545 Each expression is assembled into the next byte.
4547 @node CFI directives
4548 @section @code{.cfi_sections @var{section_list}}
4549 @cindex @code{cfi_sections} directive
4550 @code{.cfi_sections} may be used to specify whether CFI directives
4551 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4552 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4553 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4554 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4555 directive is not used is @code{.cfi_sections .eh_frame}.
4557 @section @code{.cfi_startproc [simple]}
4558 @cindex @code{cfi_startproc} directive
4559 @code{.cfi_startproc} is used at the beginning of each function that
4560 should have an entry in @code{.eh_frame}. It initializes some internal
4561 data structures. Don't forget to close the function by
4562 @code{.cfi_endproc}.
4564 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4565 it also emits some architecture dependent initial CFI instructions.
4567 @section @code{.cfi_endproc}
4568 @cindex @code{cfi_endproc} directive
4569 @code{.cfi_endproc} is used at the end of a function where it closes its
4570 unwind entry previously opened by
4571 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4573 @section @code{.cfi_personality @var{encoding} [, @var{exp}]}
4574 @code{.cfi_personality} defines personality routine and its encoding.
4575 @var{encoding} must be a constant determining how the personality
4576 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4577 argument is not present, otherwise second argument should be
4578 a constant or a symbol name. When using indirect encodings,
4579 the symbol provided should be the location where personality
4580 can be loaded from, not the personality routine itself.
4581 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4582 no personality routine.
4584 @section @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4585 @code{.cfi_lsda} defines LSDA and its encoding.
4586 @var{encoding} must be a constant determining how the LSDA
4587 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4588 argument is not present, otherwise second argument should be a constant
4589 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4592 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4593 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4594 address from @var{register} and add @var{offset} to it}.
4596 @section @code{.cfi_def_cfa_register @var{register}}
4597 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4598 now on @var{register} will be used instead of the old one. Offset
4601 @section @code{.cfi_def_cfa_offset @var{offset}}
4602 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4603 remains the same, but @var{offset} is new. Note that it is the
4604 absolute offset that will be added to a defined register to compute
4607 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4608 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4609 value that is added/substracted from the previous offset.
4611 @section @code{.cfi_offset @var{register}, @var{offset}}
4612 Previous value of @var{register} is saved at offset @var{offset} from
4615 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4616 Previous value of @var{register} is saved at offset @var{offset} from
4617 the current CFA register. This is transformed to @code{.cfi_offset}
4618 using the known displacement of the CFA register from the CFA.
4619 This is often easier to use, because the number will match the
4620 code it's annotating.
4622 @section @code{.cfi_register @var{register1}, @var{register2}}
4623 Previous value of @var{register1} is saved in register @var{register2}.
4625 @section @code{.cfi_restore @var{register}}
4626 @code{.cfi_restore} says that the rule for @var{register} is now the
4627 same as it was at the beginning of the function, after all initial
4628 instruction added by @code{.cfi_startproc} were executed.
4630 @section @code{.cfi_undefined @var{register}}
4631 From now on the previous value of @var{register} can't be restored anymore.
4633 @section @code{.cfi_same_value @var{register}}
4634 Current value of @var{register} is the same like in the previous frame,
4635 i.e. no restoration needed.
4637 @section @code{.cfi_remember_state},
4638 First save all current rules for all registers by @code{.cfi_remember_state},
4639 then totally screw them up by subsequent @code{.cfi_*} directives and when
4640 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4641 the previous saved state.
4643 @section @code{.cfi_return_column @var{register}}
4644 Change return column @var{register}, i.e. the return address is either
4645 directly in @var{register} or can be accessed by rules for @var{register}.
4647 @section @code{.cfi_signal_frame}
4648 Mark current function as signal trampoline.
4650 @section @code{.cfi_window_save}
4651 SPARC register window has been saved.
4653 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4654 Allows the user to add arbitrary bytes to the unwind info. One
4655 might use this to add OS-specific CFI opcodes, or generic CFI
4656 opcodes that GAS does not yet support.
4658 @section @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4659 The current value of @var{register} is @var{label}. The value of @var{label}
4660 will be encoded in the output file according to @var{encoding}; see the
4661 description of @code{.cfi_personality} for details on this encoding.
4663 The usefulness of equating a register to a fixed label is probably
4664 limited to the return address register. Here, it can be useful to
4665 mark a code segment that has only one return address which is reached
4666 by a direct branch and no copy of the return address exists in memory
4667 or another register.
4670 @section @code{.comm @var{symbol} , @var{length} }
4672 @cindex @code{comm} directive
4673 @cindex symbol, common
4674 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4675 common symbol in one object file may be merged with a defined or common symbol
4676 of the same name in another object file. If @code{@value{LD}} does not see a
4677 definition for the symbol--just one or more common symbols--then it will
4678 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4679 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4680 the same name, and they do not all have the same size, it will allocate space
4681 using the largest size.
4684 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4685 an optional third argument. This is the desired alignment of the symbol,
4686 specified for ELF as a byte boundary (for example, an alignment of 16 means
4687 that the least significant 4 bits of the address should be zero), and for PE
4688 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4689 boundary). The alignment must be an absolute expression, and it must be a
4690 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4691 common symbol, it will use the alignment when placing the symbol. If no
4692 alignment is specified, @command{@value{AS}} will set the alignment to the
4693 largest power of two less than or equal to the size of the symbol, up to a
4694 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4695 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4696 @samp{--section-alignment} option; image file sections in PE are aligned to
4697 multiples of 4096, which is far too large an alignment for ordinary variables.
4698 It is rather the default alignment for (non-debug) sections within object
4699 (@samp{*.o}) files, which are less strictly aligned.}.
4703 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4704 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4708 @section @code{.data @var{subsection}}
4710 @cindex @code{data} directive
4711 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4712 end of the data subsection numbered @var{subsection} (which is an
4713 absolute expression). If @var{subsection} is omitted, it defaults
4718 @section @code{.def @var{name}}
4720 @cindex @code{def} directive
4721 @cindex COFF symbols, debugging
4722 @cindex debugging COFF symbols
4723 Begin defining debugging information for a symbol @var{name}; the
4724 definition extends until the @code{.endef} directive is encountered.
4727 This directive is only observed when @command{@value{AS}} is configured for COFF
4728 format output; when producing @code{b.out}, @samp{.def} is recognized,
4735 @section @code{.desc @var{symbol}, @var{abs-expression}}
4737 @cindex @code{desc} directive
4738 @cindex COFF symbol descriptor
4739 @cindex symbol descriptor, COFF
4740 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4741 to the low 16 bits of an absolute expression.
4744 The @samp{.desc} directive is not available when @command{@value{AS}} is
4745 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4746 object format. For the sake of compatibility, @command{@value{AS}} accepts
4747 it, but produces no output, when configured for COFF.
4753 @section @code{.dim}
4755 @cindex @code{dim} directive
4756 @cindex COFF auxiliary symbol information
4757 @cindex auxiliary symbol information, COFF
4758 This directive is generated by compilers to include auxiliary debugging
4759 information in the symbol table. It is only permitted inside
4760 @code{.def}/@code{.endef} pairs.
4763 @samp{.dim} is only meaningful when generating COFF format output; when
4764 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4770 @section @code{.double @var{flonums}}
4772 @cindex @code{double} directive
4773 @cindex floating point numbers (double)
4774 @code{.double} expects zero or more flonums, separated by commas. It
4775 assembles floating point numbers.
4777 The exact kind of floating point numbers emitted depends on how
4778 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4782 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4783 in @sc{ieee} format.
4788 @section @code{.eject}
4790 @cindex @code{eject} directive
4791 @cindex new page, in listings
4792 @cindex page, in listings
4793 @cindex listing control: new page
4794 Force a page break at this point, when generating assembly listings.
4797 @section @code{.else}
4799 @cindex @code{else} directive
4800 @code{.else} is part of the @command{@value{AS}} support for conditional
4801 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4802 of code to be assembled if the condition for the preceding @code{.if}
4806 @section @code{.elseif}
4808 @cindex @code{elseif} directive
4809 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4810 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4811 @code{.if} block that would otherwise fill the entire @code{.else} section.
4814 @section @code{.end}
4816 @cindex @code{end} directive
4817 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4818 process anything in the file past the @code{.end} directive.
4822 @section @code{.endef}
4824 @cindex @code{endef} directive
4825 This directive flags the end of a symbol definition begun with
4829 @samp{.endef} is only meaningful when generating COFF format output; if
4830 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4831 directive but ignores it.
4836 @section @code{.endfunc}
4837 @cindex @code{endfunc} directive
4838 @code{.endfunc} marks the end of a function specified with @code{.func}.
4841 @section @code{.endif}
4843 @cindex @code{endif} directive
4844 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4845 it marks the end of a block of code that is only assembled
4846 conditionally. @xref{If,,@code{.if}}.
4849 @section @code{.equ @var{symbol}, @var{expression}}
4851 @cindex @code{equ} directive
4852 @cindex assigning values to symbols
4853 @cindex symbols, assigning values to
4854 This directive sets the value of @var{symbol} to @var{expression}.
4855 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4858 The syntax for @code{equ} on the HPPA is
4859 @samp{@var{symbol} .equ @var{expression}}.
4863 The syntax for @code{equ} on the Z80 is
4864 @samp{@var{symbol} equ @var{expression}}.
4865 On the Z80 it is an eror if @var{symbol} is already defined,
4866 but the symbol is not protected from later redefinition.
4867 Compare @ref{Equiv}.
4871 @section @code{.equiv @var{symbol}, @var{expression}}
4872 @cindex @code{equiv} directive
4873 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4874 the assembler will signal an error if @var{symbol} is already defined. Note a
4875 symbol which has been referenced but not actually defined is considered to be
4878 Except for the contents of the error message, this is roughly equivalent to
4885 plus it protects the symbol from later redefinition.
4888 @section @code{.eqv @var{symbol}, @var{expression}}
4889 @cindex @code{eqv} directive
4890 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
4891 evaluate the expression or any part of it immediately. Instead each time
4892 the resulting symbol is used in an expression, a snapshot of its current
4896 @section @code{.err}
4897 @cindex @code{err} directive
4898 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4899 message and, unless the @option{-Z} option was used, it will not generate an
4900 object file. This can be used to signal an error in conditionally compiled code.
4903 @section @code{.error "@var{string}"}
4904 @cindex error directive
4906 Similarly to @code{.err}, this directive emits an error, but you can specify a
4907 string that will be emitted as the error message. If you don't specify the
4908 message, it defaults to @code{".error directive invoked in source file"}.
4909 @xref{Errors, ,Error and Warning Messages}.
4912 .error "This code has not been assembled and tested."
4916 @section @code{.exitm}
4917 Exit early from the current macro definition. @xref{Macro}.
4920 @section @code{.extern}
4922 @cindex @code{extern} directive
4923 @code{.extern} is accepted in the source program---for compatibility
4924 with other assemblers---but it is ignored. @command{@value{AS}} treats
4925 all undefined symbols as external.
4928 @section @code{.fail @var{expression}}
4930 @cindex @code{fail} directive
4931 Generates an error or a warning. If the value of the @var{expression} is 500
4932 or more, @command{@value{AS}} will print a warning message. If the value is less
4933 than 500, @command{@value{AS}} will print an error message. The message will
4934 include the value of @var{expression}. This can occasionally be useful inside
4935 complex nested macros or conditional assembly.
4938 @section @code{.file}
4939 @cindex @code{file} directive
4941 @ifclear no-file-dir
4942 There are two different versions of the @code{.file} directive. Targets
4943 that support DWARF2 line number information use the DWARF2 version of
4944 @code{.file}. Other targets use the default version.
4946 @subheading Default Version
4948 @cindex logical file name
4949 @cindex file name, logical
4950 This version of the @code{.file} directive tells @command{@value{AS}} that we
4951 are about to start a new logical file. The syntax is:
4957 @var{string} is the new file name. In general, the filename is
4958 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4959 to specify an empty file name, you must give the quotes--@code{""}. This
4960 statement may go away in future: it is only recognized to be compatible with
4961 old @command{@value{AS}} programs.
4963 @subheading DWARF2 Version
4966 When emitting DWARF2 line number information, @code{.file} assigns filenames
4967 to the @code{.debug_line} file name table. The syntax is:
4970 .file @var{fileno} @var{filename}
4973 The @var{fileno} operand should be a unique positive integer to use as the
4974 index of the entry in the table. The @var{filename} operand is a C string
4977 The detail of filename indices is exposed to the user because the filename
4978 table is shared with the @code{.debug_info} section of the DWARF2 debugging
4979 information, and thus the user must know the exact indices that table
4983 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4985 @cindex @code{fill} directive
4986 @cindex writing patterns in memory
4987 @cindex patterns, writing in memory
4988 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4989 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4990 may be zero or more. @var{Size} may be zero or more, but if it is
4991 more than 8, then it is deemed to have the value 8, compatible with
4992 other people's assemblers. The contents of each @var{repeat} bytes
4993 is taken from an 8-byte number. The highest order 4 bytes are
4994 zero. The lowest order 4 bytes are @var{value} rendered in the
4995 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4996 Each @var{size} bytes in a repetition is taken from the lowest order
4997 @var{size} bytes of this number. Again, this bizarre behavior is
4998 compatible with other people's assemblers.
5000 @var{size} and @var{value} are optional.
5001 If the second comma and @var{value} are absent, @var{value} is
5002 assumed zero. If the first comma and following tokens are absent,
5003 @var{size} is assumed to be 1.
5006 @section @code{.float @var{flonums}}
5008 @cindex floating point numbers (single)
5009 @cindex @code{float} directive
5010 This directive assembles zero or more flonums, separated by commas. It
5011 has the same effect as @code{.single}.
5013 The exact kind of floating point numbers emitted depends on how
5014 @command{@value{AS}} is configured.
5015 @xref{Machine Dependencies}.
5019 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5020 in @sc{ieee} format.
5025 @section @code{.func @var{name}[,@var{label}]}
5026 @cindex @code{func} directive
5027 @code{.func} emits debugging information to denote function @var{name}, and
5028 is ignored unless the file is assembled with debugging enabled.
5029 Only @samp{--gstabs[+]} is currently supported.
5030 @var{label} is the entry point of the function and if omitted @var{name}
5031 prepended with the @samp{leading char} is used.
5032 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5033 All functions are currently defined to have @code{void} return type.
5034 The function must be terminated with @code{.endfunc}.
5037 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5039 @cindex @code{global} directive
5040 @cindex symbol, making visible to linker
5041 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5042 @var{symbol} in your partial program, its value is made available to
5043 other partial programs that are linked with it. Otherwise,
5044 @var{symbol} takes its attributes from a symbol of the same name
5045 from another file linked into the same program.
5047 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5048 compatibility with other assemblers.
5051 On the HPPA, @code{.global} is not always enough to make it accessible to other
5052 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5053 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5058 @section @code{.gnu_attribute @var{tag},@var{value}}
5059 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5062 @section @code{.hidden @var{names}}
5064 @cindex @code{hidden} directive
5066 This is one of the ELF visibility directives. The other two are
5067 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5068 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5070 This directive overrides the named symbols default visibility (which is set by
5071 their binding: local, global or weak). The directive sets the visibility to
5072 @code{hidden} which means that the symbols are not visible to other components.
5073 Such symbols are always considered to be @code{protected} as well.
5077 @section @code{.hword @var{expressions}}
5079 @cindex @code{hword} directive
5080 @cindex integers, 16-bit
5081 @cindex numbers, 16-bit
5082 @cindex sixteen bit integers
5083 This expects zero or more @var{expressions}, and emits
5084 a 16 bit number for each.
5087 This directive is a synonym for @samp{.short}; depending on the target
5088 architecture, it may also be a synonym for @samp{.word}.
5092 This directive is a synonym for @samp{.short}.
5095 This directive is a synonym for both @samp{.short} and @samp{.word}.
5100 @section @code{.ident}
5102 @cindex @code{ident} directive
5104 This directive is used by some assemblers to place tags in object files. The
5105 behavior of this directive varies depending on the target. When using the
5106 a.out object file format, @command{@value{AS}} simply accepts the directive for
5107 source-file compatibility with existing assemblers, but does not emit anything
5108 for it. When using COFF, comments are emitted to the @code{.comment} or
5109 @code{.rdata} section, depending on the target. When using ELF, comments are
5110 emitted to the @code{.comment} section.
5113 @section @code{.if @var{absolute expression}}
5115 @cindex conditional assembly
5116 @cindex @code{if} directive
5117 @code{.if} marks the beginning of a section of code which is only
5118 considered part of the source program being assembled if the argument
5119 (which must be an @var{absolute expression}) is non-zero. The end of
5120 the conditional section of code must be marked by @code{.endif}
5121 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5122 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5123 If you have several conditions to check, @code{.elseif} may be used to avoid
5124 nesting blocks if/else within each subsequent @code{.else} block.
5126 The following variants of @code{.if} are also supported:
5128 @cindex @code{ifdef} directive
5129 @item .ifdef @var{symbol}
5130 Assembles the following section of code if the specified @var{symbol}
5131 has been defined. Note a symbol which has been referenced but not yet defined
5132 is considered to be undefined.
5134 @cindex @code{ifb} directive
5135 @item .ifb @var{text}
5136 Assembles the following section of code if the operand is blank (empty).
5138 @cindex @code{ifc} directive
5139 @item .ifc @var{string1},@var{string2}
5140 Assembles the following section of code if the two strings are the same. The
5141 strings may be optionally quoted with single quotes. If they are not quoted,
5142 the first string stops at the first comma, and the second string stops at the
5143 end of the line. Strings which contain whitespace should be quoted. The
5144 string comparison is case sensitive.
5146 @cindex @code{ifeq} directive
5147 @item .ifeq @var{absolute expression}
5148 Assembles the following section of code if the argument is zero.
5150 @cindex @code{ifeqs} directive
5151 @item .ifeqs @var{string1},@var{string2}
5152 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5154 @cindex @code{ifge} directive
5155 @item .ifge @var{absolute expression}
5156 Assembles the following section of code if the argument is greater than or
5159 @cindex @code{ifgt} directive
5160 @item .ifgt @var{absolute expression}
5161 Assembles the following section of code if the argument is greater than zero.
5163 @cindex @code{ifle} directive
5164 @item .ifle @var{absolute expression}
5165 Assembles the following section of code if the argument is less than or equal
5168 @cindex @code{iflt} directive
5169 @item .iflt @var{absolute expression}
5170 Assembles the following section of code if the argument is less than zero.
5172 @cindex @code{ifnb} directive
5173 @item .ifnb @var{text}
5174 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5175 following section of code if the operand is non-blank (non-empty).
5177 @cindex @code{ifnc} directive
5178 @item .ifnc @var{string1},@var{string2}.
5179 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5180 following section of code if the two strings are not the same.
5182 @cindex @code{ifndef} directive
5183 @cindex @code{ifnotdef} directive
5184 @item .ifndef @var{symbol}
5185 @itemx .ifnotdef @var{symbol}
5186 Assembles the following section of code if the specified @var{symbol}
5187 has not been defined. Both spelling variants are equivalent. Note a symbol
5188 which has been referenced but not yet defined is considered to be undefined.
5190 @cindex @code{ifne} directive
5191 @item .ifne @var{absolute expression}
5192 Assembles the following section of code if the argument is not equal to zero
5193 (in other words, this is equivalent to @code{.if}).
5195 @cindex @code{ifnes} directive
5196 @item .ifnes @var{string1},@var{string2}
5197 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5198 following section of code if the two strings are not the same.
5202 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5204 @cindex @code{incbin} directive
5205 @cindex binary files, including
5206 The @code{incbin} directive includes @var{file} verbatim at the current
5207 location. You can control the search paths used with the @samp{-I} command-line
5208 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5211 The @var{skip} argument skips a number of bytes from the start of the
5212 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5213 read. Note that the data is not aligned in any way, so it is the user's
5214 responsibility to make sure that proper alignment is provided both before and
5215 after the @code{incbin} directive.
5218 @section @code{.include "@var{file}"}
5220 @cindex @code{include} directive
5221 @cindex supporting files, including
5222 @cindex files, including
5223 This directive provides a way to include supporting files at specified
5224 points in your source program. The code from @var{file} is assembled as
5225 if it followed the point of the @code{.include}; when the end of the
5226 included file is reached, assembly of the original file continues. You
5227 can control the search paths used with the @samp{-I} command-line option
5228 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5232 @section @code{.int @var{expressions}}
5234 @cindex @code{int} directive
5235 @cindex integers, 32-bit
5236 Expect zero or more @var{expressions}, of any section, separated by commas.
5237 For each expression, emit a number that, at run time, is the value of that
5238 expression. The byte order and bit size of the number depends on what kind
5239 of target the assembly is for.
5243 On most forms of the H8/300, @code{.int} emits 16-bit
5244 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5251 @section @code{.internal @var{names}}
5253 @cindex @code{internal} directive
5255 This is one of the ELF visibility directives. The other two are
5256 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5257 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5259 This directive overrides the named symbols default visibility (which is set by
5260 their binding: local, global or weak). The directive sets the visibility to
5261 @code{internal} which means that the symbols are considered to be @code{hidden}
5262 (i.e., not visible to other components), and that some extra, processor specific
5263 processing must also be performed upon the symbols as well.
5267 @section @code{.irp @var{symbol},@var{values}}@dots{}
5269 @cindex @code{irp} directive
5270 Evaluate a sequence of statements assigning different values to @var{symbol}.
5271 The sequence of statements starts at the @code{.irp} directive, and is
5272 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5273 set to @var{value}, and the sequence of statements is assembled. If no
5274 @var{value} is listed, the sequence of statements is assembled once, with
5275 @var{symbol} set to the null string. To refer to @var{symbol} within the
5276 sequence of statements, use @var{\symbol}.
5278 For example, assembling
5286 is equivalent to assembling
5294 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5297 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5299 @cindex @code{irpc} directive
5300 Evaluate a sequence of statements assigning different values to @var{symbol}.
5301 The sequence of statements starts at the @code{.irpc} directive, and is
5302 terminated by an @code{.endr} directive. For each character in @var{value},
5303 @var{symbol} is set to the character, and the sequence of statements is
5304 assembled. If no @var{value} is listed, the sequence of statements is
5305 assembled once, with @var{symbol} set to the null string. To refer to
5306 @var{symbol} within the sequence of statements, use @var{\symbol}.
5308 For example, assembling
5316 is equivalent to assembling
5324 For some caveats with the spelling of @var{symbol}, see also the discussion
5328 @section @code{.lcomm @var{symbol} , @var{length}}
5330 @cindex @code{lcomm} directive
5331 @cindex local common symbols
5332 @cindex symbols, local common
5333 Reserve @var{length} (an absolute expression) bytes for a local common
5334 denoted by @var{symbol}. The section and value of @var{symbol} are
5335 those of the new local common. The addresses are allocated in the bss
5336 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5337 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5338 not visible to @code{@value{LD}}.
5341 Some targets permit a third argument to be used with @code{.lcomm}. This
5342 argument specifies the desired alignment of the symbol in the bss section.
5346 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5347 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5351 @section @code{.lflags}
5353 @cindex @code{lflags} directive (ignored)
5354 @command{@value{AS}} accepts this directive, for compatibility with other
5355 assemblers, but ignores it.
5357 @ifclear no-line-dir
5359 @section @code{.line @var{line-number}}
5361 @cindex @code{line} directive
5362 @cindex logical line number
5364 Change the logical line number. @var{line-number} must be an absolute
5365 expression. The next line has that logical line number. Therefore any other
5366 statements on the current line (after a statement separator character) are
5367 reported as on logical line number @var{line-number} @minus{} 1. One day
5368 @command{@value{AS}} will no longer support this directive: it is recognized only
5369 for compatibility with existing assembler programs.
5372 Even though this is a directive associated with the @code{a.out} or
5373 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5374 when producing COFF output, and treats @samp{.line} as though it
5375 were the COFF @samp{.ln} @emph{if} it is found outside a
5376 @code{.def}/@code{.endef} pair.
5378 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5379 used by compilers to generate auxiliary symbol information for
5384 @section @code{.linkonce [@var{type}]}
5386 @cindex @code{linkonce} directive
5387 @cindex common sections
5388 Mark the current section so that the linker only includes a single copy of it.
5389 This may be used to include the same section in several different object files,
5390 but ensure that the linker will only include it once in the final output file.
5391 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5392 Duplicate sections are detected based on the section name, so it should be
5395 This directive is only supported by a few object file formats; as of this
5396 writing, the only object file format which supports it is the Portable
5397 Executable format used on Windows NT.
5399 The @var{type} argument is optional. If specified, it must be one of the
5400 following strings. For example:
5404 Not all types may be supported on all object file formats.
5408 Silently discard duplicate sections. This is the default.
5411 Warn if there are duplicate sections, but still keep only one copy.
5414 Warn if any of the duplicates have different sizes.
5417 Warn if any of the duplicates do not have exactly the same contents.
5421 @section @code{.list}
5423 @cindex @code{list} directive
5424 @cindex listing control, turning on
5425 Control (in conjunction with the @code{.nolist} directive) whether or
5426 not assembly listings are generated. These two directives maintain an
5427 internal counter (which is zero initially). @code{.list} increments the
5428 counter, and @code{.nolist} decrements it. Assembly listings are
5429 generated whenever the counter is greater than zero.
5431 By default, listings are disabled. When you enable them (with the
5432 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5433 the initial value of the listing counter is one.
5436 @section @code{.ln @var{line-number}}
5438 @cindex @code{ln} directive
5439 @ifclear no-line-dir
5440 @samp{.ln} is a synonym for @samp{.line}.
5443 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5444 must be an absolute expression. The next line has that logical
5445 line number, so any other statements on the current line (after a
5446 statement separator character @code{;}) are reported as on logical
5447 line number @var{line-number} @minus{} 1.
5450 This directive is accepted, but ignored, when @command{@value{AS}} is
5451 configured for @code{b.out}; its effect is only associated with COFF
5457 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5458 @cindex @code{loc} directive
5459 When emitting DWARF2 line number information,
5460 the @code{.loc} directive will add a row to the @code{.debug_line} line
5461 number matrix corresponding to the immediately following assembly
5462 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5463 arguments will be applied to the @code{.debug_line} state machine before
5466 The @var{options} are a sequence of the following tokens in any order:
5470 This option will set the @code{basic_block} register in the
5471 @code{.debug_line} state machine to @code{true}.
5474 This option will set the @code{prologue_end} register in the
5475 @code{.debug_line} state machine to @code{true}.
5477 @item epilogue_begin
5478 This option will set the @code{epilogue_begin} register in the
5479 @code{.debug_line} state machine to @code{true}.
5481 @item is_stmt @var{value}
5482 This option will set the @code{is_stmt} register in the
5483 @code{.debug_line} state machine to @code{value}, which must be
5486 @item isa @var{value}
5487 This directive will set the @code{isa} register in the @code{.debug_line}
5488 state machine to @var{value}, which must be an unsigned integer.
5490 @item discriminator @var{value}
5491 This directive will set the @code{discriminator} register in the @code{.debug_line}
5492 state machine to @var{value}, which must be an unsigned integer.
5496 @node Loc_mark_labels
5497 @section @code{.loc_mark_labels @var{enable}}
5498 @cindex @code{loc_mark_labels} directive
5499 When emitting DWARF2 line number information,
5500 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5501 to the @code{.debug_line} line number matrix with the @code{basic_block}
5502 register in the state machine set whenever a code label is seen.
5503 The @var{enable} argument should be either 1 or 0, to enable or disable
5504 this function respectively.
5508 @section @code{.local @var{names}}
5510 @cindex @code{local} directive
5511 This directive, which is available for ELF targets, marks each symbol in
5512 the comma-separated list of @code{names} as a local symbol so that it
5513 will not be externally visible. If the symbols do not already exist,
5514 they will be created.
5516 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5517 accept an alignment argument, which is the case for most ELF targets,
5518 the @code{.local} directive can be used in combination with @code{.comm}
5519 (@pxref{Comm}) to define aligned local common data.
5523 @section @code{.long @var{expressions}}
5525 @cindex @code{long} directive
5526 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5529 @c no one seems to know what this is for or whether this description is
5530 @c what it really ought to do
5532 @section @code{.lsym @var{symbol}, @var{expression}}
5534 @cindex @code{lsym} directive
5535 @cindex symbol, not referenced in assembly
5536 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5537 the hash table, ensuring it cannot be referenced by name during the
5538 rest of the assembly. This sets the attributes of the symbol to be
5539 the same as the expression value:
5541 @var{other} = @var{descriptor} = 0
5542 @var{type} = @r{(section of @var{expression})}
5543 @var{value} = @var{expression}
5546 The new symbol is not flagged as external.
5550 @section @code{.macro}
5553 The commands @code{.macro} and @code{.endm} allow you to define macros that
5554 generate assembly output. For example, this definition specifies a macro
5555 @code{sum} that puts a sequence of numbers into memory:
5558 .macro sum from=0, to=5
5567 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5579 @item .macro @var{macname}
5580 @itemx .macro @var{macname} @var{macargs} @dots{}
5581 @cindex @code{macro} directive
5582 Begin the definition of a macro called @var{macname}. If your macro
5583 definition requires arguments, specify their names after the macro name,
5584 separated by commas or spaces. You can qualify the macro argument to
5585 indicate whether all invocations must specify a non-blank value (through
5586 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5587 (through @samp{:@code{vararg}}). You can supply a default value for any
5588 macro argument by following the name with @samp{=@var{deflt}}. You
5589 cannot define two macros with the same @var{macname} unless it has been
5590 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5591 definitions. For example, these are all valid @code{.macro} statements:
5595 Begin the definition of a macro called @code{comm}, which takes no
5598 @item .macro plus1 p, p1
5599 @itemx .macro plus1 p p1
5600 Either statement begins the definition of a macro called @code{plus1},
5601 which takes two arguments; within the macro definition, write
5602 @samp{\p} or @samp{\p1} to evaluate the arguments.
5604 @item .macro reserve_str p1=0 p2
5605 Begin the definition of a macro called @code{reserve_str}, with two
5606 arguments. The first argument has a default value, but not the second.
5607 After the definition is complete, you can call the macro either as
5608 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5609 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5610 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5611 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5613 @item .macro m p1:req, p2=0, p3:vararg
5614 Begin the definition of a macro called @code{m}, with at least three
5615 arguments. The first argument must always have a value specified, but
5616 not the second, which instead has a default value. The third formal
5617 will get assigned all remaining arguments specified at invocation time.
5619 When you call a macro, you can specify the argument values either by
5620 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5621 @samp{sum to=17, from=9}.
5625 Note that since each of the @var{macargs} can be an identifier exactly
5626 as any other one permitted by the target architecture, there may be
5627 occasional problems if the target hand-crafts special meanings to certain
5628 characters when they occur in a special position. For example, if the colon
5629 (@code{:}) is generally permitted to be part of a symbol name, but the
5630 architecture specific code special-cases it when occurring as the final
5631 character of a symbol (to denote a label), then the macro parameter
5632 replacement code will have no way of knowing that and consider the whole
5633 construct (including the colon) an identifier, and check only this
5634 identifier for being the subject to parameter substitution. So for example
5635 this macro definition:
5643 might not work as expected. Invoking @samp{label foo} might not create a label
5644 called @samp{foo} but instead just insert the text @samp{\l:} into the
5645 assembler source, probably generating an error about an unrecognised
5648 Similarly problems might occur with the period character (@samp{.})
5649 which is often allowed inside opcode names (and hence identifier names). So
5650 for example constructing a macro to build an opcode from a base name and a
5651 length specifier like this:
5654 .macro opcode base length
5659 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5660 instruction but instead generate some kind of error as the assembler tries to
5661 interpret the text @samp{\base.\length}.
5663 There are several possible ways around this problem:
5666 @item Insert white space
5667 If it is possible to use white space characters then this is the simplest
5676 @item Use @samp{\()}
5677 The string @samp{\()} can be used to separate the end of a macro argument from
5678 the following text. eg:
5681 .macro opcode base length
5686 @item Use the alternate macro syntax mode
5687 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5688 used as a separator. eg:
5698 Note: this problem of correctly identifying string parameters to pseudo ops
5699 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5700 and @code{.irpc} (@pxref{Irpc}) as well.
5703 @cindex @code{endm} directive
5704 Mark the end of a macro definition.
5707 @cindex @code{exitm} directive
5708 Exit early from the current macro definition.
5710 @cindex number of macros executed
5711 @cindex macros, count executed
5713 @command{@value{AS}} maintains a counter of how many macros it has
5714 executed in this pseudo-variable; you can copy that number to your
5715 output with @samp{\@@}, but @emph{only within a macro definition}.
5717 @item LOCAL @var{name} [ , @dots{} ]
5718 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5719 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5720 @xref{Altmacro,,@code{.altmacro}}.
5724 @section @code{.mri @var{val}}
5726 @cindex @code{mri} directive
5727 @cindex MRI mode, temporarily
5728 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5729 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5730 affects code assembled until the next @code{.mri} directive, or until the end
5731 of the file. @xref{M, MRI mode, MRI mode}.
5734 @section @code{.noaltmacro}
5735 Disable alternate macro mode. @xref{Altmacro}.
5738 @section @code{.nolist}
5740 @cindex @code{nolist} directive
5741 @cindex listing control, turning off
5742 Control (in conjunction with the @code{.list} directive) whether or
5743 not assembly listings are generated. These two directives maintain an
5744 internal counter (which is zero initially). @code{.list} increments the
5745 counter, and @code{.nolist} decrements it. Assembly listings are
5746 generated whenever the counter is greater than zero.
5749 @section @code{.octa @var{bignums}}
5751 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5752 @cindex @code{octa} directive
5753 @cindex integer, 16-byte
5754 @cindex sixteen byte integer
5755 This directive expects zero or more bignums, separated by commas. For each
5756 bignum, it emits a 16-byte integer.
5758 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5759 hence @emph{octa}-word for 16 bytes.
5762 @section @code{.offset @var{loc}}
5764 @cindex @code{offset} directive
5765 Set the location counter to @var{loc} in the absolute section. @var{loc} must
5766 be an absolute expression. This directive may be useful for defining
5767 symbols with absolute values. Do not confuse it with the @code{.org}
5771 @section @code{.org @var{new-lc} , @var{fill}}
5773 @cindex @code{org} directive
5774 @cindex location counter, advancing
5775 @cindex advancing location counter
5776 @cindex current address, advancing
5777 Advance the location counter of the current section to
5778 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5779 expression with the same section as the current subsection. That is,
5780 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5781 wrong section, the @code{.org} directive is ignored. To be compatible
5782 with former assemblers, if the section of @var{new-lc} is absolute,
5783 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5784 is the same as the current subsection.
5786 @code{.org} may only increase the location counter, or leave it
5787 unchanged; you cannot use @code{.org} to move the location counter
5790 @c double negative used below "not undefined" because this is a specific
5791 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5792 @c section. doc@cygnus.com 18feb91
5793 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5794 may not be undefined. If you really detest this restriction we eagerly await
5795 a chance to share your improved assembler.
5797 Beware that the origin is relative to the start of the section, not
5798 to the start of the subsection. This is compatible with other
5799 people's assemblers.
5801 When the location counter (of the current subsection) is advanced, the
5802 intervening bytes are filled with @var{fill} which should be an
5803 absolute expression. If the comma and @var{fill} are omitted,
5804 @var{fill} defaults to zero.
5807 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5809 @cindex padding the location counter given a power of two
5810 @cindex @code{p2align} directive
5811 Pad the location counter (in the current subsection) to a particular
5812 storage boundary. The first expression (which must be absolute) is the
5813 number of low-order zero bits the location counter must have after
5814 advancement. For example @samp{.p2align 3} advances the location
5815 counter until it a multiple of 8. If the location counter is already a
5816 multiple of 8, no change is needed.
5818 The second expression (also absolute) gives the fill value to be stored in the
5819 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5820 padding bytes are normally zero. However, on some systems, if the section is
5821 marked as containing code and the fill value is omitted, the space is filled
5822 with no-op instructions.
5824 The third expression is also absolute, and is also optional. If it is present,
5825 it is the maximum number of bytes that should be skipped by this alignment
5826 directive. If doing the alignment would require skipping more bytes than the
5827 specified maximum, then the alignment is not done at all. You can omit the
5828 fill value (the second argument) entirely by simply using two commas after the
5829 required alignment; this can be useful if you want the alignment to be filled
5830 with no-op instructions when appropriate.
5832 @cindex @code{p2alignw} directive
5833 @cindex @code{p2alignl} directive
5834 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5835 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5836 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5837 fill pattern as a four byte longword value. For example, @code{.p2alignw
5838 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5839 filled in with the value 0x368d (the exact placement of the bytes depends upon
5840 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5845 @section @code{.popsection}
5847 @cindex @code{popsection} directive
5848 @cindex Section Stack
5849 This is one of the ELF section stack manipulation directives. The others are
5850 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5851 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5854 This directive replaces the current section (and subsection) with the top
5855 section (and subsection) on the section stack. This section is popped off the
5861 @section @code{.previous}
5863 @cindex @code{previous} directive
5864 @cindex Section Stack
5865 This is one of the ELF section stack manipulation directives. The others are
5866 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5867 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5868 (@pxref{PopSection}).
5870 This directive swaps the current section (and subsection) with most recently
5871 referenced section/subsection pair prior to this one. Multiple
5872 @code{.previous} directives in a row will flip between two sections (and their
5873 subsections). For example:
5885 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
5891 # Now in section A subsection 1
5895 # Now in section B subsection 0
5898 # Now in section B subsection 1
5901 # Now in section B subsection 0
5905 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
5906 section B and 0x9abc into subsection 1 of section B.
5908 In terms of the section stack, this directive swaps the current section with
5909 the top section on the section stack.
5913 @section @code{.print @var{string}}
5915 @cindex @code{print} directive
5916 @command{@value{AS}} will print @var{string} on the standard output during
5917 assembly. You must put @var{string} in double quotes.
5921 @section @code{.protected @var{names}}
5923 @cindex @code{protected} directive
5925 This is one of the ELF visibility directives. The other two are
5926 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5928 This directive overrides the named symbols default visibility (which is set by
5929 their binding: local, global or weak). The directive sets the visibility to
5930 @code{protected} which means that any references to the symbols from within the
5931 components that defines them must be resolved to the definition in that
5932 component, even if a definition in another component would normally preempt
5937 @section @code{.psize @var{lines} , @var{columns}}
5939 @cindex @code{psize} directive
5940 @cindex listing control: paper size
5941 @cindex paper size, for listings
5942 Use this directive to declare the number of lines---and, optionally, the
5943 number of columns---to use for each page, when generating listings.
5945 If you do not use @code{.psize}, listings use a default line-count
5946 of 60. You may omit the comma and @var{columns} specification; the
5947 default width is 200 columns.
5949 @command{@value{AS}} generates formfeeds whenever the specified number of
5950 lines is exceeded (or whenever you explicitly request one, using
5953 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5954 those explicitly specified with @code{.eject}.
5957 @section @code{.purgem @var{name}}
5959 @cindex @code{purgem} directive
5960 Undefine the macro @var{name}, so that later uses of the string will not be
5961 expanded. @xref{Macro}.
5965 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
5967 @cindex @code{pushsection} directive
5968 @cindex Section Stack
5969 This is one of the ELF section stack manipulation directives. The others are
5970 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5971 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5974 This directive pushes the current section (and subsection) onto the
5975 top of the section stack, and then replaces the current section and
5976 subsection with @code{name} and @code{subsection}. The optional
5977 @code{flags}, @code{type} and @code{arguments} are treated the same
5978 as in the @code{.section} (@pxref{Section}) directive.
5982 @section @code{.quad @var{bignums}}
5984 @cindex @code{quad} directive
5985 @code{.quad} expects zero or more bignums, separated by commas. For
5986 each bignum, it emits
5988 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5989 warning message; and just takes the lowest order 8 bytes of the bignum.
5990 @cindex eight-byte integer
5991 @cindex integer, 8-byte
5993 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5994 hence @emph{quad}-word for 8 bytes.
5997 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5998 warning message; and just takes the lowest order 16 bytes of the bignum.
5999 @cindex sixteen-byte integer
6000 @cindex integer, 16-byte
6004 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6006 @cindex @code{reloc} directive
6007 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6008 @var{expression}. If @var{offset} is a number, the relocation is generated in
6009 the current section. If @var{offset} is an expression that resolves to a
6010 symbol plus offset, the relocation is generated in the given symbol's section.
6011 @var{expression}, if present, must resolve to a symbol plus addend or to an
6012 absolute value, but note that not all targets support an addend. e.g. ELF REL
6013 targets such as i386 store an addend in the section contents rather than in the
6014 relocation. This low level interface does not support addends stored in the
6018 @section @code{.rept @var{count}}
6020 @cindex @code{rept} directive
6021 Repeat the sequence of lines between the @code{.rept} directive and the next
6022 @code{.endr} directive @var{count} times.
6024 For example, assembling
6032 is equivalent to assembling
6041 @section @code{.sbttl "@var{subheading}"}
6043 @cindex @code{sbttl} directive
6044 @cindex subtitles for listings
6045 @cindex listing control: subtitle
6046 Use @var{subheading} as the title (third line, immediately after the
6047 title line) when generating assembly listings.
6049 This directive affects subsequent pages, as well as the current page if
6050 it appears within ten lines of the top of a page.
6054 @section @code{.scl @var{class}}
6056 @cindex @code{scl} directive
6057 @cindex symbol storage class (COFF)
6058 @cindex COFF symbol storage class
6059 Set the storage-class value for a symbol. This directive may only be
6060 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6061 whether a symbol is static or external, or it may record further
6062 symbolic debugging information.
6065 The @samp{.scl} directive is primarily associated with COFF output; when
6066 configured to generate @code{b.out} output format, @command{@value{AS}}
6067 accepts this directive but ignores it.
6073 @section @code{.section @var{name}}
6075 @cindex named section
6076 Use the @code{.section} directive to assemble the following code into a section
6079 This directive is only supported for targets that actually support arbitrarily
6080 named sections; on @code{a.out} targets, for example, it is not accepted, even
6081 with a standard @code{a.out} section name.
6085 @c only print the extra heading if both COFF and ELF are set
6086 @subheading COFF Version
6089 @cindex @code{section} directive (COFF version)
6090 For COFF targets, the @code{.section} directive is used in one of the following
6094 .section @var{name}[, "@var{flags}"]
6095 .section @var{name}[, @var{subsection}]
6098 If the optional argument is quoted, it is taken as flags to use for the
6099 section. Each flag is a single character. The following flags are recognized:
6102 bss section (uninitialized data)
6104 section is not loaded
6110 exclude section from linking
6116 shared section (meaningful for PE targets)
6118 ignored. (For compatibility with the ELF version)
6120 section is not readable (meaningful for PE targets)
6122 single-digit power-of-two section alignment (GNU extension)
6125 If no flags are specified, the default flags depend upon the section name. If
6126 the section name is not recognized, the default will be for the section to be
6127 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6128 from the section, rather than adding them, so if they are used on their own it
6129 will be as if no flags had been specified at all.
6131 If the optional argument to the @code{.section} directive is not quoted, it is
6132 taken as a subsection number (@pxref{Sub-Sections}).
6137 @c only print the extra heading if both COFF and ELF are set
6138 @subheading ELF Version
6141 @cindex Section Stack
6142 This is one of the ELF section stack manipulation directives. The others are
6143 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6144 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6145 @code{.previous} (@pxref{Previous}).
6147 @cindex @code{section} directive (ELF version)
6148 For ELF targets, the @code{.section} directive is used like this:
6151 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6154 The optional @var{flags} argument is a quoted string which may contain any
6155 combination of the following characters:
6158 section is allocatable
6160 section is excluded from executable and shared library.
6164 section is executable
6166 section is mergeable
6168 section contains zero terminated strings
6170 section is a member of a section group
6172 section is used for thread-local-storage
6174 section is a member of the previously-current section's group, if any
6177 The optional @var{type} argument may contain one of the following constants:
6180 section contains data
6182 section does not contain data (i.e., section only occupies space)
6184 section contains data which is used by things other than the program
6186 section contains an array of pointers to init functions
6188 section contains an array of pointers to finish functions
6189 @item @@preinit_array
6190 section contains an array of pointers to pre-init functions
6193 Many targets only support the first three section types.
6195 Note on targets where the @code{@@} character is the start of a comment (eg
6196 ARM) then another character is used instead. For example the ARM port uses the
6199 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6200 be specified as well as an extra argument---@var{entsize}---like this:
6203 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6206 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6207 constants, each @var{entsize} octets long. Sections with both @code{M} and
6208 @code{S} must contain zero terminated strings where each character is
6209 @var{entsize} bytes long. The linker may remove duplicates within sections with
6210 the same name, same entity size and same flags. @var{entsize} must be an
6211 absolute expression. For sections with both @code{M} and @code{S}, a string
6212 which is a suffix of a larger string is considered a duplicate. Thus
6213 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6214 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6216 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6217 be present along with an additional field like this:
6220 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6223 The @var{GroupName} field specifies the name of the section group to which this
6224 particular section belongs. The optional linkage field can contain:
6227 indicates that only one copy of this section should be retained
6232 Note: if both the @var{M} and @var{G} flags are present then the fields for
6233 the Merge flag should come first, like this:
6236 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6239 If @var{flags} contains the @code{?} symbol then it may not also contain the
6240 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6241 present. Instead, @code{?} says to consider the section that's current before
6242 this directive. If that section used @code{G}, then the new section will use
6243 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6244 If not, then the @code{?} symbol has no effect.
6246 If no flags are specified, the default flags depend upon the section name. If
6247 the section name is not recognized, the default will be for the section to have
6248 none of the above flags: it will not be allocated in memory, nor writable, nor
6249 executable. The section will contain data.
6251 For ELF targets, the assembler supports another type of @code{.section}
6252 directive for compatibility with the Solaris assembler:
6255 .section "@var{name}"[, @var{flags}...]
6258 Note that the section name is quoted. There may be a sequence of comma
6262 section is allocatable
6266 section is executable
6268 section is excluded from executable and shared library.
6270 section is used for thread local storage
6273 This directive replaces the current section and subsection. See the
6274 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6275 some examples of how this directive and the other section stack directives
6281 @section @code{.set @var{symbol}, @var{expression}}
6283 @cindex @code{set} directive
6284 @cindex symbol value, setting
6285 Set the value of @var{symbol} to @var{expression}. This
6286 changes @var{symbol}'s value and type to conform to
6287 @var{expression}. If @var{symbol} was flagged as external, it remains
6288 flagged (@pxref{Symbol Attributes}).
6290 You may @code{.set} a symbol many times in the same assembly.
6292 If you @code{.set} a global symbol, the value stored in the object
6293 file is the last value stored into it.
6296 On Z80 @code{set} is a real instruction, use
6297 @samp{@var{symbol} defl @var{expression}} instead.
6301 @section @code{.short @var{expressions}}
6303 @cindex @code{short} directive
6305 @code{.short} is normally the same as @samp{.word}.
6306 @xref{Word,,@code{.word}}.
6308 In some configurations, however, @code{.short} and @code{.word} generate
6309 numbers of different lengths. @xref{Machine Dependencies}.
6313 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6316 This expects zero or more @var{expressions}, and emits
6317 a 16 bit number for each.
6322 @section @code{.single @var{flonums}}
6324 @cindex @code{single} directive
6325 @cindex floating point numbers (single)
6326 This directive assembles zero or more flonums, separated by commas. It
6327 has the same effect as @code{.float}.
6329 The exact kind of floating point numbers emitted depends on how
6330 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6334 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6335 numbers in @sc{ieee} format.
6341 @section @code{.size}
6343 This directive is used to set the size associated with a symbol.
6347 @c only print the extra heading if both COFF and ELF are set
6348 @subheading COFF Version
6351 @cindex @code{size} directive (COFF version)
6352 For COFF targets, the @code{.size} directive is only permitted inside
6353 @code{.def}/@code{.endef} pairs. It is used like this:
6356 .size @var{expression}
6360 @samp{.size} is only meaningful when generating COFF format output; when
6361 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6368 @c only print the extra heading if both COFF and ELF are set
6369 @subheading ELF Version
6372 @cindex @code{size} directive (ELF version)
6373 For ELF targets, the @code{.size} directive is used like this:
6376 .size @var{name} , @var{expression}
6379 This directive sets the size associated with a symbol @var{name}.
6380 The size in bytes is computed from @var{expression} which can make use of label
6381 arithmetic. This directive is typically used to set the size of function
6386 @ifclear no-space-dir
6388 @section @code{.skip @var{size} , @var{fill}}
6390 @cindex @code{skip} directive
6391 @cindex filling memory
6392 This directive emits @var{size} bytes, each of value @var{fill}. Both
6393 @var{size} and @var{fill} are absolute expressions. If the comma and
6394 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6399 @section @code{.sleb128 @var{expressions}}
6401 @cindex @code{sleb128} directive
6402 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6403 compact, variable length representation of numbers used by the DWARF
6404 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6406 @ifclear no-space-dir
6408 @section @code{.space @var{size} , @var{fill}}
6410 @cindex @code{space} directive
6411 @cindex filling memory
6412 This directive emits @var{size} bytes, each of value @var{fill}. Both
6413 @var{size} and @var{fill} are absolute expressions. If the comma
6414 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6419 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6420 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6421 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6422 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6430 @section @code{.stabd, .stabn, .stabs}
6432 @cindex symbolic debuggers, information for
6433 @cindex @code{stab@var{x}} directives
6434 There are three directives that begin @samp{.stab}.
6435 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6436 The symbols are not entered in the @command{@value{AS}} hash table: they
6437 cannot be referenced elsewhere in the source file.
6438 Up to five fields are required:
6442 This is the symbol's name. It may contain any character except
6443 @samp{\000}, so is more general than ordinary symbol names. Some
6444 debuggers used to code arbitrarily complex structures into symbol names
6448 An absolute expression. The symbol's type is set to the low 8 bits of
6449 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6450 and debuggers choke on silly bit patterns.
6453 An absolute expression. The symbol's ``other'' attribute is set to the
6454 low 8 bits of this expression.
6457 An absolute expression. The symbol's descriptor is set to the low 16
6458 bits of this expression.
6461 An absolute expression which becomes the symbol's value.
6464 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6465 or @code{.stabs} statement, the symbol has probably already been created;
6466 you get a half-formed symbol in your object file. This is
6467 compatible with earlier assemblers!
6470 @cindex @code{stabd} directive
6471 @item .stabd @var{type} , @var{other} , @var{desc}
6473 The ``name'' of the symbol generated is not even an empty string.
6474 It is a null pointer, for compatibility. Older assemblers used a
6475 null pointer so they didn't waste space in object files with empty
6478 The symbol's value is set to the location counter,
6479 relocatably. When your program is linked, the value of this symbol
6480 is the address of the location counter when the @code{.stabd} was
6483 @cindex @code{stabn} directive
6484 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6485 The name of the symbol is set to the empty string @code{""}.
6487 @cindex @code{stabs} directive
6488 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6489 All five fields are specified.
6495 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6496 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6498 @cindex string, copying to object file
6499 @cindex string8, copying to object file
6500 @cindex string16, copying to object file
6501 @cindex string32, copying to object file
6502 @cindex string64, copying to object file
6503 @cindex @code{string} directive
6504 @cindex @code{string8} directive
6505 @cindex @code{string16} directive
6506 @cindex @code{string32} directive
6507 @cindex @code{string64} directive
6509 Copy the characters in @var{str} to the object file. You may specify more than
6510 one string to copy, separated by commas. Unless otherwise specified for a
6511 particular machine, the assembler marks the end of each string with a 0 byte.
6512 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6514 The variants @code{string16}, @code{string32} and @code{string64} differ from
6515 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6516 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6517 are stored in target endianness byte order.
6523 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6524 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6529 @section @code{.struct @var{expression}}
6531 @cindex @code{struct} directive
6532 Switch to the absolute section, and set the section offset to @var{expression},
6533 which must be an absolute expression. You might use this as follows:
6542 This would define the symbol @code{field1} to have the value 0, the symbol
6543 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6544 value 8. Assembly would be left in the absolute section, and you would need to
6545 use a @code{.section} directive of some sort to change to some other section
6546 before further assembly.
6550 @section @code{.subsection @var{name}}
6552 @cindex @code{subsection} directive
6553 @cindex Section Stack
6554 This is one of the ELF section stack manipulation directives. The others are
6555 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6556 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6559 This directive replaces the current subsection with @code{name}. The current
6560 section is not changed. The replaced subsection is put onto the section stack
6561 in place of the then current top of stack subsection.
6566 @section @code{.symver}
6567 @cindex @code{symver} directive
6568 @cindex symbol versioning
6569 @cindex versions of symbols
6570 Use the @code{.symver} directive to bind symbols to specific version nodes
6571 within a source file. This is only supported on ELF platforms, and is
6572 typically used when assembling files to be linked into a shared library.
6573 There are cases where it may make sense to use this in objects to be bound
6574 into an application itself so as to override a versioned symbol from a
6577 For ELF targets, the @code{.symver} directive can be used like this:
6579 .symver @var{name}, @var{name2@@nodename}
6581 If the symbol @var{name} is defined within the file
6582 being assembled, the @code{.symver} directive effectively creates a symbol
6583 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6584 just don't try and create a regular alias is that the @var{@@} character isn't
6585 permitted in symbol names. The @var{name2} part of the name is the actual name
6586 of the symbol by which it will be externally referenced. The name @var{name}
6587 itself is merely a name of convenience that is used so that it is possible to
6588 have definitions for multiple versions of a function within a single source
6589 file, and so that the compiler can unambiguously know which version of a
6590 function is being mentioned. The @var{nodename} portion of the alias should be
6591 the name of a node specified in the version script supplied to the linker when
6592 building a shared library. If you are attempting to override a versioned
6593 symbol from a shared library, then @var{nodename} should correspond to the
6594 nodename of the symbol you are trying to override.
6596 If the symbol @var{name} is not defined within the file being assembled, all
6597 references to @var{name} will be changed to @var{name2@@nodename}. If no
6598 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6601 Another usage of the @code{.symver} directive is:
6603 .symver @var{name}, @var{name2@@@@nodename}
6605 In this case, the symbol @var{name} must exist and be defined within
6606 the file being assembled. It is similar to @var{name2@@nodename}. The
6607 difference is @var{name2@@@@nodename} will also be used to resolve
6608 references to @var{name2} by the linker.
6610 The third usage of the @code{.symver} directive is:
6612 .symver @var{name}, @var{name2@@@@@@nodename}
6614 When @var{name} is not defined within the
6615 file being assembled, it is treated as @var{name2@@nodename}. When
6616 @var{name} is defined within the file being assembled, the symbol
6617 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6622 @section @code{.tag @var{structname}}
6624 @cindex COFF structure debugging
6625 @cindex structure debugging, COFF
6626 @cindex @code{tag} directive
6627 This directive is generated by compilers to include auxiliary debugging
6628 information in the symbol table. It is only permitted inside
6629 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6630 definitions in the symbol table with instances of those structures.
6633 @samp{.tag} is only used when generating COFF format output; when
6634 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6640 @section @code{.text @var{subsection}}
6642 @cindex @code{text} directive
6643 Tells @command{@value{AS}} to assemble the following statements onto the end of
6644 the text subsection numbered @var{subsection}, which is an absolute
6645 expression. If @var{subsection} is omitted, subsection number zero
6649 @section @code{.title "@var{heading}"}
6651 @cindex @code{title} directive
6652 @cindex listing control: title line
6653 Use @var{heading} as the title (second line, immediately after the
6654 source file name and pagenumber) when generating assembly listings.
6656 This directive affects subsequent pages, as well as the current page if
6657 it appears within ten lines of the top of a page.
6661 @section @code{.type}
6663 This directive is used to set the type of a symbol.
6667 @c only print the extra heading if both COFF and ELF are set
6668 @subheading COFF Version
6671 @cindex COFF symbol type
6672 @cindex symbol type, COFF
6673 @cindex @code{type} directive (COFF version)
6674 For COFF targets, this directive is permitted only within
6675 @code{.def}/@code{.endef} pairs. It is used like this:
6681 This records the integer @var{int} as the type attribute of a symbol table
6685 @samp{.type} is associated only with COFF format output; when
6686 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6687 directive but ignores it.
6693 @c only print the extra heading if both COFF and ELF are set
6694 @subheading ELF Version
6697 @cindex ELF symbol type
6698 @cindex symbol type, ELF
6699 @cindex @code{type} directive (ELF version)
6700 For ELF targets, the @code{.type} directive is used like this:
6703 .type @var{name} , @var{type description}
6706 This sets the type of symbol @var{name} to be either a
6707 function symbol or an object symbol. There are five different syntaxes
6708 supported for the @var{type description} field, in order to provide
6709 compatibility with various other assemblers.
6711 Because some of the characters used in these syntaxes (such as @samp{@@} and
6712 @samp{#}) are comment characters for some architectures, some of the syntaxes
6713 below do not work on all architectures. The first variant will be accepted by
6714 the GNU assembler on all architectures so that variant should be used for
6715 maximum portability, if you do not need to assemble your code with other
6718 The syntaxes supported are:
6721 .type <name> STT_<TYPE_IN_UPPER_CASE>
6722 .type <name>,#<type>
6723 .type <name>,@@<type>
6724 .type <name>,%<type>
6725 .type <name>,"<type>"
6728 The types supported are:
6733 Mark the symbol as being a function name.
6736 @itemx gnu_indirect_function
6737 Mark the symbol as an indirect function when evaluated during reloc
6738 processing. (This is only supported on assemblers targeting GNU systems).
6742 Mark the symbol as being a data object.
6746 Mark the symbol as being a thead-local data object.
6750 Mark the symbol as being a common data object.
6754 Does not mark the symbol in any way. It is supported just for completeness.
6756 @item gnu_unique_object
6757 Marks the symbol as being a globally unique data object. The dynamic linker
6758 will make sure that in the entire process there is just one symbol with this
6759 name and type in use. (This is only supported on assemblers targeting GNU
6764 Note: Some targets support extra types in addition to those listed above.
6770 @section @code{.uleb128 @var{expressions}}
6772 @cindex @code{uleb128} directive
6773 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6774 compact, variable length representation of numbers used by the DWARF
6775 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6779 @section @code{.val @var{addr}}
6781 @cindex @code{val} directive
6782 @cindex COFF value attribute
6783 @cindex value attribute, COFF
6784 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6785 records the address @var{addr} as the value attribute of a symbol table
6789 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6790 configured for @code{b.out}, it accepts this directive but ignores it.
6796 @section @code{.version "@var{string}"}
6798 @cindex @code{version} directive
6799 This directive creates a @code{.note} section and places into it an ELF
6800 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6805 @section @code{.vtable_entry @var{table}, @var{offset}}
6807 @cindex @code{vtable_entry} directive
6808 This directive finds or creates a symbol @code{table} and creates a
6809 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6812 @section @code{.vtable_inherit @var{child}, @var{parent}}
6814 @cindex @code{vtable_inherit} directive
6815 This directive finds the symbol @code{child} and finds or creates the symbol
6816 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6817 parent whose addend is the value of the child symbol. As a special case the
6818 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6822 @section @code{.warning "@var{string}"}
6823 @cindex warning directive
6824 Similar to the directive @code{.error}
6825 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6828 @section @code{.weak @var{names}}
6830 @cindex @code{weak} directive
6831 This directive sets the weak attribute on the comma separated list of symbol
6832 @code{names}. If the symbols do not already exist, they will be created.
6834 On COFF targets other than PE, weak symbols are a GNU extension. This
6835 directive sets the weak attribute on the comma separated list of symbol
6836 @code{names}. If the symbols do not already exist, they will be created.
6838 On the PE target, weak symbols are supported natively as weak aliases.
6839 When a weak symbol is created that is not an alias, GAS creates an
6840 alternate symbol to hold the default value.
6843 @section @code{.weakref @var{alias}, @var{target}}
6845 @cindex @code{weakref} directive
6846 This directive creates an alias to the target symbol that enables the symbol to
6847 be referenced with weak-symbol semantics, but without actually making it weak.
6848 If direct references or definitions of the symbol are present, then the symbol
6849 will not be weak, but if all references to it are through weak references, the
6850 symbol will be marked as weak in the symbol table.
6852 The effect is equivalent to moving all references to the alias to a separate
6853 assembly source file, renaming the alias to the symbol in it, declaring the
6854 symbol as weak there, and running a reloadable link to merge the object files
6855 resulting from the assembly of the new source file and the old source file that
6856 had the references to the alias removed.
6858 The alias itself never makes to the symbol table, and is entirely handled
6859 within the assembler.
6862 @section @code{.word @var{expressions}}
6864 @cindex @code{word} directive
6865 This directive expects zero or more @var{expressions}, of any section,
6866 separated by commas.
6869 For each expression, @command{@value{AS}} emits a 32-bit number.
6872 For each expression, @command{@value{AS}} emits a 16-bit number.
6877 The size of the number emitted, and its byte order,
6878 depend on what target computer the assembly is for.
6881 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
6882 @c happen---32-bit addressability, period; no long/short jumps.
6883 @ifset DIFF-TBL-KLUGE
6884 @cindex difference tables altered
6885 @cindex altered difference tables
6887 @emph{Warning: Special Treatment to support Compilers}
6891 Machines with a 32-bit address space, but that do less than 32-bit
6892 addressing, require the following special treatment. If the machine of
6893 interest to you does 32-bit addressing (or doesn't require it;
6894 @pxref{Machine Dependencies}), you can ignore this issue.
6897 In order to assemble compiler output into something that works,
6898 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6899 Directives of the form @samp{.word sym1-sym2} are often emitted by
6900 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6901 directive of the form @samp{.word sym1-sym2}, and the difference between
6902 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6903 creates a @dfn{secondary jump table}, immediately before the next label.
6904 This secondary jump table is preceded by a short-jump to the
6905 first byte after the secondary table. This short-jump prevents the flow
6906 of control from accidentally falling into the new table. Inside the
6907 table is a long-jump to @code{sym2}. The original @samp{.word}
6908 contains @code{sym1} minus the address of the long-jump to
6911 If there were several occurrences of @samp{.word sym1-sym2} before the
6912 secondary jump table, all of them are adjusted. If there was a
6913 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6914 long-jump to @code{sym4} is included in the secondary jump table,
6915 and the @code{.word} directives are adjusted to contain @code{sym3}
6916 minus the address of the long-jump to @code{sym4}; and so on, for as many
6917 entries in the original jump table as necessary.
6920 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6921 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6922 assembly language programmers.
6925 @c end DIFF-TBL-KLUGE
6928 @section Deprecated Directives
6930 @cindex deprecated directives
6931 @cindex obsolescent directives
6932 One day these directives won't work.
6933 They are included for compatibility with older assemblers.
6940 @node Object Attributes
6941 @chapter Object Attributes
6942 @cindex object attributes
6944 @command{@value{AS}} assembles source files written for a specific architecture
6945 into object files for that architecture. But not all object files are alike.
6946 Many architectures support incompatible variations. For instance, floating
6947 point arguments might be passed in floating point registers if the object file
6948 requires hardware floating point support---or floating point arguments might be
6949 passed in integer registers if the object file supports processors with no
6950 hardware floating point unit. Or, if two objects are built for different
6951 generations of the same architecture, the combination may require the
6952 newer generation at run-time.
6954 This information is useful during and after linking. At link time,
6955 @command{@value{LD}} can warn about incompatible object files. After link
6956 time, tools like @command{gdb} can use it to process the linked file
6959 Compatibility information is recorded as a series of object attributes. Each
6960 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
6961 string, and indicates who sets the meaning of the tag. The tag is an integer,
6962 and indicates what property the attribute describes. The value may be a string
6963 or an integer, and indicates how the property affects this object. Missing
6964 attributes are the same as attributes with a zero value or empty string value.
6966 Object attributes were developed as part of the ABI for the ARM Architecture.
6967 The file format is documented in @cite{ELF for the ARM Architecture}.
6970 * GNU Object Attributes:: @sc{gnu} Object Attributes
6971 * Defining New Object Attributes:: Defining New Object Attributes
6974 @node GNU Object Attributes
6975 @section @sc{gnu} Object Attributes
6977 The @code{.gnu_attribute} directive records an object attribute
6978 with vendor @samp{gnu}.
6980 Except for @samp{Tag_compatibility}, which has both an integer and a string for
6981 its value, @sc{gnu} attributes have a string value if the tag number is odd and
6982 an integer value if the tag number is even. The second bit (@code{@var{tag} &
6983 2} is set for architecture-independent attributes and clear for
6984 architecture-dependent ones.
6986 @subsection Common @sc{gnu} attributes
6988 These attributes are valid on all architectures.
6991 @item Tag_compatibility (32)
6992 The compatibility attribute takes an integer flag value and a vendor name. If
6993 the flag value is 0, the file is compatible with other toolchains. If it is 1,
6994 then the file is only compatible with the named toolchain. If it is greater
6995 than 1, the file can only be processed by other toolchains under some private
6996 arrangement indicated by the flag value and the vendor name.
6999 @subsection MIPS Attributes
7002 @item Tag_GNU_MIPS_ABI_FP (4)
7003 The floating-point ABI used by this object file. The value will be:
7007 0 for files not affected by the floating-point ABI.
7009 1 for files using the hardware floating-point with a standard double-precision
7012 2 for files using the hardware floating-point ABI with a single-precision FPU.
7014 3 for files using the software floating-point ABI.
7016 4 for files using the hardware floating-point ABI with 64-bit wide
7017 double-precision floating-point registers and 32-bit wide general
7022 @subsection PowerPC Attributes
7025 @item Tag_GNU_Power_ABI_FP (4)
7026 The floating-point ABI used by this object file. The value will be:
7030 0 for files not affected by the floating-point ABI.
7032 1 for files using double-precision hardware floating-point ABI.
7034 2 for files using the software floating-point ABI.
7036 3 for files using single-precision hardware floating-point ABI.
7039 @item Tag_GNU_Power_ABI_Vector (8)
7040 The vector ABI used by this object file. The value will be:
7044 0 for files not affected by the vector ABI.
7046 1 for files using general purpose registers to pass vectors.
7048 2 for files using AltiVec registers to pass vectors.
7050 3 for files using SPE registers to pass vectors.
7054 @node Defining New Object Attributes
7055 @section Defining New Object Attributes
7057 If you want to define a new @sc{gnu} object attribute, here are the places you
7058 will need to modify. New attributes should be discussed on the @samp{binutils}
7063 This manual, which is the official register of attributes.
7065 The header for your architecture @file{include/elf}, to define the tag.
7067 The @file{bfd} support file for your architecture, to merge the attribute
7068 and issue any appropriate link warnings.
7070 Test cases in @file{ld/testsuite} for merging and link warnings.
7072 @file{binutils/readelf.c} to display your attribute.
7074 GCC, if you want the compiler to mark the attribute automatically.
7080 @node Machine Dependencies
7081 @chapter Machine Dependent Features
7083 @cindex machine dependencies
7084 The machine instruction sets are (almost by definition) different on
7085 each machine where @command{@value{AS}} runs. Floating point representations
7086 vary as well, and @command{@value{AS}} often supports a few additional
7087 directives or command-line options for compatibility with other
7088 assemblers on a particular platform. Finally, some versions of
7089 @command{@value{AS}} support special pseudo-instructions for branch
7092 This chapter discusses most of these differences, though it does not
7093 include details on any machine's instruction set. For details on that
7094 subject, see the hardware manufacturer's manual.
7098 * AArch64-Dependent:: AArch64 Dependent Features
7101 * Alpha-Dependent:: Alpha Dependent Features
7104 * ARC-Dependent:: ARC Dependent Features
7107 * ARM-Dependent:: ARM Dependent Features
7110 * AVR-Dependent:: AVR Dependent Features
7113 * Blackfin-Dependent:: Blackfin Dependent Features
7116 * CR16-Dependent:: CR16 Dependent Features
7119 * CRIS-Dependent:: CRIS Dependent Features
7122 * D10V-Dependent:: D10V Dependent Features
7125 * D30V-Dependent:: D30V Dependent Features
7128 * Epiphany-Dependent:: EPIPHANY Dependent Features
7131 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7134 * HPPA-Dependent:: HPPA Dependent Features
7137 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
7140 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7143 * i860-Dependent:: Intel 80860 Dependent Features
7146 * i960-Dependent:: Intel 80960 Dependent Features
7149 * IA-64-Dependent:: Intel IA-64 Dependent Features
7152 * IP2K-Dependent:: IP2K Dependent Features
7155 * LM32-Dependent:: LM32 Dependent Features
7158 * M32C-Dependent:: M32C Dependent Features
7161 * M32R-Dependent:: M32R Dependent Features
7164 * M68K-Dependent:: M680x0 Dependent Features
7167 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7170 * Meta-Dependent :: Meta Dependent Features
7173 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7176 * MIPS-Dependent:: MIPS Dependent Features
7179 * MMIX-Dependent:: MMIX Dependent Features
7182 * MSP430-Dependent:: MSP430 Dependent Features
7185 * NDS32-Dependent:: Andes NDS32 Dependent Features
7188 * NiosII-Dependent:: Altera Nios II Dependent Features
7191 * NS32K-Dependent:: NS32K Dependent Features
7194 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7195 * SH64-Dependent:: SuperH SH64 Dependent Features
7198 * PDP-11-Dependent:: PDP-11 Dependent Features
7201 * PJ-Dependent:: picoJava Dependent Features
7204 * PPC-Dependent:: PowerPC Dependent Features
7207 * RL78-Dependent:: RL78 Dependent Features
7210 * RX-Dependent:: RX Dependent Features
7213 * S/390-Dependent:: IBM S/390 Dependent Features
7216 * SCORE-Dependent:: SCORE Dependent Features
7219 * Sparc-Dependent:: SPARC Dependent Features
7222 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7225 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7228 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7231 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7234 * V850-Dependent:: V850 Dependent Features
7237 * XGATE-Dependent:: XGATE Features
7240 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7243 * Xtensa-Dependent:: Xtensa Dependent Features
7246 * Z80-Dependent:: Z80 Dependent Features
7249 * Z8000-Dependent:: Z8000 Dependent Features
7252 * Vax-Dependent:: VAX Dependent Features
7259 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7260 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7261 @c peculiarity: to preserve cross-references, there must be a node called
7262 @c "Machine Dependencies". Hence the conditional nodenames in each
7263 @c major node below. Node defaulting in makeinfo requires adjacency of
7264 @c node and sectioning commands; hence the repetition of @chapter BLAH
7265 @c in both conditional blocks.
7268 @include c-aarch64.texi
7272 @include c-alpha.texi
7288 @include c-bfin.texi
7292 @include c-cr16.texi
7296 @include c-cris.texi
7301 @node Machine Dependencies
7302 @chapter Machine Dependent Features
7304 The machine instruction sets are different on each Renesas chip family,
7305 and there are also some syntax differences among the families. This
7306 chapter describes the specific @command{@value{AS}} features for each
7310 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7311 * SH-Dependent:: Renesas SH Dependent Features
7318 @include c-d10v.texi
7322 @include c-d30v.texi
7326 @include c-epiphany.texi
7330 @include c-h8300.texi
7334 @include c-hppa.texi
7338 @include c-i370.texi
7342 @include c-i386.texi
7346 @include c-i860.texi
7350 @include c-i960.texi
7354 @include c-ia64.texi
7358 @include c-ip2k.texi
7362 @include c-lm32.texi
7366 @include c-m32c.texi
7370 @include c-m32r.texi
7374 @include c-m68k.texi
7378 @include c-m68hc11.texi
7382 @include c-metag.texi
7386 @include c-microblaze.texi
7390 @include c-mips.texi
7394 @include c-mmix.texi
7398 @include c-msp430.texi
7402 @include c-nds32.texi
7406 @include c-nios2.texi
7410 @include c-ns32k.texi
7414 @include c-pdp11.texi
7426 @include c-rl78.texi
7434 @include c-s390.texi
7438 @include c-score.texi
7443 @include c-sh64.texi
7447 @include c-sparc.texi
7451 @include c-tic54x.texi
7455 @include c-tic6x.texi
7459 @include c-tilegx.texi
7463 @include c-tilepro.texi
7479 @include c-v850.texi
7483 @include c-xgate.texi
7487 @include c-xstormy16.texi
7491 @include c-xtensa.texi
7495 @c reverse effect of @down at top of generic Machine-Dep chapter
7499 @node Reporting Bugs
7500 @chapter Reporting Bugs
7501 @cindex bugs in assembler
7502 @cindex reporting bugs in assembler
7504 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7506 Reporting a bug may help you by bringing a solution to your problem, or it may
7507 not. But in any case the principal function of a bug report is to help the
7508 entire community by making the next version of @command{@value{AS}} work better.
7509 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7511 In order for a bug report to serve its purpose, you must include the
7512 information that enables us to fix the bug.
7515 * Bug Criteria:: Have you found a bug?
7516 * Bug Reporting:: How to report bugs
7520 @section Have You Found a Bug?
7521 @cindex bug criteria
7523 If you are not sure whether you have found a bug, here are some guidelines:
7526 @cindex fatal signal
7527 @cindex assembler crash
7528 @cindex crash of assembler
7530 If the assembler gets a fatal signal, for any input whatever, that is a
7531 @command{@value{AS}} bug. Reliable assemblers never crash.
7533 @cindex error on valid input
7535 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7537 @cindex invalid input
7539 If @command{@value{AS}} does not produce an error message for invalid input, that
7540 is a bug. However, you should note that your idea of ``invalid input'' might
7541 be our idea of ``an extension'' or ``support for traditional practice''.
7544 If you are an experienced user of assemblers, your suggestions for improvement
7545 of @command{@value{AS}} are welcome in any case.
7549 @section How to Report Bugs
7551 @cindex assembler bugs, reporting
7553 A number of companies and individuals offer support for @sc{gnu} products. If
7554 you obtained @command{@value{AS}} from a support organization, we recommend you
7555 contact that organization first.
7557 You can find contact information for many support companies and
7558 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7562 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7566 The fundamental principle of reporting bugs usefully is this:
7567 @strong{report all the facts}. If you are not sure whether to state a
7568 fact or leave it out, state it!
7570 Often people omit facts because they think they know what causes the problem
7571 and assume that some details do not matter. Thus, you might assume that the
7572 name of a symbol you use in an example does not matter. Well, probably it does
7573 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7574 happens to fetch from the location where that name is stored in memory;
7575 perhaps, if the name were different, the contents of that location would fool
7576 the assembler into doing the right thing despite the bug. Play it safe and
7577 give a specific, complete example. That is the easiest thing for you to do,
7578 and the most helpful.
7580 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7581 it is new to us. Therefore, always write your bug reports on the assumption
7582 that the bug has not been reported previously.
7584 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7585 bell?'' This cannot help us fix a bug, so it is basically useless. We
7586 respond by asking for enough details to enable us to investigate.
7587 You might as well expedite matters by sending them to begin with.
7589 To enable us to fix the bug, you should include all these things:
7593 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
7594 it with the @samp{--version} argument.
7596 Without this, we will not know whether there is any point in looking for
7597 the bug in the current version of @command{@value{AS}}.
7600 Any patches you may have applied to the @command{@value{AS}} source.
7603 The type of machine you are using, and the operating system name and
7607 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
7611 The command arguments you gave the assembler to assemble your example and
7612 observe the bug. To guarantee you will not omit something important, list them
7613 all. A copy of the Makefile (or the output from make) is sufficient.
7615 If we were to try to guess the arguments, we would probably guess wrong
7616 and then we might not encounter the bug.
7619 A complete input file that will reproduce the bug. If the bug is observed when
7620 the assembler is invoked via a compiler, send the assembler source, not the
7621 high level language source. Most compilers will produce the assembler source
7622 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
7623 the options @samp{-v --save-temps}; this will save the assembler source in a
7624 file with an extension of @file{.s}, and also show you exactly how
7625 @command{@value{AS}} is being run.
7628 A description of what behavior you observe that you believe is
7629 incorrect. For example, ``It gets a fatal signal.''
7631 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
7632 will certainly notice it. But if the bug is incorrect output, we might not
7633 notice unless it is glaringly wrong. You might as well not give us a chance to
7636 Even if the problem you experience is a fatal signal, you should still say so
7637 explicitly. Suppose something strange is going on, such as, your copy of
7638 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
7639 library on your system. (This has happened!) Your copy might crash and ours
7640 would not. If you told us to expect a crash, then when ours fails to crash, we
7641 would know that the bug was not happening for us. If you had not told us to
7642 expect a crash, then we would not be able to draw any conclusion from our
7646 If you wish to suggest changes to the @command{@value{AS}} source, send us context
7647 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
7648 option. Always send diffs from the old file to the new file. If you even
7649 discuss something in the @command{@value{AS}} source, refer to it by context, not
7652 The line numbers in our development sources will not match those in your
7653 sources. Your line numbers would convey no useful information to us.
7656 Here are some things that are not necessary:
7660 A description of the envelope of the bug.
7662 Often people who encounter a bug spend a lot of time investigating
7663 which changes to the input file will make the bug go away and which
7664 changes will not affect it.
7666 This is often time consuming and not very useful, because the way we
7667 will find the bug is by running a single example under the debugger
7668 with breakpoints, not by pure deduction from a series of examples.
7669 We recommend that you save your time for something else.
7671 Of course, if you can find a simpler example to report @emph{instead}
7672 of the original one, that is a convenience for us. Errors in the
7673 output will be easier to spot, running under the debugger will take
7674 less time, and so on.
7676 However, simplification is not vital; if you do not want to do this,
7677 report the bug anyway and send us the entire test case you used.
7680 A patch for the bug.
7682 A patch for the bug does help us if it is a good one. But do not omit
7683 the necessary information, such as the test case, on the assumption that
7684 a patch is all we need. We might see problems with your patch and decide
7685 to fix the problem another way, or we might not understand it at all.
7687 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
7688 construct an example that will make the program follow a certain path through
7689 the code. If you do not send us the example, we will not be able to construct
7690 one, so we will not be able to verify that the bug is fixed.
7692 And if we cannot understand what bug you are trying to fix, or why your
7693 patch should be an improvement, we will not install it. A test case will
7694 help us to understand.
7697 A guess about what the bug is or what it depends on.
7699 Such guesses are usually wrong. Even we cannot guess right about such
7700 things without first using the debugger to find the facts.
7703 @node Acknowledgements
7704 @chapter Acknowledgements
7706 If you have contributed to GAS and your name isn't listed here,
7707 it is not meant as a slight. We just don't know about it. Send mail to the
7708 maintainer, and we'll correct the situation. Currently
7710 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
7712 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
7715 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
7716 information and the 68k series machines, most of the preprocessing pass, and
7717 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
7719 K. Richard Pixley maintained GAS for a while, adding various enhancements and
7720 many bug fixes, including merging support for several processors, breaking GAS
7721 up to handle multiple object file format back ends (including heavy rewrite,
7722 testing, an integration of the coff and b.out back ends), adding configuration
7723 including heavy testing and verification of cross assemblers and file splits
7724 and renaming, converted GAS to strictly ANSI C including full prototypes, added
7725 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
7726 port (including considerable amounts of reverse engineering), a SPARC opcode
7727 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
7728 assertions and made them work, much other reorganization, cleanup, and lint.
7730 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
7731 in format-specific I/O modules.
7733 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
7734 has done much work with it since.
7736 The Intel 80386 machine description was written by Eliot Dresselhaus.
7738 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
7740 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
7741 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
7743 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
7744 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
7745 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
7746 support a.out format.
7748 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
7749 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
7750 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
7751 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
7754 John Gilmore built the AMD 29000 support, added @code{.include} support, and
7755 simplified the configuration of which versions accept which directives. He
7756 updated the 68k machine description so that Motorola's opcodes always produced
7757 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
7758 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
7759 cross-compilation support, and one bug in relaxation that took a week and
7760 required the proverbial one-bit fix.
7762 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
7763 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
7764 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
7765 PowerPC assembler, and made a few other minor patches.
7767 Steve Chamberlain made GAS able to generate listings.
7769 Hewlett-Packard contributed support for the HP9000/300.
7771 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
7772 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
7773 formats). This work was supported by both the Center for Software Science at
7774 the University of Utah and Cygnus Support.
7776 Support for ELF format files has been worked on by Mark Eichin of Cygnus
7777 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
7778 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
7779 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
7780 and some initial 64-bit support).
7782 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
7784 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
7785 support for openVMS/Alpha.
7787 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
7790 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
7791 Inc.@: added support for Xtensa processors.
7793 Several engineers at Cygnus Support have also provided many small bug fixes and
7794 configuration enhancements.
7796 Jon Beniston added support for the Lattice Mico32 architecture.
7798 Many others have contributed large or small bugfixes and enhancements. If
7799 you have contributed significant work and are not mentioned on this list, and
7800 want to be, let us know. Some of the history has been lost; we are not
7801 intentionally leaving anyone out.
7803 @node GNU Free Documentation License
7804 @appendix GNU Free Documentation License
7808 @unnumbered AS Index
projects / deliverable / binutils-gdb.git / blob