1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000,
3 @c 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009
4 @c Free Software Foundation, Inc.
5 @c UPDATE!! On future updates--
6 @c (1) check for new machine-dep cmdline options in
7 @c md_parse_option definitions in config/tc-*.c
8 @c (2) for platform-specific directives, examine md_pseudo_op
10 @c (3) for object-format specific directives, examine obj_pseudo_op
12 @c (4) portable directives in potable[] in read.c
16 @macro gcctabopt{body}
19 @c defaults, config file may override:
24 @include asconfig.texi
29 @c common OR combinations of conditions
52 @set abnormal-separator
56 @settitle Using @value{AS}
59 @settitle Using @value{AS} (@value{TARGET})
61 @setchapternewpage odd
66 @c WARE! Some of the machine-dependent sections contain tables of machine
67 @c instructions. Except in multi-column format, these tables look silly.
68 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
69 @c the multi-col format is faked within @example sections.
71 @c Again unfortunately, the natural size that fits on a page, for these tables,
72 @c is different depending on whether or not smallbook is turned on.
73 @c This matters, because of order: text flow switches columns at each page
76 @c The format faked in this source works reasonably well for smallbook,
77 @c not well for the default large-page format. This manual expects that if you
78 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
79 @c tables in question. You can turn on one without the other at your
80 @c discretion, of course.
83 @c the insn tables look just as silly in info files regardless of smallbook,
84 @c might as well show 'em anyways.
90 * As: (as). The GNU assembler.
91 * Gas: (as). The GNU assembler.
100 This file documents the GNU Assembler "@value{AS}".
102 @c man begin COPYRIGHT
103 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002,
104 2006, 2007, 2008, 2009 Free Software Foundation, Inc.
106 Permission is granted to copy, distribute and/or modify this document
107 under the terms of the GNU Free Documentation License, Version 1.3
108 or any later version published by the Free Software Foundation;
109 with no Invariant Sections, with no Front-Cover Texts, and with no
110 Back-Cover Texts. A copy of the license is included in the
111 section entitled ``GNU Free Documentation License''.
117 @title Using @value{AS}
118 @subtitle The @sc{gnu} Assembler
120 @subtitle for the @value{TARGET} family
122 @ifset VERSION_PACKAGE
124 @subtitle @value{VERSION_PACKAGE}
127 @subtitle Version @value{VERSION}
130 The Free Software Foundation Inc.@: thanks The Nice Computer
131 Company of Australia for loaning Dean Elsner to write the
132 first (Vax) version of @command{as} for Project @sc{gnu}.
133 The proprietors, management and staff of TNCCA thank FSF for
134 distracting the boss while they got some work
137 @author Dean Elsner, Jay Fenlason & friends
141 \hfill {\it Using {\tt @value{AS}}}\par
142 \hfill Edited by Cygnus Support\par
144 %"boxit" macro for figures:
145 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
146 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
147 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
148 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
149 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
152 @vskip 0pt plus 1filll
153 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002,
154 2006, 2007, 2008, 2009 Free Software Foundation, Inc.
156 Permission is granted to copy, distribute and/or modify this document
157 under the terms of the GNU Free Documentation License, Version 1.3
158 or any later version published by the Free Software Foundation;
159 with no Invariant Sections, with no Front-Cover Texts, and with no
160 Back-Cover Texts. A copy of the license is included in the
161 section entitled ``GNU Free Documentation License''.
168 @top Using @value{AS}
170 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}}
171 @ifset VERSION_PACKAGE
172 @value{VERSION_PACKAGE}
174 version @value{VERSION}.
176 This version of the file describes @command{@value{AS}} configured to generate
177 code for @value{TARGET} architectures.
180 This document is distributed under the terms of the GNU Free
181 Documentation License. A copy of the license is included in the
182 section entitled ``GNU Free Documentation License''.
185 * Overview:: Overview
186 * Invoking:: Command-Line Options
188 * Sections:: Sections and Relocation
190 * Expressions:: Expressions
191 * Pseudo Ops:: Assembler Directives
193 * Object Attributes:: Object Attributes
195 * Machine Dependencies:: Machine Dependent Features
196 * Reporting Bugs:: Reporting Bugs
197 * Acknowledgements:: Who Did What
198 * GNU Free Documentation License:: GNU Free Documentation License
199 * AS Index:: AS Index
206 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
208 This version of the manual describes @command{@value{AS}} configured to generate
209 code for @value{TARGET} architectures.
213 @cindex invocation summary
214 @cindex option summary
215 @cindex summary of options
216 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
217 see @ref{Invoking,,Command-Line Options}.
219 @c man title AS the portable GNU assembler.
223 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
227 @c We don't use deffn and friends for the following because they seem
228 @c to be limited to one line for the header.
230 @c man begin SYNOPSIS
231 @value{AS} [@b{-a}[@b{cdghlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
232 [@b{--debug-prefix-map} @var{old}=@var{new}]
233 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
234 [@b{--gstabs+}] [@b{--gdwarf-2}] [@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{--target-help}] [@var{target-options}]
242 [@b{--}|@var{files} @dots{}]
244 @c Target dependent options are listed below. Keep the list sorted.
245 @c Add an empty line for separation.
248 @emph{Target Alpha options:}
250 [@b{-mdebug} | @b{-no-mdebug}]
251 [@b{-replace} | @b{-noreplace}]
252 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
253 [@b{-F}] [@b{-32addr}]
257 @emph{Target ARC options:}
263 @emph{Target ARM options:}
264 @c Don't document the deprecated options
265 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
266 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
267 [@b{-mfpu}=@var{floating-point-format}]
268 [@b{-mfloat-abi}=@var{abi}]
269 [@b{-meabi}=@var{ver}]
272 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
273 @b{-mapcs-reentrant}]
274 [@b{-mthumb-interwork}] [@b{-k}]
278 @emph{Target CRIS options:}
279 [@b{--underscore} | @b{--no-underscore}]
281 [@b{--emulation=criself} | @b{--emulation=crisaout}]
282 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
283 @c Deprecated -- deliberately not documented.
288 @emph{Target D10V options:}
293 @emph{Target D30V options:}
294 [@b{-O}|@b{-n}|@b{-N}]
298 @emph{Target H8/300 options:}
302 @c HPPA has no machine-dependent assembler options (yet).
306 @emph{Target i386 options:}
307 [@b{--32}|@b{--64}] [@b{-n}]
308 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
312 @emph{Target i960 options:}
313 @c see md_parse_option in tc-i960.c
314 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
316 [@b{-b}] [@b{-no-relax}]
320 @emph{Target IA-64 options:}
321 [@b{-mconstant-gp}|@b{-mauto-pic}]
322 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
324 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
325 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
326 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
327 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
331 @emph{Target IP2K options:}
332 [@b{-mip2022}|@b{-mip2022ext}]
336 @emph{Target M32C options:}
337 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
341 @emph{Target M32R options:}
342 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
347 @emph{Target M680X0 options:}
348 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
352 @emph{Target M68HC11 options:}
353 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}]
354 [@b{-mshort}|@b{-mlong}]
355 [@b{-mshort-double}|@b{-mlong-double}]
356 [@b{--force-long-branches}] [@b{--short-branches}]
357 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
358 [@b{--print-opcodes}] [@b{--generate-example}]
362 @emph{Target MCORE options:}
363 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
364 [@b{-mcpu=[210|340]}]
367 @emph{Target MICROBLAZE options:}
368 @c MicroBlaze has no machine-dependent assembler options.
372 @emph{Target MIPS options:}
373 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
374 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
375 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
376 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
377 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
378 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
379 [@b{-mips64}] [@b{-mips64r2}]
380 [@b{-construct-floats}] [@b{-no-construct-floats}]
381 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
382 [@b{-mips16}] [@b{-no-mips16}]
383 [@b{-msmartmips}] [@b{-mno-smartmips}]
384 [@b{-mips3d}] [@b{-no-mips3d}]
385 [@b{-mdmx}] [@b{-no-mdmx}]
386 [@b{-mdsp}] [@b{-mno-dsp}]
387 [@b{-mdspr2}] [@b{-mno-dspr2}]
388 [@b{-mmt}] [@b{-mno-mt}]
389 [@b{-mfix7000}] [@b{-mno-fix7000}]
390 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
391 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
392 [@b{-mdebug}] [@b{-no-mdebug}]
393 [@b{-mpdr}] [@b{-mno-pdr}]
397 @emph{Target MMIX options:}
398 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
399 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
400 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
401 [@b{--linker-allocated-gregs}]
405 @emph{Target PDP11 options:}
406 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
407 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
408 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
412 @emph{Target picoJava options:}
417 @emph{Target PowerPC options:}
418 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|
419 @b{-m403}|@b{-m405}|@b{-mppc64}|@b{-m620}|@b{-mppc64bridge}|@b{-mbooke}]
420 [@b{-mcom}|@b{-many}|@b{-maltivec}|@b{-mvsx}] [@b{-memb}]
421 [@b{-mregnames}|@b{-mno-regnames}]
422 [@b{-mrelocatable}|@b{-mrelocatable-lib}]
423 [@b{-mlittle}|@b{-mlittle-endian}|@b{-mbig}|@b{-mbig-endian}]
424 [@b{-msolaris}|@b{-mno-solaris}]
428 @emph{Target RX options:}
429 [@b{-mlittle-endian}|@b{-mbig-endian}]
430 [@b{-m32bit-ints}|@b{-m16bit-ints}]
431 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
435 @emph{Target s390 options:}
436 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
437 [@b{-mregnames}|@b{-mno-regnames}]
438 [@b{-mwarn-areg-zero}]
442 @emph{Target SCORE options:}
443 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
444 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
445 [@b{-march=score7}][@b{-march=score3}]
446 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
450 @emph{Target SPARC options:}
451 @c The order here is important. See c-sparc.texi.
452 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
453 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
454 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
459 @emph{Target TIC54X options:}
460 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
461 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
466 @emph{Target Z80 options:}
467 [@b{-z80}] [@b{-r800}]
468 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
469 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
470 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
471 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
472 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
473 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
477 @c Z8000 has no machine-dependent assembler options
481 @emph{Target Xtensa options:}
482 [@b{--[no-]text-section-literals}] [@b{--[no-]absolute-literals}]
483 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
484 [@b{--[no-]transform}]
485 [@b{--rename-section} @var{oldname}=@var{newname}]
493 @include at-file.texi
496 Turn on listings, in any of a variety of ways:
500 omit false conditionals
503 omit debugging directives
506 include general information, like @value{AS} version and options passed
509 include high-level source
515 include macro expansions
518 omit forms processing
524 set the name of the listing file
527 You may combine these options; for example, use @samp{-aln} for assembly
528 listing without forms processing. The @samp{=file} option, if used, must be
529 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
532 Begin in alternate macro mode.
534 @xref{Altmacro,,@code{.altmacro}}.
538 Ignored. This option is accepted for script compatibility with calls to
541 @item --debug-prefix-map @var{old}=@var{new}
542 When assembling files in directory @file{@var{old}}, record debugging
543 information describing them as in @file{@var{new}} instead.
545 @item --defsym @var{sym}=@var{value}
546 Define the symbol @var{sym} to be @var{value} before assembling the input file.
547 @var{value} must be an integer constant. As in C, a leading @samp{0x}
548 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
549 value. The value of the symbol can be overridden inside a source file via the
550 use of a @code{.set} pseudo-op.
553 ``fast''---skip whitespace and comment preprocessing (assume source is
558 Generate debugging information for each assembler source line using whichever
559 debug format is preferred by the target. This currently means either STABS,
563 Generate stabs debugging information for each assembler line. This
564 may help debugging assembler code, if the debugger can handle it.
567 Generate stabs debugging information for each assembler line, with GNU
568 extensions that probably only gdb can handle, and that could make other
569 debuggers crash or refuse to read your program. This
570 may help debugging assembler code. Currently the only GNU extension is
571 the location of the current working directory at assembling time.
574 Generate DWARF2 debugging information for each assembler line. This
575 may help debugging assembler code, if the debugger can handle it. Note---this
576 option is only supported by some targets, not all of them.
579 Print a summary of the command line options and exit.
582 Print a summary of all target specific options and exit.
585 Add directory @var{dir} to the search list for @code{.include} directives.
588 Don't warn about signed overflow.
591 @ifclear DIFF-TBL-KLUGE
592 This option is accepted but has no effect on the @value{TARGET} family.
594 @ifset DIFF-TBL-KLUGE
595 Issue warnings when difference tables altered for long displacements.
600 Keep (in the symbol table) local symbols. These symbols start with
601 system-specific local label prefixes, typically @samp{.L} for ELF systems
602 or @samp{L} for traditional a.out systems.
607 @item --listing-lhs-width=@var{number}
608 Set the maximum width, in words, of the output data column for an assembler
609 listing to @var{number}.
611 @item --listing-lhs-width2=@var{number}
612 Set the maximum width, in words, of the output data column for continuation
613 lines in an assembler listing to @var{number}.
615 @item --listing-rhs-width=@var{number}
616 Set the maximum width of an input source line, as displayed in a listing, to
619 @item --listing-cont-lines=@var{number}
620 Set the maximum number of lines printed in a listing for a single line of input
623 @item -o @var{objfile}
624 Name the object-file output from @command{@value{AS}} @var{objfile}.
627 Fold the data section into the text section.
629 @kindex --hash-size=@var{number}
630 Set the default size of GAS's hash tables to a prime number close to
631 @var{number}. Increasing this value can reduce the length of time it takes the
632 assembler to perform its tasks, at the expense of increasing the assembler's
633 memory requirements. Similarly reducing this value can reduce the memory
634 requirements at the expense of speed.
636 @item --reduce-memory-overheads
637 This option reduces GAS's memory requirements, at the expense of making the
638 assembly processes slower. Currently this switch is a synonym for
639 @samp{--hash-size=4051}, but in the future it may have other effects as well.
642 Print the maximum space (in bytes) and total time (in seconds) used by
645 @item --strip-local-absolute
646 Remove local absolute symbols from the outgoing symbol table.
650 Print the @command{as} version.
653 Print the @command{as} version and exit.
657 Suppress warning messages.
659 @item --fatal-warnings
660 Treat warnings as errors.
663 Don't suppress warning messages or treat them as errors.
672 Generate an object file even after errors.
674 @item -- | @var{files} @dots{}
675 Standard input, or source files to assemble.
680 The following options are available when @value{AS} is configured for
685 This option selects the core processor variant.
687 Select either big-endian (-EB) or little-endian (-EL) output.
692 The following options are available when @value{AS} is configured for the ARM
696 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
697 Specify which ARM processor variant is the target.
698 @item -march=@var{architecture}[+@var{extension}@dots{}]
699 Specify which ARM architecture variant is used by the target.
700 @item -mfpu=@var{floating-point-format}
701 Select which Floating Point architecture is the target.
702 @item -mfloat-abi=@var{abi}
703 Select which floating point ABI is in use.
705 Enable Thumb only instruction decoding.
706 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
707 Select which procedure calling convention is in use.
709 Select either big-endian (-EB) or little-endian (-EL) output.
710 @item -mthumb-interwork
711 Specify that the code has been generated with interworking between Thumb and
714 Specify that PIC code has been generated.
719 See the info pages for documentation of the CRIS-specific options.
723 The following options are available when @value{AS} is configured for
726 @cindex D10V optimization
727 @cindex optimization, D10V
729 Optimize output by parallelizing instructions.
734 The following options are available when @value{AS} is configured for a D30V
737 @cindex D30V optimization
738 @cindex optimization, D30V
740 Optimize output by parallelizing instructions.
744 Warn when nops are generated.
746 @cindex D30V nops after 32-bit multiply
748 Warn when a nop after a 32-bit multiply instruction is generated.
753 The following options are available when @value{AS} is configured for the
754 Intel 80960 processor.
757 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
758 Specify which variant of the 960 architecture is the target.
761 Add code to collect statistics about branches taken.
764 Do not alter compare-and-branch instructions for long displacements;
771 The following options are available when @value{AS} is configured for the
777 Specifies that the extended IP2022 instructions are allowed.
780 Restores the default behaviour, which restricts the permitted instructions to
781 just the basic IP2022 ones.
787 The following options are available when @value{AS} is configured for the
788 Renesas M32C and M16C processors.
793 Assemble M32C instructions.
796 Assemble M16C instructions (the default).
799 Enable support for link-time relaxations.
802 Support H'00 style hex constants in addition to 0x00 style.
808 The following options are available when @value{AS} is configured for the
809 Renesas M32R (formerly Mitsubishi M32R) series.
814 Specify which processor in the M32R family is the target. The default
815 is normally the M32R, but this option changes it to the M32RX.
817 @item --warn-explicit-parallel-conflicts or --Wp
818 Produce warning messages when questionable parallel constructs are
821 @item --no-warn-explicit-parallel-conflicts or --Wnp
822 Do not produce warning messages when questionable parallel constructs are
829 The following options are available when @value{AS} is configured for the
830 Motorola 68000 series.
835 Shorten references to undefined symbols, to one word instead of two.
837 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
838 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
839 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
840 Specify what processor in the 68000 family is the target. The default
841 is normally the 68020, but this can be changed at configuration time.
843 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
844 The target machine does (or does not) have a floating-point coprocessor.
845 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
846 the basic 68000 is not compatible with the 68881, a combination of the
847 two can be specified, since it's possible to do emulation of the
848 coprocessor instructions with the main processor.
850 @item -m68851 | -mno-68851
851 The target machine does (or does not) have a memory-management
852 unit coprocessor. The default is to assume an MMU for 68020 and up.
859 For details about the PDP-11 machine dependent features options,
860 see @ref{PDP-11-Options}.
863 @item -mpic | -mno-pic
864 Generate position-independent (or position-dependent) code. The
865 default is @option{-mpic}.
868 @itemx -mall-extensions
869 Enable all instruction set extensions. This is the default.
871 @item -mno-extensions
872 Disable all instruction set extensions.
874 @item -m@var{extension} | -mno-@var{extension}
875 Enable (or disable) a particular instruction set extension.
878 Enable the instruction set extensions supported by a particular CPU, and
879 disable all other extensions.
881 @item -m@var{machine}
882 Enable the instruction set extensions supported by a particular machine
883 model, and disable all other extensions.
889 The following options are available when @value{AS} is configured for
890 a picoJava processor.
894 @cindex PJ endianness
895 @cindex endianness, PJ
896 @cindex big endian output, PJ
898 Generate ``big endian'' format output.
900 @cindex little endian output, PJ
902 Generate ``little endian'' format output.
908 The following options are available when @value{AS} is configured for the
909 Motorola 68HC11 or 68HC12 series.
913 @item -m68hc11 | -m68hc12 | -m68hcs12
914 Specify what processor is the target. The default is
915 defined by the configuration option when building the assembler.
918 Specify to use the 16-bit integer ABI.
921 Specify to use the 32-bit integer ABI.
924 Specify to use the 32-bit double ABI.
927 Specify to use the 64-bit double ABI.
929 @item --force-long-branches
930 Relative branches are turned into absolute ones. This concerns
931 conditional branches, unconditional branches and branches to a
934 @item -S | --short-branches
935 Do not turn relative branches into absolute ones
936 when the offset is out of range.
938 @item --strict-direct-mode
939 Do not turn the direct addressing mode into extended addressing mode
940 when the instruction does not support direct addressing mode.
942 @item --print-insn-syntax
943 Print the syntax of instruction in case of error.
945 @item --print-opcodes
946 print the list of instructions with syntax and then exit.
948 @item --generate-example
949 print an example of instruction for each possible instruction and then exit.
950 This option is only useful for testing @command{@value{AS}}.
956 The following options are available when @command{@value{AS}} is configured
957 for the SPARC architecture:
960 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
961 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
962 Explicitly select a variant of the SPARC architecture.
964 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
965 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
967 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
968 UltraSPARC extensions.
970 @item -xarch=v8plus | -xarch=v8plusa
971 For compatibility with the Solaris v9 assembler. These options are
972 equivalent to -Av8plus and -Av8plusa, respectively.
975 Warn when the assembler switches to another architecture.
980 The following options are available when @value{AS} is configured for the 'c54x
985 Enable extended addressing mode. All addresses and relocations will assume
986 extended addressing (usually 23 bits).
987 @item -mcpu=@var{CPU_VERSION}
988 Sets the CPU version being compiled for.
989 @item -merrors-to-file @var{FILENAME}
990 Redirect error output to a file, for broken systems which don't support such
991 behaviour in the shell.
996 The following options are available when @value{AS} is configured for
997 a @sc{mips} processor.
1001 This option sets the largest size of an object that can be referenced
1002 implicitly with the @code{gp} register. It is only accepted for targets that
1003 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1005 @cindex MIPS endianness
1006 @cindex endianness, MIPS
1007 @cindex big endian output, MIPS
1009 Generate ``big endian'' format output.
1011 @cindex little endian output, MIPS
1013 Generate ``little endian'' format output.
1025 Generate code for a particular @sc{mips} Instruction Set Architecture level.
1026 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1027 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1028 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1029 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
1031 correspond to generic
1032 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
1033 and @samp{MIPS64 Release 2}
1034 ISA processors, respectively.
1036 @item -march=@var{CPU}
1037 Generate code for a particular @sc{mips} cpu.
1039 @item -mtune=@var{cpu}
1040 Schedule and tune for a particular @sc{mips} cpu.
1044 Cause nops to be inserted if the read of the destination register
1045 of an mfhi or mflo instruction occurs in the following two instructions.
1049 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1050 section instead of the standard ELF .stabs sections.
1054 Control generation of @code{.pdr} sections.
1058 The register sizes are normally inferred from the ISA and ABI, but these
1059 flags force a certain group of registers to be treated as 32 bits wide at
1060 all times. @samp{-mgp32} controls the size of general-purpose registers
1061 and @samp{-mfp32} controls the size of floating-point registers.
1065 Generate code for the MIPS 16 processor. This is equivalent to putting
1066 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1067 turns off this option.
1070 @itemx -mno-smartmips
1071 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1072 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1073 @samp{-mno-smartmips} turns off this option.
1077 Generate code for the MIPS-3D Application Specific Extension.
1078 This tells the assembler to accept MIPS-3D instructions.
1079 @samp{-no-mips3d} turns off this option.
1083 Generate code for the MDMX Application Specific Extension.
1084 This tells the assembler to accept MDMX instructions.
1085 @samp{-no-mdmx} turns off this option.
1089 Generate code for the DSP Release 1 Application Specific Extension.
1090 This tells the assembler to accept DSP Release 1 instructions.
1091 @samp{-mno-dsp} turns off this option.
1095 Generate code for the DSP Release 2 Application Specific Extension.
1096 This option implies -mdsp.
1097 This tells the assembler to accept DSP Release 2 instructions.
1098 @samp{-mno-dspr2} turns off this option.
1102 Generate code for the MT Application Specific Extension.
1103 This tells the assembler to accept MT instructions.
1104 @samp{-mno-mt} turns off this option.
1106 @item --construct-floats
1107 @itemx --no-construct-floats
1108 The @samp{--no-construct-floats} option disables the construction of
1109 double width floating point constants by loading the two halves of the
1110 value into the two single width floating point registers that make up
1111 the double width register. By default @samp{--construct-floats} is
1112 selected, allowing construction of these floating point constants.
1115 @item --emulation=@var{name}
1116 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
1117 for some other target, in all respects, including output format (choosing
1118 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
1119 debugging information or store symbol table information, and default
1120 endianness. The available configuration names are: @samp{mipsecoff},
1121 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
1122 @samp{mipsbelf}. The first two do not alter the default endianness from that
1123 of the primary target for which the assembler was configured; the others change
1124 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
1125 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
1126 selection in any case.
1128 This option is currently supported only when the primary target
1129 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
1130 Furthermore, the primary target or others specified with
1131 @samp{--enable-targets=@dots{}} at configuration time must include support for
1132 the other format, if both are to be available. For example, the Irix 5
1133 configuration includes support for both.
1135 Eventually, this option will support more configurations, with more
1136 fine-grained control over the assembler's behavior, and will be supported for
1140 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1147 Control how to deal with multiplication overflow and division by zero.
1148 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1149 (and only work for Instruction Set Architecture level 2 and higher);
1150 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1154 When this option is used, @command{@value{AS}} will issue a warning every
1155 time it generates a nop instruction from a macro.
1160 The following options are available when @value{AS} is configured for
1166 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1167 The command line option @samp{-nojsri2bsr} can be used to disable it.
1171 Enable or disable the silicon filter behaviour. By default this is disabled.
1172 The default can be overridden by the @samp{-sifilter} command line option.
1175 Alter jump instructions for long displacements.
1177 @item -mcpu=[210|340]
1178 Select the cpu type on the target hardware. This controls which instructions
1182 Assemble for a big endian target.
1185 Assemble for a little endian target.
1191 See the info pages for documentation of the MMIX-specific options.
1195 See the info pages for documentation of the RX-specific options.
1199 The following options are available when @value{AS} is configured for the s390
1205 Select the word size, either 31/32 bits or 64 bits.
1208 Select the architecture mode, either the Enterprise System
1209 Architecture (esa) or the z/Architecture mode (zarch).
1210 @item -march=@var{processor}
1211 Specify which s390 processor variant is the target, @samp{g6}, @samp{g6},
1212 @samp{z900}, @samp{z990}, @samp{z9-109}, @samp{z9-ec}, or @samp{z10}.
1214 @itemx -mno-regnames
1215 Allow or disallow symbolic names for registers.
1216 @item -mwarn-areg-zero
1217 Warn whenever the operand for a base or index register has been specified
1218 but evaluates to zero.
1223 The following options are available when @value{AS} is configured for
1224 an Xtensa processor.
1227 @item --text-section-literals | --no-text-section-literals
1228 With @option{--text-@-section-@-literals}, literal pools are interspersed
1229 in the text section. The default is
1230 @option{--no-@-text-@-section-@-literals}, which places literals in a
1231 separate section in the output file. These options only affect literals
1232 referenced via PC-relative @code{L32R} instructions; literals for
1233 absolute mode @code{L32R} instructions are handled separately.
1235 @item --absolute-literals | --no-absolute-literals
1236 Indicate to the assembler whether @code{L32R} instructions use absolute
1237 or PC-relative addressing. The default is to assume absolute addressing
1238 if the Xtensa processor includes the absolute @code{L32R} addressing
1239 option. Otherwise, only the PC-relative @code{L32R} mode can be used.
1241 @item --target-align | --no-target-align
1242 Enable or disable automatic alignment to reduce branch penalties at the
1243 expense of some code density. The default is @option{--target-@-align}.
1245 @item --longcalls | --no-longcalls
1246 Enable or disable transformation of call instructions to allow calls
1247 across a greater range of addresses. The default is
1248 @option{--no-@-longcalls}.
1250 @item --transform | --no-transform
1251 Enable or disable all assembler transformations of Xtensa instructions.
1252 The default is @option{--transform};
1253 @option{--no-transform} should be used only in the rare cases when the
1254 instructions must be exactly as specified in the assembly source.
1256 @item --rename-section @var{oldname}=@var{newname}
1257 When generating output sections, rename the @var{oldname} section to
1263 The following options are available when @value{AS} is configured for
1264 a Z80 family processor.
1267 Assemble for Z80 processor.
1269 Assemble for R800 processor.
1270 @item -ignore-undocumented-instructions
1272 Assemble undocumented Z80 instructions that also work on R800 without warning.
1273 @item -ignore-unportable-instructions
1275 Assemble all undocumented Z80 instructions without warning.
1276 @item -warn-undocumented-instructions
1278 Issue a warning for undocumented Z80 instructions that also work on R800.
1279 @item -warn-unportable-instructions
1281 Issue a warning for undocumented Z80 instructions that do not work on R800.
1282 @item -forbid-undocumented-instructions
1284 Treat all undocumented instructions as errors.
1285 @item -forbid-unportable-instructions
1287 Treat undocumented Z80 instructions that do not work on R800 as errors.
1294 * Manual:: Structure of this Manual
1295 * GNU Assembler:: The GNU Assembler
1296 * Object Formats:: Object File Formats
1297 * Command Line:: Command Line
1298 * Input Files:: Input Files
1299 * Object:: Output (Object) File
1300 * Errors:: Error and Warning Messages
1304 @section Structure of this Manual
1306 @cindex manual, structure and purpose
1307 This manual is intended to describe what you need to know to use
1308 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1309 notation for symbols, constants, and expressions; the directives that
1310 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1313 We also cover special features in the @value{TARGET}
1314 configuration of @command{@value{AS}}, including assembler directives.
1317 This manual also describes some of the machine-dependent features of
1318 various flavors of the assembler.
1321 @cindex machine instructions (not covered)
1322 On the other hand, this manual is @emph{not} intended as an introduction
1323 to programming in assembly language---let alone programming in general!
1324 In a similar vein, we make no attempt to introduce the machine
1325 architecture; we do @emph{not} describe the instruction set, standard
1326 mnemonics, registers or addressing modes that are standard to a
1327 particular architecture.
1329 You may want to consult the manufacturer's
1330 machine architecture manual for this information.
1334 For information on the H8/300 machine instruction set, see @cite{H8/300
1335 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1336 Programming Manual} (Renesas).
1339 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1340 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1341 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1342 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1345 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1349 @c I think this is premature---doc@cygnus.com, 17jan1991
1351 Throughout this manual, we assume that you are running @dfn{GNU},
1352 the portable operating system from the @dfn{Free Software
1353 Foundation, Inc.}. This restricts our attention to certain kinds of
1354 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1355 once this assumption is granted examples and definitions need less
1358 @command{@value{AS}} is part of a team of programs that turn a high-level
1359 human-readable series of instructions into a low-level
1360 computer-readable series of instructions. Different versions of
1361 @command{@value{AS}} are used for different kinds of computer.
1364 @c There used to be a section "Terminology" here, which defined
1365 @c "contents", "byte", "word", and "long". Defining "word" to any
1366 @c particular size is confusing when the .word directive may generate 16
1367 @c bits on one machine and 32 bits on another; in general, for the user
1368 @c version of this manual, none of these terms seem essential to define.
1369 @c They were used very little even in the former draft of the manual;
1370 @c this draft makes an effort to avoid them (except in names of
1374 @section The GNU Assembler
1376 @c man begin DESCRIPTION
1378 @sc{gnu} @command{as} is really a family of assemblers.
1380 This manual describes @command{@value{AS}}, a member of that family which is
1381 configured for the @value{TARGET} architectures.
1383 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1384 should find a fairly similar environment when you use it on another
1385 architecture. Each version has much in common with the others,
1386 including object file formats, most assembler directives (often called
1387 @dfn{pseudo-ops}) and assembler syntax.@refill
1389 @cindex purpose of @sc{gnu} assembler
1390 @command{@value{AS}} is primarily intended to assemble the output of the
1391 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1392 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1393 assemble correctly everything that other assemblers for the same
1394 machine would assemble.
1396 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1399 @c This remark should appear in generic version of manual; assumption
1400 @c here is that generic version sets M680x0.
1401 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1402 assembler for the same architecture; for example, we know of several
1403 incompatible versions of 680x0 assembly language syntax.
1408 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1409 program in one pass of the source file. This has a subtle impact on the
1410 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1412 @node Object Formats
1413 @section Object File Formats
1415 @cindex object file format
1416 The @sc{gnu} assembler can be configured to produce several alternative
1417 object file formats. For the most part, this does not affect how you
1418 write assembly language programs; but directives for debugging symbols
1419 are typically different in different file formats. @xref{Symbol
1420 Attributes,,Symbol Attributes}.
1423 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1424 @value{OBJ-NAME} format object files.
1426 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1428 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1429 @code{b.out} or COFF format object files.
1432 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1433 SOM or ELF format object files.
1438 @section Command Line
1440 @cindex command line conventions
1442 After the program name @command{@value{AS}}, the command line may contain
1443 options and file names. Options may appear in any order, and may be
1444 before, after, or between file names. The order of file names is
1447 @cindex standard input, as input file
1449 @file{--} (two hyphens) by itself names the standard input file
1450 explicitly, as one of the files for @command{@value{AS}} to assemble.
1452 @cindex options, command line
1453 Except for @samp{--} any command line argument that begins with a
1454 hyphen (@samp{-}) is an option. Each option changes the behavior of
1455 @command{@value{AS}}. No option changes the way another option works. An
1456 option is a @samp{-} followed by one or more letters; the case of
1457 the letter is important. All options are optional.
1459 Some options expect exactly one file name to follow them. The file
1460 name may either immediately follow the option's letter (compatible
1461 with older assemblers) or it may be the next command argument (@sc{gnu}
1462 standard). These two command lines are equivalent:
1465 @value{AS} -o my-object-file.o mumble.s
1466 @value{AS} -omy-object-file.o mumble.s
1470 @section Input Files
1473 @cindex source program
1474 @cindex files, input
1475 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1476 describe the program input to one run of @command{@value{AS}}. The program may
1477 be in one or more files; how the source is partitioned into files
1478 doesn't change the meaning of the source.
1480 @c I added "con" prefix to "catenation" just to prove I can overcome my
1481 @c APL training... doc@cygnus.com
1482 The source program is a concatenation of the text in all the files, in the
1485 @c man begin DESCRIPTION
1486 Each time you run @command{@value{AS}} it assembles exactly one source
1487 program. The source program is made up of one or more files.
1488 (The standard input is also a file.)
1490 You give @command{@value{AS}} a command line that has zero or more input file
1491 names. The input files are read (from left file name to right). A
1492 command line argument (in any position) that has no special meaning
1493 is taken to be an input file name.
1495 If you give @command{@value{AS}} no file names it attempts to read one input file
1496 from the @command{@value{AS}} standard input, which is normally your terminal. You
1497 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1500 Use @samp{--} if you need to explicitly name the standard input file
1501 in your command line.
1503 If the source is empty, @command{@value{AS}} produces a small, empty object
1508 @subheading Filenames and Line-numbers
1510 @cindex input file linenumbers
1511 @cindex line numbers, in input files
1512 There are two ways of locating a line in the input file (or files) and
1513 either may be used in reporting error messages. One way refers to a line
1514 number in a physical file; the other refers to a line number in a
1515 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1517 @dfn{Physical files} are those files named in the command line given
1518 to @command{@value{AS}}.
1520 @dfn{Logical files} are simply names declared explicitly by assembler
1521 directives; they bear no relation to physical files. Logical file names help
1522 error messages reflect the original source file, when @command{@value{AS}} source
1523 is itself synthesized from other files. @command{@value{AS}} understands the
1524 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1525 @ref{File,,@code{.file}}.
1528 @section Output (Object) File
1534 Every time you run @command{@value{AS}} it produces an output file, which is
1535 your assembly language program translated into numbers. This file
1536 is the object file. Its default name is
1544 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1546 You can give it another name by using the @option{-o} option. Conventionally,
1547 object file names end with @file{.o}. The default name is used for historical
1548 reasons: older assemblers were capable of assembling self-contained programs
1549 directly into a runnable program. (For some formats, this isn't currently
1550 possible, but it can be done for the @code{a.out} format.)
1554 The object file is meant for input to the linker @code{@value{LD}}. It contains
1555 assembled program code, information to help @code{@value{LD}} integrate
1556 the assembled program into a runnable file, and (optionally) symbolic
1557 information for the debugger.
1559 @c link above to some info file(s) like the description of a.out.
1560 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1563 @section Error and Warning Messages
1565 @c man begin DESCRIPTION
1567 @cindex error messages
1568 @cindex warning messages
1569 @cindex messages from assembler
1570 @command{@value{AS}} may write warnings and error messages to the standard error
1571 file (usually your terminal). This should not happen when a compiler
1572 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1573 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1574 grave problem that stops the assembly.
1578 @cindex format of warning messages
1579 Warning messages have the format
1582 file_name:@b{NNN}:Warning Message Text
1586 @cindex line numbers, in warnings/errors
1587 (where @b{NNN} is a line number). If a logical file name has been given
1588 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1589 the current input file is used. If a logical line number was given
1591 (@pxref{Line,,@code{.line}})
1593 then it is used to calculate the number printed,
1594 otherwise the actual line in the current source file is printed. The
1595 message text is intended to be self explanatory (in the grand Unix
1598 @cindex format of error messages
1599 Error messages have the format
1601 file_name:@b{NNN}:FATAL:Error Message Text
1603 The file name and line number are derived as for warning
1604 messages. The actual message text may be rather less explanatory
1605 because many of them aren't supposed to happen.
1608 @chapter Command-Line Options
1610 @cindex options, all versions of assembler
1611 This chapter describes command-line options available in @emph{all}
1612 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
1613 for options specific
1615 to the @value{TARGET} target.
1618 to particular machine architectures.
1621 @c man begin DESCRIPTION
1623 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1624 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1625 The assembler arguments must be separated from each other (and the @samp{-Wa})
1626 by commas. For example:
1629 gcc -c -g -O -Wa,-alh,-L file.c
1633 This passes two options to the assembler: @samp{-alh} (emit a listing to
1634 standard output with high-level and assembly source) and @samp{-L} (retain
1635 local symbols in the symbol table).
1637 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1638 command-line options are automatically passed to the assembler by the compiler.
1639 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1640 precisely what options it passes to each compilation pass, including the
1646 * a:: -a[cdghlns] enable listings
1647 * alternate:: --alternate enable alternate macro syntax
1648 * D:: -D for compatibility
1649 * f:: -f to work faster
1650 * I:: -I for .include search path
1651 @ifclear DIFF-TBL-KLUGE
1652 * K:: -K for compatibility
1654 @ifset DIFF-TBL-KLUGE
1655 * K:: -K for difference tables
1658 * L:: -L to retain local symbols
1659 * listing:: --listing-XXX to configure listing output
1660 * M:: -M or --mri to assemble in MRI compatibility mode
1661 * MD:: --MD for dependency tracking
1662 * o:: -o to name the object file
1663 * R:: -R to join data and text sections
1664 * statistics:: --statistics to see statistics about assembly
1665 * traditional-format:: --traditional-format for compatible output
1666 * v:: -v to announce version
1667 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1668 * Z:: -Z to make object file even after errors
1672 @section Enable Listings: @option{-a[cdghlns]}
1682 @cindex listings, enabling
1683 @cindex assembly listings, enabling
1685 These options enable listing output from the assembler. By itself,
1686 @samp{-a} requests high-level, assembly, and symbols listing.
1687 You can use other letters to select specific options for the list:
1688 @samp{-ah} requests a high-level language listing,
1689 @samp{-al} requests an output-program assembly listing, and
1690 @samp{-as} requests a symbol table listing.
1691 High-level listings require that a compiler debugging option like
1692 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1695 Use the @samp{-ag} option to print a first section with general assembly
1696 information, like @value{AS} version, switches passed, or time stamp.
1698 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1699 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1700 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1701 omitted from the listing.
1703 Use the @samp{-ad} option to omit debugging directives from the
1706 Once you have specified one of these options, you can further control
1707 listing output and its appearance using the directives @code{.list},
1708 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1710 The @samp{-an} option turns off all forms processing.
1711 If you do not request listing output with one of the @samp{-a} options, the
1712 listing-control directives have no effect.
1714 The letters after @samp{-a} may be combined into one option,
1715 @emph{e.g.}, @samp{-aln}.
1717 Note if the assembler source is coming from the standard input (e.g.,
1719 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1720 is being used) then the listing will not contain any comments or preprocessor
1721 directives. This is because the listing code buffers input source lines from
1722 stdin only after they have been preprocessed by the assembler. This reduces
1723 memory usage and makes the code more efficient.
1726 @section @option{--alternate}
1729 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
1732 @section @option{-D}
1735 This option has no effect whatsoever, but it is accepted to make it more
1736 likely that scripts written for other assemblers also work with
1737 @command{@value{AS}}.
1740 @section Work Faster: @option{-f}
1743 @cindex trusted compiler
1744 @cindex faster processing (@option{-f})
1745 @samp{-f} should only be used when assembling programs written by a
1746 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1747 and comment preprocessing on
1748 the input file(s) before assembling them. @xref{Preprocessing,
1752 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1753 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1758 @section @code{.include} Search Path: @option{-I} @var{path}
1760 @kindex -I @var{path}
1761 @cindex paths for @code{.include}
1762 @cindex search path for @code{.include}
1763 @cindex @code{include} directive search path
1764 Use this option to add a @var{path} to the list of directories
1765 @command{@value{AS}} searches for files specified in @code{.include}
1766 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1767 many times as necessary to include a variety of paths. The current
1768 working directory is always searched first; after that, @command{@value{AS}}
1769 searches any @samp{-I} directories in the same order as they were
1770 specified (left to right) on the command line.
1773 @section Difference Tables: @option{-K}
1776 @ifclear DIFF-TBL-KLUGE
1777 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1778 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1779 where it can be used to warn when the assembler alters the machine code
1780 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1781 family does not have the addressing limitations that sometimes lead to this
1782 alteration on other platforms.
1785 @ifset DIFF-TBL-KLUGE
1786 @cindex difference tables, warning
1787 @cindex warning for altered difference tables
1788 @command{@value{AS}} sometimes alters the code emitted for directives of the
1789 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
1790 You can use the @samp{-K} option if you want a warning issued when this
1795 @section Include Local Symbols: @option{-L}
1798 @cindex local symbols, retaining in output
1799 Symbols beginning with system-specific local label prefixes, typically
1800 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
1801 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
1802 such symbols when debugging, because they are intended for the use of
1803 programs (like compilers) that compose assembler programs, not for your
1804 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
1805 such symbols, so you do not normally debug with them.
1807 This option tells @command{@value{AS}} to retain those local symbols
1808 in the object file. Usually if you do this you also tell the linker
1809 @code{@value{LD}} to preserve those symbols.
1812 @section Configuring listing output: @option{--listing}
1814 The listing feature of the assembler can be enabled via the command line switch
1815 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1816 hex dump of the corresponding locations in the output object file, and displays
1817 them as a listing file. The format of this listing can be controlled by
1818 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
1819 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
1820 @code{.psize} (@pxref{Psize}), and
1821 @code{.eject} (@pxref{Eject}) and also by the following switches:
1824 @item --listing-lhs-width=@samp{number}
1825 @kindex --listing-lhs-width
1826 @cindex Width of first line disassembly output
1827 Sets the maximum width, in words, of the first line of the hex byte dump. This
1828 dump appears on the left hand side of the listing output.
1830 @item --listing-lhs-width2=@samp{number}
1831 @kindex --listing-lhs-width2
1832 @cindex Width of continuation lines of disassembly output
1833 Sets the maximum width, in words, of any further lines of the hex byte dump for
1834 a given input source line. If this value is not specified, it defaults to being
1835 the same as the value specified for @samp{--listing-lhs-width}. If neither
1836 switch is used the default is to one.
1838 @item --listing-rhs-width=@samp{number}
1839 @kindex --listing-rhs-width
1840 @cindex Width of source line output
1841 Sets the maximum width, in characters, of the source line that is displayed
1842 alongside the hex dump. The default value for this parameter is 100. The
1843 source line is displayed on the right hand side of the listing output.
1845 @item --listing-cont-lines=@samp{number}
1846 @kindex --listing-cont-lines
1847 @cindex Maximum number of continuation lines
1848 Sets the maximum number of continuation lines of hex dump that will be
1849 displayed for a given single line of source input. The default value is 4.
1853 @section Assemble in MRI Compatibility Mode: @option{-M}
1856 @cindex MRI compatibility mode
1857 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
1858 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
1859 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1860 configured target) assembler from Microtec Research. The exact nature of the
1861 MRI syntax will not be documented here; see the MRI manuals for more
1862 information. Note in particular that the handling of macros and macro
1863 arguments is somewhat different. The purpose of this option is to permit
1864 assembling existing MRI assembler code using @command{@value{AS}}.
1866 The MRI compatibility is not complete. Certain operations of the MRI assembler
1867 depend upon its object file format, and can not be supported using other object
1868 file formats. Supporting these would require enhancing each object file format
1869 individually. These are:
1872 @item global symbols in common section
1874 The m68k MRI assembler supports common sections which are merged by the linker.
1875 Other object file formats do not support this. @command{@value{AS}} handles
1876 common sections by treating them as a single common symbol. It permits local
1877 symbols to be defined within a common section, but it can not support global
1878 symbols, since it has no way to describe them.
1880 @item complex relocations
1882 The MRI assemblers support relocations against a negated section address, and
1883 relocations which combine the start addresses of two or more sections. These
1884 are not support by other object file formats.
1886 @item @code{END} pseudo-op specifying start address
1888 The MRI @code{END} pseudo-op permits the specification of a start address.
1889 This is not supported by other object file formats. The start address may
1890 instead be specified using the @option{-e} option to the linker, or in a linker
1893 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1895 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1896 name to the output file. This is not supported by other object file formats.
1898 @item @code{ORG} pseudo-op
1900 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1901 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
1902 which changes the location within the current section. Absolute sections are
1903 not supported by other object file formats. The address of a section may be
1904 assigned within a linker script.
1907 There are some other features of the MRI assembler which are not supported by
1908 @command{@value{AS}}, typically either because they are difficult or because they
1909 seem of little consequence. Some of these may be supported in future releases.
1913 @item EBCDIC strings
1915 EBCDIC strings are not supported.
1917 @item packed binary coded decimal
1919 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1920 and @code{DCB.P} pseudo-ops are not supported.
1922 @item @code{FEQU} pseudo-op
1924 The m68k @code{FEQU} pseudo-op is not supported.
1926 @item @code{NOOBJ} pseudo-op
1928 The m68k @code{NOOBJ} pseudo-op is not supported.
1930 @item @code{OPT} branch control options
1932 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1933 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
1934 relaxes all branches, whether forward or backward, to an appropriate size, so
1935 these options serve no purpose.
1937 @item @code{OPT} list control options
1939 The following m68k @code{OPT} list control options are ignored: @code{C},
1940 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1941 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1943 @item other @code{OPT} options
1945 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1946 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1948 @item @code{OPT} @code{D} option is default
1950 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1951 @code{OPT NOD} may be used to turn it off.
1953 @item @code{XREF} pseudo-op.
1955 The m68k @code{XREF} pseudo-op is ignored.
1957 @item @code{.debug} pseudo-op
1959 The i960 @code{.debug} pseudo-op is not supported.
1961 @item @code{.extended} pseudo-op
1963 The i960 @code{.extended} pseudo-op is not supported.
1965 @item @code{.list} pseudo-op.
1967 The various options of the i960 @code{.list} pseudo-op are not supported.
1969 @item @code{.optimize} pseudo-op
1971 The i960 @code{.optimize} pseudo-op is not supported.
1973 @item @code{.output} pseudo-op
1975 The i960 @code{.output} pseudo-op is not supported.
1977 @item @code{.setreal} pseudo-op
1979 The i960 @code{.setreal} pseudo-op is not supported.
1984 @section Dependency Tracking: @option{--MD}
1987 @cindex dependency tracking
1990 @command{@value{AS}} can generate a dependency file for the file it creates. This
1991 file consists of a single rule suitable for @code{make} describing the
1992 dependencies of the main source file.
1994 The rule is written to the file named in its argument.
1996 This feature is used in the automatic updating of makefiles.
1999 @section Name the Object File: @option{-o}
2002 @cindex naming object file
2003 @cindex object file name
2004 There is always one object file output when you run @command{@value{AS}}. By
2005 default it has the name
2008 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2022 You use this option (which takes exactly one filename) to give the
2023 object file a different name.
2025 Whatever the object file is called, @command{@value{AS}} overwrites any
2026 existing file of the same name.
2029 @section Join Data and Text Sections: @option{-R}
2032 @cindex data and text sections, joining
2033 @cindex text and data sections, joining
2034 @cindex joining text and data sections
2035 @cindex merging text and data sections
2036 @option{-R} tells @command{@value{AS}} to write the object file as if all
2037 data-section data lives in the text section. This is only done at
2038 the very last moment: your binary data are the same, but data
2039 section parts are relocated differently. The data section part of
2040 your object file is zero bytes long because all its bytes are
2041 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2043 When you specify @option{-R} it would be possible to generate shorter
2044 address displacements (because we do not have to cross between text and
2045 data section). We refrain from doing this simply for compatibility with
2046 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2049 When @command{@value{AS}} is configured for COFF or ELF output,
2050 this option is only useful if you use sections named @samp{.text} and
2055 @option{-R} is not supported for any of the HPPA targets. Using
2056 @option{-R} generates a warning from @command{@value{AS}}.
2060 @section Display Assembly Statistics: @option{--statistics}
2062 @kindex --statistics
2063 @cindex statistics, about assembly
2064 @cindex time, total for assembly
2065 @cindex space used, maximum for assembly
2066 Use @samp{--statistics} to display two statistics about the resources used by
2067 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2068 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2071 @node traditional-format
2072 @section Compatible Output: @option{--traditional-format}
2074 @kindex --traditional-format
2075 For some targets, the output of @command{@value{AS}} is different in some ways
2076 from the output of some existing assembler. This switch requests
2077 @command{@value{AS}} to use the traditional format instead.
2079 For example, it disables the exception frame optimizations which
2080 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2083 @section Announce Version: @option{-v}
2087 @cindex assembler version
2088 @cindex version of assembler
2089 You can find out what version of as is running by including the
2090 option @samp{-v} (which you can also spell as @samp{-version}) on the
2094 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2096 @command{@value{AS}} should never give a warning or error message when
2097 assembling compiler output. But programs written by people often
2098 cause @command{@value{AS}} to give a warning that a particular assumption was
2099 made. All such warnings are directed to the standard error file.
2103 @cindex suppressing warnings
2104 @cindex warnings, suppressing
2105 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2106 This only affects the warning messages: it does not change any particular of
2107 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2110 @kindex --fatal-warnings
2111 @cindex errors, caused by warnings
2112 @cindex warnings, causing error
2113 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2114 files that generate warnings to be in error.
2117 @cindex warnings, switching on
2118 You can switch these options off again by specifying @option{--warn}, which
2119 causes warnings to be output as usual.
2122 @section Generate Object File in Spite of Errors: @option{-Z}
2123 @cindex object file, after errors
2124 @cindex errors, continuing after
2125 After an error message, @command{@value{AS}} normally produces no output. If for
2126 some reason you are interested in object file output even after
2127 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2128 option. If there are any errors, @command{@value{AS}} continues anyways, and
2129 writes an object file after a final warning message of the form @samp{@var{n}
2130 errors, @var{m} warnings, generating bad object file.}
2135 @cindex machine-independent syntax
2136 @cindex syntax, machine-independent
2137 This chapter describes the machine-independent syntax allowed in a
2138 source file. @command{@value{AS}} syntax is similar to what many other
2139 assemblers use; it is inspired by the BSD 4.2
2144 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2148 * Preprocessing:: Preprocessing
2149 * Whitespace:: Whitespace
2150 * Comments:: Comments
2151 * Symbol Intro:: Symbols
2152 * Statements:: Statements
2153 * Constants:: Constants
2157 @section Preprocessing
2159 @cindex preprocessing
2160 The @command{@value{AS}} internal preprocessor:
2162 @cindex whitespace, removed by preprocessor
2164 adjusts and removes extra whitespace. It leaves one space or tab before
2165 the keywords on a line, and turns any other whitespace on the line into
2168 @cindex comments, removed by preprocessor
2170 removes all comments, replacing them with a single space, or an
2171 appropriate number of newlines.
2173 @cindex constants, converted by preprocessor
2175 converts character constants into the appropriate numeric values.
2178 It does not do macro processing, include file handling, or
2179 anything else you may get from your C compiler's preprocessor. You can
2180 do include file processing with the @code{.include} directive
2181 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2182 to get other ``CPP'' style preprocessing by giving the input file a
2183 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2184 Output, gcc.info, Using GNU CC}.
2186 Excess whitespace, comments, and character constants
2187 cannot be used in the portions of the input text that are not
2190 @cindex turning preprocessing on and off
2191 @cindex preprocessing, turning on and off
2194 If the first line of an input file is @code{#NO_APP} or if you use the
2195 @samp{-f} option, whitespace and comments are not removed from the input file.
2196 Within an input file, you can ask for whitespace and comment removal in
2197 specific portions of the by putting a line that says @code{#APP} before the
2198 text that may contain whitespace or comments, and putting a line that says
2199 @code{#NO_APP} after this text. This feature is mainly intend to support
2200 @code{asm} statements in compilers whose output is otherwise free of comments
2207 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2208 Whitespace is used to separate symbols, and to make programs neater for
2209 people to read. Unless within character constants
2210 (@pxref{Characters,,Character Constants}), any whitespace means the same
2211 as exactly one space.
2217 There are two ways of rendering comments to @command{@value{AS}}. In both
2218 cases the comment is equivalent to one space.
2220 Anything from @samp{/*} through the next @samp{*/} is a comment.
2221 This means you may not nest these comments.
2225 The only way to include a newline ('\n') in a comment
2226 is to use this sort of comment.
2229 /* This sort of comment does not nest. */
2232 @cindex line comment character
2233 Anything from the @dfn{line comment} character to the next newline
2234 is considered a comment and is ignored. The line comment character is
2236 @samp{;} on the ARC;
2239 @samp{@@} on the ARM;
2242 @samp{;} for the H8/300 family;
2245 @samp{;} for the HPPA;
2248 @samp{#} on the i386 and x86-64;
2251 @samp{#} on the i960;
2254 @samp{;} for the PDP-11;
2257 @samp{;} for picoJava;
2260 @samp{#} for Motorola PowerPC;
2263 @samp{#} for IBM S/390;
2266 @samp{#} for the Sunplus SCORE;
2269 @samp{!} for the Renesas / SuperH SH;
2272 @samp{!} on the SPARC;
2275 @samp{#} on the ip2k;
2278 @samp{#} on the m32c;
2281 @samp{#} on the m32r;
2284 @samp{|} on the 680x0;
2287 @samp{#} on the 68HC11 and 68HC12;
2293 @samp{#} on the Vax;
2296 @samp{;} for the Z80;
2299 @samp{!} for the Z8000;
2302 @samp{#} on the V850;
2305 @samp{#} for Xtensa systems;
2307 see @ref{Machine Dependencies}. @refill
2308 @c FIXME What about i860?
2311 On some machines there are two different line comment characters. One
2312 character only begins a comment if it is the first non-whitespace character on
2313 a line, while the other always begins a comment.
2317 The V850 assembler also supports a double dash as starting a comment that
2318 extends to the end of the line.
2324 @cindex lines starting with @code{#}
2325 @cindex logical line numbers
2326 To be compatible with past assemblers, lines that begin with @samp{#} have a
2327 special interpretation. Following the @samp{#} should be an absolute
2328 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2329 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2330 new logical file name. The rest of the line, if any, should be whitespace.
2332 If the first non-whitespace characters on the line are not numeric,
2333 the line is ignored. (Just like a comment.)
2336 # This is an ordinary comment.
2337 # 42-6 "new_file_name" # New logical file name
2338 # This is logical line # 36.
2340 This feature is deprecated, and may disappear from future versions
2341 of @command{@value{AS}}.
2346 @cindex characters used in symbols
2347 @ifclear SPECIAL-SYMS
2348 A @dfn{symbol} is one or more characters chosen from the set of all
2349 letters (both upper and lower case), digits and the three characters
2355 A @dfn{symbol} is one or more characters chosen from the set of all
2356 letters (both upper and lower case), digits and the three characters
2357 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2363 On most machines, you can also use @code{$} in symbol names; exceptions
2364 are noted in @ref{Machine Dependencies}.
2366 No symbol may begin with a digit. Case is significant.
2367 There is no length limit: all characters are significant. Symbols are
2368 delimited by characters not in that set, or by the beginning of a file
2369 (since the source program must end with a newline, the end of a file is
2370 not a possible symbol delimiter). @xref{Symbols}.
2371 @cindex length of symbols
2376 @cindex statements, structure of
2377 @cindex line separator character
2378 @cindex statement separator character
2380 @ifclear abnormal-separator
2381 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
2382 semicolon (@samp{;}). The newline or semicolon is considered part of
2383 the preceding statement. Newlines and semicolons within character
2384 constants are an exception: they do not end statements.
2386 @ifset abnormal-separator
2388 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
2389 point (@samp{!}). The newline or exclamation point is considered part of the
2390 preceding statement. Newlines and exclamation points within character
2391 constants are an exception: they do not end statements.
2394 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
2395 H8/300) a dollar sign (@samp{$}); or (for the Renesas-SH) a semicolon
2396 (@samp{;}). The newline or separator character is considered part of
2397 the preceding statement. Newlines and separators within character
2398 constants are an exception: they do not end statements.
2403 A @dfn{statement} ends at a newline character (@samp{\n}) or line
2404 separator character. (The line separator is usually @samp{;}, unless this
2405 conflicts with the comment character; see @ref{Machine Dependencies}.) The
2406 newline or separator character is considered part of the preceding
2407 statement. Newlines and separators within character constants are an
2408 exception: they do not end statements.
2411 @cindex newline, required at file end
2412 @cindex EOF, newline must precede
2413 It is an error to end any statement with end-of-file: the last
2414 character of any input file should be a newline.@refill
2416 An empty statement is allowed, and may include whitespace. It is ignored.
2418 @cindex instructions and directives
2419 @cindex directives and instructions
2420 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2421 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2423 A statement begins with zero or more labels, optionally followed by a
2424 key symbol which determines what kind of statement it is. The key
2425 symbol determines the syntax of the rest of the statement. If the
2426 symbol begins with a dot @samp{.} then the statement is an assembler
2427 directive: typically valid for any computer. If the symbol begins with
2428 a letter the statement is an assembly language @dfn{instruction}: it
2429 assembles into a machine language instruction.
2431 Different versions of @command{@value{AS}} for different computers
2432 recognize different instructions. In fact, the same symbol may
2433 represent a different instruction in a different computer's assembly
2437 @cindex @code{:} (label)
2438 @cindex label (@code{:})
2439 A label is a symbol immediately followed by a colon (@code{:}).
2440 Whitespace before a label or after a colon is permitted, but you may not
2441 have whitespace between a label's symbol and its colon. @xref{Labels}.
2444 For HPPA targets, labels need not be immediately followed by a colon, but
2445 the definition of a label must begin in column zero. This also implies that
2446 only one label may be defined on each line.
2450 label: .directive followed by something
2451 another_label: # This is an empty statement.
2452 instruction operand_1, operand_2, @dots{}
2459 A constant is a number, written so that its value is known by
2460 inspection, without knowing any context. Like this:
2463 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2464 .ascii "Ring the bell\7" # A string constant.
2465 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2466 .float 0f-314159265358979323846264338327\
2467 95028841971.693993751E-40 # - pi, a flonum.
2472 * Characters:: Character Constants
2473 * Numbers:: Number Constants
2477 @subsection Character Constants
2479 @cindex character constants
2480 @cindex constants, character
2481 There are two kinds of character constants. A @dfn{character} stands
2482 for one character in one byte and its value may be used in
2483 numeric expressions. String constants (properly called string
2484 @emph{literals}) are potentially many bytes and their values may not be
2485 used in arithmetic expressions.
2489 * Chars:: Characters
2493 @subsubsection Strings
2495 @cindex string constants
2496 @cindex constants, string
2497 A @dfn{string} is written between double-quotes. It may contain
2498 double-quotes or null characters. The way to get special characters
2499 into a string is to @dfn{escape} these characters: precede them with
2500 a backslash @samp{\} character. For example @samp{\\} represents
2501 one backslash: the first @code{\} is an escape which tells
2502 @command{@value{AS}} to interpret the second character literally as a backslash
2503 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2504 escape character). The complete list of escapes follows.
2506 @cindex escape codes, character
2507 @cindex character escape codes
2510 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2512 @cindex @code{\b} (backspace character)
2513 @cindex backspace (@code{\b})
2515 Mnemonic for backspace; for ASCII this is octal code 010.
2518 @c Mnemonic for EOText; for ASCII this is octal code 004.
2520 @cindex @code{\f} (formfeed character)
2521 @cindex formfeed (@code{\f})
2523 Mnemonic for FormFeed; for ASCII this is octal code 014.
2525 @cindex @code{\n} (newline character)
2526 @cindex newline (@code{\n})
2528 Mnemonic for newline; for ASCII this is octal code 012.
2531 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2533 @cindex @code{\r} (carriage return character)
2534 @cindex carriage return (@code{\r})
2536 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2539 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2540 @c other assemblers.
2542 @cindex @code{\t} (tab)
2543 @cindex tab (@code{\t})
2545 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2548 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2549 @c @item \x @var{digit} @var{digit} @var{digit}
2550 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2552 @cindex @code{\@var{ddd}} (octal character code)
2553 @cindex octal character code (@code{\@var{ddd}})
2554 @item \ @var{digit} @var{digit} @var{digit}
2555 An octal character code. The numeric code is 3 octal digits.
2556 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2557 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2559 @cindex @code{\@var{xd...}} (hex character code)
2560 @cindex hex character code (@code{\@var{xd...}})
2561 @item \@code{x} @var{hex-digits...}
2562 A hex character code. All trailing hex digits are combined. Either upper or
2563 lower case @code{x} works.
2565 @cindex @code{\\} (@samp{\} character)
2566 @cindex backslash (@code{\\})
2568 Represents one @samp{\} character.
2571 @c Represents one @samp{'} (accent acute) character.
2572 @c This is needed in single character literals
2573 @c (@xref{Characters,,Character Constants}.) to represent
2576 @cindex @code{\"} (doublequote character)
2577 @cindex doublequote (@code{\"})
2579 Represents one @samp{"} character. Needed in strings to represent
2580 this character, because an unescaped @samp{"} would end the string.
2582 @item \ @var{anything-else}
2583 Any other character when escaped by @kbd{\} gives a warning, but
2584 assembles as if the @samp{\} was not present. The idea is that if
2585 you used an escape sequence you clearly didn't want the literal
2586 interpretation of the following character. However @command{@value{AS}} has no
2587 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2588 code and warns you of the fact.
2591 Which characters are escapable, and what those escapes represent,
2592 varies widely among assemblers. The current set is what we think
2593 the BSD 4.2 assembler recognizes, and is a subset of what most C
2594 compilers recognize. If you are in doubt, do not use an escape
2598 @subsubsection Characters
2600 @cindex single character constant
2601 @cindex character, single
2602 @cindex constant, single character
2603 A single character may be written as a single quote immediately
2604 followed by that character. The same escapes apply to characters as
2605 to strings. So if you want to write the character backslash, you
2606 must write @kbd{'\\} where the first @code{\} escapes the second
2607 @code{\}. As you can see, the quote is an acute accent, not a
2608 grave accent. A newline
2610 @ifclear abnormal-separator
2611 (or semicolon @samp{;})
2613 @ifset abnormal-separator
2615 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2620 immediately following an acute accent is taken as a literal character
2621 and does not count as the end of a statement. The value of a character
2622 constant in a numeric expression is the machine's byte-wide code for
2623 that character. @command{@value{AS}} assumes your character code is ASCII:
2624 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2627 @subsection Number Constants
2629 @cindex constants, number
2630 @cindex number constants
2631 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2632 are stored in the target machine. @emph{Integers} are numbers that
2633 would fit into an @code{int} in the C language. @emph{Bignums} are
2634 integers, but they are stored in more than 32 bits. @emph{Flonums}
2635 are floating point numbers, described below.
2638 * Integers:: Integers
2643 * Bit Fields:: Bit Fields
2649 @subsubsection Integers
2651 @cindex constants, integer
2653 @cindex binary integers
2654 @cindex integers, binary
2655 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2656 the binary digits @samp{01}.
2658 @cindex octal integers
2659 @cindex integers, octal
2660 An octal integer is @samp{0} followed by zero or more of the octal
2661 digits (@samp{01234567}).
2663 @cindex decimal integers
2664 @cindex integers, decimal
2665 A decimal integer starts with a non-zero digit followed by zero or
2666 more digits (@samp{0123456789}).
2668 @cindex hexadecimal integers
2669 @cindex integers, hexadecimal
2670 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2671 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2673 Integers have the usual values. To denote a negative integer, use
2674 the prefix operator @samp{-} discussed under expressions
2675 (@pxref{Prefix Ops,,Prefix Operators}).
2678 @subsubsection Bignums
2681 @cindex constants, bignum
2682 A @dfn{bignum} has the same syntax and semantics as an integer
2683 except that the number (or its negative) takes more than 32 bits to
2684 represent in binary. The distinction is made because in some places
2685 integers are permitted while bignums are not.
2688 @subsubsection Flonums
2690 @cindex floating point numbers
2691 @cindex constants, floating point
2693 @cindex precision, floating point
2694 A @dfn{flonum} represents a floating point number. The translation is
2695 indirect: a decimal floating point number from the text is converted by
2696 @command{@value{AS}} to a generic binary floating point number of more than
2697 sufficient precision. This generic floating point number is converted
2698 to a particular computer's floating point format (or formats) by a
2699 portion of @command{@value{AS}} specialized to that computer.
2701 A flonum is written by writing (in order)
2706 (@samp{0} is optional on the HPPA.)
2710 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2712 @kbd{e} is recommended. Case is not important.
2714 @c FIXME: verify if flonum syntax really this vague for most cases
2715 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2716 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2719 On the H8/300, Renesas / SuperH SH,
2720 and AMD 29K architectures, the letter must be
2721 one of the letters @samp{DFPRSX} (in upper or lower case).
2723 On the ARC, the letter must be one of the letters @samp{DFRS}
2724 (in upper or lower case).
2726 On the Intel 960 architecture, the letter must be
2727 one of the letters @samp{DFT} (in upper or lower case).
2729 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2733 One of the letters @samp{DFRS} (in upper or lower case).
2736 One of the letters @samp{DFPRSX} (in upper or lower case).
2739 The letter @samp{E} (upper case only).
2742 One of the letters @samp{DFT} (in upper or lower case).
2747 An optional sign: either @samp{+} or @samp{-}.
2750 An optional @dfn{integer part}: zero or more decimal digits.
2753 An optional @dfn{fractional part}: @samp{.} followed by zero
2754 or more decimal digits.
2757 An optional exponent, consisting of:
2761 An @samp{E} or @samp{e}.
2762 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2763 @c principle this can perfectly well be different on different targets.
2765 Optional sign: either @samp{+} or @samp{-}.
2767 One or more decimal digits.
2772 At least one of the integer part or the fractional part must be
2773 present. The floating point number has the usual base-10 value.
2775 @command{@value{AS}} does all processing using integers. Flonums are computed
2776 independently of any floating point hardware in the computer running
2777 @command{@value{AS}}.
2781 @c Bit fields are written as a general facility but are also controlled
2782 @c by a conditional-compilation flag---which is as of now (21mar91)
2783 @c turned on only by the i960 config of GAS.
2785 @subsubsection Bit Fields
2788 @cindex constants, bit field
2789 You can also define numeric constants as @dfn{bit fields}.
2790 Specify two numbers separated by a colon---
2792 @var{mask}:@var{value}
2795 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2798 The resulting number is then packed
2800 @c this conditional paren in case bit fields turned on elsewhere than 960
2801 (in host-dependent byte order)
2803 into a field whose width depends on which assembler directive has the
2804 bit-field as its argument. Overflow (a result from the bitwise and
2805 requiring more binary digits to represent) is not an error; instead,
2806 more constants are generated, of the specified width, beginning with the
2807 least significant digits.@refill
2809 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2810 @code{.short}, and @code{.word} accept bit-field arguments.
2815 @chapter Sections and Relocation
2820 * Secs Background:: Background
2821 * Ld Sections:: Linker Sections
2822 * As Sections:: Assembler Internal Sections
2823 * Sub-Sections:: Sub-Sections
2827 @node Secs Background
2830 Roughly, a section is a range of addresses, with no gaps; all data
2831 ``in'' those addresses is treated the same for some particular purpose.
2832 For example there may be a ``read only'' section.
2834 @cindex linker, and assembler
2835 @cindex assembler, and linker
2836 The linker @code{@value{LD}} reads many object files (partial programs) and
2837 combines their contents to form a runnable program. When @command{@value{AS}}
2838 emits an object file, the partial program is assumed to start at address 0.
2839 @code{@value{LD}} assigns the final addresses for the partial program, so that
2840 different partial programs do not overlap. This is actually an
2841 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2844 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2845 addresses. These blocks slide to their run-time addresses as rigid
2846 units; their length does not change and neither does the order of bytes
2847 within them. Such a rigid unit is called a @emph{section}. Assigning
2848 run-time addresses to sections is called @dfn{relocation}. It includes
2849 the task of adjusting mentions of object-file addresses so they refer to
2850 the proper run-time addresses.
2852 For the H8/300, and for the Renesas / SuperH SH,
2853 @command{@value{AS}} pads sections if needed to
2854 ensure they end on a word (sixteen bit) boundary.
2857 @cindex standard assembler sections
2858 An object file written by @command{@value{AS}} has at least three sections, any
2859 of which may be empty. These are named @dfn{text}, @dfn{data} and
2864 When it generates COFF or ELF output,
2866 @command{@value{AS}} can also generate whatever other named sections you specify
2867 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2868 If you do not use any directives that place output in the @samp{.text}
2869 or @samp{.data} sections, these sections still exist, but are empty.
2874 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2876 @command{@value{AS}} can also generate whatever other named sections you
2877 specify using the @samp{.space} and @samp{.subspace} directives. See
2878 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2879 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2880 assembler directives.
2883 Additionally, @command{@value{AS}} uses different names for the standard
2884 text, data, and bss sections when generating SOM output. Program text
2885 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2886 BSS into @samp{$BSS$}.
2890 Within the object file, the text section starts at address @code{0}, the
2891 data section follows, and the bss section follows the data section.
2894 When generating either SOM or ELF output files on the HPPA, the text
2895 section starts at address @code{0}, the data section at address
2896 @code{0x4000000}, and the bss section follows the data section.
2899 To let @code{@value{LD}} know which data changes when the sections are
2900 relocated, and how to change that data, @command{@value{AS}} also writes to the
2901 object file details of the relocation needed. To perform relocation
2902 @code{@value{LD}} must know, each time an address in the object
2906 Where in the object file is the beginning of this reference to
2909 How long (in bytes) is this reference?
2911 Which section does the address refer to? What is the numeric value of
2913 (@var{address}) @minus{} (@var{start-address of section})?
2916 Is the reference to an address ``Program-Counter relative''?
2919 @cindex addresses, format of
2920 @cindex section-relative addressing
2921 In fact, every address @command{@value{AS}} ever uses is expressed as
2923 (@var{section}) + (@var{offset into section})
2926 Further, most expressions @command{@value{AS}} computes have this section-relative
2929 (For some object formats, such as SOM for the HPPA, some expressions are
2930 symbol-relative instead.)
2933 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2934 @var{N} into section @var{secname}.''
2936 Apart from text, data and bss sections you need to know about the
2937 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2938 addresses in the absolute section remain unchanged. For example, address
2939 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2940 @code{@value{LD}}. Although the linker never arranges two partial programs'
2941 data sections with overlapping addresses after linking, @emph{by definition}
2942 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2943 part of a program is always the same address when the program is running as
2944 address @code{@{absolute@ 239@}} in any other part of the program.
2946 The idea of sections is extended to the @dfn{undefined} section. Any
2947 address whose section is unknown at assembly time is by definition
2948 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2949 Since numbers are always defined, the only way to generate an undefined
2950 address is to mention an undefined symbol. A reference to a named
2951 common block would be such a symbol: its value is unknown at assembly
2952 time so it has section @emph{undefined}.
2954 By analogy the word @emph{section} is used to describe groups of sections in
2955 the linked program. @code{@value{LD}} puts all partial programs' text
2956 sections in contiguous addresses in the linked program. It is
2957 customary to refer to the @emph{text section} of a program, meaning all
2958 the addresses of all partial programs' text sections. Likewise for
2959 data and bss sections.
2961 Some sections are manipulated by @code{@value{LD}}; others are invented for
2962 use of @command{@value{AS}} and have no meaning except during assembly.
2965 @section Linker Sections
2966 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2971 @cindex named sections
2972 @cindex sections, named
2973 @item named sections
2976 @cindex text section
2977 @cindex data section
2981 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
2982 separate but equal sections. Anything you can say of one section is
2985 When the program is running, however, it is
2986 customary for the text section to be unalterable. The
2987 text section is often shared among processes: it contains
2988 instructions, constants and the like. The data section of a running
2989 program is usually alterable: for example, C variables would be stored
2990 in the data section.
2995 This section contains zeroed bytes when your program begins running. It
2996 is used to hold uninitialized variables or common storage. The length of
2997 each partial program's bss section is important, but because it starts
2998 out containing zeroed bytes there is no need to store explicit zero
2999 bytes in the object file. The bss section was invented to eliminate
3000 those explicit zeros from object files.
3002 @cindex absolute section
3003 @item absolute section
3004 Address 0 of this section is always ``relocated'' to runtime address 0.
3005 This is useful if you want to refer to an address that @code{@value{LD}} must
3006 not change when relocating. In this sense we speak of absolute
3007 addresses being ``unrelocatable'': they do not change during relocation.
3009 @cindex undefined section
3010 @item undefined section
3011 This ``section'' is a catch-all for address references to objects not in
3012 the preceding sections.
3013 @c FIXME: ref to some other doc on obj-file formats could go here.
3016 @cindex relocation example
3017 An idealized example of three relocatable sections follows.
3019 The example uses the traditional section names @samp{.text} and @samp{.data}.
3021 Memory addresses are on the horizontal axis.
3025 @c END TEXI2ROFF-KILL
3028 partial program # 1: |ttttt|dddd|00|
3035 partial program # 2: |TTT|DDD|000|
3038 +--+---+-----+--+----+---+-----+~~
3039 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3040 +--+---+-----+--+----+---+-----+~~
3042 addresses: 0 @dots{}
3049 \line{\it Partial program \#1: \hfil}
3050 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3051 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3053 \line{\it Partial program \#2: \hfil}
3054 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3055 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3057 \line{\it linked program: \hfil}
3058 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3059 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3060 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3061 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3063 \line{\it addresses: \hfil}
3067 @c END TEXI2ROFF-KILL
3070 @section Assembler Internal Sections
3072 @cindex internal assembler sections
3073 @cindex sections in messages, internal
3074 These sections are meant only for the internal use of @command{@value{AS}}. They
3075 have no meaning at run-time. You do not really need to know about these
3076 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3077 warning messages, so it might be helpful to have an idea of their
3078 meanings to @command{@value{AS}}. These sections are used to permit the
3079 value of every expression in your assembly language program to be a
3080 section-relative address.
3083 @cindex assembler internal logic error
3084 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3085 An internal assembler logic error has been found. This means there is a
3086 bug in the assembler.
3088 @cindex expr (internal section)
3090 The assembler stores complex expression internally as combinations of
3091 symbols. When it needs to represent an expression as a symbol, it puts
3092 it in the expr section.
3094 @c FIXME item transfer[t] vector preload
3095 @c FIXME item transfer[t] vector postload
3096 @c FIXME item register
3100 @section Sub-Sections
3102 @cindex numbered subsections
3103 @cindex grouping data
3109 fall into two sections: text and data.
3111 You may have separate groups of
3113 data in named sections
3117 data in named sections
3123 that you want to end up near to each other in the object file, even though they
3124 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3125 use @dfn{subsections} for this purpose. Within each section, there can be
3126 numbered subsections with values from 0 to 8192. Objects assembled into the
3127 same subsection go into the object file together with other objects in the same
3128 subsection. For example, a compiler might want to store constants in the text
3129 section, but might not want to have them interspersed with the program being
3130 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3131 section of code being output, and a @samp{.text 1} before each group of
3132 constants being output.
3134 Subsections are optional. If you do not use subsections, everything
3135 goes in subsection number zero.
3138 Each subsection is zero-padded up to a multiple of four bytes.
3139 (Subsections may be padded a different amount on different flavors
3140 of @command{@value{AS}}.)
3144 On the H8/300 platform, each subsection is zero-padded to a word
3145 boundary (two bytes).
3146 The same is true on the Renesas SH.
3149 @c FIXME section padding (alignment)?
3150 @c Rich Pixley says padding here depends on target obj code format; that
3151 @c doesn't seem particularly useful to say without further elaboration,
3152 @c so for now I say nothing about it. If this is a generic BFD issue,
3153 @c these paragraphs might need to vanish from this manual, and be
3154 @c discussed in BFD chapter of binutils (or some such).
3158 Subsections appear in your object file in numeric order, lowest numbered
3159 to highest. (All this to be compatible with other people's assemblers.)
3160 The object file contains no representation of subsections; @code{@value{LD}} and
3161 other programs that manipulate object files see no trace of them.
3162 They just see all your text subsections as a text section, and all your
3163 data subsections as a data section.
3165 To specify which subsection you want subsequent statements assembled
3166 into, use a numeric argument to specify it, in a @samp{.text
3167 @var{expression}} or a @samp{.data @var{expression}} statement.
3170 When generating COFF output, you
3175 can also use an extra subsection
3176 argument with arbitrary named sections: @samp{.section @var{name},
3181 When generating ELF output, you
3186 can also use the @code{.subsection} directive (@pxref{SubSection})
3187 to specify a subsection: @samp{.subsection @var{expression}}.
3189 @var{Expression} should be an absolute expression
3190 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3191 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3192 begins in @code{text 0}. For instance:
3194 .text 0 # The default subsection is text 0 anyway.
3195 .ascii "This lives in the first text subsection. *"
3197 .ascii "But this lives in the second text subsection."
3199 .ascii "This lives in the data section,"
3200 .ascii "in the first data subsection."
3202 .ascii "This lives in the first text section,"
3203 .ascii "immediately following the asterisk (*)."
3206 Each section has a @dfn{location counter} incremented by one for every byte
3207 assembled into that section. Because subsections are merely a convenience
3208 restricted to @command{@value{AS}} there is no concept of a subsection location
3209 counter. There is no way to directly manipulate a location counter---but the
3210 @code{.align} directive changes it, and any label definition captures its
3211 current value. The location counter of the section where statements are being
3212 assembled is said to be the @dfn{active} location counter.
3215 @section bss Section
3218 @cindex common variable storage
3219 The bss section is used for local common variable storage.
3220 You may allocate address space in the bss section, but you may
3221 not dictate data to load into it before your program executes. When
3222 your program starts running, all the contents of the bss
3223 section are zeroed bytes.
3225 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3226 @ref{Lcomm,,@code{.lcomm}}.
3228 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3229 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3232 When assembling for a target which supports multiple sections, such as ELF or
3233 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3234 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3235 section. Typically the section will only contain symbol definitions and
3236 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3243 Symbols are a central concept: the programmer uses symbols to name
3244 things, the linker uses symbols to link, and the debugger uses symbols
3248 @cindex debuggers, and symbol order
3249 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3250 the same order they were declared. This may break some debuggers.
3255 * Setting Symbols:: Giving Symbols Other Values
3256 * Symbol Names:: Symbol Names
3257 * Dot:: The Special Dot Symbol
3258 * Symbol Attributes:: Symbol Attributes
3265 A @dfn{label} is written as a symbol immediately followed by a colon
3266 @samp{:}. The symbol then represents the current value of the
3267 active location counter, and is, for example, a suitable instruction
3268 operand. You are warned if you use the same symbol to represent two
3269 different locations: the first definition overrides any other
3273 On the HPPA, the usual form for a label need not be immediately followed by a
3274 colon, but instead must start in column zero. Only one label may be defined on
3275 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3276 provides a special directive @code{.label} for defining labels more flexibly.
3279 @node Setting Symbols
3280 @section Giving Symbols Other Values
3282 @cindex assigning values to symbols
3283 @cindex symbol values, assigning
3284 A symbol can be given an arbitrary value by writing a symbol, followed
3285 by an equals sign @samp{=}, followed by an expression
3286 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3287 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3288 equals sign @samp{=}@samp{=} here represents an equivalent of the
3289 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3292 Blackfin does not support symbol assignment with @samp{=}.
3296 @section Symbol Names
3298 @cindex symbol names
3299 @cindex names, symbol
3300 @ifclear SPECIAL-SYMS
3301 Symbol names begin with a letter or with one of @samp{._}. On most
3302 machines, you can also use @code{$} in symbol names; exceptions are
3303 noted in @ref{Machine Dependencies}. That character may be followed by any
3304 string of digits, letters, dollar signs (unless otherwise noted for a
3305 particular target machine), and underscores.
3309 Symbol names begin with a letter or with one of @samp{._}. On the
3310 Renesas SH you can also use @code{$} in symbol names. That
3311 character may be followed by any string of digits, letters, dollar signs (save
3312 on the H8/300), and underscores.
3316 Case of letters is significant: @code{foo} is a different symbol name
3319 Each symbol has exactly one name. Each name in an assembly language program
3320 refers to exactly one symbol. You may use that symbol name any number of times
3323 @subheading Local Symbol Names
3325 @cindex local symbol names
3326 @cindex symbol names, local
3327 A local symbol is any symbol beginning with certain local label prefixes.
3328 By default, the local label prefix is @samp{.L} for ELF systems or
3329 @samp{L} for traditional a.out systems, but each target may have its own
3330 set of local label prefixes.
3332 On the HPPA local symbols begin with @samp{L$}.
3335 Local symbols are defined and used within the assembler, but they are
3336 normally not saved in object files. Thus, they are not visible when debugging.
3337 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols:
3338 @option{-L}}) to retain the local symbols in the object files.
3340 @subheading Local Labels
3342 @cindex local labels
3343 @cindex temporary symbol names
3344 @cindex symbol names, temporary
3345 Local labels help compilers and programmers use names temporarily.
3346 They create symbols which are guaranteed to be unique over the entire scope of
3347 the input source code and which can be referred to by a simple notation.
3348 To define a local label, write a label of the form @samp{@b{N}:} (where @b{N}
3349 represents any positive integer). To refer to the most recent previous
3350 definition of that label write @samp{@b{N}b}, using the same number as when
3351 you defined the label. To refer to the next definition of a local label, write
3352 @samp{@b{N}f}---the @samp{b} stands for ``backwards'' and the @samp{f} stands
3355 There is no restriction on how you can use these labels, and you can reuse them
3356 too. So that it is possible to repeatedly define the same local label (using
3357 the same number @samp{@b{N}}), although you can only refer to the most recently
3358 defined local label of that number (for a backwards reference) or the next
3359 definition of a specific local label for a forward reference. It is also worth
3360 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3361 implemented in a slightly more efficient manner than the others.
3372 Which is the equivalent of:
3375 label_1: branch label_3
3376 label_2: branch label_1
3377 label_3: branch label_4
3378 label_4: branch label_3
3381 Local label names are only a notational device. They are immediately
3382 transformed into more conventional symbol names before the assembler uses them.
3383 The symbol names are stored in the symbol table, appear in error messages, and
3384 are optionally emitted to the object file. The names are constructed using
3388 @item @emph{local label prefix}
3389 All local symbols begin with the system-specific local label prefix.
3390 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3391 that start with the local label prefix. These labels are
3392 used for symbols you are never intended to see. If you use the
3393 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3394 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3395 you may use them in debugging.
3398 This is the number that was used in the local label definition. So if the
3399 label is written @samp{55:} then the number is @samp{55}.
3402 This unusual character is included so you do not accidentally invent a symbol
3403 of the same name. The character has ASCII value of @samp{\002} (control-B).
3405 @item @emph{ordinal number}
3406 This is a serial number to keep the labels distinct. The first definition of
3407 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3408 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3409 the number @samp{1} and its 15th definition gets @samp{15} as well.
3412 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3413 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3415 @subheading Dollar Local Labels
3416 @cindex dollar local symbols
3418 @code{@value{AS}} also supports an even more local form of local labels called
3419 dollar labels. These labels go out of scope (i.e., they become undefined) as
3420 soon as a non-local label is defined. Thus they remain valid for only a small
3421 region of the input source code. Normal local labels, by contrast, remain in
3422 scope for the entire file, or until they are redefined by another occurrence of
3423 the same local label.
3425 Dollar labels are defined in exactly the same way as ordinary local labels,
3426 except that they have a dollar sign suffix to their numeric value, e.g.,
3429 They can also be distinguished from ordinary local labels by their transformed
3430 names which use ASCII character @samp{\001} (control-A) as the magic character
3431 to distinguish them from ordinary labels. For example, the fifth definition of
3432 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3435 @section The Special Dot Symbol
3437 @cindex dot (symbol)
3438 @cindex @code{.} (symbol)
3439 @cindex current address
3440 @cindex location counter
3441 The special symbol @samp{.} refers to the current address that
3442 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3443 .long .} defines @code{melvin} to contain its own address.
3444 Assigning a value to @code{.} is treated the same as a @code{.org}
3446 @ifclear no-space-dir
3447 Thus, the expression @samp{.=.+4} is the same as saying
3451 @node Symbol Attributes
3452 @section Symbol Attributes
3454 @cindex symbol attributes
3455 @cindex attributes, symbol
3456 Every symbol has, as well as its name, the attributes ``Value'' and
3457 ``Type''. Depending on output format, symbols can also have auxiliary
3460 The detailed definitions are in @file{a.out.h}.
3463 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3464 all these attributes, and probably won't warn you. This makes the
3465 symbol an externally defined symbol, which is generally what you
3469 * Symbol Value:: Value
3470 * Symbol Type:: Type
3473 * a.out Symbols:: Symbol Attributes: @code{a.out}
3477 * a.out Symbols:: Symbol Attributes: @code{a.out}
3480 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3485 * COFF Symbols:: Symbol Attributes for COFF
3488 * SOM Symbols:: Symbol Attributes for SOM
3495 @cindex value of a symbol
3496 @cindex symbol value
3497 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3498 location in the text, data, bss or absolute sections the value is the
3499 number of addresses from the start of that section to the label.
3500 Naturally for text, data and bss sections the value of a symbol changes
3501 as @code{@value{LD}} changes section base addresses during linking. Absolute
3502 symbols' values do not change during linking: that is why they are
3505 The value of an undefined symbol is treated in a special way. If it is
3506 0 then the symbol is not defined in this assembler source file, and
3507 @code{@value{LD}} tries to determine its value from other files linked into the
3508 same program. You make this kind of symbol simply by mentioning a symbol
3509 name without defining it. A non-zero value represents a @code{.comm}
3510 common declaration. The value is how much common storage to reserve, in
3511 bytes (addresses). The symbol refers to the first address of the
3517 @cindex type of a symbol
3519 The type attribute of a symbol contains relocation (section)
3520 information, any flag settings indicating that a symbol is external, and
3521 (optionally), other information for linkers and debuggers. The exact
3522 format depends on the object-code output format in use.
3527 @c The following avoids a "widow" subsection title. @group would be
3528 @c better if it were available outside examples.
3531 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3533 @cindex @code{b.out} symbol attributes
3534 @cindex symbol attributes, @code{b.out}
3535 These symbol attributes appear only when @command{@value{AS}} is configured for
3536 one of the Berkeley-descended object output formats---@code{a.out} or
3542 @subsection Symbol Attributes: @code{a.out}
3544 @cindex @code{a.out} symbol attributes
3545 @cindex symbol attributes, @code{a.out}
3551 @subsection Symbol Attributes: @code{a.out}
3553 @cindex @code{a.out} symbol attributes
3554 @cindex symbol attributes, @code{a.out}
3558 * Symbol Desc:: Descriptor
3559 * Symbol Other:: Other
3563 @subsubsection Descriptor
3565 @cindex descriptor, of @code{a.out} symbol
3566 This is an arbitrary 16-bit value. You may establish a symbol's
3567 descriptor value by using a @code{.desc} statement
3568 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3569 @command{@value{AS}}.
3572 @subsubsection Other
3574 @cindex other attribute, of @code{a.out} symbol
3575 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3580 @subsection Symbol Attributes for COFF
3582 @cindex COFF symbol attributes
3583 @cindex symbol attributes, COFF
3585 The COFF format supports a multitude of auxiliary symbol attributes;
3586 like the primary symbol attributes, they are set between @code{.def} and
3587 @code{.endef} directives.
3589 @subsubsection Primary Attributes
3591 @cindex primary attributes, COFF symbols
3592 The symbol name is set with @code{.def}; the value and type,
3593 respectively, with @code{.val} and @code{.type}.
3595 @subsubsection Auxiliary Attributes
3597 @cindex auxiliary attributes, COFF symbols
3598 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3599 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3600 table information for COFF.
3605 @subsection Symbol Attributes for SOM
3607 @cindex SOM symbol attributes
3608 @cindex symbol attributes, SOM
3610 The SOM format for the HPPA supports a multitude of symbol attributes set with
3611 the @code{.EXPORT} and @code{.IMPORT} directives.
3613 The attributes are described in @cite{HP9000 Series 800 Assembly
3614 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3615 @code{EXPORT} assembler directive documentation.
3619 @chapter Expressions
3623 @cindex numeric values
3624 An @dfn{expression} specifies an address or numeric value.
3625 Whitespace may precede and/or follow an expression.
3627 The result of an expression must be an absolute number, or else an offset into
3628 a particular section. If an expression is not absolute, and there is not
3629 enough information when @command{@value{AS}} sees the expression to know its
3630 section, a second pass over the source program might be necessary to interpret
3631 the expression---but the second pass is currently not implemented.
3632 @command{@value{AS}} aborts with an error message in this situation.
3635 * Empty Exprs:: Empty Expressions
3636 * Integer Exprs:: Integer Expressions
3640 @section Empty Expressions
3642 @cindex empty expressions
3643 @cindex expressions, empty
3644 An empty expression has no value: it is just whitespace or null.
3645 Wherever an absolute expression is required, you may omit the
3646 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3647 is compatible with other assemblers.
3650 @section Integer Expressions
3652 @cindex integer expressions
3653 @cindex expressions, integer
3654 An @dfn{integer expression} is one or more @emph{arguments} delimited
3655 by @emph{operators}.
3658 * Arguments:: Arguments
3659 * Operators:: Operators
3660 * Prefix Ops:: Prefix Operators
3661 * Infix Ops:: Infix Operators
3665 @subsection Arguments
3667 @cindex expression arguments
3668 @cindex arguments in expressions
3669 @cindex operands in expressions
3670 @cindex arithmetic operands
3671 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3672 contexts arguments are sometimes called ``arithmetic operands''. In
3673 this manual, to avoid confusing them with the ``instruction operands'' of
3674 the machine language, we use the term ``argument'' to refer to parts of
3675 expressions only, reserving the word ``operand'' to refer only to machine
3676 instruction operands.
3678 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3679 @var{section} is one of text, data, bss, absolute,
3680 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3683 Numbers are usually integers.
3685 A number can be a flonum or bignum. In this case, you are warned
3686 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3687 these 32 bits are an integer. You may write integer-manipulating
3688 instructions that act on exotic constants, compatible with other
3691 @cindex subexpressions
3692 Subexpressions are a left parenthesis @samp{(} followed by an integer
3693 expression, followed by a right parenthesis @samp{)}; or a prefix
3694 operator followed by an argument.
3697 @subsection Operators
3699 @cindex operators, in expressions
3700 @cindex arithmetic functions
3701 @cindex functions, in expressions
3702 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3703 operators are followed by an argument. Infix operators appear
3704 between their arguments. Operators may be preceded and/or followed by
3708 @subsection Prefix Operator
3710 @cindex prefix operators
3711 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3712 one argument, which must be absolute.
3714 @c the tex/end tex stuff surrounding this small table is meant to make
3715 @c it align, on the printed page, with the similar table in the next
3716 @c section (which is inside an enumerate).
3718 \global\advance\leftskip by \itemindent
3723 @dfn{Negation}. Two's complement negation.
3725 @dfn{Complementation}. Bitwise not.
3729 \global\advance\leftskip by -\itemindent
3733 @subsection Infix Operators
3735 @cindex infix operators
3736 @cindex operators, permitted arguments
3737 @dfn{Infix operators} take two arguments, one on either side. Operators
3738 have precedence, but operations with equal precedence are performed left
3739 to right. Apart from @code{+} or @option{-}, both arguments must be
3740 absolute, and the result is absolute.
3743 @cindex operator precedence
3744 @cindex precedence of operators
3751 @dfn{Multiplication}.
3754 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3760 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3763 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3767 Intermediate precedence
3772 @dfn{Bitwise Inclusive Or}.
3778 @dfn{Bitwise Exclusive Or}.
3781 @dfn{Bitwise Or Not}.
3788 @cindex addition, permitted arguments
3789 @cindex plus, permitted arguments
3790 @cindex arguments for addition
3792 @dfn{Addition}. If either argument is absolute, the result has the section of
3793 the other argument. You may not add together arguments from different
3796 @cindex subtraction, permitted arguments
3797 @cindex minus, permitted arguments
3798 @cindex arguments for subtraction
3800 @dfn{Subtraction}. If the right argument is absolute, the
3801 result has the section of the left argument.
3802 If both arguments are in the same section, the result is absolute.
3803 You may not subtract arguments from different sections.
3804 @c FIXME is there still something useful to say about undefined - undefined ?
3806 @cindex comparison expressions
3807 @cindex expressions, comparison
3812 @dfn{Is Not Equal To}
3816 @dfn{Is Greater Than}
3818 @dfn{Is Greater Than Or Equal To}
3820 @dfn{Is Less Than Or Equal To}
3822 The comparison operators can be used as infix operators. A true results has a
3823 value of -1 whereas a false result has a value of 0. Note, these operators
3824 perform signed comparisons.
3827 @item Lowest Precedence
3836 These two logical operations can be used to combine the results of sub
3837 expressions. Note, unlike the comparison operators a true result returns a
3838 value of 1 but a false results does still return 0. Also note that the logical
3839 or operator has a slightly lower precedence than logical and.
3844 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3845 address; you can only have a defined section in one of the two arguments.
3848 @chapter Assembler Directives
3850 @cindex directives, machine independent
3851 @cindex pseudo-ops, machine independent
3852 @cindex machine independent directives
3853 All assembler directives have names that begin with a period (@samp{.}).
3854 The rest of the name is letters, usually in lower case.
3856 This chapter discusses directives that are available regardless of the
3857 target machine configuration for the @sc{gnu} assembler.
3859 Some machine configurations provide additional directives.
3860 @xref{Machine Dependencies}.
3863 @ifset machine-directives
3864 @xref{Machine Dependencies}, for additional directives.
3869 * Abort:: @code{.abort}
3871 * ABORT (COFF):: @code{.ABORT}
3874 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3875 * Altmacro:: @code{.altmacro}
3876 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3877 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3878 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3879 * Byte:: @code{.byte @var{expressions}}
3880 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
3881 * Comm:: @code{.comm @var{symbol} , @var{length} }
3882 * Data:: @code{.data @var{subsection}}
3884 * Def:: @code{.def @var{name}}
3887 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3893 * Double:: @code{.double @var{flonums}}
3894 * Eject:: @code{.eject}
3895 * Else:: @code{.else}
3896 * Elseif:: @code{.elseif}
3899 * Endef:: @code{.endef}
3902 * Endfunc:: @code{.endfunc}
3903 * Endif:: @code{.endif}
3904 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3905 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3906 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
3908 * Error:: @code{.error @var{string}}
3909 * Exitm:: @code{.exitm}
3910 * Extern:: @code{.extern}
3911 * Fail:: @code{.fail}
3912 * File:: @code{.file}
3913 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3914 * Float:: @code{.float @var{flonums}}
3915 * Func:: @code{.func}
3916 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3918 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
3919 * Hidden:: @code{.hidden @var{names}}
3922 * hword:: @code{.hword @var{expressions}}
3923 * Ident:: @code{.ident}
3924 * If:: @code{.if @var{absolute expression}}
3925 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
3926 * Include:: @code{.include "@var{file}"}
3927 * Int:: @code{.int @var{expressions}}
3929 * Internal:: @code{.internal @var{names}}
3932 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3933 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3934 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3935 * Lflags:: @code{.lflags}
3936 @ifclear no-line-dir
3937 * Line:: @code{.line @var{line-number}}
3940 * Linkonce:: @code{.linkonce [@var{type}]}
3941 * List:: @code{.list}
3942 * Ln:: @code{.ln @var{line-number}}
3943 * Loc:: @code{.loc @var{fileno} @var{lineno}}
3944 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
3946 * Local:: @code{.local @var{names}}
3949 * Long:: @code{.long @var{expressions}}
3951 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3954 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3955 * MRI:: @code{.mri @var{val}}
3956 * Noaltmacro:: @code{.noaltmacro}
3957 * Nolist:: @code{.nolist}
3958 * Octa:: @code{.octa @var{bignums}}
3959 * Org:: @code{.org @var{new-lc}, @var{fill}}
3960 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3962 * PopSection:: @code{.popsection}
3963 * Previous:: @code{.previous}
3966 * Print:: @code{.print @var{string}}
3968 * Protected:: @code{.protected @var{names}}
3971 * Psize:: @code{.psize @var{lines}, @var{columns}}
3972 * Purgem:: @code{.purgem @var{name}}
3974 * PushSection:: @code{.pushsection @var{name}}
3977 * Quad:: @code{.quad @var{bignums}}
3978 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
3979 * Rept:: @code{.rept @var{count}}
3980 * Sbttl:: @code{.sbttl "@var{subheading}"}
3982 * Scl:: @code{.scl @var{class}}
3985 * Section:: @code{.section @var{name}[, @var{flags}]}
3988 * Set:: @code{.set @var{symbol}, @var{expression}}
3989 * Short:: @code{.short @var{expressions}}
3990 * Single:: @code{.single @var{flonums}}
3992 * Size:: @code{.size [@var{name} , @var{expression}]}
3994 @ifclear no-space-dir
3995 * Skip:: @code{.skip @var{size} , @var{fill}}
3998 * Sleb128:: @code{.sleb128 @var{expressions}}
3999 @ifclear no-space-dir
4000 * Space:: @code{.space @var{size} , @var{fill}}
4003 * Stab:: @code{.stabd, .stabn, .stabs}
4006 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4007 * Struct:: @code{.struct @var{expression}}
4009 * SubSection:: @code{.subsection}
4010 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4014 * Tag:: @code{.tag @var{structname}}
4017 * Text:: @code{.text @var{subsection}}
4018 * Title:: @code{.title "@var{heading}"}
4020 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4023 * Uleb128:: @code{.uleb128 @var{expressions}}
4025 * Val:: @code{.val @var{addr}}
4029 * Version:: @code{.version "@var{string}"}
4030 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4031 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4034 * Warning:: @code{.warning @var{string}}
4035 * Weak:: @code{.weak @var{names}}
4036 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4037 * Word:: @code{.word @var{expressions}}
4038 * Deprecated:: Deprecated Directives
4042 @section @code{.abort}
4044 @cindex @code{abort} directive
4045 @cindex stopping the assembly
4046 This directive stops the assembly immediately. It is for
4047 compatibility with other assemblers. The original idea was that the
4048 assembly language source would be piped into the assembler. If the sender
4049 of the source quit, it could use this directive tells @command{@value{AS}} to
4050 quit also. One day @code{.abort} will not be supported.
4054 @section @code{.ABORT} (COFF)
4056 @cindex @code{ABORT} directive
4057 When producing COFF output, @command{@value{AS}} accepts this directive as a
4058 synonym for @samp{.abort}.
4061 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4067 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4069 @cindex padding the location counter
4070 @cindex @code{align} directive
4071 Pad the location counter (in the current subsection) to a particular storage
4072 boundary. The first expression (which must be absolute) is the alignment
4073 required, as described below.
4075 The second expression (also absolute) gives the fill value to be stored in the
4076 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4077 padding bytes are normally zero. However, on some systems, if the section is
4078 marked as containing code and the fill value is omitted, the space is filled
4079 with no-op instructions.
4081 The third expression is also absolute, and is also optional. If it is present,
4082 it is the maximum number of bytes that should be skipped by this alignment
4083 directive. If doing the alignment would require skipping more bytes than the
4084 specified maximum, then the alignment is not done at all. You can omit the
4085 fill value (the second argument) entirely by simply using two commas after the
4086 required alignment; this can be useful if you want the alignment to be filled
4087 with no-op instructions when appropriate.
4089 The way the required alignment is specified varies from system to system.
4090 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or32,
4091 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4092 alignment request in bytes. For example @samp{.align 8} advances
4093 the location counter until it is a multiple of 8. If the location counter
4094 is already a multiple of 8, no change is needed. For the tic54x, the
4095 first expression is the alignment request in words.
4097 For other systems, including ppc, i386 using a.out format, arm and
4098 strongarm, it is the
4099 number of low-order zero bits the location counter must have after
4100 advancement. For example @samp{.align 3} advances the location
4101 counter until it a multiple of 8. If the location counter is already a
4102 multiple of 8, no change is needed.
4104 This inconsistency is due to the different behaviors of the various
4105 native assemblers for these systems which GAS must emulate.
4106 GAS also provides @code{.balign} and @code{.p2align} directives,
4107 described later, which have a consistent behavior across all
4108 architectures (but are specific to GAS).
4111 @section @code{.altmacro}
4112 Enable alternate macro mode, enabling:
4115 @item LOCAL @var{name} [ , @dots{} ]
4116 One additional directive, @code{LOCAL}, is available. It is used to
4117 generate a string replacement for each of the @var{name} arguments, and
4118 replace any instances of @var{name} in each macro expansion. The
4119 replacement string is unique in the assembly, and different for each
4120 separate macro expansion. @code{LOCAL} allows you to write macros that
4121 define symbols, without fear of conflict between separate macro expansions.
4123 @item String delimiters
4124 You can write strings delimited in these other ways besides
4125 @code{"@var{string}"}:
4128 @item '@var{string}'
4129 You can delimit strings with single-quote characters.
4131 @item <@var{string}>
4132 You can delimit strings with matching angle brackets.
4135 @item single-character string escape
4136 To include any single character literally in a string (even if the
4137 character would otherwise have some special meaning), you can prefix the
4138 character with @samp{!} (an exclamation mark). For example, you can
4139 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4141 @item Expression results as strings
4142 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4143 and use the result as a string.
4147 @section @code{.ascii "@var{string}"}@dots{}
4149 @cindex @code{ascii} directive
4150 @cindex string literals
4151 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4152 separated by commas. It assembles each string (with no automatic
4153 trailing zero byte) into consecutive addresses.
4156 @section @code{.asciz "@var{string}"}@dots{}
4158 @cindex @code{asciz} directive
4159 @cindex zero-terminated strings
4160 @cindex null-terminated strings
4161 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4162 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4165 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4167 @cindex padding the location counter given number of bytes
4168 @cindex @code{balign} directive
4169 Pad the location counter (in the current subsection) to a particular
4170 storage boundary. The first expression (which must be absolute) is the
4171 alignment request in bytes. For example @samp{.balign 8} advances
4172 the location counter until it is a multiple of 8. If the location counter
4173 is already a multiple of 8, no change is needed.
4175 The second expression (also absolute) gives the fill value to be stored in the
4176 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4177 padding bytes are normally zero. However, on some systems, if the section is
4178 marked as containing code and the fill value is omitted, the space is filled
4179 with no-op instructions.
4181 The third expression is also absolute, and is also optional. If it is present,
4182 it is the maximum number of bytes that should be skipped by this alignment
4183 directive. If doing the alignment would require skipping more bytes than the
4184 specified maximum, then the alignment is not done at all. You can omit the
4185 fill value (the second argument) entirely by simply using two commas after the
4186 required alignment; this can be useful if you want the alignment to be filled
4187 with no-op instructions when appropriate.
4189 @cindex @code{balignw} directive
4190 @cindex @code{balignl} directive
4191 The @code{.balignw} and @code{.balignl} directives are variants of the
4192 @code{.balign} directive. The @code{.balignw} directive treats the fill
4193 pattern as a two byte word value. The @code{.balignl} directives treats the
4194 fill pattern as a four byte longword value. For example, @code{.balignw
4195 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4196 filled in with the value 0x368d (the exact placement of the bytes depends upon
4197 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4201 @section @code{.byte @var{expressions}}
4203 @cindex @code{byte} directive
4204 @cindex integers, one byte
4205 @code{.byte} expects zero or more expressions, separated by commas.
4206 Each expression is assembled into the next byte.
4208 @node CFI directives
4209 @section @code{.cfi_sections @var{section_list}}
4210 @cindex @code{cfi_sections} directive
4211 @code{.cfi_sections} may be used to specify whether CFI directives
4212 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4213 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4214 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4215 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4216 directive is not used is @code{.cfi_sections .eh_frame}.
4218 @section @code{.cfi_startproc [simple]}
4219 @cindex @code{cfi_startproc} directive
4220 @code{.cfi_startproc} is used at the beginning of each function that
4221 should have an entry in @code{.eh_frame}. It initializes some internal
4222 data structures. Don't forget to close the function by
4223 @code{.cfi_endproc}.
4225 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4226 it also emits some architecture dependent initial CFI instructions.
4228 @section @code{.cfi_endproc}
4229 @cindex @code{cfi_endproc} directive
4230 @code{.cfi_endproc} is used at the end of a function where it closes its
4231 unwind entry previously opened by
4232 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4234 @section @code{.cfi_personality @var{encoding} [, @var{exp}]}
4235 @code{.cfi_personality} defines personality routine and its encoding.
4236 @var{encoding} must be a constant determining how the personality
4237 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4238 argument is not present, otherwise second argument should be
4239 a constant or a symbol name. When using indirect encodings,
4240 the symbol provided should be the location where personality
4241 can be loaded from, not the personality routine itself.
4242 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4243 no personality routine.
4245 @section @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4246 @code{.cfi_lsda} defines LSDA and its encoding.
4247 @var{encoding} must be a constant determining how the LSDA
4248 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4249 argument is not present, otherwise second argument should be a constant
4250 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4253 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4254 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4255 address from @var{register} and add @var{offset} to it}.
4257 @section @code{.cfi_def_cfa_register @var{register}}
4258 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4259 now on @var{register} will be used instead of the old one. Offset
4262 @section @code{.cfi_def_cfa_offset @var{offset}}
4263 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4264 remains the same, but @var{offset} is new. Note that it is the
4265 absolute offset that will be added to a defined register to compute
4268 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4269 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4270 value that is added/substracted from the previous offset.
4272 @section @code{.cfi_offset @var{register}, @var{offset}}
4273 Previous value of @var{register} is saved at offset @var{offset} from
4276 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4277 Previous value of @var{register} is saved at offset @var{offset} from
4278 the current CFA register. This is transformed to @code{.cfi_offset}
4279 using the known displacement of the CFA register from the CFA.
4280 This is often easier to use, because the number will match the
4281 code it's annotating.
4283 @section @code{.cfi_register @var{register1}, @var{register2}}
4284 Previous value of @var{register1} is saved in register @var{register2}.
4286 @section @code{.cfi_restore @var{register}}
4287 @code{.cfi_restore} says that the rule for @var{register} is now the
4288 same as it was at the beginning of the function, after all initial
4289 instruction added by @code{.cfi_startproc} were executed.
4291 @section @code{.cfi_undefined @var{register}}
4292 From now on the previous value of @var{register} can't be restored anymore.
4294 @section @code{.cfi_same_value @var{register}}
4295 Current value of @var{register} is the same like in the previous frame,
4296 i.e. no restoration needed.
4298 @section @code{.cfi_remember_state},
4299 First save all current rules for all registers by @code{.cfi_remember_state},
4300 then totally screw them up by subsequent @code{.cfi_*} directives and when
4301 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4302 the previous saved state.
4304 @section @code{.cfi_return_column @var{register}}
4305 Change return column @var{register}, i.e. the return address is either
4306 directly in @var{register} or can be accessed by rules for @var{register}.
4308 @section @code{.cfi_signal_frame}
4309 Mark current function as signal trampoline.
4311 @section @code{.cfi_window_save}
4312 SPARC register window has been saved.
4314 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4315 Allows the user to add arbitrary bytes to the unwind info. One
4316 might use this to add OS-specific CFI opcodes, or generic CFI
4317 opcodes that GAS does not yet support.
4319 @section @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4320 The current value of @var{register} is @var{label}. The value of @var{label}
4321 will be encoded in the output file according to @var{encoding}; see the
4322 description of @code{.cfi_personality} for details on this encoding.
4324 The usefulness of equating a register to a fixed label is probably
4325 limited to the return address register. Here, it can be useful to
4326 mark a code segment that has only one return address which is reached
4327 by a direct branch and no copy of the return address exists in memory
4328 or another register.
4331 @section @code{.comm @var{symbol} , @var{length} }
4333 @cindex @code{comm} directive
4334 @cindex symbol, common
4335 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4336 common symbol in one object file may be merged with a defined or common symbol
4337 of the same name in another object file. If @code{@value{LD}} does not see a
4338 definition for the symbol--just one or more common symbols--then it will
4339 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4340 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4341 the same name, and they do not all have the same size, it will allocate space
4342 using the largest size.
4345 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4346 an optional third argument. This is the desired alignment of the symbol,
4347 specified for ELF as a byte boundary (for example, an alignment of 16 means
4348 that the least significant 4 bits of the address should be zero), and for PE
4349 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4350 boundary). The alignment must be an absolute expression, and it must be a
4351 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4352 common symbol, it will use the alignment when placing the symbol. If no
4353 alignment is specified, @command{@value{AS}} will set the alignment to the
4354 largest power of two less than or equal to the size of the symbol, up to a
4355 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4356 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4357 @samp{--section-alignment} option; image file sections in PE are aligned to
4358 multiples of 4096, which is far too large an alignment for ordinary variables.
4359 It is rather the default alignment for (non-debug) sections within object
4360 (@samp{*.o}) files, which are less strictly aligned.}.
4364 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4365 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4369 @section @code{.data @var{subsection}}
4371 @cindex @code{data} directive
4372 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4373 end of the data subsection numbered @var{subsection} (which is an
4374 absolute expression). If @var{subsection} is omitted, it defaults
4379 @section @code{.def @var{name}}
4381 @cindex @code{def} directive
4382 @cindex COFF symbols, debugging
4383 @cindex debugging COFF symbols
4384 Begin defining debugging information for a symbol @var{name}; the
4385 definition extends until the @code{.endef} directive is encountered.
4388 This directive is only observed when @command{@value{AS}} is configured for COFF
4389 format output; when producing @code{b.out}, @samp{.def} is recognized,
4396 @section @code{.desc @var{symbol}, @var{abs-expression}}
4398 @cindex @code{desc} directive
4399 @cindex COFF symbol descriptor
4400 @cindex symbol descriptor, COFF
4401 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4402 to the low 16 bits of an absolute expression.
4405 The @samp{.desc} directive is not available when @command{@value{AS}} is
4406 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4407 object format. For the sake of compatibility, @command{@value{AS}} accepts
4408 it, but produces no output, when configured for COFF.
4414 @section @code{.dim}
4416 @cindex @code{dim} directive
4417 @cindex COFF auxiliary symbol information
4418 @cindex auxiliary symbol information, COFF
4419 This directive is generated by compilers to include auxiliary debugging
4420 information in the symbol table. It is only permitted inside
4421 @code{.def}/@code{.endef} pairs.
4424 @samp{.dim} is only meaningful when generating COFF format output; when
4425 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4431 @section @code{.double @var{flonums}}
4433 @cindex @code{double} directive
4434 @cindex floating point numbers (double)
4435 @code{.double} expects zero or more flonums, separated by commas. It
4436 assembles floating point numbers.
4438 The exact kind of floating point numbers emitted depends on how
4439 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4443 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4444 in @sc{ieee} format.
4449 @section @code{.eject}
4451 @cindex @code{eject} directive
4452 @cindex new page, in listings
4453 @cindex page, in listings
4454 @cindex listing control: new page
4455 Force a page break at this point, when generating assembly listings.
4458 @section @code{.else}
4460 @cindex @code{else} directive
4461 @code{.else} is part of the @command{@value{AS}} support for conditional
4462 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4463 of code to be assembled if the condition for the preceding @code{.if}
4467 @section @code{.elseif}
4469 @cindex @code{elseif} directive
4470 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4471 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4472 @code{.if} block that would otherwise fill the entire @code{.else} section.
4475 @section @code{.end}
4477 @cindex @code{end} directive
4478 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4479 process anything in the file past the @code{.end} directive.
4483 @section @code{.endef}
4485 @cindex @code{endef} directive
4486 This directive flags the end of a symbol definition begun with
4490 @samp{.endef} is only meaningful when generating COFF format output; if
4491 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4492 directive but ignores it.
4497 @section @code{.endfunc}
4498 @cindex @code{endfunc} directive
4499 @code{.endfunc} marks the end of a function specified with @code{.func}.
4502 @section @code{.endif}
4504 @cindex @code{endif} directive
4505 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4506 it marks the end of a block of code that is only assembled
4507 conditionally. @xref{If,,@code{.if}}.
4510 @section @code{.equ @var{symbol}, @var{expression}}
4512 @cindex @code{equ} directive
4513 @cindex assigning values to symbols
4514 @cindex symbols, assigning values to
4515 This directive sets the value of @var{symbol} to @var{expression}.
4516 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4519 The syntax for @code{equ} on the HPPA is
4520 @samp{@var{symbol} .equ @var{expression}}.
4524 The syntax for @code{equ} on the Z80 is
4525 @samp{@var{symbol} equ @var{expression}}.
4526 On the Z80 it is an eror if @var{symbol} is already defined,
4527 but the symbol is not protected from later redefinition.
4528 Compare @ref{Equiv}.
4532 @section @code{.equiv @var{symbol}, @var{expression}}
4533 @cindex @code{equiv} directive
4534 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4535 the assembler will signal an error if @var{symbol} is already defined. Note a
4536 symbol which has been referenced but not actually defined is considered to be
4539 Except for the contents of the error message, this is roughly equivalent to
4546 plus it protects the symbol from later redefinition.
4549 @section @code{.eqv @var{symbol}, @var{expression}}
4550 @cindex @code{eqv} directive
4551 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
4552 evaluate the expression or any part of it immediately. Instead each time
4553 the resulting symbol is used in an expression, a snapshot of its current
4557 @section @code{.err}
4558 @cindex @code{err} directive
4559 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4560 message and, unless the @option{-Z} option was used, it will not generate an
4561 object file. This can be used to signal an error in conditionally compiled code.
4564 @section @code{.error "@var{string}"}
4565 @cindex error directive
4567 Similarly to @code{.err}, this directive emits an error, but you can specify a
4568 string that will be emitted as the error message. If you don't specify the
4569 message, it defaults to @code{".error directive invoked in source file"}.
4570 @xref{Errors, ,Error and Warning Messages}.
4573 .error "This code has not been assembled and tested."
4577 @section @code{.exitm}
4578 Exit early from the current macro definition. @xref{Macro}.
4581 @section @code{.extern}
4583 @cindex @code{extern} directive
4584 @code{.extern} is accepted in the source program---for compatibility
4585 with other assemblers---but it is ignored. @command{@value{AS}} treats
4586 all undefined symbols as external.
4589 @section @code{.fail @var{expression}}
4591 @cindex @code{fail} directive
4592 Generates an error or a warning. If the value of the @var{expression} is 500
4593 or more, @command{@value{AS}} will print a warning message. If the value is less
4594 than 500, @command{@value{AS}} will print an error message. The message will
4595 include the value of @var{expression}. This can occasionally be useful inside
4596 complex nested macros or conditional assembly.
4599 @section @code{.file}
4600 @cindex @code{file} directive
4602 @ifclear no-file-dir
4603 There are two different versions of the @code{.file} directive. Targets
4604 that support DWARF2 line number information use the DWARF2 version of
4605 @code{.file}. Other targets use the default version.
4607 @subheading Default Version
4609 @cindex logical file name
4610 @cindex file name, logical
4611 This version of the @code{.file} directive tells @command{@value{AS}} that we
4612 are about to start a new logical file. The syntax is:
4618 @var{string} is the new file name. In general, the filename is
4619 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4620 to specify an empty file name, you must give the quotes--@code{""}. This
4621 statement may go away in future: it is only recognized to be compatible with
4622 old @command{@value{AS}} programs.
4624 @subheading DWARF2 Version
4627 When emitting DWARF2 line number information, @code{.file} assigns filenames
4628 to the @code{.debug_line} file name table. The syntax is:
4631 .file @var{fileno} @var{filename}
4634 The @var{fileno} operand should be a unique positive integer to use as the
4635 index of the entry in the table. The @var{filename} operand is a C string
4638 The detail of filename indices is exposed to the user because the filename
4639 table is shared with the @code{.debug_info} section of the DWARF2 debugging
4640 information, and thus the user must know the exact indices that table
4644 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4646 @cindex @code{fill} directive
4647 @cindex writing patterns in memory
4648 @cindex patterns, writing in memory
4649 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4650 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4651 may be zero or more. @var{Size} may be zero or more, but if it is
4652 more than 8, then it is deemed to have the value 8, compatible with
4653 other people's assemblers. The contents of each @var{repeat} bytes
4654 is taken from an 8-byte number. The highest order 4 bytes are
4655 zero. The lowest order 4 bytes are @var{value} rendered in the
4656 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4657 Each @var{size} bytes in a repetition is taken from the lowest order
4658 @var{size} bytes of this number. Again, this bizarre behavior is
4659 compatible with other people's assemblers.
4661 @var{size} and @var{value} are optional.
4662 If the second comma and @var{value} are absent, @var{value} is
4663 assumed zero. If the first comma and following tokens are absent,
4664 @var{size} is assumed to be 1.
4667 @section @code{.float @var{flonums}}
4669 @cindex floating point numbers (single)
4670 @cindex @code{float} directive
4671 This directive assembles zero or more flonums, separated by commas. It
4672 has the same effect as @code{.single}.
4674 The exact kind of floating point numbers emitted depends on how
4675 @command{@value{AS}} is configured.
4676 @xref{Machine Dependencies}.
4680 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4681 in @sc{ieee} format.
4686 @section @code{.func @var{name}[,@var{label}]}
4687 @cindex @code{func} directive
4688 @code{.func} emits debugging information to denote function @var{name}, and
4689 is ignored unless the file is assembled with debugging enabled.
4690 Only @samp{--gstabs[+]} is currently supported.
4691 @var{label} is the entry point of the function and if omitted @var{name}
4692 prepended with the @samp{leading char} is used.
4693 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4694 All functions are currently defined to have @code{void} return type.
4695 The function must be terminated with @code{.endfunc}.
4698 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4700 @cindex @code{global} directive
4701 @cindex symbol, making visible to linker
4702 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4703 @var{symbol} in your partial program, its value is made available to
4704 other partial programs that are linked with it. Otherwise,
4705 @var{symbol} takes its attributes from a symbol of the same name
4706 from another file linked into the same program.
4708 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4709 compatibility with other assemblers.
4712 On the HPPA, @code{.global} is not always enough to make it accessible to other
4713 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4714 @xref{HPPA Directives, ,HPPA Assembler Directives}.
4719 @section @code{.gnu_attribute @var{tag},@var{value}}
4720 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
4723 @section @code{.hidden @var{names}}
4725 @cindex @code{hidden} directive
4727 This is one of the ELF visibility directives. The other two are
4728 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4729 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4731 This directive overrides the named symbols default visibility (which is set by
4732 their binding: local, global or weak). The directive sets the visibility to
4733 @code{hidden} which means that the symbols are not visible to other components.
4734 Such symbols are always considered to be @code{protected} as well.
4738 @section @code{.hword @var{expressions}}
4740 @cindex @code{hword} directive
4741 @cindex integers, 16-bit
4742 @cindex numbers, 16-bit
4743 @cindex sixteen bit integers
4744 This expects zero or more @var{expressions}, and emits
4745 a 16 bit number for each.
4748 This directive is a synonym for @samp{.short}; depending on the target
4749 architecture, it may also be a synonym for @samp{.word}.
4753 This directive is a synonym for @samp{.short}.
4756 This directive is a synonym for both @samp{.short} and @samp{.word}.
4761 @section @code{.ident}
4763 @cindex @code{ident} directive
4765 This directive is used by some assemblers to place tags in object files. The
4766 behavior of this directive varies depending on the target. When using the
4767 a.out object file format, @command{@value{AS}} simply accepts the directive for
4768 source-file compatibility with existing assemblers, but does not emit anything
4769 for it. When using COFF, comments are emitted to the @code{.comment} or
4770 @code{.rdata} section, depending on the target. When using ELF, comments are
4771 emitted to the @code{.comment} section.
4774 @section @code{.if @var{absolute expression}}
4776 @cindex conditional assembly
4777 @cindex @code{if} directive
4778 @code{.if} marks the beginning of a section of code which is only
4779 considered part of the source program being assembled if the argument
4780 (which must be an @var{absolute expression}) is non-zero. The end of
4781 the conditional section of code must be marked by @code{.endif}
4782 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4783 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4784 If you have several conditions to check, @code{.elseif} may be used to avoid
4785 nesting blocks if/else within each subsequent @code{.else} block.
4787 The following variants of @code{.if} are also supported:
4789 @cindex @code{ifdef} directive
4790 @item .ifdef @var{symbol}
4791 Assembles the following section of code if the specified @var{symbol}
4792 has been defined. Note a symbol which has been referenced but not yet defined
4793 is considered to be undefined.
4795 @cindex @code{ifb} directive
4796 @item .ifb @var{text}
4797 Assembles the following section of code if the operand is blank (empty).
4799 @cindex @code{ifc} directive
4800 @item .ifc @var{string1},@var{string2}
4801 Assembles the following section of code if the two strings are the same. The
4802 strings may be optionally quoted with single quotes. If they are not quoted,
4803 the first string stops at the first comma, and the second string stops at the
4804 end of the line. Strings which contain whitespace should be quoted. The
4805 string comparison is case sensitive.
4807 @cindex @code{ifeq} directive
4808 @item .ifeq @var{absolute expression}
4809 Assembles the following section of code if the argument is zero.
4811 @cindex @code{ifeqs} directive
4812 @item .ifeqs @var{string1},@var{string2}
4813 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4815 @cindex @code{ifge} directive
4816 @item .ifge @var{absolute expression}
4817 Assembles the following section of code if the argument is greater than or
4820 @cindex @code{ifgt} directive
4821 @item .ifgt @var{absolute expression}
4822 Assembles the following section of code if the argument is greater than zero.
4824 @cindex @code{ifle} directive
4825 @item .ifle @var{absolute expression}
4826 Assembles the following section of code if the argument is less than or equal
4829 @cindex @code{iflt} directive
4830 @item .iflt @var{absolute expression}
4831 Assembles the following section of code if the argument is less than zero.
4833 @cindex @code{ifnb} directive
4834 @item .ifnb @var{text}
4835 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
4836 following section of code if the operand is non-blank (non-empty).
4838 @cindex @code{ifnc} directive
4839 @item .ifnc @var{string1},@var{string2}.
4840 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4841 following section of code if the two strings are not the same.
4843 @cindex @code{ifndef} directive
4844 @cindex @code{ifnotdef} directive
4845 @item .ifndef @var{symbol}
4846 @itemx .ifnotdef @var{symbol}
4847 Assembles the following section of code if the specified @var{symbol}
4848 has not been defined. Both spelling variants are equivalent. Note a symbol
4849 which has been referenced but not yet defined is considered to be undefined.
4851 @cindex @code{ifne} directive
4852 @item .ifne @var{absolute expression}
4853 Assembles the following section of code if the argument is not equal to zero
4854 (in other words, this is equivalent to @code{.if}).
4856 @cindex @code{ifnes} directive
4857 @item .ifnes @var{string1},@var{string2}
4858 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4859 following section of code if the two strings are not the same.
4863 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4865 @cindex @code{incbin} directive
4866 @cindex binary files, including
4867 The @code{incbin} directive includes @var{file} verbatim at the current
4868 location. You can control the search paths used with the @samp{-I} command-line
4869 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4872 The @var{skip} argument skips a number of bytes from the start of the
4873 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4874 read. Note that the data is not aligned in any way, so it is the user's
4875 responsibility to make sure that proper alignment is provided both before and
4876 after the @code{incbin} directive.
4879 @section @code{.include "@var{file}"}
4881 @cindex @code{include} directive
4882 @cindex supporting files, including
4883 @cindex files, including
4884 This directive provides a way to include supporting files at specified
4885 points in your source program. The code from @var{file} is assembled as
4886 if it followed the point of the @code{.include}; when the end of the
4887 included file is reached, assembly of the original file continues. You
4888 can control the search paths used with the @samp{-I} command-line option
4889 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4893 @section @code{.int @var{expressions}}
4895 @cindex @code{int} directive
4896 @cindex integers, 32-bit
4897 Expect zero or more @var{expressions}, of any section, separated by commas.
4898 For each expression, emit a number that, at run time, is the value of that
4899 expression. The byte order and bit size of the number depends on what kind
4900 of target the assembly is for.
4904 On most forms of the H8/300, @code{.int} emits 16-bit
4905 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
4912 @section @code{.internal @var{names}}
4914 @cindex @code{internal} directive
4916 This is one of the ELF visibility directives. The other two are
4917 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
4918 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4920 This directive overrides the named symbols default visibility (which is set by
4921 their binding: local, global or weak). The directive sets the visibility to
4922 @code{internal} which means that the symbols are considered to be @code{hidden}
4923 (i.e., not visible to other components), and that some extra, processor specific
4924 processing must also be performed upon the symbols as well.
4928 @section @code{.irp @var{symbol},@var{values}}@dots{}
4930 @cindex @code{irp} directive
4931 Evaluate a sequence of statements assigning different values to @var{symbol}.
4932 The sequence of statements starts at the @code{.irp} directive, and is
4933 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
4934 set to @var{value}, and the sequence of statements is assembled. If no
4935 @var{value} is listed, the sequence of statements is assembled once, with
4936 @var{symbol} set to the null string. To refer to @var{symbol} within the
4937 sequence of statements, use @var{\symbol}.
4939 For example, assembling
4947 is equivalent to assembling
4955 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
4958 @section @code{.irpc @var{symbol},@var{values}}@dots{}
4960 @cindex @code{irpc} directive
4961 Evaluate a sequence of statements assigning different values to @var{symbol}.
4962 The sequence of statements starts at the @code{.irpc} directive, and is
4963 terminated by an @code{.endr} directive. For each character in @var{value},
4964 @var{symbol} is set to the character, and the sequence of statements is
4965 assembled. If no @var{value} is listed, the sequence of statements is
4966 assembled once, with @var{symbol} set to the null string. To refer to
4967 @var{symbol} within the sequence of statements, use @var{\symbol}.
4969 For example, assembling
4977 is equivalent to assembling
4985 For some caveats with the spelling of @var{symbol}, see also the discussion
4989 @section @code{.lcomm @var{symbol} , @var{length}}
4991 @cindex @code{lcomm} directive
4992 @cindex local common symbols
4993 @cindex symbols, local common
4994 Reserve @var{length} (an absolute expression) bytes for a local common
4995 denoted by @var{symbol}. The section and value of @var{symbol} are
4996 those of the new local common. The addresses are allocated in the bss
4997 section, so that at run-time the bytes start off zeroed. @var{Symbol}
4998 is not declared global (@pxref{Global,,@code{.global}}), so is normally
4999 not visible to @code{@value{LD}}.
5002 Some targets permit a third argument to be used with @code{.lcomm}. This
5003 argument specifies the desired alignment of the symbol in the bss section.
5007 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5008 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5012 @section @code{.lflags}
5014 @cindex @code{lflags} directive (ignored)
5015 @command{@value{AS}} accepts this directive, for compatibility with other
5016 assemblers, but ignores it.
5018 @ifclear no-line-dir
5020 @section @code{.line @var{line-number}}
5022 @cindex @code{line} directive
5023 @cindex logical line number
5025 Change the logical line number. @var{line-number} must be an absolute
5026 expression. The next line has that logical line number. Therefore any other
5027 statements on the current line (after a statement separator character) are
5028 reported as on logical line number @var{line-number} @minus{} 1. One day
5029 @command{@value{AS}} will no longer support this directive: it is recognized only
5030 for compatibility with existing assembler programs.
5033 Even though this is a directive associated with the @code{a.out} or
5034 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5035 when producing COFF output, and treats @samp{.line} as though it
5036 were the COFF @samp{.ln} @emph{if} it is found outside a
5037 @code{.def}/@code{.endef} pair.
5039 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5040 used by compilers to generate auxiliary symbol information for
5045 @section @code{.linkonce [@var{type}]}
5047 @cindex @code{linkonce} directive
5048 @cindex common sections
5049 Mark the current section so that the linker only includes a single copy of it.
5050 This may be used to include the same section in several different object files,
5051 but ensure that the linker will only include it once in the final output file.
5052 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5053 Duplicate sections are detected based on the section name, so it should be
5056 This directive is only supported by a few object file formats; as of this
5057 writing, the only object file format which supports it is the Portable
5058 Executable format used on Windows NT.
5060 The @var{type} argument is optional. If specified, it must be one of the
5061 following strings. For example:
5065 Not all types may be supported on all object file formats.
5069 Silently discard duplicate sections. This is the default.
5072 Warn if there are duplicate sections, but still keep only one copy.
5075 Warn if any of the duplicates have different sizes.
5078 Warn if any of the duplicates do not have exactly the same contents.
5082 @section @code{.list}
5084 @cindex @code{list} directive
5085 @cindex listing control, turning on
5086 Control (in conjunction with the @code{.nolist} directive) whether or
5087 not assembly listings are generated. These two directives maintain an
5088 internal counter (which is zero initially). @code{.list} increments the
5089 counter, and @code{.nolist} decrements it. Assembly listings are
5090 generated whenever the counter is greater than zero.
5092 By default, listings are disabled. When you enable them (with the
5093 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5094 the initial value of the listing counter is one.
5097 @section @code{.ln @var{line-number}}
5099 @cindex @code{ln} directive
5100 @ifclear no-line-dir
5101 @samp{.ln} is a synonym for @samp{.line}.
5104 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5105 must be an absolute expression. The next line has that logical
5106 line number, so any other statements on the current line (after a
5107 statement separator character @code{;}) are reported as on logical
5108 line number @var{line-number} @minus{} 1.
5111 This directive is accepted, but ignored, when @command{@value{AS}} is
5112 configured for @code{b.out}; its effect is only associated with COFF
5118 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5119 @cindex @code{loc} directive
5120 When emitting DWARF2 line number information,
5121 the @code{.loc} directive will add a row to the @code{.debug_line} line
5122 number matrix corresponding to the immediately following assembly
5123 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5124 arguments will be applied to the @code{.debug_line} state machine before
5127 The @var{options} are a sequence of the following tokens in any order:
5131 This option will set the @code{basic_block} register in the
5132 @code{.debug_line} state machine to @code{true}.
5135 This option will set the @code{prologue_end} register in the
5136 @code{.debug_line} state machine to @code{true}.
5138 @item epilogue_begin
5139 This option will set the @code{epilogue_begin} register in the
5140 @code{.debug_line} state machine to @code{true}.
5142 @item is_stmt @var{value}
5143 This option will set the @code{is_stmt} register in the
5144 @code{.debug_line} state machine to @code{value}, which must be
5147 @item isa @var{value}
5148 This directive will set the @code{isa} register in the @code{.debug_line}
5149 state machine to @var{value}, which must be an unsigned integer.
5151 @item discriminator @var{value}
5152 This directive will set the @code{discriminator} register in the @code{.debug_line}
5153 state machine to @var{value}, which must be an unsigned integer.
5157 @node Loc_mark_labels
5158 @section @code{.loc_mark_labels @var{enable}}
5159 @cindex @code{loc_mark_labels} directive
5160 When emitting DWARF2 line number information,
5161 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5162 to the @code{.debug_line} line number matrix with the @code{basic_block}
5163 register in the state machine set whenever a code label is seen.
5164 The @var{enable} argument should be either 1 or 0, to enable or disable
5165 this function respectively.
5169 @section @code{.local @var{names}}
5171 @cindex @code{local} directive
5172 This directive, which is available for ELF targets, marks each symbol in
5173 the comma-separated list of @code{names} as a local symbol so that it
5174 will not be externally visible. If the symbols do not already exist,
5175 they will be created.
5177 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5178 accept an alignment argument, which is the case for most ELF targets,
5179 the @code{.local} directive can be used in combination with @code{.comm}
5180 (@pxref{Comm}) to define aligned local common data.
5184 @section @code{.long @var{expressions}}
5186 @cindex @code{long} directive
5187 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5190 @c no one seems to know what this is for or whether this description is
5191 @c what it really ought to do
5193 @section @code{.lsym @var{symbol}, @var{expression}}
5195 @cindex @code{lsym} directive
5196 @cindex symbol, not referenced in assembly
5197 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5198 the hash table, ensuring it cannot be referenced by name during the
5199 rest of the assembly. This sets the attributes of the symbol to be
5200 the same as the expression value:
5202 @var{other} = @var{descriptor} = 0
5203 @var{type} = @r{(section of @var{expression})}
5204 @var{value} = @var{expression}
5207 The new symbol is not flagged as external.
5211 @section @code{.macro}
5214 The commands @code{.macro} and @code{.endm} allow you to define macros that
5215 generate assembly output. For example, this definition specifies a macro
5216 @code{sum} that puts a sequence of numbers into memory:
5219 .macro sum from=0, to=5
5228 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5240 @item .macro @var{macname}
5241 @itemx .macro @var{macname} @var{macargs} @dots{}
5242 @cindex @code{macro} directive
5243 Begin the definition of a macro called @var{macname}. If your macro
5244 definition requires arguments, specify their names after the macro name,
5245 separated by commas or spaces. You can qualify the macro argument to
5246 indicate whether all invocations must specify a non-blank value (through
5247 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5248 (through @samp{:@code{vararg}}). You can supply a default value for any
5249 macro argument by following the name with @samp{=@var{deflt}}. You
5250 cannot define two macros with the same @var{macname} unless it has been
5251 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5252 definitions. For example, these are all valid @code{.macro} statements:
5256 Begin the definition of a macro called @code{comm}, which takes no
5259 @item .macro plus1 p, p1
5260 @itemx .macro plus1 p p1
5261 Either statement begins the definition of a macro called @code{plus1},
5262 which takes two arguments; within the macro definition, write
5263 @samp{\p} or @samp{\p1} to evaluate the arguments.
5265 @item .macro reserve_str p1=0 p2
5266 Begin the definition of a macro called @code{reserve_str}, with two
5267 arguments. The first argument has a default value, but not the second.
5268 After the definition is complete, you can call the macro either as
5269 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5270 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5271 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5272 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5274 @item .macro m p1:req, p2=0, p3:vararg
5275 Begin the definition of a macro called @code{m}, with at least three
5276 arguments. The first argument must always have a value specified, but
5277 not the second, which instead has a default value. The third formal
5278 will get assigned all remaining arguments specified at invocation time.
5280 When you call a macro, you can specify the argument values either by
5281 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5282 @samp{sum to=17, from=9}.
5286 Note that since each of the @var{macargs} can be an identifier exactly
5287 as any other one permitted by the target architecture, there may be
5288 occasional problems if the target hand-crafts special meanings to certain
5289 characters when they occur in a special position. For example, if the colon
5290 (@code{:}) is generally permitted to be part of a symbol name, but the
5291 architecture specific code special-cases it when occurring as the final
5292 character of a symbol (to denote a label), then the macro parameter
5293 replacement code will have no way of knowing that and consider the whole
5294 construct (including the colon) an identifier, and check only this
5295 identifier for being the subject to parameter substitution. So for example
5296 this macro definition:
5304 might not work as expected. Invoking @samp{label foo} might not create a label
5305 called @samp{foo} but instead just insert the text @samp{\l:} into the
5306 assembler source, probably generating an error about an unrecognised
5309 Similarly problems might occur with the period character (@samp{.})
5310 which is often allowed inside opcode names (and hence identifier names). So
5311 for example constructing a macro to build an opcode from a base name and a
5312 length specifier like this:
5315 .macro opcode base length
5320 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5321 instruction but instead generate some kind of error as the assembler tries to
5322 interpret the text @samp{\base.\length}.
5324 There are several possible ways around this problem:
5327 @item Insert white space
5328 If it is possible to use white space characters then this is the simplest
5337 @item Use @samp{\()}
5338 The string @samp{\()} can be used to separate the end of a macro argument from
5339 the following text. eg:
5342 .macro opcode base length
5347 @item Use the alternate macro syntax mode
5348 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5349 used as a separator. eg:
5359 Note: this problem of correctly identifying string parameters to pseudo ops
5360 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5361 and @code{.irpc} (@pxref{Irpc}) as well.
5364 @cindex @code{endm} directive
5365 Mark the end of a macro definition.
5368 @cindex @code{exitm} directive
5369 Exit early from the current macro definition.
5371 @cindex number of macros executed
5372 @cindex macros, count executed
5374 @command{@value{AS}} maintains a counter of how many macros it has
5375 executed in this pseudo-variable; you can copy that number to your
5376 output with @samp{\@@}, but @emph{only within a macro definition}.
5378 @item LOCAL @var{name} [ , @dots{} ]
5379 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5380 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5381 @xref{Altmacro,,@code{.altmacro}}.
5385 @section @code{.mri @var{val}}
5387 @cindex @code{mri} directive
5388 @cindex MRI mode, temporarily
5389 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5390 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5391 affects code assembled until the next @code{.mri} directive, or until the end
5392 of the file. @xref{M, MRI mode, MRI mode}.
5395 @section @code{.noaltmacro}
5396 Disable alternate macro mode. @xref{Altmacro}.
5399 @section @code{.nolist}
5401 @cindex @code{nolist} directive
5402 @cindex listing control, turning off
5403 Control (in conjunction with the @code{.list} directive) whether or
5404 not assembly listings are generated. These two directives maintain an
5405 internal counter (which is zero initially). @code{.list} increments the
5406 counter, and @code{.nolist} decrements it. Assembly listings are
5407 generated whenever the counter is greater than zero.
5410 @section @code{.octa @var{bignums}}
5412 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5413 @cindex @code{octa} directive
5414 @cindex integer, 16-byte
5415 @cindex sixteen byte integer
5416 This directive expects zero or more bignums, separated by commas. For each
5417 bignum, it emits a 16-byte integer.
5419 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5420 hence @emph{octa}-word for 16 bytes.
5423 @section @code{.org @var{new-lc} , @var{fill}}
5425 @cindex @code{org} directive
5426 @cindex location counter, advancing
5427 @cindex advancing location counter
5428 @cindex current address, advancing
5429 Advance the location counter of the current section to
5430 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5431 expression with the same section as the current subsection. That is,
5432 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5433 wrong section, the @code{.org} directive is ignored. To be compatible
5434 with former assemblers, if the section of @var{new-lc} is absolute,
5435 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5436 is the same as the current subsection.
5438 @code{.org} may only increase the location counter, or leave it
5439 unchanged; you cannot use @code{.org} to move the location counter
5442 @c double negative used below "not undefined" because this is a specific
5443 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5444 @c section. doc@cygnus.com 18feb91
5445 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5446 may not be undefined. If you really detest this restriction we eagerly await
5447 a chance to share your improved assembler.
5449 Beware that the origin is relative to the start of the section, not
5450 to the start of the subsection. This is compatible with other
5451 people's assemblers.
5453 When the location counter (of the current subsection) is advanced, the
5454 intervening bytes are filled with @var{fill} which should be an
5455 absolute expression. If the comma and @var{fill} are omitted,
5456 @var{fill} defaults to zero.
5459 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5461 @cindex padding the location counter given a power of two
5462 @cindex @code{p2align} directive
5463 Pad the location counter (in the current subsection) to a particular
5464 storage boundary. The first expression (which must be absolute) is the
5465 number of low-order zero bits the location counter must have after
5466 advancement. For example @samp{.p2align 3} advances the location
5467 counter until it a multiple of 8. If the location counter is already a
5468 multiple of 8, no change is needed.
5470 The second expression (also absolute) gives the fill value to be stored in the
5471 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5472 padding bytes are normally zero. However, on some systems, if the section is
5473 marked as containing code and the fill value is omitted, the space is filled
5474 with no-op instructions.
5476 The third expression is also absolute, and is also optional. If it is present,
5477 it is the maximum number of bytes that should be skipped by this alignment
5478 directive. If doing the alignment would require skipping more bytes than the
5479 specified maximum, then the alignment is not done at all. You can omit the
5480 fill value (the second argument) entirely by simply using two commas after the
5481 required alignment; this can be useful if you want the alignment to be filled
5482 with no-op instructions when appropriate.
5484 @cindex @code{p2alignw} directive
5485 @cindex @code{p2alignl} directive
5486 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5487 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5488 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5489 fill pattern as a four byte longword value. For example, @code{.p2alignw
5490 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5491 filled in with the value 0x368d (the exact placement of the bytes depends upon
5492 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5497 @section @code{.popsection}
5499 @cindex @code{popsection} directive
5500 @cindex Section Stack
5501 This is one of the ELF section stack manipulation directives. The others are
5502 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5503 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5506 This directive replaces the current section (and subsection) with the top
5507 section (and subsection) on the section stack. This section is popped off the
5513 @section @code{.previous}
5515 @cindex @code{previous} directive
5516 @cindex Section Stack
5517 This is one of the ELF section stack manipulation directives. The others are
5518 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5519 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5520 (@pxref{PopSection}).
5522 This directive swaps the current section (and subsection) with most recently
5523 referenced section/subsection pair prior to this one. Multiple
5524 @code{.previous} directives in a row will flip between two sections (and their
5525 subsections). For example:
5537 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
5543 # Now in section A subsection 1
5547 # Now in section B subsection 0
5550 # Now in section B subsection 1
5553 # Now in section B subsection 0
5557 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
5558 section B and 0x9abc into subsection 1 of section B.
5560 In terms of the section stack, this directive swaps the current section with
5561 the top section on the section stack.
5565 @section @code{.print @var{string}}
5567 @cindex @code{print} directive
5568 @command{@value{AS}} will print @var{string} on the standard output during
5569 assembly. You must put @var{string} in double quotes.
5573 @section @code{.protected @var{names}}
5575 @cindex @code{protected} directive
5577 This is one of the ELF visibility directives. The other two are
5578 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5580 This directive overrides the named symbols default visibility (which is set by
5581 their binding: local, global or weak). The directive sets the visibility to
5582 @code{protected} which means that any references to the symbols from within the
5583 components that defines them must be resolved to the definition in that
5584 component, even if a definition in another component would normally preempt
5589 @section @code{.psize @var{lines} , @var{columns}}
5591 @cindex @code{psize} directive
5592 @cindex listing control: paper size
5593 @cindex paper size, for listings
5594 Use this directive to declare the number of lines---and, optionally, the
5595 number of columns---to use for each page, when generating listings.
5597 If you do not use @code{.psize}, listings use a default line-count
5598 of 60. You may omit the comma and @var{columns} specification; the
5599 default width is 200 columns.
5601 @command{@value{AS}} generates formfeeds whenever the specified number of
5602 lines is exceeded (or whenever you explicitly request one, using
5605 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5606 those explicitly specified with @code{.eject}.
5609 @section @code{.purgem @var{name}}
5611 @cindex @code{purgem} directive
5612 Undefine the macro @var{name}, so that later uses of the string will not be
5613 expanded. @xref{Macro}.
5617 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
5619 @cindex @code{pushsection} directive
5620 @cindex Section Stack
5621 This is one of the ELF section stack manipulation directives. The others are
5622 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5623 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5626 This directive pushes the current section (and subsection) onto the
5627 top of the section stack, and then replaces the current section and
5628 subsection with @code{name} and @code{subsection}. The optional
5629 @code{flags}, @code{type} and @code{arguments} are treated the same
5630 as in the @code{.section} (@pxref{Section}) directive.
5634 @section @code{.quad @var{bignums}}
5636 @cindex @code{quad} directive
5637 @code{.quad} expects zero or more bignums, separated by commas. For
5638 each bignum, it emits
5640 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5641 warning message; and just takes the lowest order 8 bytes of the bignum.
5642 @cindex eight-byte integer
5643 @cindex integer, 8-byte
5645 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5646 hence @emph{quad}-word for 8 bytes.
5649 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5650 warning message; and just takes the lowest order 16 bytes of the bignum.
5651 @cindex sixteen-byte integer
5652 @cindex integer, 16-byte
5656 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
5658 @cindex @code{reloc} directive
5659 Generate a relocation at @var{offset} of type @var{reloc_name} with value
5660 @var{expression}. If @var{offset} is a number, the relocation is generated in
5661 the current section. If @var{offset} is an expression that resolves to a
5662 symbol plus offset, the relocation is generated in the given symbol's section.
5663 @var{expression}, if present, must resolve to a symbol plus addend or to an
5664 absolute value, but note that not all targets support an addend. e.g. ELF REL
5665 targets such as i386 store an addend in the section contents rather than in the
5666 relocation. This low level interface does not support addends stored in the
5670 @section @code{.rept @var{count}}
5672 @cindex @code{rept} directive
5673 Repeat the sequence of lines between the @code{.rept} directive and the next
5674 @code{.endr} directive @var{count} times.
5676 For example, assembling
5684 is equivalent to assembling
5693 @section @code{.sbttl "@var{subheading}"}
5695 @cindex @code{sbttl} directive
5696 @cindex subtitles for listings
5697 @cindex listing control: subtitle
5698 Use @var{subheading} as the title (third line, immediately after the
5699 title line) when generating assembly listings.
5701 This directive affects subsequent pages, as well as the current page if
5702 it appears within ten lines of the top of a page.
5706 @section @code{.scl @var{class}}
5708 @cindex @code{scl} directive
5709 @cindex symbol storage class (COFF)
5710 @cindex COFF symbol storage class
5711 Set the storage-class value for a symbol. This directive may only be
5712 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5713 whether a symbol is static or external, or it may record further
5714 symbolic debugging information.
5717 The @samp{.scl} directive is primarily associated with COFF output; when
5718 configured to generate @code{b.out} output format, @command{@value{AS}}
5719 accepts this directive but ignores it.
5725 @section @code{.section @var{name}}
5727 @cindex named section
5728 Use the @code{.section} directive to assemble the following code into a section
5731 This directive is only supported for targets that actually support arbitrarily
5732 named sections; on @code{a.out} targets, for example, it is not accepted, even
5733 with a standard @code{a.out} section name.
5737 @c only print the extra heading if both COFF and ELF are set
5738 @subheading COFF Version
5741 @cindex @code{section} directive (COFF version)
5742 For COFF targets, the @code{.section} directive is used in one of the following
5746 .section @var{name}[, "@var{flags}"]
5747 .section @var{name}[, @var{subsection}]
5750 If the optional argument is quoted, it is taken as flags to use for the
5751 section. Each flag is a single character. The following flags are recognized:
5754 bss section (uninitialized data)
5756 section is not loaded
5766 shared section (meaningful for PE targets)
5768 ignored. (For compatibility with the ELF version)
5770 section is not readable (meaningful for PE targets)
5773 If no flags are specified, the default flags depend upon the section name. If
5774 the section name is not recognized, the default will be for the section to be
5775 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
5776 from the section, rather than adding them, so if they are used on their own it
5777 will be as if no flags had been specified at all.
5779 If the optional argument to the @code{.section} directive is not quoted, it is
5780 taken as a subsection number (@pxref{Sub-Sections}).
5785 @c only print the extra heading if both COFF and ELF are set
5786 @subheading ELF Version
5789 @cindex Section Stack
5790 This is one of the ELF section stack manipulation directives. The others are
5791 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
5792 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
5793 @code{.previous} (@pxref{Previous}).
5795 @cindex @code{section} directive (ELF version)
5796 For ELF targets, the @code{.section} directive is used like this:
5799 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
5802 The optional @var{flags} argument is a quoted string which may contain any
5803 combination of the following characters:
5806 section is allocatable
5810 section is executable
5812 section is mergeable
5814 section contains zero terminated strings
5816 section is a member of a section group
5818 section is used for thread-local-storage
5821 The optional @var{type} argument may contain one of the following constants:
5824 section contains data
5826 section does not contain data (i.e., section only occupies space)
5828 section contains data which is used by things other than the program
5830 section contains an array of pointers to init functions
5832 section contains an array of pointers to finish functions
5833 @item @@preinit_array
5834 section contains an array of pointers to pre-init functions
5837 Many targets only support the first three section types.
5839 Note on targets where the @code{@@} character is the start of a comment (eg
5840 ARM) then another character is used instead. For example the ARM port uses the
5843 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
5844 be specified as well as an extra argument---@var{entsize}---like this:
5847 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
5850 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
5851 constants, each @var{entsize} octets long. Sections with both @code{M} and
5852 @code{S} must contain zero terminated strings where each character is
5853 @var{entsize} bytes long. The linker may remove duplicates within sections with
5854 the same name, same entity size and same flags. @var{entsize} must be an
5855 absolute expression. For sections with both @code{M} and @code{S}, a string
5856 which is a suffix of a larger string is considered a duplicate. Thus
5857 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
5858 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
5860 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
5861 be present along with an additional field like this:
5864 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
5867 The @var{GroupName} field specifies the name of the section group to which this
5868 particular section belongs. The optional linkage field can contain:
5871 indicates that only one copy of this section should be retained
5876 Note: if both the @var{M} and @var{G} flags are present then the fields for
5877 the Merge flag should come first, like this:
5880 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
5883 If no flags are specified, the default flags depend upon the section name. If
5884 the section name is not recognized, the default will be for the section to have
5885 none of the above flags: it will not be allocated in memory, nor writable, nor
5886 executable. The section will contain data.
5888 For ELF targets, the assembler supports another type of @code{.section}
5889 directive for compatibility with the Solaris assembler:
5892 .section "@var{name}"[, @var{flags}...]
5895 Note that the section name is quoted. There may be a sequence of comma
5899 section is allocatable
5903 section is executable
5905 section is used for thread local storage
5908 This directive replaces the current section and subsection. See the
5909 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
5910 some examples of how this directive and the other section stack directives
5916 @section @code{.set @var{symbol}, @var{expression}}
5918 @cindex @code{set} directive
5919 @cindex symbol value, setting
5920 Set the value of @var{symbol} to @var{expression}. This
5921 changes @var{symbol}'s value and type to conform to
5922 @var{expression}. If @var{symbol} was flagged as external, it remains
5923 flagged (@pxref{Symbol Attributes}).
5925 You may @code{.set} a symbol many times in the same assembly.
5927 If you @code{.set} a global symbol, the value stored in the object
5928 file is the last value stored into it.
5931 On Z80 @code{set} is a real instruction, use
5932 @samp{@var{symbol} defl @var{expression}} instead.
5936 @section @code{.short @var{expressions}}
5938 @cindex @code{short} directive
5940 @code{.short} is normally the same as @samp{.word}.
5941 @xref{Word,,@code{.word}}.
5943 In some configurations, however, @code{.short} and @code{.word} generate
5944 numbers of different lengths. @xref{Machine Dependencies}.
5948 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
5951 This expects zero or more @var{expressions}, and emits
5952 a 16 bit number for each.
5957 @section @code{.single @var{flonums}}
5959 @cindex @code{single} directive
5960 @cindex floating point numbers (single)
5961 This directive assembles zero or more flonums, separated by commas. It
5962 has the same effect as @code{.float}.
5964 The exact kind of floating point numbers emitted depends on how
5965 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5969 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
5970 numbers in @sc{ieee} format.
5976 @section @code{.size}
5978 This directive is used to set the size associated with a symbol.
5982 @c only print the extra heading if both COFF and ELF are set
5983 @subheading COFF Version
5986 @cindex @code{size} directive (COFF version)
5987 For COFF targets, the @code{.size} directive is only permitted inside
5988 @code{.def}/@code{.endef} pairs. It is used like this:
5991 .size @var{expression}
5995 @samp{.size} is only meaningful when generating COFF format output; when
5996 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6003 @c only print the extra heading if both COFF and ELF are set
6004 @subheading ELF Version
6007 @cindex @code{size} directive (ELF version)
6008 For ELF targets, the @code{.size} directive is used like this:
6011 .size @var{name} , @var{expression}
6014 This directive sets the size associated with a symbol @var{name}.
6015 The size in bytes is computed from @var{expression} which can make use of label
6016 arithmetic. This directive is typically used to set the size of function
6021 @ifclear no-space-dir
6023 @section @code{.skip @var{size} , @var{fill}}
6025 @cindex @code{skip} directive
6026 @cindex filling memory
6027 This directive emits @var{size} bytes, each of value @var{fill}. Both
6028 @var{size} and @var{fill} are absolute expressions. If the comma and
6029 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6034 @section @code{.sleb128 @var{expressions}}
6036 @cindex @code{sleb128} directive
6037 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6038 compact, variable length representation of numbers used by the DWARF
6039 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6041 @ifclear no-space-dir
6043 @section @code{.space @var{size} , @var{fill}}
6045 @cindex @code{space} directive
6046 @cindex filling memory
6047 This directive emits @var{size} bytes, each of value @var{fill}. Both
6048 @var{size} and @var{fill} are absolute expressions. If the comma
6049 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6054 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6055 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6056 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6057 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6065 @section @code{.stabd, .stabn, .stabs}
6067 @cindex symbolic debuggers, information for
6068 @cindex @code{stab@var{x}} directives
6069 There are three directives that begin @samp{.stab}.
6070 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6071 The symbols are not entered in the @command{@value{AS}} hash table: they
6072 cannot be referenced elsewhere in the source file.
6073 Up to five fields are required:
6077 This is the symbol's name. It may contain any character except
6078 @samp{\000}, so is more general than ordinary symbol names. Some
6079 debuggers used to code arbitrarily complex structures into symbol names
6083 An absolute expression. The symbol's type is set to the low 8 bits of
6084 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6085 and debuggers choke on silly bit patterns.
6088 An absolute expression. The symbol's ``other'' attribute is set to the
6089 low 8 bits of this expression.
6092 An absolute expression. The symbol's descriptor is set to the low 16
6093 bits of this expression.
6096 An absolute expression which becomes the symbol's value.
6099 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6100 or @code{.stabs} statement, the symbol has probably already been created;
6101 you get a half-formed symbol in your object file. This is
6102 compatible with earlier assemblers!
6105 @cindex @code{stabd} directive
6106 @item .stabd @var{type} , @var{other} , @var{desc}
6108 The ``name'' of the symbol generated is not even an empty string.
6109 It is a null pointer, for compatibility. Older assemblers used a
6110 null pointer so they didn't waste space in object files with empty
6113 The symbol's value is set to the location counter,
6114 relocatably. When your program is linked, the value of this symbol
6115 is the address of the location counter when the @code{.stabd} was
6118 @cindex @code{stabn} directive
6119 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6120 The name of the symbol is set to the empty string @code{""}.
6122 @cindex @code{stabs} directive
6123 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6124 All five fields are specified.
6130 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6131 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6133 @cindex string, copying to object file
6134 @cindex string8, copying to object file
6135 @cindex string16, copying to object file
6136 @cindex string32, copying to object file
6137 @cindex string64, copying to object file
6138 @cindex @code{string} directive
6139 @cindex @code{string8} directive
6140 @cindex @code{string16} directive
6141 @cindex @code{string32} directive
6142 @cindex @code{string64} directive
6144 Copy the characters in @var{str} to the object file. You may specify more than
6145 one string to copy, separated by commas. Unless otherwise specified for a
6146 particular machine, the assembler marks the end of each string with a 0 byte.
6147 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6149 The variants @code{string16}, @code{string32} and @code{string64} differ from
6150 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6151 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6152 are stored in target endianness byte order.
6158 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6159 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6164 @section @code{.struct @var{expression}}
6166 @cindex @code{struct} directive
6167 Switch to the absolute section, and set the section offset to @var{expression},
6168 which must be an absolute expression. You might use this as follows:
6177 This would define the symbol @code{field1} to have the value 0, the symbol
6178 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6179 value 8. Assembly would be left in the absolute section, and you would need to
6180 use a @code{.section} directive of some sort to change to some other section
6181 before further assembly.
6185 @section @code{.subsection @var{name}}
6187 @cindex @code{subsection} directive
6188 @cindex Section Stack
6189 This is one of the ELF section stack manipulation directives. The others are
6190 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6191 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6194 This directive replaces the current subsection with @code{name}. The current
6195 section is not changed. The replaced subsection is put onto the section stack
6196 in place of the then current top of stack subsection.
6201 @section @code{.symver}
6202 @cindex @code{symver} directive
6203 @cindex symbol versioning
6204 @cindex versions of symbols
6205 Use the @code{.symver} directive to bind symbols to specific version nodes
6206 within a source file. This is only supported on ELF platforms, and is
6207 typically used when assembling files to be linked into a shared library.
6208 There are cases where it may make sense to use this in objects to be bound
6209 into an application itself so as to override a versioned symbol from a
6212 For ELF targets, the @code{.symver} directive can be used like this:
6214 .symver @var{name}, @var{name2@@nodename}
6216 If the symbol @var{name} is defined within the file
6217 being assembled, the @code{.symver} directive effectively creates a symbol
6218 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6219 just don't try and create a regular alias is that the @var{@@} character isn't
6220 permitted in symbol names. The @var{name2} part of the name is the actual name
6221 of the symbol by which it will be externally referenced. The name @var{name}
6222 itself is merely a name of convenience that is used so that it is possible to
6223 have definitions for multiple versions of a function within a single source
6224 file, and so that the compiler can unambiguously know which version of a
6225 function is being mentioned. The @var{nodename} portion of the alias should be
6226 the name of a node specified in the version script supplied to the linker when
6227 building a shared library. If you are attempting to override a versioned
6228 symbol from a shared library, then @var{nodename} should correspond to the
6229 nodename of the symbol you are trying to override.
6231 If the symbol @var{name} is not defined within the file being assembled, all
6232 references to @var{name} will be changed to @var{name2@@nodename}. If no
6233 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6236 Another usage of the @code{.symver} directive is:
6238 .symver @var{name}, @var{name2@@@@nodename}
6240 In this case, the symbol @var{name} must exist and be defined within
6241 the file being assembled. It is similar to @var{name2@@nodename}. The
6242 difference is @var{name2@@@@nodename} will also be used to resolve
6243 references to @var{name2} by the linker.
6245 The third usage of the @code{.symver} directive is:
6247 .symver @var{name}, @var{name2@@@@@@nodename}
6249 When @var{name} is not defined within the
6250 file being assembled, it is treated as @var{name2@@nodename}. When
6251 @var{name} is defined within the file being assembled, the symbol
6252 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6257 @section @code{.tag @var{structname}}
6259 @cindex COFF structure debugging
6260 @cindex structure debugging, COFF
6261 @cindex @code{tag} directive
6262 This directive is generated by compilers to include auxiliary debugging
6263 information in the symbol table. It is only permitted inside
6264 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6265 definitions in the symbol table with instances of those structures.
6268 @samp{.tag} is only used when generating COFF format output; when
6269 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6275 @section @code{.text @var{subsection}}
6277 @cindex @code{text} directive
6278 Tells @command{@value{AS}} to assemble the following statements onto the end of
6279 the text subsection numbered @var{subsection}, which is an absolute
6280 expression. If @var{subsection} is omitted, subsection number zero
6284 @section @code{.title "@var{heading}"}
6286 @cindex @code{title} directive
6287 @cindex listing control: title line
6288 Use @var{heading} as the title (second line, immediately after the
6289 source file name and pagenumber) when generating assembly listings.
6291 This directive affects subsequent pages, as well as the current page if
6292 it appears within ten lines of the top of a page.
6296 @section @code{.type}
6298 This directive is used to set the type of a symbol.
6302 @c only print the extra heading if both COFF and ELF are set
6303 @subheading COFF Version
6306 @cindex COFF symbol type
6307 @cindex symbol type, COFF
6308 @cindex @code{type} directive (COFF version)
6309 For COFF targets, this directive is permitted only within
6310 @code{.def}/@code{.endef} pairs. It is used like this:
6316 This records the integer @var{int} as the type attribute of a symbol table
6320 @samp{.type} is associated only with COFF format output; when
6321 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6322 directive but ignores it.
6328 @c only print the extra heading if both COFF and ELF are set
6329 @subheading ELF Version
6332 @cindex ELF symbol type
6333 @cindex symbol type, ELF
6334 @cindex @code{type} directive (ELF version)
6335 For ELF targets, the @code{.type} directive is used like this:
6338 .type @var{name} , @var{type description}
6341 This sets the type of symbol @var{name} to be either a
6342 function symbol or an object symbol. There are five different syntaxes
6343 supported for the @var{type description} field, in order to provide
6344 compatibility with various other assemblers.
6346 Because some of the characters used in these syntaxes (such as @samp{@@} and
6347 @samp{#}) are comment characters for some architectures, some of the syntaxes
6348 below do not work on all architectures. The first variant will be accepted by
6349 the GNU assembler on all architectures so that variant should be used for
6350 maximum portability, if you do not need to assemble your code with other
6353 The syntaxes supported are:
6356 .type <name> STT_<TYPE_IN_UPPER_CASE>
6357 .type <name>,#<type>
6358 .type <name>,@@<type>
6359 .type <name>,%<type>
6360 .type <name>,"<type>"
6363 The types supported are:
6368 Mark the symbol as being a function name.
6371 @itemx gnu_indirect_function
6372 Mark the symbol as an indirect function when evaluated during reloc
6373 processing. (This is only supported on Linux targeted assemblers).
6377 Mark the symbol as being a data object.
6381 Mark the symbol as being a thead-local data object.
6385 Mark the symbol as being a common data object.
6389 Does not mark the symbol in any way. It is supported just for completeness.
6391 @item gnu_unique_object
6392 Marks the symbol as being a globally unique data object. The dynamic linker
6393 will make sure that in the entire process there is just one symbol with this
6394 name and type in use. (This is only supported on Linux targeted assemblers).
6398 Note: Some targets support extra types in addition to those listed above.
6404 @section @code{.uleb128 @var{expressions}}
6406 @cindex @code{uleb128} directive
6407 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6408 compact, variable length representation of numbers used by the DWARF
6409 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6413 @section @code{.val @var{addr}}
6415 @cindex @code{val} directive
6416 @cindex COFF value attribute
6417 @cindex value attribute, COFF
6418 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6419 records the address @var{addr} as the value attribute of a symbol table
6423 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6424 configured for @code{b.out}, it accepts this directive but ignores it.
6430 @section @code{.version "@var{string}"}
6432 @cindex @code{version} directive
6433 This directive creates a @code{.note} section and places into it an ELF
6434 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6439 @section @code{.vtable_entry @var{table}, @var{offset}}
6441 @cindex @code{vtable_entry} directive
6442 This directive finds or creates a symbol @code{table} and creates a
6443 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6446 @section @code{.vtable_inherit @var{child}, @var{parent}}
6448 @cindex @code{vtable_inherit} directive
6449 This directive finds the symbol @code{child} and finds or creates the symbol
6450 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6451 parent whose addend is the value of the child symbol. As a special case the
6452 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6456 @section @code{.warning "@var{string}"}
6457 @cindex warning directive
6458 Similar to the directive @code{.error}
6459 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6462 @section @code{.weak @var{names}}
6464 @cindex @code{weak} directive
6465 This directive sets the weak attribute on the comma separated list of symbol
6466 @code{names}. If the symbols do not already exist, they will be created.
6468 On COFF targets other than PE, weak symbols are a GNU extension. This
6469 directive sets the weak attribute on the comma separated list of symbol
6470 @code{names}. If the symbols do not already exist, they will be created.
6472 On the PE target, weak symbols are supported natively as weak aliases.
6473 When a weak symbol is created that is not an alias, GAS creates an
6474 alternate symbol to hold the default value.
6477 @section @code{.weakref @var{alias}, @var{target}}
6479 @cindex @code{weakref} directive
6480 This directive creates an alias to the target symbol that enables the symbol to
6481 be referenced with weak-symbol semantics, but without actually making it weak.
6482 If direct references or definitions of the symbol are present, then the symbol
6483 will not be weak, but if all references to it are through weak references, the
6484 symbol will be marked as weak in the symbol table.
6486 The effect is equivalent to moving all references to the alias to a separate
6487 assembly source file, renaming the alias to the symbol in it, declaring the
6488 symbol as weak there, and running a reloadable link to merge the object files
6489 resulting from the assembly of the new source file and the old source file that
6490 had the references to the alias removed.
6492 The alias itself never makes to the symbol table, and is entirely handled
6493 within the assembler.
6496 @section @code{.word @var{expressions}}
6498 @cindex @code{word} directive
6499 This directive expects zero or more @var{expressions}, of any section,
6500 separated by commas.
6503 For each expression, @command{@value{AS}} emits a 32-bit number.
6506 For each expression, @command{@value{AS}} emits a 16-bit number.
6511 The size of the number emitted, and its byte order,
6512 depend on what target computer the assembly is for.
6515 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
6516 @c happen---32-bit addressability, period; no long/short jumps.
6517 @ifset DIFF-TBL-KLUGE
6518 @cindex difference tables altered
6519 @cindex altered difference tables
6521 @emph{Warning: Special Treatment to support Compilers}
6525 Machines with a 32-bit address space, but that do less than 32-bit
6526 addressing, require the following special treatment. If the machine of
6527 interest to you does 32-bit addressing (or doesn't require it;
6528 @pxref{Machine Dependencies}), you can ignore this issue.
6531 In order to assemble compiler output into something that works,
6532 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6533 Directives of the form @samp{.word sym1-sym2} are often emitted by
6534 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6535 directive of the form @samp{.word sym1-sym2}, and the difference between
6536 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6537 creates a @dfn{secondary jump table}, immediately before the next label.
6538 This secondary jump table is preceded by a short-jump to the
6539 first byte after the secondary table. This short-jump prevents the flow
6540 of control from accidentally falling into the new table. Inside the
6541 table is a long-jump to @code{sym2}. The original @samp{.word}
6542 contains @code{sym1} minus the address of the long-jump to
6545 If there were several occurrences of @samp{.word sym1-sym2} before the
6546 secondary jump table, all of them are adjusted. If there was a
6547 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6548 long-jump to @code{sym4} is included in the secondary jump table,
6549 and the @code{.word} directives are adjusted to contain @code{sym3}
6550 minus the address of the long-jump to @code{sym4}; and so on, for as many
6551 entries in the original jump table as necessary.
6554 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6555 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6556 assembly language programmers.
6559 @c end DIFF-TBL-KLUGE
6562 @section Deprecated Directives
6564 @cindex deprecated directives
6565 @cindex obsolescent directives
6566 One day these directives won't work.
6567 They are included for compatibility with older assemblers.
6574 @node Object Attributes
6575 @chapter Object Attributes
6576 @cindex object attributes
6578 @command{@value{AS}} assembles source files written for a specific architecture
6579 into object files for that architecture. But not all object files are alike.
6580 Many architectures support incompatible variations. For instance, floating
6581 point arguments might be passed in floating point registers if the object file
6582 requires hardware floating point support---or floating point arguments might be
6583 passed in integer registers if the object file supports processors with no
6584 hardware floating point unit. Or, if two objects are built for different
6585 generations of the same architecture, the combination may require the
6586 newer generation at run-time.
6588 This information is useful during and after linking. At link time,
6589 @command{@value{LD}} can warn about incompatible object files. After link
6590 time, tools like @command{gdb} can use it to process the linked file
6593 Compatibility information is recorded as a series of object attributes. Each
6594 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
6595 string, and indicates who sets the meaning of the tag. The tag is an integer,
6596 and indicates what property the attribute describes. The value may be a string
6597 or an integer, and indicates how the property affects this object. Missing
6598 attributes are the same as attributes with a zero value or empty string value.
6600 Object attributes were developed as part of the ABI for the ARM Architecture.
6601 The file format is documented in @cite{ELF for the ARM Architecture}.
6604 * GNU Object Attributes:: @sc{gnu} Object Attributes
6605 * Defining New Object Attributes:: Defining New Object Attributes
6608 @node GNU Object Attributes
6609 @section @sc{gnu} Object Attributes
6611 The @code{.gnu_attribute} directive records an object attribute
6612 with vendor @samp{gnu}.
6614 Except for @samp{Tag_compatibility}, which has both an integer and a string for
6615 its value, @sc{gnu} attributes have a string value if the tag number is odd and
6616 an integer value if the tag number is even. The second bit (@code{@var{tag} &
6617 2} is set for architecture-independent attributes and clear for
6618 architecture-dependent ones.
6620 @subsection Common @sc{gnu} attributes
6622 These attributes are valid on all architectures.
6625 @item Tag_compatibility (32)
6626 The compatibility attribute takes an integer flag value and a vendor name. If
6627 the flag value is 0, the file is compatible with other toolchains. If it is 1,
6628 then the file is only compatible with the named toolchain. If it is greater
6629 than 1, the file can only be processed by other toolchains under some private
6630 arrangement indicated by the flag value and the vendor name.
6633 @subsection MIPS Attributes
6636 @item Tag_GNU_MIPS_ABI_FP (4)
6637 The floating-point ABI used by this object file. The value will be:
6641 0 for files not affected by the floating-point ABI.
6643 1 for files using the hardware floating-point with a standard double-precision
6646 2 for files using the hardware floating-point ABI with a single-precision FPU.
6648 3 for files using the software floating-point ABI.
6650 4 for files using the hardware floating-point ABI with 64-bit wide
6651 double-precision floating-point registers and 32-bit wide general
6656 @subsection PowerPC Attributes
6659 @item Tag_GNU_Power_ABI_FP (4)
6660 The floating-point ABI used by this object file. The value will be:
6664 0 for files not affected by the floating-point ABI.
6666 1 for files using double-precision hardware floating-point ABI.
6668 2 for files using the software floating-point ABI.
6670 3 for files using single-precision hardware floating-point ABI.
6673 @item Tag_GNU_Power_ABI_Vector (8)
6674 The vector ABI used by this object file. The value will be:
6678 0 for files not affected by the vector ABI.
6680 1 for files using general purpose registers to pass vectors.
6682 2 for files using AltiVec registers to pass vectors.
6684 3 for files using SPE registers to pass vectors.
6688 @node Defining New Object Attributes
6689 @section Defining New Object Attributes
6691 If you want to define a new @sc{gnu} object attribute, here are the places you
6692 will need to modify. New attributes should be discussed on the @samp{binutils}
6697 This manual, which is the official register of attributes.
6699 The header for your architecture @file{include/elf}, to define the tag.
6701 The @file{bfd} support file for your architecture, to merge the attribute
6702 and issue any appropriate link warnings.
6704 Test cases in @file{ld/testsuite} for merging and link warnings.
6706 @file{binutils/readelf.c} to display your attribute.
6708 GCC, if you want the compiler to mark the attribute automatically.
6714 @node Machine Dependencies
6715 @chapter Machine Dependent Features
6717 @cindex machine dependencies
6718 The machine instruction sets are (almost by definition) different on
6719 each machine where @command{@value{AS}} runs. Floating point representations
6720 vary as well, and @command{@value{AS}} often supports a few additional
6721 directives or command-line options for compatibility with other
6722 assemblers on a particular platform. Finally, some versions of
6723 @command{@value{AS}} support special pseudo-instructions for branch
6726 This chapter discusses most of these differences, though it does not
6727 include details on any machine's instruction set. For details on that
6728 subject, see the hardware manufacturer's manual.
6732 * Alpha-Dependent:: Alpha Dependent Features
6735 * ARC-Dependent:: ARC Dependent Features
6738 * ARM-Dependent:: ARM Dependent Features
6741 * AVR-Dependent:: AVR Dependent Features
6744 * Blackfin-Dependent:: Blackfin Dependent Features
6747 * CR16-Dependent:: CR16 Dependent Features
6750 * CRIS-Dependent:: CRIS Dependent Features
6753 * D10V-Dependent:: D10V Dependent Features
6756 * D30V-Dependent:: D30V Dependent Features
6759 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6762 * HPPA-Dependent:: HPPA Dependent Features
6765 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
6768 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
6771 * i860-Dependent:: Intel 80860 Dependent Features
6774 * i960-Dependent:: Intel 80960 Dependent Features
6777 * IA-64-Dependent:: Intel IA-64 Dependent Features
6780 * IP2K-Dependent:: IP2K Dependent Features
6783 * LM32-Dependent:: LM32 Dependent Features
6786 * M32C-Dependent:: M32C Dependent Features
6789 * M32R-Dependent:: M32R Dependent Features
6792 * M68K-Dependent:: M680x0 Dependent Features
6795 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
6798 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
6801 * MIPS-Dependent:: MIPS Dependent Features
6804 * MMIX-Dependent:: MMIX Dependent Features
6807 * MSP430-Dependent:: MSP430 Dependent Features
6810 * SH-Dependent:: Renesas / SuperH SH Dependent Features
6811 * SH64-Dependent:: SuperH SH64 Dependent Features
6814 * PDP-11-Dependent:: PDP-11 Dependent Features
6817 * PJ-Dependent:: picoJava Dependent Features
6820 * PPC-Dependent:: PowerPC Dependent Features
6823 * RX-Dependent:: RX Dependent Features
6826 * S/390-Dependent:: IBM S/390 Dependent Features
6829 * SCORE-Dependent:: SCORE Dependent Features
6832 * Sparc-Dependent:: SPARC Dependent Features
6835 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
6838 * V850-Dependent:: V850 Dependent Features
6841 * Xtensa-Dependent:: Xtensa Dependent Features
6844 * Z80-Dependent:: Z80 Dependent Features
6847 * Z8000-Dependent:: Z8000 Dependent Features
6850 * Vax-Dependent:: VAX Dependent Features
6857 @c The following major nodes are *sections* in the GENERIC version, *chapters*
6858 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
6859 @c peculiarity: to preserve cross-references, there must be a node called
6860 @c "Machine Dependencies". Hence the conditional nodenames in each
6861 @c major node below. Node defaulting in makeinfo requires adjacency of
6862 @c node and sectioning commands; hence the repetition of @chapter BLAH
6863 @c in both conditional blocks.
6866 @include c-alpha.texi
6882 @include c-bfin.texi
6886 @include c-cr16.texi
6890 @include c-cris.texi
6895 @node Machine Dependencies
6896 @chapter Machine Dependent Features
6898 The machine instruction sets are different on each Renesas chip family,
6899 and there are also some syntax differences among the families. This
6900 chapter describes the specific @command{@value{AS}} features for each
6904 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6905 * SH-Dependent:: Renesas SH Dependent Features
6912 @include c-d10v.texi
6916 @include c-d30v.texi
6920 @include c-h8300.texi
6924 @include c-hppa.texi
6928 @include c-i370.texi
6932 @include c-i386.texi
6936 @include c-i860.texi
6940 @include c-i960.texi
6944 @include c-ia64.texi
6948 @include c-ip2k.texi
6952 @include c-lm32.texi
6956 @include c-m32c.texi
6960 @include c-m32r.texi
6964 @include c-m68k.texi
6968 @include c-m68hc11.texi
6972 @include c-microblaze.texi
6976 @include c-mips.texi
6980 @include c-mmix.texi
6984 @include c-msp430.texi
6988 @include c-ns32k.texi
6992 @include c-pdp11.texi
7008 @include c-s390.texi
7012 @include c-score.texi
7017 @include c-sh64.texi
7021 @include c-sparc.texi
7025 @include c-tic54x.texi
7041 @include c-v850.texi
7045 @include c-xtensa.texi
7049 @c reverse effect of @down at top of generic Machine-Dep chapter
7053 @node Reporting Bugs
7054 @chapter Reporting Bugs
7055 @cindex bugs in assembler
7056 @cindex reporting bugs in assembler
7058 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7060 Reporting a bug may help you by bringing a solution to your problem, or it may
7061 not. But in any case the principal function of a bug report is to help the
7062 entire community by making the next version of @command{@value{AS}} work better.
7063 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7065 In order for a bug report to serve its purpose, you must include the
7066 information that enables us to fix the bug.
7069 * Bug Criteria:: Have you found a bug?
7070 * Bug Reporting:: How to report bugs
7074 @section Have You Found a Bug?
7075 @cindex bug criteria
7077 If you are not sure whether you have found a bug, here are some guidelines:
7080 @cindex fatal signal
7081 @cindex assembler crash
7082 @cindex crash of assembler
7084 If the assembler gets a fatal signal, for any input whatever, that is a
7085 @command{@value{AS}} bug. Reliable assemblers never crash.
7087 @cindex error on valid input
7089 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7091 @cindex invalid input
7093 If @command{@value{AS}} does not produce an error message for invalid input, that
7094 is a bug. However, you should note that your idea of ``invalid input'' might
7095 be our idea of ``an extension'' or ``support for traditional practice''.
7098 If you are an experienced user of assemblers, your suggestions for improvement
7099 of @command{@value{AS}} are welcome in any case.
7103 @section How to Report Bugs
7105 @cindex assembler bugs, reporting
7107 A number of companies and individuals offer support for @sc{gnu} products. If
7108 you obtained @command{@value{AS}} from a support organization, we recommend you
7109 contact that organization first.
7111 You can find contact information for many support companies and
7112 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7116 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7120 The fundamental principle of reporting bugs usefully is this:
7121 @strong{report all the facts}. If you are not sure whether to state a
7122 fact or leave it out, state it!
7124 Often people omit facts because they think they know what causes the problem
7125 and assume that some details do not matter. Thus, you might assume that the
7126 name of a symbol you use in an example does not matter. Well, probably it does
7127 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7128 happens to fetch from the location where that name is stored in memory;
7129 perhaps, if the name were different, the contents of that location would fool
7130 the assembler into doing the right thing despite the bug. Play it safe and
7131 give a specific, complete example. That is the easiest thing for you to do,
7132 and the most helpful.
7134 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7135 it is new to us. Therefore, always write your bug reports on the assumption
7136 that the bug has not been reported previously.
7138 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7139 bell?'' This cannot help us fix a bug, so it is basically useless. We
7140 respond by asking for enough details to enable us to investigate.
7141 You might as well expedite matters by sending them to begin with.
7143 To enable us to fix the bug, you should include all these things:
7147 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
7148 it with the @samp{--version} argument.
7150 Without this, we will not know whether there is any point in looking for
7151 the bug in the current version of @command{@value{AS}}.
7154 Any patches you may have applied to the @command{@value{AS}} source.
7157 The type of machine you are using, and the operating system name and
7161 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
7165 The command arguments you gave the assembler to assemble your example and
7166 observe the bug. To guarantee you will not omit something important, list them
7167 all. A copy of the Makefile (or the output from make) is sufficient.
7169 If we were to try to guess the arguments, we would probably guess wrong
7170 and then we might not encounter the bug.
7173 A complete input file that will reproduce the bug. If the bug is observed when
7174 the assembler is invoked via a compiler, send the assembler source, not the
7175 high level language source. Most compilers will produce the assembler source
7176 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
7177 the options @samp{-v --save-temps}; this will save the assembler source in a
7178 file with an extension of @file{.s}, and also show you exactly how
7179 @command{@value{AS}} is being run.
7182 A description of what behavior you observe that you believe is
7183 incorrect. For example, ``It gets a fatal signal.''
7185 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
7186 will certainly notice it. But if the bug is incorrect output, we might not
7187 notice unless it is glaringly wrong. You might as well not give us a chance to
7190 Even if the problem you experience is a fatal signal, you should still say so
7191 explicitly. Suppose something strange is going on, such as, your copy of
7192 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
7193 library on your system. (This has happened!) Your copy might crash and ours
7194 would not. If you told us to expect a crash, then when ours fails to crash, we
7195 would know that the bug was not happening for us. If you had not told us to
7196 expect a crash, then we would not be able to draw any conclusion from our
7200 If you wish to suggest changes to the @command{@value{AS}} source, send us context
7201 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
7202 option. Always send diffs from the old file to the new file. If you even
7203 discuss something in the @command{@value{AS}} source, refer to it by context, not
7206 The line numbers in our development sources will not match those in your
7207 sources. Your line numbers would convey no useful information to us.
7210 Here are some things that are not necessary:
7214 A description of the envelope of the bug.
7216 Often people who encounter a bug spend a lot of time investigating
7217 which changes to the input file will make the bug go away and which
7218 changes will not affect it.
7220 This is often time consuming and not very useful, because the way we
7221 will find the bug is by running a single example under the debugger
7222 with breakpoints, not by pure deduction from a series of examples.
7223 We recommend that you save your time for something else.
7225 Of course, if you can find a simpler example to report @emph{instead}
7226 of the original one, that is a convenience for us. Errors in the
7227 output will be easier to spot, running under the debugger will take
7228 less time, and so on.
7230 However, simplification is not vital; if you do not want to do this,
7231 report the bug anyway and send us the entire test case you used.
7234 A patch for the bug.
7236 A patch for the bug does help us if it is a good one. But do not omit
7237 the necessary information, such as the test case, on the assumption that
7238 a patch is all we need. We might see problems with your patch and decide
7239 to fix the problem another way, or we might not understand it at all.
7241 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
7242 construct an example that will make the program follow a certain path through
7243 the code. If you do not send us the example, we will not be able to construct
7244 one, so we will not be able to verify that the bug is fixed.
7246 And if we cannot understand what bug you are trying to fix, or why your
7247 patch should be an improvement, we will not install it. A test case will
7248 help us to understand.
7251 A guess about what the bug is or what it depends on.
7253 Such guesses are usually wrong. Even we cannot guess right about such
7254 things without first using the debugger to find the facts.
7257 @node Acknowledgements
7258 @chapter Acknowledgements
7260 If you have contributed to GAS and your name isn't listed here,
7261 it is not meant as a slight. We just don't know about it. Send mail to the
7262 maintainer, and we'll correct the situation. Currently
7264 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
7266 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
7269 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
7270 information and the 68k series machines, most of the preprocessing pass, and
7271 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
7273 K. Richard Pixley maintained GAS for a while, adding various enhancements and
7274 many bug fixes, including merging support for several processors, breaking GAS
7275 up to handle multiple object file format back ends (including heavy rewrite,
7276 testing, an integration of the coff and b.out back ends), adding configuration
7277 including heavy testing and verification of cross assemblers and file splits
7278 and renaming, converted GAS to strictly ANSI C including full prototypes, added
7279 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
7280 port (including considerable amounts of reverse engineering), a SPARC opcode
7281 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
7282 assertions and made them work, much other reorganization, cleanup, and lint.
7284 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
7285 in format-specific I/O modules.
7287 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
7288 has done much work with it since.
7290 The Intel 80386 machine description was written by Eliot Dresselhaus.
7292 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
7294 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
7295 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
7297 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
7298 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
7299 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
7300 support a.out format.
7302 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
7303 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
7304 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
7305 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
7308 John Gilmore built the AMD 29000 support, added @code{.include} support, and
7309 simplified the configuration of which versions accept which directives. He
7310 updated the 68k machine description so that Motorola's opcodes always produced
7311 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
7312 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
7313 cross-compilation support, and one bug in relaxation that took a week and
7314 required the proverbial one-bit fix.
7316 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
7317 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
7318 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
7319 PowerPC assembler, and made a few other minor patches.
7321 Steve Chamberlain made GAS able to generate listings.
7323 Hewlett-Packard contributed support for the HP9000/300.
7325 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
7326 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
7327 formats). This work was supported by both the Center for Software Science at
7328 the University of Utah and Cygnus Support.
7330 Support for ELF format files has been worked on by Mark Eichin of Cygnus
7331 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
7332 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
7333 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
7334 and some initial 64-bit support).
7336 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
7338 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
7339 support for openVMS/Alpha.
7341 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
7344 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
7345 Inc.@: added support for Xtensa processors.
7347 Several engineers at Cygnus Support have also provided many small bug fixes and
7348 configuration enhancements.
7350 Jon Beniston added support for the Lattice Mico32 architecture.
7352 Many others have contributed large or small bugfixes and enhancements. If
7353 you have contributed significant work and are not mentioned on this list, and
7354 want to be, let us know. Some of the history has been lost; we are not
7355 intentionally leaving anyone out.
7357 @node GNU Free Documentation License
7358 @appendix GNU Free Documentation License
7362 @unnumbered AS Index