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, 2010
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
55 @set abnormal-separator
59 @settitle Using @value{AS}
62 @settitle Using @value{AS} (@value{TARGET})
64 @setchapternewpage odd
69 @c WARE! Some of the machine-dependent sections contain tables of machine
70 @c instructions. Except in multi-column format, these tables look silly.
71 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
72 @c the multi-col format is faked within @example sections.
74 @c Again unfortunately, the natural size that fits on a page, for these tables,
75 @c is different depending on whether or not smallbook is turned on.
76 @c This matters, because of order: text flow switches columns at each page
79 @c The format faked in this source works reasonably well for smallbook,
80 @c not well for the default large-page format. This manual expects that if you
81 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
82 @c tables in question. You can turn on one without the other at your
83 @c discretion, of course.
86 @c the insn tables look just as silly in info files regardless of smallbook,
87 @c might as well show 'em anyways.
93 * As: (as). The GNU assembler.
94 * Gas: (as). The GNU assembler.
103 This file documents the GNU Assembler "@value{AS}".
105 @c man begin COPYRIGHT
106 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002,
107 2006, 2007, 2008, 2009, 2010 Free Software Foundation, Inc.
109 Permission is granted to copy, distribute and/or modify this document
110 under the terms of the GNU Free Documentation License, Version 1.3
111 or any later version published by the Free Software Foundation;
112 with no Invariant Sections, with no Front-Cover Texts, and with no
113 Back-Cover Texts. A copy of the license is included in the
114 section entitled ``GNU Free Documentation License''.
120 @title Using @value{AS}
121 @subtitle The @sc{gnu} Assembler
123 @subtitle for the @value{TARGET} family
125 @ifset VERSION_PACKAGE
127 @subtitle @value{VERSION_PACKAGE}
130 @subtitle Version @value{VERSION}
133 The Free Software Foundation Inc.@: thanks The Nice Computer
134 Company of Australia for loaning Dean Elsner to write the
135 first (Vax) version of @command{as} for Project @sc{gnu}.
136 The proprietors, management and staff of TNCCA thank FSF for
137 distracting the boss while they got some work
140 @author Dean Elsner, Jay Fenlason & friends
144 \hfill {\it Using {\tt @value{AS}}}\par
145 \hfill Edited by Cygnus Support\par
147 %"boxit" macro for figures:
148 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
149 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
150 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
151 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
152 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
155 @vskip 0pt plus 1filll
156 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002,
157 2006, 2007, 2008, 2009, 2010 Free Software Foundation, Inc.
159 Permission is granted to copy, distribute and/or modify this document
160 under the terms of the GNU Free Documentation License, Version 1.3
161 or any later version published by the Free Software Foundation;
162 with no Invariant Sections, with no Front-Cover Texts, and with no
163 Back-Cover Texts. A copy of the license is included in the
164 section entitled ``GNU Free Documentation License''.
171 @top Using @value{AS}
173 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}}
174 @ifset VERSION_PACKAGE
175 @value{VERSION_PACKAGE}
177 version @value{VERSION}.
179 This version of the file describes @command{@value{AS}} configured to generate
180 code for @value{TARGET} architectures.
183 This document is distributed under the terms of the GNU Free
184 Documentation License. A copy of the license is included in the
185 section entitled ``GNU Free Documentation License''.
188 * Overview:: Overview
189 * Invoking:: Command-Line Options
191 * Sections:: Sections and Relocation
193 * Expressions:: Expressions
194 * Pseudo Ops:: Assembler Directives
196 * Object Attributes:: Object Attributes
198 * Machine Dependencies:: Machine Dependent Features
199 * Reporting Bugs:: Reporting Bugs
200 * Acknowledgements:: Who Did What
201 * GNU Free Documentation License:: GNU Free Documentation License
202 * AS Index:: AS Index
209 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
211 This version of the manual describes @command{@value{AS}} configured to generate
212 code for @value{TARGET} architectures.
216 @cindex invocation summary
217 @cindex option summary
218 @cindex summary of options
219 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
220 see @ref{Invoking,,Command-Line Options}.
222 @c man title AS the portable GNU assembler.
226 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
230 @c We don't use deffn and friends for the following because they seem
231 @c to be limited to one line for the header.
233 @c man begin SYNOPSIS
234 @value{AS} [@b{-a}[@b{cdghlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
235 [@b{--debug-prefix-map} @var{old}=@var{new}]
236 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
237 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
238 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
239 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
240 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}] [@b{-o}
241 @var{objfile}] [@b{-R}] [@b{--reduce-memory-overheads}] [@b{--statistics}]
242 [@b{-v}] [@b{-version}] [@b{--version}] [@b{-W}] [@b{--warn}]
243 [@b{--fatal-warnings}] [@b{-w}] [@b{-x}] [@b{-Z}] [@b{@@@var{FILE}}]
244 [@b{--target-help}] [@var{target-options}]
245 [@b{--}|@var{files} @dots{}]
247 @c Target dependent options are listed below. Keep the list sorted.
248 @c Add an empty line for separation.
251 @emph{Target Alpha options:}
253 [@b{-mdebug} | @b{-no-mdebug}]
254 [@b{-replace} | @b{-noreplace}]
255 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
256 [@b{-F}] [@b{-32addr}]
260 @emph{Target ARC options:}
266 @emph{Target ARM options:}
267 @c Don't document the deprecated options
268 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
269 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
270 [@b{-mfpu}=@var{floating-point-format}]
271 [@b{-mfloat-abi}=@var{abi}]
272 [@b{-meabi}=@var{ver}]
275 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
276 @b{-mapcs-reentrant}]
277 [@b{-mthumb-interwork}] [@b{-k}]
281 @emph{Target Blackfin options:}
282 [@b{-mcpu}=@var{processor}[-@var{sirevision}]]
289 @emph{Target CRIS options:}
290 [@b{--underscore} | @b{--no-underscore}]
292 [@b{--emulation=criself} | @b{--emulation=crisaout}]
293 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
294 @c Deprecated -- deliberately not documented.
299 @emph{Target D10V options:}
304 @emph{Target D30V options:}
305 [@b{-O}|@b{-n}|@b{-N}]
309 @emph{Target H8/300 options:}
313 @c HPPA has no machine-dependent assembler options (yet).
317 @emph{Target i386 options:}
318 [@b{--32}|@b{--64}] [@b{-n}]
319 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
323 @emph{Target i960 options:}
324 @c see md_parse_option in tc-i960.c
325 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
327 [@b{-b}] [@b{-no-relax}]
331 @emph{Target IA-64 options:}
332 [@b{-mconstant-gp}|@b{-mauto-pic}]
333 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
335 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
336 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
337 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
338 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
342 @emph{Target IP2K options:}
343 [@b{-mip2022}|@b{-mip2022ext}]
347 @emph{Target M32C options:}
348 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
352 @emph{Target M32R options:}
353 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
358 @emph{Target M680X0 options:}
359 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
363 @emph{Target M68HC11 options:}
364 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}]
365 [@b{-mshort}|@b{-mlong}]
366 [@b{-mshort-double}|@b{-mlong-double}]
367 [@b{--force-long-branches}] [@b{--short-branches}]
368 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
369 [@b{--print-opcodes}] [@b{--generate-example}]
373 @emph{Target MCORE options:}
374 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
375 [@b{-mcpu=[210|340]}]
378 @emph{Target MICROBLAZE options:}
379 @c MicroBlaze has no machine-dependent assembler options.
383 @emph{Target MIPS options:}
384 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
385 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
386 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
387 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
388 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
389 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
390 [@b{-mips64}] [@b{-mips64r2}]
391 [@b{-construct-floats}] [@b{-no-construct-floats}]
392 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
393 [@b{-mips16}] [@b{-no-mips16}]
394 [@b{-msmartmips}] [@b{-mno-smartmips}]
395 [@b{-mips3d}] [@b{-no-mips3d}]
396 [@b{-mdmx}] [@b{-no-mdmx}]
397 [@b{-mdsp}] [@b{-mno-dsp}]
398 [@b{-mdspr2}] [@b{-mno-dspr2}]
399 [@b{-mmt}] [@b{-mno-mt}]
400 [@b{-mfix7000}] [@b{-mno-fix7000}]
401 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
402 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
403 [@b{-mdebug}] [@b{-no-mdebug}]
404 [@b{-mpdr}] [@b{-mno-pdr}]
408 @emph{Target MMIX options:}
409 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
410 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
411 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
412 [@b{--linker-allocated-gregs}]
416 @emph{Target PDP11 options:}
417 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
418 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
419 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
423 @emph{Target picoJava options:}
428 @emph{Target PowerPC options:}
429 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|
430 @b{-m403}|@b{-m405}|@b{-mppc64}|@b{-m620}|@b{-mppc64bridge}|@b{-mbooke}]
431 [@b{-mcom}|@b{-many}|@b{-maltivec}|@b{-mvsx}] [@b{-memb}]
432 [@b{-mregnames}|@b{-mno-regnames}]
433 [@b{-mrelocatable}|@b{-mrelocatable-lib}]
434 [@b{-mlittle}|@b{-mlittle-endian}|@b{-mbig}|@b{-mbig-endian}]
435 [@b{-msolaris}|@b{-mno-solaris}]
439 @emph{Target RX options:}
440 [@b{-mlittle-endian}|@b{-mbig-endian}]
441 [@b{-m32bit-ints}|@b{-m16bit-ints}]
442 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
446 @emph{Target s390 options:}
447 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
448 [@b{-mregnames}|@b{-mno-regnames}]
449 [@b{-mwarn-areg-zero}]
453 @emph{Target SCORE options:}
454 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
455 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
456 [@b{-march=score7}][@b{-march=score3}]
457 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
461 @emph{Target SPARC options:}
462 @c The order here is important. See c-sparc.texi.
463 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
464 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
465 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
470 @emph{Target TIC54X options:}
471 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
472 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
477 @emph{Target TIC6X options:}
478 [@b{-march=@var{arch}}] [@b{-matomic}|@b{-mno-atomic}]
479 [@b{-mbig-endian}|@b{-mlittle-endian}]
484 @emph{Target Z80 options:}
485 [@b{-z80}] [@b{-r800}]
486 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
487 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
488 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
489 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
490 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
491 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
495 @c Z8000 has no machine-dependent assembler options
499 @emph{Target Xtensa options:}
500 [@b{--[no-]text-section-literals}] [@b{--[no-]absolute-literals}]
501 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
502 [@b{--[no-]transform}]
503 [@b{--rename-section} @var{oldname}=@var{newname}]
511 @include at-file.texi
514 Turn on listings, in any of a variety of ways:
518 omit false conditionals
521 omit debugging directives
524 include general information, like @value{AS} version and options passed
527 include high-level source
533 include macro expansions
536 omit forms processing
542 set the name of the listing file
545 You may combine these options; for example, use @samp{-aln} for assembly
546 listing without forms processing. The @samp{=file} option, if used, must be
547 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
550 Begin in alternate macro mode.
552 @xref{Altmacro,,@code{.altmacro}}.
556 Ignored. This option is accepted for script compatibility with calls to
559 @item --debug-prefix-map @var{old}=@var{new}
560 When assembling files in directory @file{@var{old}}, record debugging
561 information describing them as in @file{@var{new}} instead.
563 @item --defsym @var{sym}=@var{value}
564 Define the symbol @var{sym} to be @var{value} before assembling the input file.
565 @var{value} must be an integer constant. As in C, a leading @samp{0x}
566 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
567 value. The value of the symbol can be overridden inside a source file via the
568 use of a @code{.set} pseudo-op.
571 ``fast''---skip whitespace and comment preprocessing (assume source is
576 Generate debugging information for each assembler source line using whichever
577 debug format is preferred by the target. This currently means either STABS,
581 Generate stabs debugging information for each assembler line. This
582 may help debugging assembler code, if the debugger can handle it.
585 Generate stabs debugging information for each assembler line, with GNU
586 extensions that probably only gdb can handle, and that could make other
587 debuggers crash or refuse to read your program. This
588 may help debugging assembler code. Currently the only GNU extension is
589 the location of the current working directory at assembling time.
592 Generate DWARF2 debugging information for each assembler line. This
593 may help debugging assembler code, if the debugger can handle it. Note---this
594 option is only supported by some targets, not all of them.
597 Print a summary of the command line options and exit.
600 Print a summary of all target specific options and exit.
603 Add directory @var{dir} to the search list for @code{.include} directives.
606 Don't warn about signed overflow.
609 @ifclear DIFF-TBL-KLUGE
610 This option is accepted but has no effect on the @value{TARGET} family.
612 @ifset DIFF-TBL-KLUGE
613 Issue warnings when difference tables altered for long displacements.
618 Keep (in the symbol table) local symbols. These symbols start with
619 system-specific local label prefixes, typically @samp{.L} for ELF systems
620 or @samp{L} for traditional a.out systems.
625 @item --listing-lhs-width=@var{number}
626 Set the maximum width, in words, of the output data column for an assembler
627 listing to @var{number}.
629 @item --listing-lhs-width2=@var{number}
630 Set the maximum width, in words, of the output data column for continuation
631 lines in an assembler listing to @var{number}.
633 @item --listing-rhs-width=@var{number}
634 Set the maximum width of an input source line, as displayed in a listing, to
637 @item --listing-cont-lines=@var{number}
638 Set the maximum number of lines printed in a listing for a single line of input
641 @item -o @var{objfile}
642 Name the object-file output from @command{@value{AS}} @var{objfile}.
645 Fold the data section into the text section.
647 @kindex --hash-size=@var{number}
648 Set the default size of GAS's hash tables to a prime number close to
649 @var{number}. Increasing this value can reduce the length of time it takes the
650 assembler to perform its tasks, at the expense of increasing the assembler's
651 memory requirements. Similarly reducing this value can reduce the memory
652 requirements at the expense of speed.
654 @item --reduce-memory-overheads
655 This option reduces GAS's memory requirements, at the expense of making the
656 assembly processes slower. Currently this switch is a synonym for
657 @samp{--hash-size=4051}, but in the future it may have other effects as well.
660 Print the maximum space (in bytes) and total time (in seconds) used by
663 @item --strip-local-absolute
664 Remove local absolute symbols from the outgoing symbol table.
668 Print the @command{as} version.
671 Print the @command{as} version and exit.
675 Suppress warning messages.
677 @item --fatal-warnings
678 Treat warnings as errors.
681 Don't suppress warning messages or treat them as errors.
690 Generate an object file even after errors.
692 @item -- | @var{files} @dots{}
693 Standard input, or source files to assemble.
698 The following options are available when @value{AS} is configured for
703 This option selects the core processor variant.
705 Select either big-endian (-EB) or little-endian (-EL) output.
710 The following options are available when @value{AS} is configured for the ARM
714 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
715 Specify which ARM processor variant is the target.
716 @item -march=@var{architecture}[+@var{extension}@dots{}]
717 Specify which ARM architecture variant is used by the target.
718 @item -mfpu=@var{floating-point-format}
719 Select which Floating Point architecture is the target.
720 @item -mfloat-abi=@var{abi}
721 Select which floating point ABI is in use.
723 Enable Thumb only instruction decoding.
724 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
725 Select which procedure calling convention is in use.
727 Select either big-endian (-EB) or little-endian (-EL) output.
728 @item -mthumb-interwork
729 Specify that the code has been generated with interworking between Thumb and
732 Specify that PIC code has been generated.
737 The following options are available when @value{AS} is configured for
738 the Blackfin processor family.
741 @item -mcpu=@var{processor}@r{[}-@var{sirevision}@r{]}
742 This option specifies the target processor. The optional @var{sirevision}
743 is not used in assembler.
745 Assemble for the FDPIC ABI.
746 @item -mno-fdpic/-mnopic
752 See the info pages for documentation of the CRIS-specific options.
756 The following options are available when @value{AS} is configured for
759 @cindex D10V optimization
760 @cindex optimization, D10V
762 Optimize output by parallelizing instructions.
767 The following options are available when @value{AS} is configured for a D30V
770 @cindex D30V optimization
771 @cindex optimization, D30V
773 Optimize output by parallelizing instructions.
777 Warn when nops are generated.
779 @cindex D30V nops after 32-bit multiply
781 Warn when a nop after a 32-bit multiply instruction is generated.
786 The following options are available when @value{AS} is configured for the
787 Intel 80960 processor.
790 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
791 Specify which variant of the 960 architecture is the target.
794 Add code to collect statistics about branches taken.
797 Do not alter compare-and-branch instructions for long displacements;
804 The following options are available when @value{AS} is configured for the
810 Specifies that the extended IP2022 instructions are allowed.
813 Restores the default behaviour, which restricts the permitted instructions to
814 just the basic IP2022 ones.
820 The following options are available when @value{AS} is configured for the
821 Renesas M32C and M16C processors.
826 Assemble M32C instructions.
829 Assemble M16C instructions (the default).
832 Enable support for link-time relaxations.
835 Support H'00 style hex constants in addition to 0x00 style.
841 The following options are available when @value{AS} is configured for the
842 Renesas M32R (formerly Mitsubishi M32R) series.
847 Specify which processor in the M32R family is the target. The default
848 is normally the M32R, but this option changes it to the M32RX.
850 @item --warn-explicit-parallel-conflicts or --Wp
851 Produce warning messages when questionable parallel constructs are
854 @item --no-warn-explicit-parallel-conflicts or --Wnp
855 Do not produce warning messages when questionable parallel constructs are
862 The following options are available when @value{AS} is configured for the
863 Motorola 68000 series.
868 Shorten references to undefined symbols, to one word instead of two.
870 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
871 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
872 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
873 Specify what processor in the 68000 family is the target. The default
874 is normally the 68020, but this can be changed at configuration time.
876 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
877 The target machine does (or does not) have a floating-point coprocessor.
878 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
879 the basic 68000 is not compatible with the 68881, a combination of the
880 two can be specified, since it's possible to do emulation of the
881 coprocessor instructions with the main processor.
883 @item -m68851 | -mno-68851
884 The target machine does (or does not) have a memory-management
885 unit coprocessor. The default is to assume an MMU for 68020 and up.
892 For details about the PDP-11 machine dependent features options,
893 see @ref{PDP-11-Options}.
896 @item -mpic | -mno-pic
897 Generate position-independent (or position-dependent) code. The
898 default is @option{-mpic}.
901 @itemx -mall-extensions
902 Enable all instruction set extensions. This is the default.
904 @item -mno-extensions
905 Disable all instruction set extensions.
907 @item -m@var{extension} | -mno-@var{extension}
908 Enable (or disable) a particular instruction set extension.
911 Enable the instruction set extensions supported by a particular CPU, and
912 disable all other extensions.
914 @item -m@var{machine}
915 Enable the instruction set extensions supported by a particular machine
916 model, and disable all other extensions.
922 The following options are available when @value{AS} is configured for
923 a picoJava processor.
927 @cindex PJ endianness
928 @cindex endianness, PJ
929 @cindex big endian output, PJ
931 Generate ``big endian'' format output.
933 @cindex little endian output, PJ
935 Generate ``little endian'' format output.
941 The following options are available when @value{AS} is configured for the
942 Motorola 68HC11 or 68HC12 series.
946 @item -m68hc11 | -m68hc12 | -m68hcs12
947 Specify what processor is the target. The default is
948 defined by the configuration option when building the assembler.
951 Specify to use the 16-bit integer ABI.
954 Specify to use the 32-bit integer ABI.
957 Specify to use the 32-bit double ABI.
960 Specify to use the 64-bit double ABI.
962 @item --force-long-branches
963 Relative branches are turned into absolute ones. This concerns
964 conditional branches, unconditional branches and branches to a
967 @item -S | --short-branches
968 Do not turn relative branches into absolute ones
969 when the offset is out of range.
971 @item --strict-direct-mode
972 Do not turn the direct addressing mode into extended addressing mode
973 when the instruction does not support direct addressing mode.
975 @item --print-insn-syntax
976 Print the syntax of instruction in case of error.
978 @item --print-opcodes
979 print the list of instructions with syntax and then exit.
981 @item --generate-example
982 print an example of instruction for each possible instruction and then exit.
983 This option is only useful for testing @command{@value{AS}}.
989 The following options are available when @command{@value{AS}} is configured
990 for the SPARC architecture:
993 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
994 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
995 Explicitly select a variant of the SPARC architecture.
997 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
998 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1000 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1001 UltraSPARC extensions.
1003 @item -xarch=v8plus | -xarch=v8plusa
1004 For compatibility with the Solaris v9 assembler. These options are
1005 equivalent to -Av8plus and -Av8plusa, respectively.
1008 Warn when the assembler switches to another architecture.
1013 The following options are available when @value{AS} is configured for the 'c54x
1018 Enable extended addressing mode. All addresses and relocations will assume
1019 extended addressing (usually 23 bits).
1020 @item -mcpu=@var{CPU_VERSION}
1021 Sets the CPU version being compiled for.
1022 @item -merrors-to-file @var{FILENAME}
1023 Redirect error output to a file, for broken systems which don't support such
1024 behaviour in the shell.
1029 The following options are available when @value{AS} is configured for
1030 a @sc{mips} processor.
1034 This option sets the largest size of an object that can be referenced
1035 implicitly with the @code{gp} register. It is only accepted for targets that
1036 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1038 @cindex MIPS endianness
1039 @cindex endianness, MIPS
1040 @cindex big endian output, MIPS
1042 Generate ``big endian'' format output.
1044 @cindex little endian output, MIPS
1046 Generate ``little endian'' format output.
1058 Generate code for a particular @sc{mips} Instruction Set Architecture level.
1059 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1060 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1061 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1062 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
1064 correspond to generic
1065 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
1066 and @samp{MIPS64 Release 2}
1067 ISA processors, respectively.
1069 @item -march=@var{CPU}
1070 Generate code for a particular @sc{mips} cpu.
1072 @item -mtune=@var{cpu}
1073 Schedule and tune for a particular @sc{mips} cpu.
1077 Cause nops to be inserted if the read of the destination register
1078 of an mfhi or mflo instruction occurs in the following two instructions.
1082 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1083 section instead of the standard ELF .stabs sections.
1087 Control generation of @code{.pdr} sections.
1091 The register sizes are normally inferred from the ISA and ABI, but these
1092 flags force a certain group of registers to be treated as 32 bits wide at
1093 all times. @samp{-mgp32} controls the size of general-purpose registers
1094 and @samp{-mfp32} controls the size of floating-point registers.
1098 Generate code for the MIPS 16 processor. This is equivalent to putting
1099 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1100 turns off this option.
1103 @itemx -mno-smartmips
1104 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1105 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1106 @samp{-mno-smartmips} turns off this option.
1110 Generate code for the MIPS-3D Application Specific Extension.
1111 This tells the assembler to accept MIPS-3D instructions.
1112 @samp{-no-mips3d} turns off this option.
1116 Generate code for the MDMX Application Specific Extension.
1117 This tells the assembler to accept MDMX instructions.
1118 @samp{-no-mdmx} turns off this option.
1122 Generate code for the DSP Release 1 Application Specific Extension.
1123 This tells the assembler to accept DSP Release 1 instructions.
1124 @samp{-mno-dsp} turns off this option.
1128 Generate code for the DSP Release 2 Application Specific Extension.
1129 This option implies -mdsp.
1130 This tells the assembler to accept DSP Release 2 instructions.
1131 @samp{-mno-dspr2} turns off this option.
1135 Generate code for the MT Application Specific Extension.
1136 This tells the assembler to accept MT instructions.
1137 @samp{-mno-mt} turns off this option.
1139 @item --construct-floats
1140 @itemx --no-construct-floats
1141 The @samp{--no-construct-floats} option disables the construction of
1142 double width floating point constants by loading the two halves of the
1143 value into the two single width floating point registers that make up
1144 the double width register. By default @samp{--construct-floats} is
1145 selected, allowing construction of these floating point constants.
1148 @item --emulation=@var{name}
1149 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
1150 for some other target, in all respects, including output format (choosing
1151 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
1152 debugging information or store symbol table information, and default
1153 endianness. The available configuration names are: @samp{mipsecoff},
1154 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
1155 @samp{mipsbelf}. The first two do not alter the default endianness from that
1156 of the primary target for which the assembler was configured; the others change
1157 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
1158 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
1159 selection in any case.
1161 This option is currently supported only when the primary target
1162 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
1163 Furthermore, the primary target or others specified with
1164 @samp{--enable-targets=@dots{}} at configuration time must include support for
1165 the other format, if both are to be available. For example, the Irix 5
1166 configuration includes support for both.
1168 Eventually, this option will support more configurations, with more
1169 fine-grained control over the assembler's behavior, and will be supported for
1173 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1180 Control how to deal with multiplication overflow and division by zero.
1181 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1182 (and only work for Instruction Set Architecture level 2 and higher);
1183 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1187 When this option is used, @command{@value{AS}} will issue a warning every
1188 time it generates a nop instruction from a macro.
1193 The following options are available when @value{AS} is configured for
1199 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1200 The command line option @samp{-nojsri2bsr} can be used to disable it.
1204 Enable or disable the silicon filter behaviour. By default this is disabled.
1205 The default can be overridden by the @samp{-sifilter} command line option.
1208 Alter jump instructions for long displacements.
1210 @item -mcpu=[210|340]
1211 Select the cpu type on the target hardware. This controls which instructions
1215 Assemble for a big endian target.
1218 Assemble for a little endian target.
1224 See the info pages for documentation of the MMIX-specific options.
1228 See the info pages for documentation of the RX-specific options.
1232 The following options are available when @value{AS} is configured for the s390
1238 Select the word size, either 31/32 bits or 64 bits.
1241 Select the architecture mode, either the Enterprise System
1242 Architecture (esa) or the z/Architecture mode (zarch).
1243 @item -march=@var{processor}
1244 Specify which s390 processor variant is the target, @samp{g6}, @samp{g6},
1245 @samp{z900}, @samp{z990}, @samp{z9-109}, @samp{z9-ec}, or @samp{z10}.
1247 @itemx -mno-regnames
1248 Allow or disallow symbolic names for registers.
1249 @item -mwarn-areg-zero
1250 Warn whenever the operand for a base or index register has been specified
1251 but evaluates to zero.
1256 The following options are available when @value{AS} is configured for a
1257 TMS320C6000 processor.
1260 @item -march=@var{arch}
1261 Enable (only) instructions from architecture @var{arch}. By default,
1262 all instructions are permitted.
1264 The following values of @var{arch} are accepted: @code{c62x},
1265 @code{c64x}, @code{c64x+}, @code{c67x}, @code{c67x+}, @code{c674x}.
1269 Enable or disable the optional C64x+ atomic operation instructions.
1270 By default, they are enabled if no @option{-march} option is given, or
1271 if an architecture is specified with @option{-march} that implies
1272 these instructions are present (currently, there are no such
1273 architectures); they are disabled if an architecture is specified with
1274 @option{-march} on which the instructions are optional or not
1275 present. This option overrides such a default from the architecture,
1276 independent of the order in which the @option{-march} or
1277 @option{-matomic} or @option{-mno-atomic} options are passed.
1280 @itemx -mlittle-endian
1281 Generate code for the specified endianness. The default is
1288 The following options are available when @value{AS} is configured for
1289 an Xtensa processor.
1292 @item --text-section-literals | --no-text-section-literals
1293 With @option{--text-@-section-@-literals}, literal pools are interspersed
1294 in the text section. The default is
1295 @option{--no-@-text-@-section-@-literals}, which places literals in a
1296 separate section in the output file. These options only affect literals
1297 referenced via PC-relative @code{L32R} instructions; literals for
1298 absolute mode @code{L32R} instructions are handled separately.
1300 @item --absolute-literals | --no-absolute-literals
1301 Indicate to the assembler whether @code{L32R} instructions use absolute
1302 or PC-relative addressing. The default is to assume absolute addressing
1303 if the Xtensa processor includes the absolute @code{L32R} addressing
1304 option. Otherwise, only the PC-relative @code{L32R} mode can be used.
1306 @item --target-align | --no-target-align
1307 Enable or disable automatic alignment to reduce branch penalties at the
1308 expense of some code density. The default is @option{--target-@-align}.
1310 @item --longcalls | --no-longcalls
1311 Enable or disable transformation of call instructions to allow calls
1312 across a greater range of addresses. The default is
1313 @option{--no-@-longcalls}.
1315 @item --transform | --no-transform
1316 Enable or disable all assembler transformations of Xtensa instructions.
1317 The default is @option{--transform};
1318 @option{--no-transform} should be used only in the rare cases when the
1319 instructions must be exactly as specified in the assembly source.
1321 @item --rename-section @var{oldname}=@var{newname}
1322 When generating output sections, rename the @var{oldname} section to
1328 The following options are available when @value{AS} is configured for
1329 a Z80 family processor.
1332 Assemble for Z80 processor.
1334 Assemble for R800 processor.
1335 @item -ignore-undocumented-instructions
1337 Assemble undocumented Z80 instructions that also work on R800 without warning.
1338 @item -ignore-unportable-instructions
1340 Assemble all undocumented Z80 instructions without warning.
1341 @item -warn-undocumented-instructions
1343 Issue a warning for undocumented Z80 instructions that also work on R800.
1344 @item -warn-unportable-instructions
1346 Issue a warning for undocumented Z80 instructions that do not work on R800.
1347 @item -forbid-undocumented-instructions
1349 Treat all undocumented instructions as errors.
1350 @item -forbid-unportable-instructions
1352 Treat undocumented Z80 instructions that do not work on R800 as errors.
1359 * Manual:: Structure of this Manual
1360 * GNU Assembler:: The GNU Assembler
1361 * Object Formats:: Object File Formats
1362 * Command Line:: Command Line
1363 * Input Files:: Input Files
1364 * Object:: Output (Object) File
1365 * Errors:: Error and Warning Messages
1369 @section Structure of this Manual
1371 @cindex manual, structure and purpose
1372 This manual is intended to describe what you need to know to use
1373 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1374 notation for symbols, constants, and expressions; the directives that
1375 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1378 We also cover special features in the @value{TARGET}
1379 configuration of @command{@value{AS}}, including assembler directives.
1382 This manual also describes some of the machine-dependent features of
1383 various flavors of the assembler.
1386 @cindex machine instructions (not covered)
1387 On the other hand, this manual is @emph{not} intended as an introduction
1388 to programming in assembly language---let alone programming in general!
1389 In a similar vein, we make no attempt to introduce the machine
1390 architecture; we do @emph{not} describe the instruction set, standard
1391 mnemonics, registers or addressing modes that are standard to a
1392 particular architecture.
1394 You may want to consult the manufacturer's
1395 machine architecture manual for this information.
1399 For information on the H8/300 machine instruction set, see @cite{H8/300
1400 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1401 Programming Manual} (Renesas).
1404 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1405 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1406 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1407 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1410 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1414 @c I think this is premature---doc@cygnus.com, 17jan1991
1416 Throughout this manual, we assume that you are running @dfn{GNU},
1417 the portable operating system from the @dfn{Free Software
1418 Foundation, Inc.}. This restricts our attention to certain kinds of
1419 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1420 once this assumption is granted examples and definitions need less
1423 @command{@value{AS}} is part of a team of programs that turn a high-level
1424 human-readable series of instructions into a low-level
1425 computer-readable series of instructions. Different versions of
1426 @command{@value{AS}} are used for different kinds of computer.
1429 @c There used to be a section "Terminology" here, which defined
1430 @c "contents", "byte", "word", and "long". Defining "word" to any
1431 @c particular size is confusing when the .word directive may generate 16
1432 @c bits on one machine and 32 bits on another; in general, for the user
1433 @c version of this manual, none of these terms seem essential to define.
1434 @c They were used very little even in the former draft of the manual;
1435 @c this draft makes an effort to avoid them (except in names of
1439 @section The GNU Assembler
1441 @c man begin DESCRIPTION
1443 @sc{gnu} @command{as} is really a family of assemblers.
1445 This manual describes @command{@value{AS}}, a member of that family which is
1446 configured for the @value{TARGET} architectures.
1448 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1449 should find a fairly similar environment when you use it on another
1450 architecture. Each version has much in common with the others,
1451 including object file formats, most assembler directives (often called
1452 @dfn{pseudo-ops}) and assembler syntax.@refill
1454 @cindex purpose of @sc{gnu} assembler
1455 @command{@value{AS}} is primarily intended to assemble the output of the
1456 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1457 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1458 assemble correctly everything that other assemblers for the same
1459 machine would assemble.
1461 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1464 @c This remark should appear in generic version of manual; assumption
1465 @c here is that generic version sets M680x0.
1466 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1467 assembler for the same architecture; for example, we know of several
1468 incompatible versions of 680x0 assembly language syntax.
1473 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1474 program in one pass of the source file. This has a subtle impact on the
1475 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1477 @node Object Formats
1478 @section Object File Formats
1480 @cindex object file format
1481 The @sc{gnu} assembler can be configured to produce several alternative
1482 object file formats. For the most part, this does not affect how you
1483 write assembly language programs; but directives for debugging symbols
1484 are typically different in different file formats. @xref{Symbol
1485 Attributes,,Symbol Attributes}.
1488 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1489 @value{OBJ-NAME} format object files.
1491 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1493 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1494 @code{b.out} or COFF format object files.
1497 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1498 SOM or ELF format object files.
1503 @section Command Line
1505 @cindex command line conventions
1507 After the program name @command{@value{AS}}, the command line may contain
1508 options and file names. Options may appear in any order, and may be
1509 before, after, or between file names. The order of file names is
1512 @cindex standard input, as input file
1514 @file{--} (two hyphens) by itself names the standard input file
1515 explicitly, as one of the files for @command{@value{AS}} to assemble.
1517 @cindex options, command line
1518 Except for @samp{--} any command line argument that begins with a
1519 hyphen (@samp{-}) is an option. Each option changes the behavior of
1520 @command{@value{AS}}. No option changes the way another option works. An
1521 option is a @samp{-} followed by one or more letters; the case of
1522 the letter is important. All options are optional.
1524 Some options expect exactly one file name to follow them. The file
1525 name may either immediately follow the option's letter (compatible
1526 with older assemblers) or it may be the next command argument (@sc{gnu}
1527 standard). These two command lines are equivalent:
1530 @value{AS} -o my-object-file.o mumble.s
1531 @value{AS} -omy-object-file.o mumble.s
1535 @section Input Files
1538 @cindex source program
1539 @cindex files, input
1540 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1541 describe the program input to one run of @command{@value{AS}}. The program may
1542 be in one or more files; how the source is partitioned into files
1543 doesn't change the meaning of the source.
1545 @c I added "con" prefix to "catenation" just to prove I can overcome my
1546 @c APL training... doc@cygnus.com
1547 The source program is a concatenation of the text in all the files, in the
1550 @c man begin DESCRIPTION
1551 Each time you run @command{@value{AS}} it assembles exactly one source
1552 program. The source program is made up of one or more files.
1553 (The standard input is also a file.)
1555 You give @command{@value{AS}} a command line that has zero or more input file
1556 names. The input files are read (from left file name to right). A
1557 command line argument (in any position) that has no special meaning
1558 is taken to be an input file name.
1560 If you give @command{@value{AS}} no file names it attempts to read one input file
1561 from the @command{@value{AS}} standard input, which is normally your terminal. You
1562 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1565 Use @samp{--} if you need to explicitly name the standard input file
1566 in your command line.
1568 If the source is empty, @command{@value{AS}} produces a small, empty object
1573 @subheading Filenames and Line-numbers
1575 @cindex input file linenumbers
1576 @cindex line numbers, in input files
1577 There are two ways of locating a line in the input file (or files) and
1578 either may be used in reporting error messages. One way refers to a line
1579 number in a physical file; the other refers to a line number in a
1580 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1582 @dfn{Physical files} are those files named in the command line given
1583 to @command{@value{AS}}.
1585 @dfn{Logical files} are simply names declared explicitly by assembler
1586 directives; they bear no relation to physical files. Logical file names help
1587 error messages reflect the original source file, when @command{@value{AS}} source
1588 is itself synthesized from other files. @command{@value{AS}} understands the
1589 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1590 @ref{File,,@code{.file}}.
1593 @section Output (Object) File
1599 Every time you run @command{@value{AS}} it produces an output file, which is
1600 your assembly language program translated into numbers. This file
1601 is the object file. Its default name is
1609 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1611 You can give it another name by using the @option{-o} option. Conventionally,
1612 object file names end with @file{.o}. The default name is used for historical
1613 reasons: older assemblers were capable of assembling self-contained programs
1614 directly into a runnable program. (For some formats, this isn't currently
1615 possible, but it can be done for the @code{a.out} format.)
1619 The object file is meant for input to the linker @code{@value{LD}}. It contains
1620 assembled program code, information to help @code{@value{LD}} integrate
1621 the assembled program into a runnable file, and (optionally) symbolic
1622 information for the debugger.
1624 @c link above to some info file(s) like the description of a.out.
1625 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1628 @section Error and Warning Messages
1630 @c man begin DESCRIPTION
1632 @cindex error messages
1633 @cindex warning messages
1634 @cindex messages from assembler
1635 @command{@value{AS}} may write warnings and error messages to the standard error
1636 file (usually your terminal). This should not happen when a compiler
1637 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1638 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1639 grave problem that stops the assembly.
1643 @cindex format of warning messages
1644 Warning messages have the format
1647 file_name:@b{NNN}:Warning Message Text
1651 @cindex line numbers, in warnings/errors
1652 (where @b{NNN} is a line number). If a logical file name has been given
1653 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1654 the current input file is used. If a logical line number was given
1656 (@pxref{Line,,@code{.line}})
1658 then it is used to calculate the number printed,
1659 otherwise the actual line in the current source file is printed. The
1660 message text is intended to be self explanatory (in the grand Unix
1663 @cindex format of error messages
1664 Error messages have the format
1666 file_name:@b{NNN}:FATAL:Error Message Text
1668 The file name and line number are derived as for warning
1669 messages. The actual message text may be rather less explanatory
1670 because many of them aren't supposed to happen.
1673 @chapter Command-Line Options
1675 @cindex options, all versions of assembler
1676 This chapter describes command-line options available in @emph{all}
1677 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
1678 for options specific
1680 to the @value{TARGET} target.
1683 to particular machine architectures.
1686 @c man begin DESCRIPTION
1688 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1689 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1690 The assembler arguments must be separated from each other (and the @samp{-Wa})
1691 by commas. For example:
1694 gcc -c -g -O -Wa,-alh,-L file.c
1698 This passes two options to the assembler: @samp{-alh} (emit a listing to
1699 standard output with high-level and assembly source) and @samp{-L} (retain
1700 local symbols in the symbol table).
1702 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1703 command-line options are automatically passed to the assembler by the compiler.
1704 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1705 precisely what options it passes to each compilation pass, including the
1711 * a:: -a[cdghlns] enable listings
1712 * alternate:: --alternate enable alternate macro syntax
1713 * D:: -D for compatibility
1714 * f:: -f to work faster
1715 * I:: -I for .include search path
1716 @ifclear DIFF-TBL-KLUGE
1717 * K:: -K for compatibility
1719 @ifset DIFF-TBL-KLUGE
1720 * K:: -K for difference tables
1723 * L:: -L to retain local symbols
1724 * listing:: --listing-XXX to configure listing output
1725 * M:: -M or --mri to assemble in MRI compatibility mode
1726 * MD:: --MD for dependency tracking
1727 * o:: -o to name the object file
1728 * R:: -R to join data and text sections
1729 * statistics:: --statistics to see statistics about assembly
1730 * traditional-format:: --traditional-format for compatible output
1731 * v:: -v to announce version
1732 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1733 * Z:: -Z to make object file even after errors
1737 @section Enable Listings: @option{-a[cdghlns]}
1747 @cindex listings, enabling
1748 @cindex assembly listings, enabling
1750 These options enable listing output from the assembler. By itself,
1751 @samp{-a} requests high-level, assembly, and symbols listing.
1752 You can use other letters to select specific options for the list:
1753 @samp{-ah} requests a high-level language listing,
1754 @samp{-al} requests an output-program assembly listing, and
1755 @samp{-as} requests a symbol table listing.
1756 High-level listings require that a compiler debugging option like
1757 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1760 Use the @samp{-ag} option to print a first section with general assembly
1761 information, like @value{AS} version, switches passed, or time stamp.
1763 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1764 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1765 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1766 omitted from the listing.
1768 Use the @samp{-ad} option to omit debugging directives from the
1771 Once you have specified one of these options, you can further control
1772 listing output and its appearance using the directives @code{.list},
1773 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1775 The @samp{-an} option turns off all forms processing.
1776 If you do not request listing output with one of the @samp{-a} options, the
1777 listing-control directives have no effect.
1779 The letters after @samp{-a} may be combined into one option,
1780 @emph{e.g.}, @samp{-aln}.
1782 Note if the assembler source is coming from the standard input (e.g.,
1784 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1785 is being used) then the listing will not contain any comments or preprocessor
1786 directives. This is because the listing code buffers input source lines from
1787 stdin only after they have been preprocessed by the assembler. This reduces
1788 memory usage and makes the code more efficient.
1791 @section @option{--alternate}
1794 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
1797 @section @option{-D}
1800 This option has no effect whatsoever, but it is accepted to make it more
1801 likely that scripts written for other assemblers also work with
1802 @command{@value{AS}}.
1805 @section Work Faster: @option{-f}
1808 @cindex trusted compiler
1809 @cindex faster processing (@option{-f})
1810 @samp{-f} should only be used when assembling programs written by a
1811 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1812 and comment preprocessing on
1813 the input file(s) before assembling them. @xref{Preprocessing,
1817 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1818 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1823 @section @code{.include} Search Path: @option{-I} @var{path}
1825 @kindex -I @var{path}
1826 @cindex paths for @code{.include}
1827 @cindex search path for @code{.include}
1828 @cindex @code{include} directive search path
1829 Use this option to add a @var{path} to the list of directories
1830 @command{@value{AS}} searches for files specified in @code{.include}
1831 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1832 many times as necessary to include a variety of paths. The current
1833 working directory is always searched first; after that, @command{@value{AS}}
1834 searches any @samp{-I} directories in the same order as they were
1835 specified (left to right) on the command line.
1838 @section Difference Tables: @option{-K}
1841 @ifclear DIFF-TBL-KLUGE
1842 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1843 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1844 where it can be used to warn when the assembler alters the machine code
1845 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1846 family does not have the addressing limitations that sometimes lead to this
1847 alteration on other platforms.
1850 @ifset DIFF-TBL-KLUGE
1851 @cindex difference tables, warning
1852 @cindex warning for altered difference tables
1853 @command{@value{AS}} sometimes alters the code emitted for directives of the
1854 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
1855 You can use the @samp{-K} option if you want a warning issued when this
1860 @section Include Local Symbols: @option{-L}
1863 @cindex local symbols, retaining in output
1864 Symbols beginning with system-specific local label prefixes, typically
1865 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
1866 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
1867 such symbols when debugging, because they are intended for the use of
1868 programs (like compilers) that compose assembler programs, not for your
1869 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
1870 such symbols, so you do not normally debug with them.
1872 This option tells @command{@value{AS}} to retain those local symbols
1873 in the object file. Usually if you do this you also tell the linker
1874 @code{@value{LD}} to preserve those symbols.
1877 @section Configuring listing output: @option{--listing}
1879 The listing feature of the assembler can be enabled via the command line switch
1880 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1881 hex dump of the corresponding locations in the output object file, and displays
1882 them as a listing file. The format of this listing can be controlled by
1883 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
1884 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
1885 @code{.psize} (@pxref{Psize}), and
1886 @code{.eject} (@pxref{Eject}) and also by the following switches:
1889 @item --listing-lhs-width=@samp{number}
1890 @kindex --listing-lhs-width
1891 @cindex Width of first line disassembly output
1892 Sets the maximum width, in words, of the first line of the hex byte dump. This
1893 dump appears on the left hand side of the listing output.
1895 @item --listing-lhs-width2=@samp{number}
1896 @kindex --listing-lhs-width2
1897 @cindex Width of continuation lines of disassembly output
1898 Sets the maximum width, in words, of any further lines of the hex byte dump for
1899 a given input source line. If this value is not specified, it defaults to being
1900 the same as the value specified for @samp{--listing-lhs-width}. If neither
1901 switch is used the default is to one.
1903 @item --listing-rhs-width=@samp{number}
1904 @kindex --listing-rhs-width
1905 @cindex Width of source line output
1906 Sets the maximum width, in characters, of the source line that is displayed
1907 alongside the hex dump. The default value for this parameter is 100. The
1908 source line is displayed on the right hand side of the listing output.
1910 @item --listing-cont-lines=@samp{number}
1911 @kindex --listing-cont-lines
1912 @cindex Maximum number of continuation lines
1913 Sets the maximum number of continuation lines of hex dump that will be
1914 displayed for a given single line of source input. The default value is 4.
1918 @section Assemble in MRI Compatibility Mode: @option{-M}
1921 @cindex MRI compatibility mode
1922 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
1923 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
1924 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1925 configured target) assembler from Microtec Research. The exact nature of the
1926 MRI syntax will not be documented here; see the MRI manuals for more
1927 information. Note in particular that the handling of macros and macro
1928 arguments is somewhat different. The purpose of this option is to permit
1929 assembling existing MRI assembler code using @command{@value{AS}}.
1931 The MRI compatibility is not complete. Certain operations of the MRI assembler
1932 depend upon its object file format, and can not be supported using other object
1933 file formats. Supporting these would require enhancing each object file format
1934 individually. These are:
1937 @item global symbols in common section
1939 The m68k MRI assembler supports common sections which are merged by the linker.
1940 Other object file formats do not support this. @command{@value{AS}} handles
1941 common sections by treating them as a single common symbol. It permits local
1942 symbols to be defined within a common section, but it can not support global
1943 symbols, since it has no way to describe them.
1945 @item complex relocations
1947 The MRI assemblers support relocations against a negated section address, and
1948 relocations which combine the start addresses of two or more sections. These
1949 are not support by other object file formats.
1951 @item @code{END} pseudo-op specifying start address
1953 The MRI @code{END} pseudo-op permits the specification of a start address.
1954 This is not supported by other object file formats. The start address may
1955 instead be specified using the @option{-e} option to the linker, or in a linker
1958 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1960 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1961 name to the output file. This is not supported by other object file formats.
1963 @item @code{ORG} pseudo-op
1965 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1966 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
1967 which changes the location within the current section. Absolute sections are
1968 not supported by other object file formats. The address of a section may be
1969 assigned within a linker script.
1972 There are some other features of the MRI assembler which are not supported by
1973 @command{@value{AS}}, typically either because they are difficult or because they
1974 seem of little consequence. Some of these may be supported in future releases.
1978 @item EBCDIC strings
1980 EBCDIC strings are not supported.
1982 @item packed binary coded decimal
1984 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1985 and @code{DCB.P} pseudo-ops are not supported.
1987 @item @code{FEQU} pseudo-op
1989 The m68k @code{FEQU} pseudo-op is not supported.
1991 @item @code{NOOBJ} pseudo-op
1993 The m68k @code{NOOBJ} pseudo-op is not supported.
1995 @item @code{OPT} branch control options
1997 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1998 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
1999 relaxes all branches, whether forward or backward, to an appropriate size, so
2000 these options serve no purpose.
2002 @item @code{OPT} list control options
2004 The following m68k @code{OPT} list control options are ignored: @code{C},
2005 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2006 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2008 @item other @code{OPT} options
2010 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2011 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2013 @item @code{OPT} @code{D} option is default
2015 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2016 @code{OPT NOD} may be used to turn it off.
2018 @item @code{XREF} pseudo-op.
2020 The m68k @code{XREF} pseudo-op is ignored.
2022 @item @code{.debug} pseudo-op
2024 The i960 @code{.debug} pseudo-op is not supported.
2026 @item @code{.extended} pseudo-op
2028 The i960 @code{.extended} pseudo-op is not supported.
2030 @item @code{.list} pseudo-op.
2032 The various options of the i960 @code{.list} pseudo-op are not supported.
2034 @item @code{.optimize} pseudo-op
2036 The i960 @code{.optimize} pseudo-op is not supported.
2038 @item @code{.output} pseudo-op
2040 The i960 @code{.output} pseudo-op is not supported.
2042 @item @code{.setreal} pseudo-op
2044 The i960 @code{.setreal} pseudo-op is not supported.
2049 @section Dependency Tracking: @option{--MD}
2052 @cindex dependency tracking
2055 @command{@value{AS}} can generate a dependency file for the file it creates. This
2056 file consists of a single rule suitable for @code{make} describing the
2057 dependencies of the main source file.
2059 The rule is written to the file named in its argument.
2061 This feature is used in the automatic updating of makefiles.
2064 @section Name the Object File: @option{-o}
2067 @cindex naming object file
2068 @cindex object file name
2069 There is always one object file output when you run @command{@value{AS}}. By
2070 default it has the name
2073 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2087 You use this option (which takes exactly one filename) to give the
2088 object file a different name.
2090 Whatever the object file is called, @command{@value{AS}} overwrites any
2091 existing file of the same name.
2094 @section Join Data and Text Sections: @option{-R}
2097 @cindex data and text sections, joining
2098 @cindex text and data sections, joining
2099 @cindex joining text and data sections
2100 @cindex merging text and data sections
2101 @option{-R} tells @command{@value{AS}} to write the object file as if all
2102 data-section data lives in the text section. This is only done at
2103 the very last moment: your binary data are the same, but data
2104 section parts are relocated differently. The data section part of
2105 your object file is zero bytes long because all its bytes are
2106 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2108 When you specify @option{-R} it would be possible to generate shorter
2109 address displacements (because we do not have to cross between text and
2110 data section). We refrain from doing this simply for compatibility with
2111 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2114 When @command{@value{AS}} is configured for COFF or ELF output,
2115 this option is only useful if you use sections named @samp{.text} and
2120 @option{-R} is not supported for any of the HPPA targets. Using
2121 @option{-R} generates a warning from @command{@value{AS}}.
2125 @section Display Assembly Statistics: @option{--statistics}
2127 @kindex --statistics
2128 @cindex statistics, about assembly
2129 @cindex time, total for assembly
2130 @cindex space used, maximum for assembly
2131 Use @samp{--statistics} to display two statistics about the resources used by
2132 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2133 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2136 @node traditional-format
2137 @section Compatible Output: @option{--traditional-format}
2139 @kindex --traditional-format
2140 For some targets, the output of @command{@value{AS}} is different in some ways
2141 from the output of some existing assembler. This switch requests
2142 @command{@value{AS}} to use the traditional format instead.
2144 For example, it disables the exception frame optimizations which
2145 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2148 @section Announce Version: @option{-v}
2152 @cindex assembler version
2153 @cindex version of assembler
2154 You can find out what version of as is running by including the
2155 option @samp{-v} (which you can also spell as @samp{-version}) on the
2159 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2161 @command{@value{AS}} should never give a warning or error message when
2162 assembling compiler output. But programs written by people often
2163 cause @command{@value{AS}} to give a warning that a particular assumption was
2164 made. All such warnings are directed to the standard error file.
2168 @cindex suppressing warnings
2169 @cindex warnings, suppressing
2170 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2171 This only affects the warning messages: it does not change any particular of
2172 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2175 @kindex --fatal-warnings
2176 @cindex errors, caused by warnings
2177 @cindex warnings, causing error
2178 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2179 files that generate warnings to be in error.
2182 @cindex warnings, switching on
2183 You can switch these options off again by specifying @option{--warn}, which
2184 causes warnings to be output as usual.
2187 @section Generate Object File in Spite of Errors: @option{-Z}
2188 @cindex object file, after errors
2189 @cindex errors, continuing after
2190 After an error message, @command{@value{AS}} normally produces no output. If for
2191 some reason you are interested in object file output even after
2192 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2193 option. If there are any errors, @command{@value{AS}} continues anyways, and
2194 writes an object file after a final warning message of the form @samp{@var{n}
2195 errors, @var{m} warnings, generating bad object file.}
2200 @cindex machine-independent syntax
2201 @cindex syntax, machine-independent
2202 This chapter describes the machine-independent syntax allowed in a
2203 source file. @command{@value{AS}} syntax is similar to what many other
2204 assemblers use; it is inspired by the BSD 4.2
2209 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2213 * Preprocessing:: Preprocessing
2214 * Whitespace:: Whitespace
2215 * Comments:: Comments
2216 * Symbol Intro:: Symbols
2217 * Statements:: Statements
2218 * Constants:: Constants
2222 @section Preprocessing
2224 @cindex preprocessing
2225 The @command{@value{AS}} internal preprocessor:
2227 @cindex whitespace, removed by preprocessor
2229 adjusts and removes extra whitespace. It leaves one space or tab before
2230 the keywords on a line, and turns any other whitespace on the line into
2233 @cindex comments, removed by preprocessor
2235 removes all comments, replacing them with a single space, or an
2236 appropriate number of newlines.
2238 @cindex constants, converted by preprocessor
2240 converts character constants into the appropriate numeric values.
2243 It does not do macro processing, include file handling, or
2244 anything else you may get from your C compiler's preprocessor. You can
2245 do include file processing with the @code{.include} directive
2246 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2247 to get other ``CPP'' style preprocessing by giving the input file a
2248 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2249 Output, gcc.info, Using GNU CC}.
2251 Excess whitespace, comments, and character constants
2252 cannot be used in the portions of the input text that are not
2255 @cindex turning preprocessing on and off
2256 @cindex preprocessing, turning on and off
2259 If the first line of an input file is @code{#NO_APP} or if you use the
2260 @samp{-f} option, whitespace and comments are not removed from the input file.
2261 Within an input file, you can ask for whitespace and comment removal in
2262 specific portions of the by putting a line that says @code{#APP} before the
2263 text that may contain whitespace or comments, and putting a line that says
2264 @code{#NO_APP} after this text. This feature is mainly intend to support
2265 @code{asm} statements in compilers whose output is otherwise free of comments
2272 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2273 Whitespace is used to separate symbols, and to make programs neater for
2274 people to read. Unless within character constants
2275 (@pxref{Characters,,Character Constants}), any whitespace means the same
2276 as exactly one space.
2282 There are two ways of rendering comments to @command{@value{AS}}. In both
2283 cases the comment is equivalent to one space.
2285 Anything from @samp{/*} through the next @samp{*/} is a comment.
2286 This means you may not nest these comments.
2290 The only way to include a newline ('\n') in a comment
2291 is to use this sort of comment.
2294 /* This sort of comment does not nest. */
2297 @cindex line comment character
2298 Anything from the @dfn{line comment} character to the next newline
2299 is considered a comment and is ignored. The line comment character is
2301 @samp{;} on the ARC;
2304 @samp{@@} on the ARM;
2307 @samp{;} for the H8/300 family;
2310 @samp{;} for the HPPA;
2313 @samp{#} on the i386 and x86-64;
2316 @samp{#} on the i960;
2319 @samp{;} for the PDP-11;
2322 @samp{;} for picoJava;
2325 @samp{#} for Motorola PowerPC;
2328 @samp{#} for IBM S/390;
2331 @samp{#} for the Sunplus SCORE;
2334 @samp{!} for the Renesas / SuperH SH;
2337 @samp{!} on the SPARC;
2340 @samp{#} on the ip2k;
2343 @samp{#} on the m32c;
2346 @samp{#} on the m32r;
2349 @samp{|} on the 680x0;
2352 @samp{#} on the 68HC11 and 68HC12;
2358 @samp{;} on the TMS320C6X;
2361 @samp{#} on the Vax;
2364 @samp{;} for the Z80;
2367 @samp{!} for the Z8000;
2370 @samp{#} on the V850;
2373 @samp{#} for Xtensa systems;
2375 see @ref{Machine Dependencies}. @refill
2376 @c FIXME What about i860?
2379 On some machines there are two different line comment characters. One
2380 character only begins a comment if it is the first non-whitespace character on
2381 a line, while the other always begins a comment.
2385 The V850 assembler also supports a double dash as starting a comment that
2386 extends to the end of the line.
2392 @cindex lines starting with @code{#}
2393 @cindex logical line numbers
2394 To be compatible with past assemblers, lines that begin with @samp{#} have a
2395 special interpretation. Following the @samp{#} should be an absolute
2396 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2397 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2398 new logical file name. The rest of the line, if any, should be whitespace.
2400 If the first non-whitespace characters on the line are not numeric,
2401 the line is ignored. (Just like a comment.)
2404 # This is an ordinary comment.
2405 # 42-6 "new_file_name" # New logical file name
2406 # This is logical line # 36.
2408 This feature is deprecated, and may disappear from future versions
2409 of @command{@value{AS}}.
2414 @cindex characters used in symbols
2415 @ifclear SPECIAL-SYMS
2416 A @dfn{symbol} is one or more characters chosen from the set of all
2417 letters (both upper and lower case), digits and the three characters
2423 A @dfn{symbol} is one or more characters chosen from the set of all
2424 letters (both upper and lower case), digits and the three characters
2425 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2431 On most machines, you can also use @code{$} in symbol names; exceptions
2432 are noted in @ref{Machine Dependencies}.
2434 No symbol may begin with a digit. Case is significant.
2435 There is no length limit: all characters are significant. Symbols are
2436 delimited by characters not in that set, or by the beginning of a file
2437 (since the source program must end with a newline, the end of a file is
2438 not a possible symbol delimiter). @xref{Symbols}.
2439 @cindex length of symbols
2444 @cindex statements, structure of
2445 @cindex line separator character
2446 @cindex statement separator character
2448 @ifclear abnormal-separator
2449 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
2450 semicolon (@samp{;}). The newline or semicolon is considered part of
2451 the preceding statement. Newlines and semicolons within character
2452 constants are an exception: they do not end statements.
2454 @ifset abnormal-separator
2456 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
2457 point (@samp{!}). The newline or exclamation point is considered part of the
2458 preceding statement. Newlines and exclamation points within character
2459 constants are an exception: they do not end statements.
2462 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
2463 H8/300) a dollar sign (@samp{$}); or (for the Renesas-SH) a semicolon
2464 (@samp{;}). The newline or separator character is considered part of
2465 the preceding statement. Newlines and separators within character
2466 constants are an exception: they do not end statements.
2471 A @dfn{statement} ends at a newline character (@samp{\n}) or line
2472 separator character. (The line separator is usually @samp{;}, unless this
2473 conflicts with the comment character; see @ref{Machine Dependencies}.) The
2474 newline or separator character is considered part of the preceding
2475 statement. Newlines and separators within character constants are an
2476 exception: they do not end statements.
2479 @cindex newline, required at file end
2480 @cindex EOF, newline must precede
2481 It is an error to end any statement with end-of-file: the last
2482 character of any input file should be a newline.@refill
2484 An empty statement is allowed, and may include whitespace. It is ignored.
2486 @cindex instructions and directives
2487 @cindex directives and instructions
2488 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2489 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2491 A statement begins with zero or more labels, optionally followed by a
2492 key symbol which determines what kind of statement it is. The key
2493 symbol determines the syntax of the rest of the statement. If the
2494 symbol begins with a dot @samp{.} then the statement is an assembler
2495 directive: typically valid for any computer. If the symbol begins with
2496 a letter the statement is an assembly language @dfn{instruction}: it
2497 assembles into a machine language instruction.
2499 Different versions of @command{@value{AS}} for different computers
2500 recognize different instructions. In fact, the same symbol may
2501 represent a different instruction in a different computer's assembly
2505 @cindex @code{:} (label)
2506 @cindex label (@code{:})
2507 A label is a symbol immediately followed by a colon (@code{:}).
2508 Whitespace before a label or after a colon is permitted, but you may not
2509 have whitespace between a label's symbol and its colon. @xref{Labels}.
2512 For HPPA targets, labels need not be immediately followed by a colon, but
2513 the definition of a label must begin in column zero. This also implies that
2514 only one label may be defined on each line.
2518 label: .directive followed by something
2519 another_label: # This is an empty statement.
2520 instruction operand_1, operand_2, @dots{}
2527 A constant is a number, written so that its value is known by
2528 inspection, without knowing any context. Like this:
2531 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2532 .ascii "Ring the bell\7" # A string constant.
2533 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2534 .float 0f-314159265358979323846264338327\
2535 95028841971.693993751E-40 # - pi, a flonum.
2540 * Characters:: Character Constants
2541 * Numbers:: Number Constants
2545 @subsection Character Constants
2547 @cindex character constants
2548 @cindex constants, character
2549 There are two kinds of character constants. A @dfn{character} stands
2550 for one character in one byte and its value may be used in
2551 numeric expressions. String constants (properly called string
2552 @emph{literals}) are potentially many bytes and their values may not be
2553 used in arithmetic expressions.
2557 * Chars:: Characters
2561 @subsubsection Strings
2563 @cindex string constants
2564 @cindex constants, string
2565 A @dfn{string} is written between double-quotes. It may contain
2566 double-quotes or null characters. The way to get special characters
2567 into a string is to @dfn{escape} these characters: precede them with
2568 a backslash @samp{\} character. For example @samp{\\} represents
2569 one backslash: the first @code{\} is an escape which tells
2570 @command{@value{AS}} to interpret the second character literally as a backslash
2571 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2572 escape character). The complete list of escapes follows.
2574 @cindex escape codes, character
2575 @cindex character escape codes
2578 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2580 @cindex @code{\b} (backspace character)
2581 @cindex backspace (@code{\b})
2583 Mnemonic for backspace; for ASCII this is octal code 010.
2586 @c Mnemonic for EOText; for ASCII this is octal code 004.
2588 @cindex @code{\f} (formfeed character)
2589 @cindex formfeed (@code{\f})
2591 Mnemonic for FormFeed; for ASCII this is octal code 014.
2593 @cindex @code{\n} (newline character)
2594 @cindex newline (@code{\n})
2596 Mnemonic for newline; for ASCII this is octal code 012.
2599 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2601 @cindex @code{\r} (carriage return character)
2602 @cindex carriage return (@code{\r})
2604 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2607 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2608 @c other assemblers.
2610 @cindex @code{\t} (tab)
2611 @cindex tab (@code{\t})
2613 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2616 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2617 @c @item \x @var{digit} @var{digit} @var{digit}
2618 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2620 @cindex @code{\@var{ddd}} (octal character code)
2621 @cindex octal character code (@code{\@var{ddd}})
2622 @item \ @var{digit} @var{digit} @var{digit}
2623 An octal character code. The numeric code is 3 octal digits.
2624 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2625 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2627 @cindex @code{\@var{xd...}} (hex character code)
2628 @cindex hex character code (@code{\@var{xd...}})
2629 @item \@code{x} @var{hex-digits...}
2630 A hex character code. All trailing hex digits are combined. Either upper or
2631 lower case @code{x} works.
2633 @cindex @code{\\} (@samp{\} character)
2634 @cindex backslash (@code{\\})
2636 Represents one @samp{\} character.
2639 @c Represents one @samp{'} (accent acute) character.
2640 @c This is needed in single character literals
2641 @c (@xref{Characters,,Character Constants}.) to represent
2644 @cindex @code{\"} (doublequote character)
2645 @cindex doublequote (@code{\"})
2647 Represents one @samp{"} character. Needed in strings to represent
2648 this character, because an unescaped @samp{"} would end the string.
2650 @item \ @var{anything-else}
2651 Any other character when escaped by @kbd{\} gives a warning, but
2652 assembles as if the @samp{\} was not present. The idea is that if
2653 you used an escape sequence you clearly didn't want the literal
2654 interpretation of the following character. However @command{@value{AS}} has no
2655 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2656 code and warns you of the fact.
2659 Which characters are escapable, and what those escapes represent,
2660 varies widely among assemblers. The current set is what we think
2661 the BSD 4.2 assembler recognizes, and is a subset of what most C
2662 compilers recognize. If you are in doubt, do not use an escape
2666 @subsubsection Characters
2668 @cindex single character constant
2669 @cindex character, single
2670 @cindex constant, single character
2671 A single character may be written as a single quote immediately
2672 followed by that character. The same escapes apply to characters as
2673 to strings. So if you want to write the character backslash, you
2674 must write @kbd{'\\} where the first @code{\} escapes the second
2675 @code{\}. As you can see, the quote is an acute accent, not a
2676 grave accent. A newline
2678 @ifclear abnormal-separator
2679 (or semicolon @samp{;})
2681 @ifset abnormal-separator
2683 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2688 immediately following an acute accent is taken as a literal character
2689 and does not count as the end of a statement. The value of a character
2690 constant in a numeric expression is the machine's byte-wide code for
2691 that character. @command{@value{AS}} assumes your character code is ASCII:
2692 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2695 @subsection Number Constants
2697 @cindex constants, number
2698 @cindex number constants
2699 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2700 are stored in the target machine. @emph{Integers} are numbers that
2701 would fit into an @code{int} in the C language. @emph{Bignums} are
2702 integers, but they are stored in more than 32 bits. @emph{Flonums}
2703 are floating point numbers, described below.
2706 * Integers:: Integers
2711 * Bit Fields:: Bit Fields
2717 @subsubsection Integers
2719 @cindex constants, integer
2721 @cindex binary integers
2722 @cindex integers, binary
2723 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2724 the binary digits @samp{01}.
2726 @cindex octal integers
2727 @cindex integers, octal
2728 An octal integer is @samp{0} followed by zero or more of the octal
2729 digits (@samp{01234567}).
2731 @cindex decimal integers
2732 @cindex integers, decimal
2733 A decimal integer starts with a non-zero digit followed by zero or
2734 more digits (@samp{0123456789}).
2736 @cindex hexadecimal integers
2737 @cindex integers, hexadecimal
2738 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2739 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2741 Integers have the usual values. To denote a negative integer, use
2742 the prefix operator @samp{-} discussed under expressions
2743 (@pxref{Prefix Ops,,Prefix Operators}).
2746 @subsubsection Bignums
2749 @cindex constants, bignum
2750 A @dfn{bignum} has the same syntax and semantics as an integer
2751 except that the number (or its negative) takes more than 32 bits to
2752 represent in binary. The distinction is made because in some places
2753 integers are permitted while bignums are not.
2756 @subsubsection Flonums
2758 @cindex floating point numbers
2759 @cindex constants, floating point
2761 @cindex precision, floating point
2762 A @dfn{flonum} represents a floating point number. The translation is
2763 indirect: a decimal floating point number from the text is converted by
2764 @command{@value{AS}} to a generic binary floating point number of more than
2765 sufficient precision. This generic floating point number is converted
2766 to a particular computer's floating point format (or formats) by a
2767 portion of @command{@value{AS}} specialized to that computer.
2769 A flonum is written by writing (in order)
2774 (@samp{0} is optional on the HPPA.)
2778 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2780 @kbd{e} is recommended. Case is not important.
2782 @c FIXME: verify if flonum syntax really this vague for most cases
2783 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2784 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2787 On the H8/300, Renesas / SuperH SH,
2788 and AMD 29K architectures, the letter must be
2789 one of the letters @samp{DFPRSX} (in upper or lower case).
2791 On the ARC, the letter must be one of the letters @samp{DFRS}
2792 (in upper or lower case).
2794 On the Intel 960 architecture, the letter must be
2795 one of the letters @samp{DFT} (in upper or lower case).
2797 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2801 One of the letters @samp{DFRS} (in upper or lower case).
2804 One of the letters @samp{DFPRSX} (in upper or lower case).
2807 The letter @samp{E} (upper case only).
2810 One of the letters @samp{DFT} (in upper or lower case).
2815 An optional sign: either @samp{+} or @samp{-}.
2818 An optional @dfn{integer part}: zero or more decimal digits.
2821 An optional @dfn{fractional part}: @samp{.} followed by zero
2822 or more decimal digits.
2825 An optional exponent, consisting of:
2829 An @samp{E} or @samp{e}.
2830 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2831 @c principle this can perfectly well be different on different targets.
2833 Optional sign: either @samp{+} or @samp{-}.
2835 One or more decimal digits.
2840 At least one of the integer part or the fractional part must be
2841 present. The floating point number has the usual base-10 value.
2843 @command{@value{AS}} does all processing using integers. Flonums are computed
2844 independently of any floating point hardware in the computer running
2845 @command{@value{AS}}.
2849 @c Bit fields are written as a general facility but are also controlled
2850 @c by a conditional-compilation flag---which is as of now (21mar91)
2851 @c turned on only by the i960 config of GAS.
2853 @subsubsection Bit Fields
2856 @cindex constants, bit field
2857 You can also define numeric constants as @dfn{bit fields}.
2858 Specify two numbers separated by a colon---
2860 @var{mask}:@var{value}
2863 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2866 The resulting number is then packed
2868 @c this conditional paren in case bit fields turned on elsewhere than 960
2869 (in host-dependent byte order)
2871 into a field whose width depends on which assembler directive has the
2872 bit-field as its argument. Overflow (a result from the bitwise and
2873 requiring more binary digits to represent) is not an error; instead,
2874 more constants are generated, of the specified width, beginning with the
2875 least significant digits.@refill
2877 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2878 @code{.short}, and @code{.word} accept bit-field arguments.
2883 @chapter Sections and Relocation
2888 * Secs Background:: Background
2889 * Ld Sections:: Linker Sections
2890 * As Sections:: Assembler Internal Sections
2891 * Sub-Sections:: Sub-Sections
2895 @node Secs Background
2898 Roughly, a section is a range of addresses, with no gaps; all data
2899 ``in'' those addresses is treated the same for some particular purpose.
2900 For example there may be a ``read only'' section.
2902 @cindex linker, and assembler
2903 @cindex assembler, and linker
2904 The linker @code{@value{LD}} reads many object files (partial programs) and
2905 combines their contents to form a runnable program. When @command{@value{AS}}
2906 emits an object file, the partial program is assumed to start at address 0.
2907 @code{@value{LD}} assigns the final addresses for the partial program, so that
2908 different partial programs do not overlap. This is actually an
2909 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2912 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2913 addresses. These blocks slide to their run-time addresses as rigid
2914 units; their length does not change and neither does the order of bytes
2915 within them. Such a rigid unit is called a @emph{section}. Assigning
2916 run-time addresses to sections is called @dfn{relocation}. It includes
2917 the task of adjusting mentions of object-file addresses so they refer to
2918 the proper run-time addresses.
2920 For the H8/300, and for the Renesas / SuperH SH,
2921 @command{@value{AS}} pads sections if needed to
2922 ensure they end on a word (sixteen bit) boundary.
2925 @cindex standard assembler sections
2926 An object file written by @command{@value{AS}} has at least three sections, any
2927 of which may be empty. These are named @dfn{text}, @dfn{data} and
2932 When it generates COFF or ELF output,
2934 @command{@value{AS}} can also generate whatever other named sections you specify
2935 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2936 If you do not use any directives that place output in the @samp{.text}
2937 or @samp{.data} sections, these sections still exist, but are empty.
2942 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2944 @command{@value{AS}} can also generate whatever other named sections you
2945 specify using the @samp{.space} and @samp{.subspace} directives. See
2946 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2947 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2948 assembler directives.
2951 Additionally, @command{@value{AS}} uses different names for the standard
2952 text, data, and bss sections when generating SOM output. Program text
2953 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2954 BSS into @samp{$BSS$}.
2958 Within the object file, the text section starts at address @code{0}, the
2959 data section follows, and the bss section follows the data section.
2962 When generating either SOM or ELF output files on the HPPA, the text
2963 section starts at address @code{0}, the data section at address
2964 @code{0x4000000}, and the bss section follows the data section.
2967 To let @code{@value{LD}} know which data changes when the sections are
2968 relocated, and how to change that data, @command{@value{AS}} also writes to the
2969 object file details of the relocation needed. To perform relocation
2970 @code{@value{LD}} must know, each time an address in the object
2974 Where in the object file is the beginning of this reference to
2977 How long (in bytes) is this reference?
2979 Which section does the address refer to? What is the numeric value of
2981 (@var{address}) @minus{} (@var{start-address of section})?
2984 Is the reference to an address ``Program-Counter relative''?
2987 @cindex addresses, format of
2988 @cindex section-relative addressing
2989 In fact, every address @command{@value{AS}} ever uses is expressed as
2991 (@var{section}) + (@var{offset into section})
2994 Further, most expressions @command{@value{AS}} computes have this section-relative
2997 (For some object formats, such as SOM for the HPPA, some expressions are
2998 symbol-relative instead.)
3001 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3002 @var{N} into section @var{secname}.''
3004 Apart from text, data and bss sections you need to know about the
3005 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3006 addresses in the absolute section remain unchanged. For example, address
3007 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3008 @code{@value{LD}}. Although the linker never arranges two partial programs'
3009 data sections with overlapping addresses after linking, @emph{by definition}
3010 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3011 part of a program is always the same address when the program is running as
3012 address @code{@{absolute@ 239@}} in any other part of the program.
3014 The idea of sections is extended to the @dfn{undefined} section. Any
3015 address whose section is unknown at assembly time is by definition
3016 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3017 Since numbers are always defined, the only way to generate an undefined
3018 address is to mention an undefined symbol. A reference to a named
3019 common block would be such a symbol: its value is unknown at assembly
3020 time so it has section @emph{undefined}.
3022 By analogy the word @emph{section} is used to describe groups of sections in
3023 the linked program. @code{@value{LD}} puts all partial programs' text
3024 sections in contiguous addresses in the linked program. It is
3025 customary to refer to the @emph{text section} of a program, meaning all
3026 the addresses of all partial programs' text sections. Likewise for
3027 data and bss sections.
3029 Some sections are manipulated by @code{@value{LD}}; others are invented for
3030 use of @command{@value{AS}} and have no meaning except during assembly.
3033 @section Linker Sections
3034 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3039 @cindex named sections
3040 @cindex sections, named
3041 @item named sections
3044 @cindex text section
3045 @cindex data section
3049 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3050 separate but equal sections. Anything you can say of one section is
3053 When the program is running, however, it is
3054 customary for the text section to be unalterable. The
3055 text section is often shared among processes: it contains
3056 instructions, constants and the like. The data section of a running
3057 program is usually alterable: for example, C variables would be stored
3058 in the data section.
3063 This section contains zeroed bytes when your program begins running. It
3064 is used to hold uninitialized variables or common storage. The length of
3065 each partial program's bss section is important, but because it starts
3066 out containing zeroed bytes there is no need to store explicit zero
3067 bytes in the object file. The bss section was invented to eliminate
3068 those explicit zeros from object files.
3070 @cindex absolute section
3071 @item absolute section
3072 Address 0 of this section is always ``relocated'' to runtime address 0.
3073 This is useful if you want to refer to an address that @code{@value{LD}} must
3074 not change when relocating. In this sense we speak of absolute
3075 addresses being ``unrelocatable'': they do not change during relocation.
3077 @cindex undefined section
3078 @item undefined section
3079 This ``section'' is a catch-all for address references to objects not in
3080 the preceding sections.
3081 @c FIXME: ref to some other doc on obj-file formats could go here.
3084 @cindex relocation example
3085 An idealized example of three relocatable sections follows.
3087 The example uses the traditional section names @samp{.text} and @samp{.data}.
3089 Memory addresses are on the horizontal axis.
3093 @c END TEXI2ROFF-KILL
3096 partial program # 1: |ttttt|dddd|00|
3103 partial program # 2: |TTT|DDD|000|
3106 +--+---+-----+--+----+---+-----+~~
3107 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3108 +--+---+-----+--+----+---+-----+~~
3110 addresses: 0 @dots{}
3117 \line{\it Partial program \#1: \hfil}
3118 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3119 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3121 \line{\it Partial program \#2: \hfil}
3122 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3123 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3125 \line{\it linked program: \hfil}
3126 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3127 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3128 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3129 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3131 \line{\it addresses: \hfil}
3135 @c END TEXI2ROFF-KILL
3138 @section Assembler Internal Sections
3140 @cindex internal assembler sections
3141 @cindex sections in messages, internal
3142 These sections are meant only for the internal use of @command{@value{AS}}. They
3143 have no meaning at run-time. You do not really need to know about these
3144 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3145 warning messages, so it might be helpful to have an idea of their
3146 meanings to @command{@value{AS}}. These sections are used to permit the
3147 value of every expression in your assembly language program to be a
3148 section-relative address.
3151 @cindex assembler internal logic error
3152 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3153 An internal assembler logic error has been found. This means there is a
3154 bug in the assembler.
3156 @cindex expr (internal section)
3158 The assembler stores complex expression internally as combinations of
3159 symbols. When it needs to represent an expression as a symbol, it puts
3160 it in the expr section.
3162 @c FIXME item transfer[t] vector preload
3163 @c FIXME item transfer[t] vector postload
3164 @c FIXME item register
3168 @section Sub-Sections
3170 @cindex numbered subsections
3171 @cindex grouping data
3177 fall into two sections: text and data.
3179 You may have separate groups of
3181 data in named sections
3185 data in named sections
3191 that you want to end up near to each other in the object file, even though they
3192 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3193 use @dfn{subsections} for this purpose. Within each section, there can be
3194 numbered subsections with values from 0 to 8192. Objects assembled into the
3195 same subsection go into the object file together with other objects in the same
3196 subsection. For example, a compiler might want to store constants in the text
3197 section, but might not want to have them interspersed with the program being
3198 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3199 section of code being output, and a @samp{.text 1} before each group of
3200 constants being output.
3202 Subsections are optional. If you do not use subsections, everything
3203 goes in subsection number zero.
3206 Each subsection is zero-padded up to a multiple of four bytes.
3207 (Subsections may be padded a different amount on different flavors
3208 of @command{@value{AS}}.)
3212 On the H8/300 platform, each subsection is zero-padded to a word
3213 boundary (two bytes).
3214 The same is true on the Renesas SH.
3217 @c FIXME section padding (alignment)?
3218 @c Rich Pixley says padding here depends on target obj code format; that
3219 @c doesn't seem particularly useful to say without further elaboration,
3220 @c so for now I say nothing about it. If this is a generic BFD issue,
3221 @c these paragraphs might need to vanish from this manual, and be
3222 @c discussed in BFD chapter of binutils (or some such).
3226 Subsections appear in your object file in numeric order, lowest numbered
3227 to highest. (All this to be compatible with other people's assemblers.)
3228 The object file contains no representation of subsections; @code{@value{LD}} and
3229 other programs that manipulate object files see no trace of them.
3230 They just see all your text subsections as a text section, and all your
3231 data subsections as a data section.
3233 To specify which subsection you want subsequent statements assembled
3234 into, use a numeric argument to specify it, in a @samp{.text
3235 @var{expression}} or a @samp{.data @var{expression}} statement.
3238 When generating COFF output, you
3243 can also use an extra subsection
3244 argument with arbitrary named sections: @samp{.section @var{name},
3249 When generating ELF output, you
3254 can also use the @code{.subsection} directive (@pxref{SubSection})
3255 to specify a subsection: @samp{.subsection @var{expression}}.
3257 @var{Expression} should be an absolute expression
3258 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3259 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3260 begins in @code{text 0}. For instance:
3262 .text 0 # The default subsection is text 0 anyway.
3263 .ascii "This lives in the first text subsection. *"
3265 .ascii "But this lives in the second text subsection."
3267 .ascii "This lives in the data section,"
3268 .ascii "in the first data subsection."
3270 .ascii "This lives in the first text section,"
3271 .ascii "immediately following the asterisk (*)."
3274 Each section has a @dfn{location counter} incremented by one for every byte
3275 assembled into that section. Because subsections are merely a convenience
3276 restricted to @command{@value{AS}} there is no concept of a subsection location
3277 counter. There is no way to directly manipulate a location counter---but the
3278 @code{.align} directive changes it, and any label definition captures its
3279 current value. The location counter of the section where statements are being
3280 assembled is said to be the @dfn{active} location counter.
3283 @section bss Section
3286 @cindex common variable storage
3287 The bss section is used for local common variable storage.
3288 You may allocate address space in the bss section, but you may
3289 not dictate data to load into it before your program executes. When
3290 your program starts running, all the contents of the bss
3291 section are zeroed bytes.
3293 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3294 @ref{Lcomm,,@code{.lcomm}}.
3296 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3297 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3300 When assembling for a target which supports multiple sections, such as ELF or
3301 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3302 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3303 section. Typically the section will only contain symbol definitions and
3304 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3311 Symbols are a central concept: the programmer uses symbols to name
3312 things, the linker uses symbols to link, and the debugger uses symbols
3316 @cindex debuggers, and symbol order
3317 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3318 the same order they were declared. This may break some debuggers.
3323 * Setting Symbols:: Giving Symbols Other Values
3324 * Symbol Names:: Symbol Names
3325 * Dot:: The Special Dot Symbol
3326 * Symbol Attributes:: Symbol Attributes
3333 A @dfn{label} is written as a symbol immediately followed by a colon
3334 @samp{:}. The symbol then represents the current value of the
3335 active location counter, and is, for example, a suitable instruction
3336 operand. You are warned if you use the same symbol to represent two
3337 different locations: the first definition overrides any other
3341 On the HPPA, the usual form for a label need not be immediately followed by a
3342 colon, but instead must start in column zero. Only one label may be defined on
3343 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3344 provides a special directive @code{.label} for defining labels more flexibly.
3347 @node Setting Symbols
3348 @section Giving Symbols Other Values
3350 @cindex assigning values to symbols
3351 @cindex symbol values, assigning
3352 A symbol can be given an arbitrary value by writing a symbol, followed
3353 by an equals sign @samp{=}, followed by an expression
3354 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3355 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3356 equals sign @samp{=}@samp{=} here represents an equivalent of the
3357 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3360 Blackfin does not support symbol assignment with @samp{=}.
3364 @section Symbol Names
3366 @cindex symbol names
3367 @cindex names, symbol
3368 @ifclear SPECIAL-SYMS
3369 Symbol names begin with a letter or with one of @samp{._}. On most
3370 machines, you can also use @code{$} in symbol names; exceptions are
3371 noted in @ref{Machine Dependencies}. That character may be followed by any
3372 string of digits, letters, dollar signs (unless otherwise noted for a
3373 particular target machine), and underscores.
3377 Symbol names begin with a letter or with one of @samp{._}. On the
3378 Renesas SH you can also use @code{$} in symbol names. That
3379 character may be followed by any string of digits, letters, dollar signs (save
3380 on the H8/300), and underscores.
3384 Case of letters is significant: @code{foo} is a different symbol name
3387 Each symbol has exactly one name. Each name in an assembly language program
3388 refers to exactly one symbol. You may use that symbol name any number of times
3391 @subheading Local Symbol Names
3393 @cindex local symbol names
3394 @cindex symbol names, local
3395 A local symbol is any symbol beginning with certain local label prefixes.
3396 By default, the local label prefix is @samp{.L} for ELF systems or
3397 @samp{L} for traditional a.out systems, but each target may have its own
3398 set of local label prefixes.
3400 On the HPPA local symbols begin with @samp{L$}.
3403 Local symbols are defined and used within the assembler, but they are
3404 normally not saved in object files. Thus, they are not visible when debugging.
3405 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols:
3406 @option{-L}}) to retain the local symbols in the object files.
3408 @subheading Local Labels
3410 @cindex local labels
3411 @cindex temporary symbol names
3412 @cindex symbol names, temporary
3413 Local labels help compilers and programmers use names temporarily.
3414 They create symbols which are guaranteed to be unique over the entire scope of
3415 the input source code and which can be referred to by a simple notation.
3416 To define a local label, write a label of the form @samp{@b{N}:} (where @b{N}
3417 represents any positive integer). To refer to the most recent previous
3418 definition of that label write @samp{@b{N}b}, using the same number as when
3419 you defined the label. To refer to the next definition of a local label, write
3420 @samp{@b{N}f}---the @samp{b} stands for ``backwards'' and the @samp{f} stands
3423 There is no restriction on how you can use these labels, and you can reuse them
3424 too. So that it is possible to repeatedly define the same local label (using
3425 the same number @samp{@b{N}}), although you can only refer to the most recently
3426 defined local label of that number (for a backwards reference) or the next
3427 definition of a specific local label for a forward reference. It is also worth
3428 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3429 implemented in a slightly more efficient manner than the others.
3440 Which is the equivalent of:
3443 label_1: branch label_3
3444 label_2: branch label_1
3445 label_3: branch label_4
3446 label_4: branch label_3
3449 Local label names are only a notational device. They are immediately
3450 transformed into more conventional symbol names before the assembler uses them.
3451 The symbol names are stored in the symbol table, appear in error messages, and
3452 are optionally emitted to the object file. The names are constructed using
3456 @item @emph{local label prefix}
3457 All local symbols begin with the system-specific local label prefix.
3458 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3459 that start with the local label prefix. These labels are
3460 used for symbols you are never intended to see. If you use the
3461 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3462 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3463 you may use them in debugging.
3466 This is the number that was used in the local label definition. So if the
3467 label is written @samp{55:} then the number is @samp{55}.
3470 This unusual character is included so you do not accidentally invent a symbol
3471 of the same name. The character has ASCII value of @samp{\002} (control-B).
3473 @item @emph{ordinal number}
3474 This is a serial number to keep the labels distinct. The first definition of
3475 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3476 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3477 the number @samp{1} and its 15th definition gets @samp{15} as well.
3480 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3481 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3483 @subheading Dollar Local Labels
3484 @cindex dollar local symbols
3486 @code{@value{AS}} also supports an even more local form of local labels called
3487 dollar labels. These labels go out of scope (i.e., they become undefined) as
3488 soon as a non-local label is defined. Thus they remain valid for only a small
3489 region of the input source code. Normal local labels, by contrast, remain in
3490 scope for the entire file, or until they are redefined by another occurrence of
3491 the same local label.
3493 Dollar labels are defined in exactly the same way as ordinary local labels,
3494 except that they have a dollar sign suffix to their numeric value, e.g.,
3497 They can also be distinguished from ordinary local labels by their transformed
3498 names which use ASCII character @samp{\001} (control-A) as the magic character
3499 to distinguish them from ordinary labels. For example, the fifth definition of
3500 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3503 @section The Special Dot Symbol
3505 @cindex dot (symbol)
3506 @cindex @code{.} (symbol)
3507 @cindex current address
3508 @cindex location counter
3509 The special symbol @samp{.} refers to the current address that
3510 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3511 .long .} defines @code{melvin} to contain its own address.
3512 Assigning a value to @code{.} is treated the same as a @code{.org}
3514 @ifclear no-space-dir
3515 Thus, the expression @samp{.=.+4} is the same as saying
3519 @node Symbol Attributes
3520 @section Symbol Attributes
3522 @cindex symbol attributes
3523 @cindex attributes, symbol
3524 Every symbol has, as well as its name, the attributes ``Value'' and
3525 ``Type''. Depending on output format, symbols can also have auxiliary
3528 The detailed definitions are in @file{a.out.h}.
3531 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3532 all these attributes, and probably won't warn you. This makes the
3533 symbol an externally defined symbol, which is generally what you
3537 * Symbol Value:: Value
3538 * Symbol Type:: Type
3541 * a.out Symbols:: Symbol Attributes: @code{a.out}
3545 * a.out Symbols:: Symbol Attributes: @code{a.out}
3548 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3553 * COFF Symbols:: Symbol Attributes for COFF
3556 * SOM Symbols:: Symbol Attributes for SOM
3563 @cindex value of a symbol
3564 @cindex symbol value
3565 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3566 location in the text, data, bss or absolute sections the value is the
3567 number of addresses from the start of that section to the label.
3568 Naturally for text, data and bss sections the value of a symbol changes
3569 as @code{@value{LD}} changes section base addresses during linking. Absolute
3570 symbols' values do not change during linking: that is why they are
3573 The value of an undefined symbol is treated in a special way. If it is
3574 0 then the symbol is not defined in this assembler source file, and
3575 @code{@value{LD}} tries to determine its value from other files linked into the
3576 same program. You make this kind of symbol simply by mentioning a symbol
3577 name without defining it. A non-zero value represents a @code{.comm}
3578 common declaration. The value is how much common storage to reserve, in
3579 bytes (addresses). The symbol refers to the first address of the
3585 @cindex type of a symbol
3587 The type attribute of a symbol contains relocation (section)
3588 information, any flag settings indicating that a symbol is external, and
3589 (optionally), other information for linkers and debuggers. The exact
3590 format depends on the object-code output format in use.
3595 @c The following avoids a "widow" subsection title. @group would be
3596 @c better if it were available outside examples.
3599 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3601 @cindex @code{b.out} symbol attributes
3602 @cindex symbol attributes, @code{b.out}
3603 These symbol attributes appear only when @command{@value{AS}} is configured for
3604 one of the Berkeley-descended object output formats---@code{a.out} or
3610 @subsection Symbol Attributes: @code{a.out}
3612 @cindex @code{a.out} symbol attributes
3613 @cindex symbol attributes, @code{a.out}
3619 @subsection Symbol Attributes: @code{a.out}
3621 @cindex @code{a.out} symbol attributes
3622 @cindex symbol attributes, @code{a.out}
3626 * Symbol Desc:: Descriptor
3627 * Symbol Other:: Other
3631 @subsubsection Descriptor
3633 @cindex descriptor, of @code{a.out} symbol
3634 This is an arbitrary 16-bit value. You may establish a symbol's
3635 descriptor value by using a @code{.desc} statement
3636 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3637 @command{@value{AS}}.
3640 @subsubsection Other
3642 @cindex other attribute, of @code{a.out} symbol
3643 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3648 @subsection Symbol Attributes for COFF
3650 @cindex COFF symbol attributes
3651 @cindex symbol attributes, COFF
3653 The COFF format supports a multitude of auxiliary symbol attributes;
3654 like the primary symbol attributes, they are set between @code{.def} and
3655 @code{.endef} directives.
3657 @subsubsection Primary Attributes
3659 @cindex primary attributes, COFF symbols
3660 The symbol name is set with @code{.def}; the value and type,
3661 respectively, with @code{.val} and @code{.type}.
3663 @subsubsection Auxiliary Attributes
3665 @cindex auxiliary attributes, COFF symbols
3666 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3667 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3668 table information for COFF.
3673 @subsection Symbol Attributes for SOM
3675 @cindex SOM symbol attributes
3676 @cindex symbol attributes, SOM
3678 The SOM format for the HPPA supports a multitude of symbol attributes set with
3679 the @code{.EXPORT} and @code{.IMPORT} directives.
3681 The attributes are described in @cite{HP9000 Series 800 Assembly
3682 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3683 @code{EXPORT} assembler directive documentation.
3687 @chapter Expressions
3691 @cindex numeric values
3692 An @dfn{expression} specifies an address or numeric value.
3693 Whitespace may precede and/or follow an expression.
3695 The result of an expression must be an absolute number, or else an offset into
3696 a particular section. If an expression is not absolute, and there is not
3697 enough information when @command{@value{AS}} sees the expression to know its
3698 section, a second pass over the source program might be necessary to interpret
3699 the expression---but the second pass is currently not implemented.
3700 @command{@value{AS}} aborts with an error message in this situation.
3703 * Empty Exprs:: Empty Expressions
3704 * Integer Exprs:: Integer Expressions
3708 @section Empty Expressions
3710 @cindex empty expressions
3711 @cindex expressions, empty
3712 An empty expression has no value: it is just whitespace or null.
3713 Wherever an absolute expression is required, you may omit the
3714 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3715 is compatible with other assemblers.
3718 @section Integer Expressions
3720 @cindex integer expressions
3721 @cindex expressions, integer
3722 An @dfn{integer expression} is one or more @emph{arguments} delimited
3723 by @emph{operators}.
3726 * Arguments:: Arguments
3727 * Operators:: Operators
3728 * Prefix Ops:: Prefix Operators
3729 * Infix Ops:: Infix Operators
3733 @subsection Arguments
3735 @cindex expression arguments
3736 @cindex arguments in expressions
3737 @cindex operands in expressions
3738 @cindex arithmetic operands
3739 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3740 contexts arguments are sometimes called ``arithmetic operands''. In
3741 this manual, to avoid confusing them with the ``instruction operands'' of
3742 the machine language, we use the term ``argument'' to refer to parts of
3743 expressions only, reserving the word ``operand'' to refer only to machine
3744 instruction operands.
3746 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3747 @var{section} is one of text, data, bss, absolute,
3748 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3751 Numbers are usually integers.
3753 A number can be a flonum or bignum. In this case, you are warned
3754 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3755 these 32 bits are an integer. You may write integer-manipulating
3756 instructions that act on exotic constants, compatible with other
3759 @cindex subexpressions
3760 Subexpressions are a left parenthesis @samp{(} followed by an integer
3761 expression, followed by a right parenthesis @samp{)}; or a prefix
3762 operator followed by an argument.
3765 @subsection Operators
3767 @cindex operators, in expressions
3768 @cindex arithmetic functions
3769 @cindex functions, in expressions
3770 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3771 operators are followed by an argument. Infix operators appear
3772 between their arguments. Operators may be preceded and/or followed by
3776 @subsection Prefix Operator
3778 @cindex prefix operators
3779 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3780 one argument, which must be absolute.
3782 @c the tex/end tex stuff surrounding this small table is meant to make
3783 @c it align, on the printed page, with the similar table in the next
3784 @c section (which is inside an enumerate).
3786 \global\advance\leftskip by \itemindent
3791 @dfn{Negation}. Two's complement negation.
3793 @dfn{Complementation}. Bitwise not.
3797 \global\advance\leftskip by -\itemindent
3801 @subsection Infix Operators
3803 @cindex infix operators
3804 @cindex operators, permitted arguments
3805 @dfn{Infix operators} take two arguments, one on either side. Operators
3806 have precedence, but operations with equal precedence are performed left
3807 to right. Apart from @code{+} or @option{-}, both arguments must be
3808 absolute, and the result is absolute.
3811 @cindex operator precedence
3812 @cindex precedence of operators
3819 @dfn{Multiplication}.
3822 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3828 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3831 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3835 Intermediate precedence
3840 @dfn{Bitwise Inclusive Or}.
3846 @dfn{Bitwise Exclusive Or}.
3849 @dfn{Bitwise Or Not}.
3856 @cindex addition, permitted arguments
3857 @cindex plus, permitted arguments
3858 @cindex arguments for addition
3860 @dfn{Addition}. If either argument is absolute, the result has the section of
3861 the other argument. You may not add together arguments from different
3864 @cindex subtraction, permitted arguments
3865 @cindex minus, permitted arguments
3866 @cindex arguments for subtraction
3868 @dfn{Subtraction}. If the right argument is absolute, the
3869 result has the section of the left argument.
3870 If both arguments are in the same section, the result is absolute.
3871 You may not subtract arguments from different sections.
3872 @c FIXME is there still something useful to say about undefined - undefined ?
3874 @cindex comparison expressions
3875 @cindex expressions, comparison
3880 @dfn{Is Not Equal To}
3884 @dfn{Is Greater Than}
3886 @dfn{Is Greater Than Or Equal To}
3888 @dfn{Is Less Than Or Equal To}
3890 The comparison operators can be used as infix operators. A true results has a
3891 value of -1 whereas a false result has a value of 0. Note, these operators
3892 perform signed comparisons.
3895 @item Lowest Precedence
3904 These two logical operations can be used to combine the results of sub
3905 expressions. Note, unlike the comparison operators a true result returns a
3906 value of 1 but a false results does still return 0. Also note that the logical
3907 or operator has a slightly lower precedence than logical and.
3912 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3913 address; you can only have a defined section in one of the two arguments.
3916 @chapter Assembler Directives
3918 @cindex directives, machine independent
3919 @cindex pseudo-ops, machine independent
3920 @cindex machine independent directives
3921 All assembler directives have names that begin with a period (@samp{.}).
3922 The rest of the name is letters, usually in lower case.
3924 This chapter discusses directives that are available regardless of the
3925 target machine configuration for the @sc{gnu} assembler.
3927 Some machine configurations provide additional directives.
3928 @xref{Machine Dependencies}.
3931 @ifset machine-directives
3932 @xref{Machine Dependencies}, for additional directives.
3937 * Abort:: @code{.abort}
3939 * ABORT (COFF):: @code{.ABORT}
3942 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3943 * Altmacro:: @code{.altmacro}
3944 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3945 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3946 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3947 * Byte:: @code{.byte @var{expressions}}
3948 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
3949 * Comm:: @code{.comm @var{symbol} , @var{length} }
3950 * Data:: @code{.data @var{subsection}}
3952 * Def:: @code{.def @var{name}}
3955 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3961 * Double:: @code{.double @var{flonums}}
3962 * Eject:: @code{.eject}
3963 * Else:: @code{.else}
3964 * Elseif:: @code{.elseif}
3967 * Endef:: @code{.endef}
3970 * Endfunc:: @code{.endfunc}
3971 * Endif:: @code{.endif}
3972 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3973 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3974 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
3976 * Error:: @code{.error @var{string}}
3977 * Exitm:: @code{.exitm}
3978 * Extern:: @code{.extern}
3979 * Fail:: @code{.fail}
3980 * File:: @code{.file}
3981 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3982 * Float:: @code{.float @var{flonums}}
3983 * Func:: @code{.func}
3984 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3986 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
3987 * Hidden:: @code{.hidden @var{names}}
3990 * hword:: @code{.hword @var{expressions}}
3991 * Ident:: @code{.ident}
3992 * If:: @code{.if @var{absolute expression}}
3993 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
3994 * Include:: @code{.include "@var{file}"}
3995 * Int:: @code{.int @var{expressions}}
3997 * Internal:: @code{.internal @var{names}}
4000 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4001 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4002 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4003 * Lflags:: @code{.lflags}
4004 @ifclear no-line-dir
4005 * Line:: @code{.line @var{line-number}}
4008 * Linkonce:: @code{.linkonce [@var{type}]}
4009 * List:: @code{.list}
4010 * Ln:: @code{.ln @var{line-number}}
4011 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4012 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4014 * Local:: @code{.local @var{names}}
4017 * Long:: @code{.long @var{expressions}}
4019 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4022 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4023 * MRI:: @code{.mri @var{val}}
4024 * Noaltmacro:: @code{.noaltmacro}
4025 * Nolist:: @code{.nolist}
4026 * Octa:: @code{.octa @var{bignums}}
4027 * Org:: @code{.org @var{new-lc}, @var{fill}}
4028 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4030 * PopSection:: @code{.popsection}
4031 * Previous:: @code{.previous}
4034 * Print:: @code{.print @var{string}}
4036 * Protected:: @code{.protected @var{names}}
4039 * Psize:: @code{.psize @var{lines}, @var{columns}}
4040 * Purgem:: @code{.purgem @var{name}}
4042 * PushSection:: @code{.pushsection @var{name}}
4045 * Quad:: @code{.quad @var{bignums}}
4046 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4047 * Rept:: @code{.rept @var{count}}
4048 * Sbttl:: @code{.sbttl "@var{subheading}"}
4050 * Scl:: @code{.scl @var{class}}
4053 * Section:: @code{.section @var{name}[, @var{flags}]}
4056 * Set:: @code{.set @var{symbol}, @var{expression}}
4057 * Short:: @code{.short @var{expressions}}
4058 * Single:: @code{.single @var{flonums}}
4060 * Size:: @code{.size [@var{name} , @var{expression}]}
4062 @ifclear no-space-dir
4063 * Skip:: @code{.skip @var{size} , @var{fill}}
4066 * Sleb128:: @code{.sleb128 @var{expressions}}
4067 @ifclear no-space-dir
4068 * Space:: @code{.space @var{size} , @var{fill}}
4071 * Stab:: @code{.stabd, .stabn, .stabs}
4074 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4075 * Struct:: @code{.struct @var{expression}}
4077 * SubSection:: @code{.subsection}
4078 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4082 * Tag:: @code{.tag @var{structname}}
4085 * Text:: @code{.text @var{subsection}}
4086 * Title:: @code{.title "@var{heading}"}
4088 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4091 * Uleb128:: @code{.uleb128 @var{expressions}}
4093 * Val:: @code{.val @var{addr}}
4097 * Version:: @code{.version "@var{string}"}
4098 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4099 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4102 * Warning:: @code{.warning @var{string}}
4103 * Weak:: @code{.weak @var{names}}
4104 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4105 * Word:: @code{.word @var{expressions}}
4106 * Deprecated:: Deprecated Directives
4110 @section @code{.abort}
4112 @cindex @code{abort} directive
4113 @cindex stopping the assembly
4114 This directive stops the assembly immediately. It is for
4115 compatibility with other assemblers. The original idea was that the
4116 assembly language source would be piped into the assembler. If the sender
4117 of the source quit, it could use this directive tells @command{@value{AS}} to
4118 quit also. One day @code{.abort} will not be supported.
4122 @section @code{.ABORT} (COFF)
4124 @cindex @code{ABORT} directive
4125 When producing COFF output, @command{@value{AS}} accepts this directive as a
4126 synonym for @samp{.abort}.
4129 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4135 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4137 @cindex padding the location counter
4138 @cindex @code{align} directive
4139 Pad the location counter (in the current subsection) to a particular storage
4140 boundary. The first expression (which must be absolute) is the alignment
4141 required, as described below.
4143 The second expression (also absolute) gives the fill value to be stored in the
4144 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4145 padding bytes are normally zero. However, on some systems, if the section is
4146 marked as containing code and the fill value is omitted, the space is filled
4147 with no-op instructions.
4149 The third expression is also absolute, and is also optional. If it is present,
4150 it is the maximum number of bytes that should be skipped by this alignment
4151 directive. If doing the alignment would require skipping more bytes than the
4152 specified maximum, then the alignment is not done at all. You can omit the
4153 fill value (the second argument) entirely by simply using two commas after the
4154 required alignment; this can be useful if you want the alignment to be filled
4155 with no-op instructions when appropriate.
4157 The way the required alignment is specified varies from system to system.
4158 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or32,
4159 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4160 alignment request in bytes. For example @samp{.align 8} advances
4161 the location counter until it is a multiple of 8. If the location counter
4162 is already a multiple of 8, no change is needed. For the tic54x, the
4163 first expression is the alignment request in words.
4165 For other systems, including ppc, i386 using a.out format, arm and
4166 strongarm, it is the
4167 number of low-order zero bits the location counter must have after
4168 advancement. For example @samp{.align 3} advances the location
4169 counter until it a multiple of 8. If the location counter is already a
4170 multiple of 8, no change is needed.
4172 This inconsistency is due to the different behaviors of the various
4173 native assemblers for these systems which GAS must emulate.
4174 GAS also provides @code{.balign} and @code{.p2align} directives,
4175 described later, which have a consistent behavior across all
4176 architectures (but are specific to GAS).
4179 @section @code{.altmacro}
4180 Enable alternate macro mode, enabling:
4183 @item LOCAL @var{name} [ , @dots{} ]
4184 One additional directive, @code{LOCAL}, is available. It is used to
4185 generate a string replacement for each of the @var{name} arguments, and
4186 replace any instances of @var{name} in each macro expansion. The
4187 replacement string is unique in the assembly, and different for each
4188 separate macro expansion. @code{LOCAL} allows you to write macros that
4189 define symbols, without fear of conflict between separate macro expansions.
4191 @item String delimiters
4192 You can write strings delimited in these other ways besides
4193 @code{"@var{string}"}:
4196 @item '@var{string}'
4197 You can delimit strings with single-quote characters.
4199 @item <@var{string}>
4200 You can delimit strings with matching angle brackets.
4203 @item single-character string escape
4204 To include any single character literally in a string (even if the
4205 character would otherwise have some special meaning), you can prefix the
4206 character with @samp{!} (an exclamation mark). For example, you can
4207 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4209 @item Expression results as strings
4210 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4211 and use the result as a string.
4215 @section @code{.ascii "@var{string}"}@dots{}
4217 @cindex @code{ascii} directive
4218 @cindex string literals
4219 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4220 separated by commas. It assembles each string (with no automatic
4221 trailing zero byte) into consecutive addresses.
4224 @section @code{.asciz "@var{string}"}@dots{}
4226 @cindex @code{asciz} directive
4227 @cindex zero-terminated strings
4228 @cindex null-terminated strings
4229 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4230 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4233 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4235 @cindex padding the location counter given number of bytes
4236 @cindex @code{balign} directive
4237 Pad the location counter (in the current subsection) to a particular
4238 storage boundary. The first expression (which must be absolute) is the
4239 alignment request in bytes. For example @samp{.balign 8} advances
4240 the location counter until it is a multiple of 8. If the location counter
4241 is already a multiple of 8, no change is needed.
4243 The second expression (also absolute) gives the fill value to be stored in the
4244 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4245 padding bytes are normally zero. However, on some systems, if the section is
4246 marked as containing code and the fill value is omitted, the space is filled
4247 with no-op instructions.
4249 The third expression is also absolute, and is also optional. If it is present,
4250 it is the maximum number of bytes that should be skipped by this alignment
4251 directive. If doing the alignment would require skipping more bytes than the
4252 specified maximum, then the alignment is not done at all. You can omit the
4253 fill value (the second argument) entirely by simply using two commas after the
4254 required alignment; this can be useful if you want the alignment to be filled
4255 with no-op instructions when appropriate.
4257 @cindex @code{balignw} directive
4258 @cindex @code{balignl} directive
4259 The @code{.balignw} and @code{.balignl} directives are variants of the
4260 @code{.balign} directive. The @code{.balignw} directive treats the fill
4261 pattern as a two byte word value. The @code{.balignl} directives treats the
4262 fill pattern as a four byte longword value. For example, @code{.balignw
4263 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4264 filled in with the value 0x368d (the exact placement of the bytes depends upon
4265 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4269 @section @code{.byte @var{expressions}}
4271 @cindex @code{byte} directive
4272 @cindex integers, one byte
4273 @code{.byte} expects zero or more expressions, separated by commas.
4274 Each expression is assembled into the next byte.
4276 @node CFI directives
4277 @section @code{.cfi_sections @var{section_list}}
4278 @cindex @code{cfi_sections} directive
4279 @code{.cfi_sections} may be used to specify whether CFI directives
4280 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4281 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4282 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4283 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4284 directive is not used is @code{.cfi_sections .eh_frame}.
4286 @section @code{.cfi_startproc [simple]}
4287 @cindex @code{cfi_startproc} directive
4288 @code{.cfi_startproc} is used at the beginning of each function that
4289 should have an entry in @code{.eh_frame}. It initializes some internal
4290 data structures. Don't forget to close the function by
4291 @code{.cfi_endproc}.
4293 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4294 it also emits some architecture dependent initial CFI instructions.
4296 @section @code{.cfi_endproc}
4297 @cindex @code{cfi_endproc} directive
4298 @code{.cfi_endproc} is used at the end of a function where it closes its
4299 unwind entry previously opened by
4300 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4302 @section @code{.cfi_personality @var{encoding} [, @var{exp}]}
4303 @code{.cfi_personality} defines personality routine and its encoding.
4304 @var{encoding} must be a constant determining how the personality
4305 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4306 argument is not present, otherwise second argument should be
4307 a constant or a symbol name. When using indirect encodings,
4308 the symbol provided should be the location where personality
4309 can be loaded from, not the personality routine itself.
4310 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4311 no personality routine.
4313 @section @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4314 @code{.cfi_lsda} defines LSDA and its encoding.
4315 @var{encoding} must be a constant determining how the LSDA
4316 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4317 argument is not present, otherwise second argument should be a constant
4318 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4321 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4322 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4323 address from @var{register} and add @var{offset} to it}.
4325 @section @code{.cfi_def_cfa_register @var{register}}
4326 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4327 now on @var{register} will be used instead of the old one. Offset
4330 @section @code{.cfi_def_cfa_offset @var{offset}}
4331 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4332 remains the same, but @var{offset} is new. Note that it is the
4333 absolute offset that will be added to a defined register to compute
4336 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4337 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4338 value that is added/substracted from the previous offset.
4340 @section @code{.cfi_offset @var{register}, @var{offset}}
4341 Previous value of @var{register} is saved at offset @var{offset} from
4344 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4345 Previous value of @var{register} is saved at offset @var{offset} from
4346 the current CFA register. This is transformed to @code{.cfi_offset}
4347 using the known displacement of the CFA register from the CFA.
4348 This is often easier to use, because the number will match the
4349 code it's annotating.
4351 @section @code{.cfi_register @var{register1}, @var{register2}}
4352 Previous value of @var{register1} is saved in register @var{register2}.
4354 @section @code{.cfi_restore @var{register}}
4355 @code{.cfi_restore} says that the rule for @var{register} is now the
4356 same as it was at the beginning of the function, after all initial
4357 instruction added by @code{.cfi_startproc} were executed.
4359 @section @code{.cfi_undefined @var{register}}
4360 From now on the previous value of @var{register} can't be restored anymore.
4362 @section @code{.cfi_same_value @var{register}}
4363 Current value of @var{register} is the same like in the previous frame,
4364 i.e. no restoration needed.
4366 @section @code{.cfi_remember_state},
4367 First save all current rules for all registers by @code{.cfi_remember_state},
4368 then totally screw them up by subsequent @code{.cfi_*} directives and when
4369 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4370 the previous saved state.
4372 @section @code{.cfi_return_column @var{register}}
4373 Change return column @var{register}, i.e. the return address is either
4374 directly in @var{register} or can be accessed by rules for @var{register}.
4376 @section @code{.cfi_signal_frame}
4377 Mark current function as signal trampoline.
4379 @section @code{.cfi_window_save}
4380 SPARC register window has been saved.
4382 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4383 Allows the user to add arbitrary bytes to the unwind info. One
4384 might use this to add OS-specific CFI opcodes, or generic CFI
4385 opcodes that GAS does not yet support.
4387 @section @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4388 The current value of @var{register} is @var{label}. The value of @var{label}
4389 will be encoded in the output file according to @var{encoding}; see the
4390 description of @code{.cfi_personality} for details on this encoding.
4392 The usefulness of equating a register to a fixed label is probably
4393 limited to the return address register. Here, it can be useful to
4394 mark a code segment that has only one return address which is reached
4395 by a direct branch and no copy of the return address exists in memory
4396 or another register.
4399 @section @code{.comm @var{symbol} , @var{length} }
4401 @cindex @code{comm} directive
4402 @cindex symbol, common
4403 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4404 common symbol in one object file may be merged with a defined or common symbol
4405 of the same name in another object file. If @code{@value{LD}} does not see a
4406 definition for the symbol--just one or more common symbols--then it will
4407 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4408 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4409 the same name, and they do not all have the same size, it will allocate space
4410 using the largest size.
4413 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4414 an optional third argument. This is the desired alignment of the symbol,
4415 specified for ELF as a byte boundary (for example, an alignment of 16 means
4416 that the least significant 4 bits of the address should be zero), and for PE
4417 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4418 boundary). The alignment must be an absolute expression, and it must be a
4419 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4420 common symbol, it will use the alignment when placing the symbol. If no
4421 alignment is specified, @command{@value{AS}} will set the alignment to the
4422 largest power of two less than or equal to the size of the symbol, up to a
4423 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4424 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4425 @samp{--section-alignment} option; image file sections in PE are aligned to
4426 multiples of 4096, which is far too large an alignment for ordinary variables.
4427 It is rather the default alignment for (non-debug) sections within object
4428 (@samp{*.o}) files, which are less strictly aligned.}.
4432 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4433 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4437 @section @code{.data @var{subsection}}
4439 @cindex @code{data} directive
4440 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4441 end of the data subsection numbered @var{subsection} (which is an
4442 absolute expression). If @var{subsection} is omitted, it defaults
4447 @section @code{.def @var{name}}
4449 @cindex @code{def} directive
4450 @cindex COFF symbols, debugging
4451 @cindex debugging COFF symbols
4452 Begin defining debugging information for a symbol @var{name}; the
4453 definition extends until the @code{.endef} directive is encountered.
4456 This directive is only observed when @command{@value{AS}} is configured for COFF
4457 format output; when producing @code{b.out}, @samp{.def} is recognized,
4464 @section @code{.desc @var{symbol}, @var{abs-expression}}
4466 @cindex @code{desc} directive
4467 @cindex COFF symbol descriptor
4468 @cindex symbol descriptor, COFF
4469 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4470 to the low 16 bits of an absolute expression.
4473 The @samp{.desc} directive is not available when @command{@value{AS}} is
4474 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4475 object format. For the sake of compatibility, @command{@value{AS}} accepts
4476 it, but produces no output, when configured for COFF.
4482 @section @code{.dim}
4484 @cindex @code{dim} directive
4485 @cindex COFF auxiliary symbol information
4486 @cindex auxiliary symbol information, COFF
4487 This directive is generated by compilers to include auxiliary debugging
4488 information in the symbol table. It is only permitted inside
4489 @code{.def}/@code{.endef} pairs.
4492 @samp{.dim} is only meaningful when generating COFF format output; when
4493 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4499 @section @code{.double @var{flonums}}
4501 @cindex @code{double} directive
4502 @cindex floating point numbers (double)
4503 @code{.double} expects zero or more flonums, separated by commas. It
4504 assembles floating point numbers.
4506 The exact kind of floating point numbers emitted depends on how
4507 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4511 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4512 in @sc{ieee} format.
4517 @section @code{.eject}
4519 @cindex @code{eject} directive
4520 @cindex new page, in listings
4521 @cindex page, in listings
4522 @cindex listing control: new page
4523 Force a page break at this point, when generating assembly listings.
4526 @section @code{.else}
4528 @cindex @code{else} directive
4529 @code{.else} is part of the @command{@value{AS}} support for conditional
4530 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4531 of code to be assembled if the condition for the preceding @code{.if}
4535 @section @code{.elseif}
4537 @cindex @code{elseif} directive
4538 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4539 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4540 @code{.if} block that would otherwise fill the entire @code{.else} section.
4543 @section @code{.end}
4545 @cindex @code{end} directive
4546 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4547 process anything in the file past the @code{.end} directive.
4551 @section @code{.endef}
4553 @cindex @code{endef} directive
4554 This directive flags the end of a symbol definition begun with
4558 @samp{.endef} is only meaningful when generating COFF format output; if
4559 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4560 directive but ignores it.
4565 @section @code{.endfunc}
4566 @cindex @code{endfunc} directive
4567 @code{.endfunc} marks the end of a function specified with @code{.func}.
4570 @section @code{.endif}
4572 @cindex @code{endif} directive
4573 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4574 it marks the end of a block of code that is only assembled
4575 conditionally. @xref{If,,@code{.if}}.
4578 @section @code{.equ @var{symbol}, @var{expression}}
4580 @cindex @code{equ} directive
4581 @cindex assigning values to symbols
4582 @cindex symbols, assigning values to
4583 This directive sets the value of @var{symbol} to @var{expression}.
4584 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4587 The syntax for @code{equ} on the HPPA is
4588 @samp{@var{symbol} .equ @var{expression}}.
4592 The syntax for @code{equ} on the Z80 is
4593 @samp{@var{symbol} equ @var{expression}}.
4594 On the Z80 it is an eror if @var{symbol} is already defined,
4595 but the symbol is not protected from later redefinition.
4596 Compare @ref{Equiv}.
4600 @section @code{.equiv @var{symbol}, @var{expression}}
4601 @cindex @code{equiv} directive
4602 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4603 the assembler will signal an error if @var{symbol} is already defined. Note a
4604 symbol which has been referenced but not actually defined is considered to be
4607 Except for the contents of the error message, this is roughly equivalent to
4614 plus it protects the symbol from later redefinition.
4617 @section @code{.eqv @var{symbol}, @var{expression}}
4618 @cindex @code{eqv} directive
4619 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
4620 evaluate the expression or any part of it immediately. Instead each time
4621 the resulting symbol is used in an expression, a snapshot of its current
4625 @section @code{.err}
4626 @cindex @code{err} directive
4627 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4628 message and, unless the @option{-Z} option was used, it will not generate an
4629 object file. This can be used to signal an error in conditionally compiled code.
4632 @section @code{.error "@var{string}"}
4633 @cindex error directive
4635 Similarly to @code{.err}, this directive emits an error, but you can specify a
4636 string that will be emitted as the error message. If you don't specify the
4637 message, it defaults to @code{".error directive invoked in source file"}.
4638 @xref{Errors, ,Error and Warning Messages}.
4641 .error "This code has not been assembled and tested."
4645 @section @code{.exitm}
4646 Exit early from the current macro definition. @xref{Macro}.
4649 @section @code{.extern}
4651 @cindex @code{extern} directive
4652 @code{.extern} is accepted in the source program---for compatibility
4653 with other assemblers---but it is ignored. @command{@value{AS}} treats
4654 all undefined symbols as external.
4657 @section @code{.fail @var{expression}}
4659 @cindex @code{fail} directive
4660 Generates an error or a warning. If the value of the @var{expression} is 500
4661 or more, @command{@value{AS}} will print a warning message. If the value is less
4662 than 500, @command{@value{AS}} will print an error message. The message will
4663 include the value of @var{expression}. This can occasionally be useful inside
4664 complex nested macros or conditional assembly.
4667 @section @code{.file}
4668 @cindex @code{file} directive
4670 @ifclear no-file-dir
4671 There are two different versions of the @code{.file} directive. Targets
4672 that support DWARF2 line number information use the DWARF2 version of
4673 @code{.file}. Other targets use the default version.
4675 @subheading Default Version
4677 @cindex logical file name
4678 @cindex file name, logical
4679 This version of the @code{.file} directive tells @command{@value{AS}} that we
4680 are about to start a new logical file. The syntax is:
4686 @var{string} is the new file name. In general, the filename is
4687 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4688 to specify an empty file name, you must give the quotes--@code{""}. This
4689 statement may go away in future: it is only recognized to be compatible with
4690 old @command{@value{AS}} programs.
4692 @subheading DWARF2 Version
4695 When emitting DWARF2 line number information, @code{.file} assigns filenames
4696 to the @code{.debug_line} file name table. The syntax is:
4699 .file @var{fileno} @var{filename}
4702 The @var{fileno} operand should be a unique positive integer to use as the
4703 index of the entry in the table. The @var{filename} operand is a C string
4706 The detail of filename indices is exposed to the user because the filename
4707 table is shared with the @code{.debug_info} section of the DWARF2 debugging
4708 information, and thus the user must know the exact indices that table
4712 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4714 @cindex @code{fill} directive
4715 @cindex writing patterns in memory
4716 @cindex patterns, writing in memory
4717 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4718 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4719 may be zero or more. @var{Size} may be zero or more, but if it is
4720 more than 8, then it is deemed to have the value 8, compatible with
4721 other people's assemblers. The contents of each @var{repeat} bytes
4722 is taken from an 8-byte number. The highest order 4 bytes are
4723 zero. The lowest order 4 bytes are @var{value} rendered in the
4724 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4725 Each @var{size} bytes in a repetition is taken from the lowest order
4726 @var{size} bytes of this number. Again, this bizarre behavior is
4727 compatible with other people's assemblers.
4729 @var{size} and @var{value} are optional.
4730 If the second comma and @var{value} are absent, @var{value} is
4731 assumed zero. If the first comma and following tokens are absent,
4732 @var{size} is assumed to be 1.
4735 @section @code{.float @var{flonums}}
4737 @cindex floating point numbers (single)
4738 @cindex @code{float} directive
4739 This directive assembles zero or more flonums, separated by commas. It
4740 has the same effect as @code{.single}.
4742 The exact kind of floating point numbers emitted depends on how
4743 @command{@value{AS}} is configured.
4744 @xref{Machine Dependencies}.
4748 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4749 in @sc{ieee} format.
4754 @section @code{.func @var{name}[,@var{label}]}
4755 @cindex @code{func} directive
4756 @code{.func} emits debugging information to denote function @var{name}, and
4757 is ignored unless the file is assembled with debugging enabled.
4758 Only @samp{--gstabs[+]} is currently supported.
4759 @var{label} is the entry point of the function and if omitted @var{name}
4760 prepended with the @samp{leading char} is used.
4761 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4762 All functions are currently defined to have @code{void} return type.
4763 The function must be terminated with @code{.endfunc}.
4766 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4768 @cindex @code{global} directive
4769 @cindex symbol, making visible to linker
4770 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4771 @var{symbol} in your partial program, its value is made available to
4772 other partial programs that are linked with it. Otherwise,
4773 @var{symbol} takes its attributes from a symbol of the same name
4774 from another file linked into the same program.
4776 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4777 compatibility with other assemblers.
4780 On the HPPA, @code{.global} is not always enough to make it accessible to other
4781 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4782 @xref{HPPA Directives, ,HPPA Assembler Directives}.
4787 @section @code{.gnu_attribute @var{tag},@var{value}}
4788 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
4791 @section @code{.hidden @var{names}}
4793 @cindex @code{hidden} directive
4795 This is one of the ELF visibility directives. The other two are
4796 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4797 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4799 This directive overrides the named symbols default visibility (which is set by
4800 their binding: local, global or weak). The directive sets the visibility to
4801 @code{hidden} which means that the symbols are not visible to other components.
4802 Such symbols are always considered to be @code{protected} as well.
4806 @section @code{.hword @var{expressions}}
4808 @cindex @code{hword} directive
4809 @cindex integers, 16-bit
4810 @cindex numbers, 16-bit
4811 @cindex sixteen bit integers
4812 This expects zero or more @var{expressions}, and emits
4813 a 16 bit number for each.
4816 This directive is a synonym for @samp{.short}; depending on the target
4817 architecture, it may also be a synonym for @samp{.word}.
4821 This directive is a synonym for @samp{.short}.
4824 This directive is a synonym for both @samp{.short} and @samp{.word}.
4829 @section @code{.ident}
4831 @cindex @code{ident} directive
4833 This directive is used by some assemblers to place tags in object files. The
4834 behavior of this directive varies depending on the target. When using the
4835 a.out object file format, @command{@value{AS}} simply accepts the directive for
4836 source-file compatibility with existing assemblers, but does not emit anything
4837 for it. When using COFF, comments are emitted to the @code{.comment} or
4838 @code{.rdata} section, depending on the target. When using ELF, comments are
4839 emitted to the @code{.comment} section.
4842 @section @code{.if @var{absolute expression}}
4844 @cindex conditional assembly
4845 @cindex @code{if} directive
4846 @code{.if} marks the beginning of a section of code which is only
4847 considered part of the source program being assembled if the argument
4848 (which must be an @var{absolute expression}) is non-zero. The end of
4849 the conditional section of code must be marked by @code{.endif}
4850 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4851 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4852 If you have several conditions to check, @code{.elseif} may be used to avoid
4853 nesting blocks if/else within each subsequent @code{.else} block.
4855 The following variants of @code{.if} are also supported:
4857 @cindex @code{ifdef} directive
4858 @item .ifdef @var{symbol}
4859 Assembles the following section of code if the specified @var{symbol}
4860 has been defined. Note a symbol which has been referenced but not yet defined
4861 is considered to be undefined.
4863 @cindex @code{ifb} directive
4864 @item .ifb @var{text}
4865 Assembles the following section of code if the operand is blank (empty).
4867 @cindex @code{ifc} directive
4868 @item .ifc @var{string1},@var{string2}
4869 Assembles the following section of code if the two strings are the same. The
4870 strings may be optionally quoted with single quotes. If they are not quoted,
4871 the first string stops at the first comma, and the second string stops at the
4872 end of the line. Strings which contain whitespace should be quoted. The
4873 string comparison is case sensitive.
4875 @cindex @code{ifeq} directive
4876 @item .ifeq @var{absolute expression}
4877 Assembles the following section of code if the argument is zero.
4879 @cindex @code{ifeqs} directive
4880 @item .ifeqs @var{string1},@var{string2}
4881 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4883 @cindex @code{ifge} directive
4884 @item .ifge @var{absolute expression}
4885 Assembles the following section of code if the argument is greater than or
4888 @cindex @code{ifgt} directive
4889 @item .ifgt @var{absolute expression}
4890 Assembles the following section of code if the argument is greater than zero.
4892 @cindex @code{ifle} directive
4893 @item .ifle @var{absolute expression}
4894 Assembles the following section of code if the argument is less than or equal
4897 @cindex @code{iflt} directive
4898 @item .iflt @var{absolute expression}
4899 Assembles the following section of code if the argument is less than zero.
4901 @cindex @code{ifnb} directive
4902 @item .ifnb @var{text}
4903 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
4904 following section of code if the operand is non-blank (non-empty).
4906 @cindex @code{ifnc} directive
4907 @item .ifnc @var{string1},@var{string2}.
4908 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4909 following section of code if the two strings are not the same.
4911 @cindex @code{ifndef} directive
4912 @cindex @code{ifnotdef} directive
4913 @item .ifndef @var{symbol}
4914 @itemx .ifnotdef @var{symbol}
4915 Assembles the following section of code if the specified @var{symbol}
4916 has not been defined. Both spelling variants are equivalent. Note a symbol
4917 which has been referenced but not yet defined is considered to be undefined.
4919 @cindex @code{ifne} directive
4920 @item .ifne @var{absolute expression}
4921 Assembles the following section of code if the argument is not equal to zero
4922 (in other words, this is equivalent to @code{.if}).
4924 @cindex @code{ifnes} directive
4925 @item .ifnes @var{string1},@var{string2}
4926 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4927 following section of code if the two strings are not the same.
4931 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4933 @cindex @code{incbin} directive
4934 @cindex binary files, including
4935 The @code{incbin} directive includes @var{file} verbatim at the current
4936 location. You can control the search paths used with the @samp{-I} command-line
4937 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4940 The @var{skip} argument skips a number of bytes from the start of the
4941 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4942 read. Note that the data is not aligned in any way, so it is the user's
4943 responsibility to make sure that proper alignment is provided both before and
4944 after the @code{incbin} directive.
4947 @section @code{.include "@var{file}"}
4949 @cindex @code{include} directive
4950 @cindex supporting files, including
4951 @cindex files, including
4952 This directive provides a way to include supporting files at specified
4953 points in your source program. The code from @var{file} is assembled as
4954 if it followed the point of the @code{.include}; when the end of the
4955 included file is reached, assembly of the original file continues. You
4956 can control the search paths used with the @samp{-I} command-line option
4957 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4961 @section @code{.int @var{expressions}}
4963 @cindex @code{int} directive
4964 @cindex integers, 32-bit
4965 Expect zero or more @var{expressions}, of any section, separated by commas.
4966 For each expression, emit a number that, at run time, is the value of that
4967 expression. The byte order and bit size of the number depends on what kind
4968 of target the assembly is for.
4972 On most forms of the H8/300, @code{.int} emits 16-bit
4973 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
4980 @section @code{.internal @var{names}}
4982 @cindex @code{internal} directive
4984 This is one of the ELF visibility directives. The other two are
4985 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
4986 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4988 This directive overrides the named symbols default visibility (which is set by
4989 their binding: local, global or weak). The directive sets the visibility to
4990 @code{internal} which means that the symbols are considered to be @code{hidden}
4991 (i.e., not visible to other components), and that some extra, processor specific
4992 processing must also be performed upon the symbols as well.
4996 @section @code{.irp @var{symbol},@var{values}}@dots{}
4998 @cindex @code{irp} directive
4999 Evaluate a sequence of statements assigning different values to @var{symbol}.
5000 The sequence of statements starts at the @code{.irp} directive, and is
5001 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5002 set to @var{value}, and the sequence of statements is assembled. If no
5003 @var{value} is listed, the sequence of statements is assembled once, with
5004 @var{symbol} set to the null string. To refer to @var{symbol} within the
5005 sequence of statements, use @var{\symbol}.
5007 For example, assembling
5015 is equivalent to assembling
5023 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5026 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5028 @cindex @code{irpc} directive
5029 Evaluate a sequence of statements assigning different values to @var{symbol}.
5030 The sequence of statements starts at the @code{.irpc} directive, and is
5031 terminated by an @code{.endr} directive. For each character in @var{value},
5032 @var{symbol} is set to the character, and the sequence of statements is
5033 assembled. If no @var{value} is listed, the sequence of statements is
5034 assembled once, with @var{symbol} set to the null string. To refer to
5035 @var{symbol} within the sequence of statements, use @var{\symbol}.
5037 For example, assembling
5045 is equivalent to assembling
5053 For some caveats with the spelling of @var{symbol}, see also the discussion
5057 @section @code{.lcomm @var{symbol} , @var{length}}
5059 @cindex @code{lcomm} directive
5060 @cindex local common symbols
5061 @cindex symbols, local common
5062 Reserve @var{length} (an absolute expression) bytes for a local common
5063 denoted by @var{symbol}. The section and value of @var{symbol} are
5064 those of the new local common. The addresses are allocated in the bss
5065 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5066 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5067 not visible to @code{@value{LD}}.
5070 Some targets permit a third argument to be used with @code{.lcomm}. This
5071 argument specifies the desired alignment of the symbol in the bss section.
5075 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5076 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5080 @section @code{.lflags}
5082 @cindex @code{lflags} directive (ignored)
5083 @command{@value{AS}} accepts this directive, for compatibility with other
5084 assemblers, but ignores it.
5086 @ifclear no-line-dir
5088 @section @code{.line @var{line-number}}
5090 @cindex @code{line} directive
5091 @cindex logical line number
5093 Change the logical line number. @var{line-number} must be an absolute
5094 expression. The next line has that logical line number. Therefore any other
5095 statements on the current line (after a statement separator character) are
5096 reported as on logical line number @var{line-number} @minus{} 1. One day
5097 @command{@value{AS}} will no longer support this directive: it is recognized only
5098 for compatibility with existing assembler programs.
5101 Even though this is a directive associated with the @code{a.out} or
5102 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5103 when producing COFF output, and treats @samp{.line} as though it
5104 were the COFF @samp{.ln} @emph{if} it is found outside a
5105 @code{.def}/@code{.endef} pair.
5107 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5108 used by compilers to generate auxiliary symbol information for
5113 @section @code{.linkonce [@var{type}]}
5115 @cindex @code{linkonce} directive
5116 @cindex common sections
5117 Mark the current section so that the linker only includes a single copy of it.
5118 This may be used to include the same section in several different object files,
5119 but ensure that the linker will only include it once in the final output file.
5120 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5121 Duplicate sections are detected based on the section name, so it should be
5124 This directive is only supported by a few object file formats; as of this
5125 writing, the only object file format which supports it is the Portable
5126 Executable format used on Windows NT.
5128 The @var{type} argument is optional. If specified, it must be one of the
5129 following strings. For example:
5133 Not all types may be supported on all object file formats.
5137 Silently discard duplicate sections. This is the default.
5140 Warn if there are duplicate sections, but still keep only one copy.
5143 Warn if any of the duplicates have different sizes.
5146 Warn if any of the duplicates do not have exactly the same contents.
5150 @section @code{.list}
5152 @cindex @code{list} directive
5153 @cindex listing control, turning on
5154 Control (in conjunction with the @code{.nolist} directive) whether or
5155 not assembly listings are generated. These two directives maintain an
5156 internal counter (which is zero initially). @code{.list} increments the
5157 counter, and @code{.nolist} decrements it. Assembly listings are
5158 generated whenever the counter is greater than zero.
5160 By default, listings are disabled. When you enable them (with the
5161 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5162 the initial value of the listing counter is one.
5165 @section @code{.ln @var{line-number}}
5167 @cindex @code{ln} directive
5168 @ifclear no-line-dir
5169 @samp{.ln} is a synonym for @samp{.line}.
5172 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5173 must be an absolute expression. The next line has that logical
5174 line number, so any other statements on the current line (after a
5175 statement separator character @code{;}) are reported as on logical
5176 line number @var{line-number} @minus{} 1.
5179 This directive is accepted, but ignored, when @command{@value{AS}} is
5180 configured for @code{b.out}; its effect is only associated with COFF
5186 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5187 @cindex @code{loc} directive
5188 When emitting DWARF2 line number information,
5189 the @code{.loc} directive will add a row to the @code{.debug_line} line
5190 number matrix corresponding to the immediately following assembly
5191 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5192 arguments will be applied to the @code{.debug_line} state machine before
5195 The @var{options} are a sequence of the following tokens in any order:
5199 This option will set the @code{basic_block} register in the
5200 @code{.debug_line} state machine to @code{true}.
5203 This option will set the @code{prologue_end} register in the
5204 @code{.debug_line} state machine to @code{true}.
5206 @item epilogue_begin
5207 This option will set the @code{epilogue_begin} register in the
5208 @code{.debug_line} state machine to @code{true}.
5210 @item is_stmt @var{value}
5211 This option will set the @code{is_stmt} register in the
5212 @code{.debug_line} state machine to @code{value}, which must be
5215 @item isa @var{value}
5216 This directive will set the @code{isa} register in the @code{.debug_line}
5217 state machine to @var{value}, which must be an unsigned integer.
5219 @item discriminator @var{value}
5220 This directive will set the @code{discriminator} register in the @code{.debug_line}
5221 state machine to @var{value}, which must be an unsigned integer.
5225 @node Loc_mark_labels
5226 @section @code{.loc_mark_labels @var{enable}}
5227 @cindex @code{loc_mark_labels} directive
5228 When emitting DWARF2 line number information,
5229 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5230 to the @code{.debug_line} line number matrix with the @code{basic_block}
5231 register in the state machine set whenever a code label is seen.
5232 The @var{enable} argument should be either 1 or 0, to enable or disable
5233 this function respectively.
5237 @section @code{.local @var{names}}
5239 @cindex @code{local} directive
5240 This directive, which is available for ELF targets, marks each symbol in
5241 the comma-separated list of @code{names} as a local symbol so that it
5242 will not be externally visible. If the symbols do not already exist,
5243 they will be created.
5245 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5246 accept an alignment argument, which is the case for most ELF targets,
5247 the @code{.local} directive can be used in combination with @code{.comm}
5248 (@pxref{Comm}) to define aligned local common data.
5252 @section @code{.long @var{expressions}}
5254 @cindex @code{long} directive
5255 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5258 @c no one seems to know what this is for or whether this description is
5259 @c what it really ought to do
5261 @section @code{.lsym @var{symbol}, @var{expression}}
5263 @cindex @code{lsym} directive
5264 @cindex symbol, not referenced in assembly
5265 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5266 the hash table, ensuring it cannot be referenced by name during the
5267 rest of the assembly. This sets the attributes of the symbol to be
5268 the same as the expression value:
5270 @var{other} = @var{descriptor} = 0
5271 @var{type} = @r{(section of @var{expression})}
5272 @var{value} = @var{expression}
5275 The new symbol is not flagged as external.
5279 @section @code{.macro}
5282 The commands @code{.macro} and @code{.endm} allow you to define macros that
5283 generate assembly output. For example, this definition specifies a macro
5284 @code{sum} that puts a sequence of numbers into memory:
5287 .macro sum from=0, to=5
5296 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5308 @item .macro @var{macname}
5309 @itemx .macro @var{macname} @var{macargs} @dots{}
5310 @cindex @code{macro} directive
5311 Begin the definition of a macro called @var{macname}. If your macro
5312 definition requires arguments, specify their names after the macro name,
5313 separated by commas or spaces. You can qualify the macro argument to
5314 indicate whether all invocations must specify a non-blank value (through
5315 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5316 (through @samp{:@code{vararg}}). You can supply a default value for any
5317 macro argument by following the name with @samp{=@var{deflt}}. You
5318 cannot define two macros with the same @var{macname} unless it has been
5319 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5320 definitions. For example, these are all valid @code{.macro} statements:
5324 Begin the definition of a macro called @code{comm}, which takes no
5327 @item .macro plus1 p, p1
5328 @itemx .macro plus1 p p1
5329 Either statement begins the definition of a macro called @code{plus1},
5330 which takes two arguments; within the macro definition, write
5331 @samp{\p} or @samp{\p1} to evaluate the arguments.
5333 @item .macro reserve_str p1=0 p2
5334 Begin the definition of a macro called @code{reserve_str}, with two
5335 arguments. The first argument has a default value, but not the second.
5336 After the definition is complete, you can call the macro either as
5337 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5338 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5339 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5340 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5342 @item .macro m p1:req, p2=0, p3:vararg
5343 Begin the definition of a macro called @code{m}, with at least three
5344 arguments. The first argument must always have a value specified, but
5345 not the second, which instead has a default value. The third formal
5346 will get assigned all remaining arguments specified at invocation time.
5348 When you call a macro, you can specify the argument values either by
5349 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5350 @samp{sum to=17, from=9}.
5354 Note that since each of the @var{macargs} can be an identifier exactly
5355 as any other one permitted by the target architecture, there may be
5356 occasional problems if the target hand-crafts special meanings to certain
5357 characters when they occur in a special position. For example, if the colon
5358 (@code{:}) is generally permitted to be part of a symbol name, but the
5359 architecture specific code special-cases it when occurring as the final
5360 character of a symbol (to denote a label), then the macro parameter
5361 replacement code will have no way of knowing that and consider the whole
5362 construct (including the colon) an identifier, and check only this
5363 identifier for being the subject to parameter substitution. So for example
5364 this macro definition:
5372 might not work as expected. Invoking @samp{label foo} might not create a label
5373 called @samp{foo} but instead just insert the text @samp{\l:} into the
5374 assembler source, probably generating an error about an unrecognised
5377 Similarly problems might occur with the period character (@samp{.})
5378 which is often allowed inside opcode names (and hence identifier names). So
5379 for example constructing a macro to build an opcode from a base name and a
5380 length specifier like this:
5383 .macro opcode base length
5388 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5389 instruction but instead generate some kind of error as the assembler tries to
5390 interpret the text @samp{\base.\length}.
5392 There are several possible ways around this problem:
5395 @item Insert white space
5396 If it is possible to use white space characters then this is the simplest
5405 @item Use @samp{\()}
5406 The string @samp{\()} can be used to separate the end of a macro argument from
5407 the following text. eg:
5410 .macro opcode base length
5415 @item Use the alternate macro syntax mode
5416 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5417 used as a separator. eg:
5427 Note: this problem of correctly identifying string parameters to pseudo ops
5428 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5429 and @code{.irpc} (@pxref{Irpc}) as well.
5432 @cindex @code{endm} directive
5433 Mark the end of a macro definition.
5436 @cindex @code{exitm} directive
5437 Exit early from the current macro definition.
5439 @cindex number of macros executed
5440 @cindex macros, count executed
5442 @command{@value{AS}} maintains a counter of how many macros it has
5443 executed in this pseudo-variable; you can copy that number to your
5444 output with @samp{\@@}, but @emph{only within a macro definition}.
5446 @item LOCAL @var{name} [ , @dots{} ]
5447 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5448 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5449 @xref{Altmacro,,@code{.altmacro}}.
5453 @section @code{.mri @var{val}}
5455 @cindex @code{mri} directive
5456 @cindex MRI mode, temporarily
5457 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5458 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5459 affects code assembled until the next @code{.mri} directive, or until the end
5460 of the file. @xref{M, MRI mode, MRI mode}.
5463 @section @code{.noaltmacro}
5464 Disable alternate macro mode. @xref{Altmacro}.
5467 @section @code{.nolist}
5469 @cindex @code{nolist} directive
5470 @cindex listing control, turning off
5471 Control (in conjunction with the @code{.list} directive) whether or
5472 not assembly listings are generated. These two directives maintain an
5473 internal counter (which is zero initially). @code{.list} increments the
5474 counter, and @code{.nolist} decrements it. Assembly listings are
5475 generated whenever the counter is greater than zero.
5478 @section @code{.octa @var{bignums}}
5480 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5481 @cindex @code{octa} directive
5482 @cindex integer, 16-byte
5483 @cindex sixteen byte integer
5484 This directive expects zero or more bignums, separated by commas. For each
5485 bignum, it emits a 16-byte integer.
5487 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5488 hence @emph{octa}-word for 16 bytes.
5491 @section @code{.org @var{new-lc} , @var{fill}}
5493 @cindex @code{org} directive
5494 @cindex location counter, advancing
5495 @cindex advancing location counter
5496 @cindex current address, advancing
5497 Advance the location counter of the current section to
5498 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5499 expression with the same section as the current subsection. That is,
5500 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5501 wrong section, the @code{.org} directive is ignored. To be compatible
5502 with former assemblers, if the section of @var{new-lc} is absolute,
5503 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5504 is the same as the current subsection.
5506 @code{.org} may only increase the location counter, or leave it
5507 unchanged; you cannot use @code{.org} to move the location counter
5510 @c double negative used below "not undefined" because this is a specific
5511 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5512 @c section. doc@cygnus.com 18feb91
5513 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5514 may not be undefined. If you really detest this restriction we eagerly await
5515 a chance to share your improved assembler.
5517 Beware that the origin is relative to the start of the section, not
5518 to the start of the subsection. This is compatible with other
5519 people's assemblers.
5521 When the location counter (of the current subsection) is advanced, the
5522 intervening bytes are filled with @var{fill} which should be an
5523 absolute expression. If the comma and @var{fill} are omitted,
5524 @var{fill} defaults to zero.
5527 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5529 @cindex padding the location counter given a power of two
5530 @cindex @code{p2align} directive
5531 Pad the location counter (in the current subsection) to a particular
5532 storage boundary. The first expression (which must be absolute) is the
5533 number of low-order zero bits the location counter must have after
5534 advancement. For example @samp{.p2align 3} advances the location
5535 counter until it a multiple of 8. If the location counter is already a
5536 multiple of 8, no change is needed.
5538 The second expression (also absolute) gives the fill value to be stored in the
5539 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5540 padding bytes are normally zero. However, on some systems, if the section is
5541 marked as containing code and the fill value is omitted, the space is filled
5542 with no-op instructions.
5544 The third expression is also absolute, and is also optional. If it is present,
5545 it is the maximum number of bytes that should be skipped by this alignment
5546 directive. If doing the alignment would require skipping more bytes than the
5547 specified maximum, then the alignment is not done at all. You can omit the
5548 fill value (the second argument) entirely by simply using two commas after the
5549 required alignment; this can be useful if you want the alignment to be filled
5550 with no-op instructions when appropriate.
5552 @cindex @code{p2alignw} directive
5553 @cindex @code{p2alignl} directive
5554 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5555 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5556 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5557 fill pattern as a four byte longword value. For example, @code{.p2alignw
5558 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5559 filled in with the value 0x368d (the exact placement of the bytes depends upon
5560 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5565 @section @code{.popsection}
5567 @cindex @code{popsection} directive
5568 @cindex Section Stack
5569 This is one of the ELF section stack manipulation directives. The others are
5570 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5571 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5574 This directive replaces the current section (and subsection) with the top
5575 section (and subsection) on the section stack. This section is popped off the
5581 @section @code{.previous}
5583 @cindex @code{previous} directive
5584 @cindex Section Stack
5585 This is one of the ELF section stack manipulation directives. The others are
5586 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5587 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5588 (@pxref{PopSection}).
5590 This directive swaps the current section (and subsection) with most recently
5591 referenced section/subsection pair prior to this one. Multiple
5592 @code{.previous} directives in a row will flip between two sections (and their
5593 subsections). For example:
5605 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
5611 # Now in section A subsection 1
5615 # Now in section B subsection 0
5618 # Now in section B subsection 1
5621 # Now in section B subsection 0
5625 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
5626 section B and 0x9abc into subsection 1 of section B.
5628 In terms of the section stack, this directive swaps the current section with
5629 the top section on the section stack.
5633 @section @code{.print @var{string}}
5635 @cindex @code{print} directive
5636 @command{@value{AS}} will print @var{string} on the standard output during
5637 assembly. You must put @var{string} in double quotes.
5641 @section @code{.protected @var{names}}
5643 @cindex @code{protected} directive
5645 This is one of the ELF visibility directives. The other two are
5646 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5648 This directive overrides the named symbols default visibility (which is set by
5649 their binding: local, global or weak). The directive sets the visibility to
5650 @code{protected} which means that any references to the symbols from within the
5651 components that defines them must be resolved to the definition in that
5652 component, even if a definition in another component would normally preempt
5657 @section @code{.psize @var{lines} , @var{columns}}
5659 @cindex @code{psize} directive
5660 @cindex listing control: paper size
5661 @cindex paper size, for listings
5662 Use this directive to declare the number of lines---and, optionally, the
5663 number of columns---to use for each page, when generating listings.
5665 If you do not use @code{.psize}, listings use a default line-count
5666 of 60. You may omit the comma and @var{columns} specification; the
5667 default width is 200 columns.
5669 @command{@value{AS}} generates formfeeds whenever the specified number of
5670 lines is exceeded (or whenever you explicitly request one, using
5673 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5674 those explicitly specified with @code{.eject}.
5677 @section @code{.purgem @var{name}}
5679 @cindex @code{purgem} directive
5680 Undefine the macro @var{name}, so that later uses of the string will not be
5681 expanded. @xref{Macro}.
5685 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
5687 @cindex @code{pushsection} directive
5688 @cindex Section Stack
5689 This is one of the ELF section stack manipulation directives. The others are
5690 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5691 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5694 This directive pushes the current section (and subsection) onto the
5695 top of the section stack, and then replaces the current section and
5696 subsection with @code{name} and @code{subsection}. The optional
5697 @code{flags}, @code{type} and @code{arguments} are treated the same
5698 as in the @code{.section} (@pxref{Section}) directive.
5702 @section @code{.quad @var{bignums}}
5704 @cindex @code{quad} directive
5705 @code{.quad} expects zero or more bignums, separated by commas. For
5706 each bignum, it emits
5708 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5709 warning message; and just takes the lowest order 8 bytes of the bignum.
5710 @cindex eight-byte integer
5711 @cindex integer, 8-byte
5713 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5714 hence @emph{quad}-word for 8 bytes.
5717 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5718 warning message; and just takes the lowest order 16 bytes of the bignum.
5719 @cindex sixteen-byte integer
5720 @cindex integer, 16-byte
5724 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
5726 @cindex @code{reloc} directive
5727 Generate a relocation at @var{offset} of type @var{reloc_name} with value
5728 @var{expression}. If @var{offset} is a number, the relocation is generated in
5729 the current section. If @var{offset} is an expression that resolves to a
5730 symbol plus offset, the relocation is generated in the given symbol's section.
5731 @var{expression}, if present, must resolve to a symbol plus addend or to an
5732 absolute value, but note that not all targets support an addend. e.g. ELF REL
5733 targets such as i386 store an addend in the section contents rather than in the
5734 relocation. This low level interface does not support addends stored in the
5738 @section @code{.rept @var{count}}
5740 @cindex @code{rept} directive
5741 Repeat the sequence of lines between the @code{.rept} directive and the next
5742 @code{.endr} directive @var{count} times.
5744 For example, assembling
5752 is equivalent to assembling
5761 @section @code{.sbttl "@var{subheading}"}
5763 @cindex @code{sbttl} directive
5764 @cindex subtitles for listings
5765 @cindex listing control: subtitle
5766 Use @var{subheading} as the title (third line, immediately after the
5767 title line) when generating assembly listings.
5769 This directive affects subsequent pages, as well as the current page if
5770 it appears within ten lines of the top of a page.
5774 @section @code{.scl @var{class}}
5776 @cindex @code{scl} directive
5777 @cindex symbol storage class (COFF)
5778 @cindex COFF symbol storage class
5779 Set the storage-class value for a symbol. This directive may only be
5780 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5781 whether a symbol is static or external, or it may record further
5782 symbolic debugging information.
5785 The @samp{.scl} directive is primarily associated with COFF output; when
5786 configured to generate @code{b.out} output format, @command{@value{AS}}
5787 accepts this directive but ignores it.
5793 @section @code{.section @var{name}}
5795 @cindex named section
5796 Use the @code{.section} directive to assemble the following code into a section
5799 This directive is only supported for targets that actually support arbitrarily
5800 named sections; on @code{a.out} targets, for example, it is not accepted, even
5801 with a standard @code{a.out} section name.
5805 @c only print the extra heading if both COFF and ELF are set
5806 @subheading COFF Version
5809 @cindex @code{section} directive (COFF version)
5810 For COFF targets, the @code{.section} directive is used in one of the following
5814 .section @var{name}[, "@var{flags}"]
5815 .section @var{name}[, @var{subsection}]
5818 If the optional argument is quoted, it is taken as flags to use for the
5819 section. Each flag is a single character. The following flags are recognized:
5822 bss section (uninitialized data)
5824 section is not loaded
5834 shared section (meaningful for PE targets)
5836 ignored. (For compatibility with the ELF version)
5838 section is not readable (meaningful for PE targets)
5840 single-digit power-of-two section alignment (GNU extension)
5843 If no flags are specified, the default flags depend upon the section name. If
5844 the section name is not recognized, the default will be for the section to be
5845 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
5846 from the section, rather than adding them, so if they are used on their own it
5847 will be as if no flags had been specified at all.
5849 If the optional argument to the @code{.section} directive is not quoted, it is
5850 taken as a subsection number (@pxref{Sub-Sections}).
5855 @c only print the extra heading if both COFF and ELF are set
5856 @subheading ELF Version
5859 @cindex Section Stack
5860 This is one of the ELF section stack manipulation directives. The others are
5861 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
5862 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
5863 @code{.previous} (@pxref{Previous}).
5865 @cindex @code{section} directive (ELF version)
5866 For ELF targets, the @code{.section} directive is used like this:
5869 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
5872 The optional @var{flags} argument is a quoted string which may contain any
5873 combination of the following characters:
5876 section is allocatable
5880 section is executable
5882 section is mergeable
5884 section contains zero terminated strings
5886 section is a member of a section group
5888 section is used for thread-local-storage
5891 The optional @var{type} argument may contain one of the following constants:
5894 section contains data
5896 section does not contain data (i.e., section only occupies space)
5898 section contains data which is used by things other than the program
5900 section contains an array of pointers to init functions
5902 section contains an array of pointers to finish functions
5903 @item @@preinit_array
5904 section contains an array of pointers to pre-init functions
5907 Many targets only support the first three section types.
5909 Note on targets where the @code{@@} character is the start of a comment (eg
5910 ARM) then another character is used instead. For example the ARM port uses the
5913 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
5914 be specified as well as an extra argument---@var{entsize}---like this:
5917 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
5920 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
5921 constants, each @var{entsize} octets long. Sections with both @code{M} and
5922 @code{S} must contain zero terminated strings where each character is
5923 @var{entsize} bytes long. The linker may remove duplicates within sections with
5924 the same name, same entity size and same flags. @var{entsize} must be an
5925 absolute expression. For sections with both @code{M} and @code{S}, a string
5926 which is a suffix of a larger string is considered a duplicate. Thus
5927 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
5928 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
5930 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
5931 be present along with an additional field like this:
5934 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
5937 The @var{GroupName} field specifies the name of the section group to which this
5938 particular section belongs. The optional linkage field can contain:
5941 indicates that only one copy of this section should be retained
5946 Note: if both the @var{M} and @var{G} flags are present then the fields for
5947 the Merge flag should come first, like this:
5950 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
5953 If no flags are specified, the default flags depend upon the section name. If
5954 the section name is not recognized, the default will be for the section to have
5955 none of the above flags: it will not be allocated in memory, nor writable, nor
5956 executable. The section will contain data.
5958 For ELF targets, the assembler supports another type of @code{.section}
5959 directive for compatibility with the Solaris assembler:
5962 .section "@var{name}"[, @var{flags}...]
5965 Note that the section name is quoted. There may be a sequence of comma
5969 section is allocatable
5973 section is executable
5975 section is used for thread local storage
5978 This directive replaces the current section and subsection. See the
5979 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
5980 some examples of how this directive and the other section stack directives
5986 @section @code{.set @var{symbol}, @var{expression}}
5988 @cindex @code{set} directive
5989 @cindex symbol value, setting
5990 Set the value of @var{symbol} to @var{expression}. This
5991 changes @var{symbol}'s value and type to conform to
5992 @var{expression}. If @var{symbol} was flagged as external, it remains
5993 flagged (@pxref{Symbol Attributes}).
5995 You may @code{.set} a symbol many times in the same assembly.
5997 If you @code{.set} a global symbol, the value stored in the object
5998 file is the last value stored into it.
6001 On Z80 @code{set} is a real instruction, use
6002 @samp{@var{symbol} defl @var{expression}} instead.
6006 @section @code{.short @var{expressions}}
6008 @cindex @code{short} directive
6010 @code{.short} is normally the same as @samp{.word}.
6011 @xref{Word,,@code{.word}}.
6013 In some configurations, however, @code{.short} and @code{.word} generate
6014 numbers of different lengths. @xref{Machine Dependencies}.
6018 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6021 This expects zero or more @var{expressions}, and emits
6022 a 16 bit number for each.
6027 @section @code{.single @var{flonums}}
6029 @cindex @code{single} directive
6030 @cindex floating point numbers (single)
6031 This directive assembles zero or more flonums, separated by commas. It
6032 has the same effect as @code{.float}.
6034 The exact kind of floating point numbers emitted depends on how
6035 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6039 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6040 numbers in @sc{ieee} format.
6046 @section @code{.size}
6048 This directive is used to set the size associated with a symbol.
6052 @c only print the extra heading if both COFF and ELF are set
6053 @subheading COFF Version
6056 @cindex @code{size} directive (COFF version)
6057 For COFF targets, the @code{.size} directive is only permitted inside
6058 @code{.def}/@code{.endef} pairs. It is used like this:
6061 .size @var{expression}
6065 @samp{.size} is only meaningful when generating COFF format output; when
6066 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6073 @c only print the extra heading if both COFF and ELF are set
6074 @subheading ELF Version
6077 @cindex @code{size} directive (ELF version)
6078 For ELF targets, the @code{.size} directive is used like this:
6081 .size @var{name} , @var{expression}
6084 This directive sets the size associated with a symbol @var{name}.
6085 The size in bytes is computed from @var{expression} which can make use of label
6086 arithmetic. This directive is typically used to set the size of function
6091 @ifclear no-space-dir
6093 @section @code{.skip @var{size} , @var{fill}}
6095 @cindex @code{skip} directive
6096 @cindex filling memory
6097 This directive emits @var{size} bytes, each of value @var{fill}. Both
6098 @var{size} and @var{fill} are absolute expressions. If the comma and
6099 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6104 @section @code{.sleb128 @var{expressions}}
6106 @cindex @code{sleb128} directive
6107 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6108 compact, variable length representation of numbers used by the DWARF
6109 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6111 @ifclear no-space-dir
6113 @section @code{.space @var{size} , @var{fill}}
6115 @cindex @code{space} directive
6116 @cindex filling memory
6117 This directive emits @var{size} bytes, each of value @var{fill}. Both
6118 @var{size} and @var{fill} are absolute expressions. If the comma
6119 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6124 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6125 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6126 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6127 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6135 @section @code{.stabd, .stabn, .stabs}
6137 @cindex symbolic debuggers, information for
6138 @cindex @code{stab@var{x}} directives
6139 There are three directives that begin @samp{.stab}.
6140 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6141 The symbols are not entered in the @command{@value{AS}} hash table: they
6142 cannot be referenced elsewhere in the source file.
6143 Up to five fields are required:
6147 This is the symbol's name. It may contain any character except
6148 @samp{\000}, so is more general than ordinary symbol names. Some
6149 debuggers used to code arbitrarily complex structures into symbol names
6153 An absolute expression. The symbol's type is set to the low 8 bits of
6154 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6155 and debuggers choke on silly bit patterns.
6158 An absolute expression. The symbol's ``other'' attribute is set to the
6159 low 8 bits of this expression.
6162 An absolute expression. The symbol's descriptor is set to the low 16
6163 bits of this expression.
6166 An absolute expression which becomes the symbol's value.
6169 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6170 or @code{.stabs} statement, the symbol has probably already been created;
6171 you get a half-formed symbol in your object file. This is
6172 compatible with earlier assemblers!
6175 @cindex @code{stabd} directive
6176 @item .stabd @var{type} , @var{other} , @var{desc}
6178 The ``name'' of the symbol generated is not even an empty string.
6179 It is a null pointer, for compatibility. Older assemblers used a
6180 null pointer so they didn't waste space in object files with empty
6183 The symbol's value is set to the location counter,
6184 relocatably. When your program is linked, the value of this symbol
6185 is the address of the location counter when the @code{.stabd} was
6188 @cindex @code{stabn} directive
6189 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6190 The name of the symbol is set to the empty string @code{""}.
6192 @cindex @code{stabs} directive
6193 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6194 All five fields are specified.
6200 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6201 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6203 @cindex string, copying to object file
6204 @cindex string8, copying to object file
6205 @cindex string16, copying to object file
6206 @cindex string32, copying to object file
6207 @cindex string64, copying to object file
6208 @cindex @code{string} directive
6209 @cindex @code{string8} directive
6210 @cindex @code{string16} directive
6211 @cindex @code{string32} directive
6212 @cindex @code{string64} directive
6214 Copy the characters in @var{str} to the object file. You may specify more than
6215 one string to copy, separated by commas. Unless otherwise specified for a
6216 particular machine, the assembler marks the end of each string with a 0 byte.
6217 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6219 The variants @code{string16}, @code{string32} and @code{string64} differ from
6220 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6221 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6222 are stored in target endianness byte order.
6228 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6229 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6234 @section @code{.struct @var{expression}}
6236 @cindex @code{struct} directive
6237 Switch to the absolute section, and set the section offset to @var{expression},
6238 which must be an absolute expression. You might use this as follows:
6247 This would define the symbol @code{field1} to have the value 0, the symbol
6248 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6249 value 8. Assembly would be left in the absolute section, and you would need to
6250 use a @code{.section} directive of some sort to change to some other section
6251 before further assembly.
6255 @section @code{.subsection @var{name}}
6257 @cindex @code{subsection} directive
6258 @cindex Section Stack
6259 This is one of the ELF section stack manipulation directives. The others are
6260 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6261 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6264 This directive replaces the current subsection with @code{name}. The current
6265 section is not changed. The replaced subsection is put onto the section stack
6266 in place of the then current top of stack subsection.
6271 @section @code{.symver}
6272 @cindex @code{symver} directive
6273 @cindex symbol versioning
6274 @cindex versions of symbols
6275 Use the @code{.symver} directive to bind symbols to specific version nodes
6276 within a source file. This is only supported on ELF platforms, and is
6277 typically used when assembling files to be linked into a shared library.
6278 There are cases where it may make sense to use this in objects to be bound
6279 into an application itself so as to override a versioned symbol from a
6282 For ELF targets, the @code{.symver} directive can be used like this:
6284 .symver @var{name}, @var{name2@@nodename}
6286 If the symbol @var{name} is defined within the file
6287 being assembled, the @code{.symver} directive effectively creates a symbol
6288 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6289 just don't try and create a regular alias is that the @var{@@} character isn't
6290 permitted in symbol names. The @var{name2} part of the name is the actual name
6291 of the symbol by which it will be externally referenced. The name @var{name}
6292 itself is merely a name of convenience that is used so that it is possible to
6293 have definitions for multiple versions of a function within a single source
6294 file, and so that the compiler can unambiguously know which version of a
6295 function is being mentioned. The @var{nodename} portion of the alias should be
6296 the name of a node specified in the version script supplied to the linker when
6297 building a shared library. If you are attempting to override a versioned
6298 symbol from a shared library, then @var{nodename} should correspond to the
6299 nodename of the symbol you are trying to override.
6301 If the symbol @var{name} is not defined within the file being assembled, all
6302 references to @var{name} will be changed to @var{name2@@nodename}. If no
6303 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6306 Another usage of the @code{.symver} directive is:
6308 .symver @var{name}, @var{name2@@@@nodename}
6310 In this case, the symbol @var{name} must exist and be defined within
6311 the file being assembled. It is similar to @var{name2@@nodename}. The
6312 difference is @var{name2@@@@nodename} will also be used to resolve
6313 references to @var{name2} by the linker.
6315 The third usage of the @code{.symver} directive is:
6317 .symver @var{name}, @var{name2@@@@@@nodename}
6319 When @var{name} is not defined within the
6320 file being assembled, it is treated as @var{name2@@nodename}. When
6321 @var{name} is defined within the file being assembled, the symbol
6322 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6327 @section @code{.tag @var{structname}}
6329 @cindex COFF structure debugging
6330 @cindex structure debugging, COFF
6331 @cindex @code{tag} directive
6332 This directive is generated by compilers to include auxiliary debugging
6333 information in the symbol table. It is only permitted inside
6334 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6335 definitions in the symbol table with instances of those structures.
6338 @samp{.tag} is only used when generating COFF format output; when
6339 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6345 @section @code{.text @var{subsection}}
6347 @cindex @code{text} directive
6348 Tells @command{@value{AS}} to assemble the following statements onto the end of
6349 the text subsection numbered @var{subsection}, which is an absolute
6350 expression. If @var{subsection} is omitted, subsection number zero
6354 @section @code{.title "@var{heading}"}
6356 @cindex @code{title} directive
6357 @cindex listing control: title line
6358 Use @var{heading} as the title (second line, immediately after the
6359 source file name and pagenumber) when generating assembly listings.
6361 This directive affects subsequent pages, as well as the current page if
6362 it appears within ten lines of the top of a page.
6366 @section @code{.type}
6368 This directive is used to set the type of a symbol.
6372 @c only print the extra heading if both COFF and ELF are set
6373 @subheading COFF Version
6376 @cindex COFF symbol type
6377 @cindex symbol type, COFF
6378 @cindex @code{type} directive (COFF version)
6379 For COFF targets, this directive is permitted only within
6380 @code{.def}/@code{.endef} pairs. It is used like this:
6386 This records the integer @var{int} as the type attribute of a symbol table
6390 @samp{.type} is associated only with COFF format output; when
6391 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6392 directive but ignores it.
6398 @c only print the extra heading if both COFF and ELF are set
6399 @subheading ELF Version
6402 @cindex ELF symbol type
6403 @cindex symbol type, ELF
6404 @cindex @code{type} directive (ELF version)
6405 For ELF targets, the @code{.type} directive is used like this:
6408 .type @var{name} , @var{type description}
6411 This sets the type of symbol @var{name} to be either a
6412 function symbol or an object symbol. There are five different syntaxes
6413 supported for the @var{type description} field, in order to provide
6414 compatibility with various other assemblers.
6416 Because some of the characters used in these syntaxes (such as @samp{@@} and
6417 @samp{#}) are comment characters for some architectures, some of the syntaxes
6418 below do not work on all architectures. The first variant will be accepted by
6419 the GNU assembler on all architectures so that variant should be used for
6420 maximum portability, if you do not need to assemble your code with other
6423 The syntaxes supported are:
6426 .type <name> STT_<TYPE_IN_UPPER_CASE>
6427 .type <name>,#<type>
6428 .type <name>,@@<type>
6429 .type <name>,%<type>
6430 .type <name>,"<type>"
6433 The types supported are:
6438 Mark the symbol as being a function name.
6441 @itemx gnu_indirect_function
6442 Mark the symbol as an indirect function when evaluated during reloc
6443 processing. (This is only supported on Linux targeted assemblers).
6447 Mark the symbol as being a data object.
6451 Mark the symbol as being a thead-local data object.
6455 Mark the symbol as being a common data object.
6459 Does not mark the symbol in any way. It is supported just for completeness.
6461 @item gnu_unique_object
6462 Marks the symbol as being a globally unique data object. The dynamic linker
6463 will make sure that in the entire process there is just one symbol with this
6464 name and type in use. (This is only supported on Linux targeted assemblers).
6468 Note: Some targets support extra types in addition to those listed above.
6474 @section @code{.uleb128 @var{expressions}}
6476 @cindex @code{uleb128} directive
6477 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6478 compact, variable length representation of numbers used by the DWARF
6479 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6483 @section @code{.val @var{addr}}
6485 @cindex @code{val} directive
6486 @cindex COFF value attribute
6487 @cindex value attribute, COFF
6488 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6489 records the address @var{addr} as the value attribute of a symbol table
6493 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6494 configured for @code{b.out}, it accepts this directive but ignores it.
6500 @section @code{.version "@var{string}"}
6502 @cindex @code{version} directive
6503 This directive creates a @code{.note} section and places into it an ELF
6504 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6509 @section @code{.vtable_entry @var{table}, @var{offset}}
6511 @cindex @code{vtable_entry} directive
6512 This directive finds or creates a symbol @code{table} and creates a
6513 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6516 @section @code{.vtable_inherit @var{child}, @var{parent}}
6518 @cindex @code{vtable_inherit} directive
6519 This directive finds the symbol @code{child} and finds or creates the symbol
6520 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6521 parent whose addend is the value of the child symbol. As a special case the
6522 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6526 @section @code{.warning "@var{string}"}
6527 @cindex warning directive
6528 Similar to the directive @code{.error}
6529 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6532 @section @code{.weak @var{names}}
6534 @cindex @code{weak} directive
6535 This directive sets the weak attribute on the comma separated list of symbol
6536 @code{names}. If the symbols do not already exist, they will be created.
6538 On COFF targets other than PE, weak symbols are a GNU extension. This
6539 directive sets the weak attribute on the comma separated list of symbol
6540 @code{names}. If the symbols do not already exist, they will be created.
6542 On the PE target, weak symbols are supported natively as weak aliases.
6543 When a weak symbol is created that is not an alias, GAS creates an
6544 alternate symbol to hold the default value.
6547 @section @code{.weakref @var{alias}, @var{target}}
6549 @cindex @code{weakref} directive
6550 This directive creates an alias to the target symbol that enables the symbol to
6551 be referenced with weak-symbol semantics, but without actually making it weak.
6552 If direct references or definitions of the symbol are present, then the symbol
6553 will not be weak, but if all references to it are through weak references, the
6554 symbol will be marked as weak in the symbol table.
6556 The effect is equivalent to moving all references to the alias to a separate
6557 assembly source file, renaming the alias to the symbol in it, declaring the
6558 symbol as weak there, and running a reloadable link to merge the object files
6559 resulting from the assembly of the new source file and the old source file that
6560 had the references to the alias removed.
6562 The alias itself never makes to the symbol table, and is entirely handled
6563 within the assembler.
6566 @section @code{.word @var{expressions}}
6568 @cindex @code{word} directive
6569 This directive expects zero or more @var{expressions}, of any section,
6570 separated by commas.
6573 For each expression, @command{@value{AS}} emits a 32-bit number.
6576 For each expression, @command{@value{AS}} emits a 16-bit number.
6581 The size of the number emitted, and its byte order,
6582 depend on what target computer the assembly is for.
6585 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
6586 @c happen---32-bit addressability, period; no long/short jumps.
6587 @ifset DIFF-TBL-KLUGE
6588 @cindex difference tables altered
6589 @cindex altered difference tables
6591 @emph{Warning: Special Treatment to support Compilers}
6595 Machines with a 32-bit address space, but that do less than 32-bit
6596 addressing, require the following special treatment. If the machine of
6597 interest to you does 32-bit addressing (or doesn't require it;
6598 @pxref{Machine Dependencies}), you can ignore this issue.
6601 In order to assemble compiler output into something that works,
6602 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6603 Directives of the form @samp{.word sym1-sym2} are often emitted by
6604 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6605 directive of the form @samp{.word sym1-sym2}, and the difference between
6606 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6607 creates a @dfn{secondary jump table}, immediately before the next label.
6608 This secondary jump table is preceded by a short-jump to the
6609 first byte after the secondary table. This short-jump prevents the flow
6610 of control from accidentally falling into the new table. Inside the
6611 table is a long-jump to @code{sym2}. The original @samp{.word}
6612 contains @code{sym1} minus the address of the long-jump to
6615 If there were several occurrences of @samp{.word sym1-sym2} before the
6616 secondary jump table, all of them are adjusted. If there was a
6617 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6618 long-jump to @code{sym4} is included in the secondary jump table,
6619 and the @code{.word} directives are adjusted to contain @code{sym3}
6620 minus the address of the long-jump to @code{sym4}; and so on, for as many
6621 entries in the original jump table as necessary.
6624 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6625 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6626 assembly language programmers.
6629 @c end DIFF-TBL-KLUGE
6632 @section Deprecated Directives
6634 @cindex deprecated directives
6635 @cindex obsolescent directives
6636 One day these directives won't work.
6637 They are included for compatibility with older assemblers.
6644 @node Object Attributes
6645 @chapter Object Attributes
6646 @cindex object attributes
6648 @command{@value{AS}} assembles source files written for a specific architecture
6649 into object files for that architecture. But not all object files are alike.
6650 Many architectures support incompatible variations. For instance, floating
6651 point arguments might be passed in floating point registers if the object file
6652 requires hardware floating point support---or floating point arguments might be
6653 passed in integer registers if the object file supports processors with no
6654 hardware floating point unit. Or, if two objects are built for different
6655 generations of the same architecture, the combination may require the
6656 newer generation at run-time.
6658 This information is useful during and after linking. At link time,
6659 @command{@value{LD}} can warn about incompatible object files. After link
6660 time, tools like @command{gdb} can use it to process the linked file
6663 Compatibility information is recorded as a series of object attributes. Each
6664 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
6665 string, and indicates who sets the meaning of the tag. The tag is an integer,
6666 and indicates what property the attribute describes. The value may be a string
6667 or an integer, and indicates how the property affects this object. Missing
6668 attributes are the same as attributes with a zero value or empty string value.
6670 Object attributes were developed as part of the ABI for the ARM Architecture.
6671 The file format is documented in @cite{ELF for the ARM Architecture}.
6674 * GNU Object Attributes:: @sc{gnu} Object Attributes
6675 * Defining New Object Attributes:: Defining New Object Attributes
6678 @node GNU Object Attributes
6679 @section @sc{gnu} Object Attributes
6681 The @code{.gnu_attribute} directive records an object attribute
6682 with vendor @samp{gnu}.
6684 Except for @samp{Tag_compatibility}, which has both an integer and a string for
6685 its value, @sc{gnu} attributes have a string value if the tag number is odd and
6686 an integer value if the tag number is even. The second bit (@code{@var{tag} &
6687 2} is set for architecture-independent attributes and clear for
6688 architecture-dependent ones.
6690 @subsection Common @sc{gnu} attributes
6692 These attributes are valid on all architectures.
6695 @item Tag_compatibility (32)
6696 The compatibility attribute takes an integer flag value and a vendor name. If
6697 the flag value is 0, the file is compatible with other toolchains. If it is 1,
6698 then the file is only compatible with the named toolchain. If it is greater
6699 than 1, the file can only be processed by other toolchains under some private
6700 arrangement indicated by the flag value and the vendor name.
6703 @subsection MIPS Attributes
6706 @item Tag_GNU_MIPS_ABI_FP (4)
6707 The floating-point ABI used by this object file. The value will be:
6711 0 for files not affected by the floating-point ABI.
6713 1 for files using the hardware floating-point with a standard double-precision
6716 2 for files using the hardware floating-point ABI with a single-precision FPU.
6718 3 for files using the software floating-point ABI.
6720 4 for files using the hardware floating-point ABI with 64-bit wide
6721 double-precision floating-point registers and 32-bit wide general
6726 @subsection PowerPC Attributes
6729 @item Tag_GNU_Power_ABI_FP (4)
6730 The floating-point ABI used by this object file. The value will be:
6734 0 for files not affected by the floating-point ABI.
6736 1 for files using double-precision hardware floating-point ABI.
6738 2 for files using the software floating-point ABI.
6740 3 for files using single-precision hardware floating-point ABI.
6743 @item Tag_GNU_Power_ABI_Vector (8)
6744 The vector ABI used by this object file. The value will be:
6748 0 for files not affected by the vector ABI.
6750 1 for files using general purpose registers to pass vectors.
6752 2 for files using AltiVec registers to pass vectors.
6754 3 for files using SPE registers to pass vectors.
6758 @node Defining New Object Attributes
6759 @section Defining New Object Attributes
6761 If you want to define a new @sc{gnu} object attribute, here are the places you
6762 will need to modify. New attributes should be discussed on the @samp{binutils}
6767 This manual, which is the official register of attributes.
6769 The header for your architecture @file{include/elf}, to define the tag.
6771 The @file{bfd} support file for your architecture, to merge the attribute
6772 and issue any appropriate link warnings.
6774 Test cases in @file{ld/testsuite} for merging and link warnings.
6776 @file{binutils/readelf.c} to display your attribute.
6778 GCC, if you want the compiler to mark the attribute automatically.
6784 @node Machine Dependencies
6785 @chapter Machine Dependent Features
6787 @cindex machine dependencies
6788 The machine instruction sets are (almost by definition) different on
6789 each machine where @command{@value{AS}} runs. Floating point representations
6790 vary as well, and @command{@value{AS}} often supports a few additional
6791 directives or command-line options for compatibility with other
6792 assemblers on a particular platform. Finally, some versions of
6793 @command{@value{AS}} support special pseudo-instructions for branch
6796 This chapter discusses most of these differences, though it does not
6797 include details on any machine's instruction set. For details on that
6798 subject, see the hardware manufacturer's manual.
6802 * Alpha-Dependent:: Alpha Dependent Features
6805 * ARC-Dependent:: ARC Dependent Features
6808 * ARM-Dependent:: ARM Dependent Features
6811 * AVR-Dependent:: AVR Dependent Features
6814 * Blackfin-Dependent:: Blackfin Dependent Features
6817 * CR16-Dependent:: CR16 Dependent Features
6820 * CRIS-Dependent:: CRIS Dependent Features
6823 * D10V-Dependent:: D10V Dependent Features
6826 * D30V-Dependent:: D30V Dependent Features
6829 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6832 * HPPA-Dependent:: HPPA Dependent Features
6835 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
6838 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
6841 * i860-Dependent:: Intel 80860 Dependent Features
6844 * i960-Dependent:: Intel 80960 Dependent Features
6847 * IA-64-Dependent:: Intel IA-64 Dependent Features
6850 * IP2K-Dependent:: IP2K Dependent Features
6853 * LM32-Dependent:: LM32 Dependent Features
6856 * M32C-Dependent:: M32C Dependent Features
6859 * M32R-Dependent:: M32R Dependent Features
6862 * M68K-Dependent:: M680x0 Dependent Features
6865 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
6868 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
6871 * MIPS-Dependent:: MIPS Dependent Features
6874 * MMIX-Dependent:: MMIX Dependent Features
6877 * MSP430-Dependent:: MSP430 Dependent Features
6880 * SH-Dependent:: Renesas / SuperH SH Dependent Features
6881 * SH64-Dependent:: SuperH SH64 Dependent Features
6884 * PDP-11-Dependent:: PDP-11 Dependent Features
6887 * PJ-Dependent:: picoJava Dependent Features
6890 * PPC-Dependent:: PowerPC Dependent Features
6893 * RX-Dependent:: RX Dependent Features
6896 * S/390-Dependent:: IBM S/390 Dependent Features
6899 * SCORE-Dependent:: SCORE Dependent Features
6902 * Sparc-Dependent:: SPARC Dependent Features
6905 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
6908 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
6911 * V850-Dependent:: V850 Dependent Features
6914 * Xtensa-Dependent:: Xtensa Dependent Features
6917 * Z80-Dependent:: Z80 Dependent Features
6920 * Z8000-Dependent:: Z8000 Dependent Features
6923 * Vax-Dependent:: VAX Dependent Features
6930 @c The following major nodes are *sections* in the GENERIC version, *chapters*
6931 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
6932 @c peculiarity: to preserve cross-references, there must be a node called
6933 @c "Machine Dependencies". Hence the conditional nodenames in each
6934 @c major node below. Node defaulting in makeinfo requires adjacency of
6935 @c node and sectioning commands; hence the repetition of @chapter BLAH
6936 @c in both conditional blocks.
6939 @include c-alpha.texi
6955 @include c-bfin.texi
6959 @include c-cr16.texi
6963 @include c-cris.texi
6968 @node Machine Dependencies
6969 @chapter Machine Dependent Features
6971 The machine instruction sets are different on each Renesas chip family,
6972 and there are also some syntax differences among the families. This
6973 chapter describes the specific @command{@value{AS}} features for each
6977 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6978 * SH-Dependent:: Renesas SH Dependent Features
6985 @include c-d10v.texi
6989 @include c-d30v.texi
6993 @include c-h8300.texi
6997 @include c-hppa.texi
7001 @include c-i370.texi
7005 @include c-i386.texi
7009 @include c-i860.texi
7013 @include c-i960.texi
7017 @include c-ia64.texi
7021 @include c-ip2k.texi
7025 @include c-lm32.texi
7029 @include c-m32c.texi
7033 @include c-m32r.texi
7037 @include c-m68k.texi
7041 @include c-m68hc11.texi
7045 @include c-microblaze.texi
7049 @include c-mips.texi
7053 @include c-mmix.texi
7057 @include c-msp430.texi
7061 @include c-ns32k.texi
7065 @include c-pdp11.texi
7081 @include c-s390.texi
7085 @include c-score.texi
7090 @include c-sh64.texi
7094 @include c-sparc.texi
7098 @include c-tic54x.texi
7102 @include c-tic6x.texi
7118 @include c-v850.texi
7122 @include c-xtensa.texi
7126 @c reverse effect of @down at top of generic Machine-Dep chapter
7130 @node Reporting Bugs
7131 @chapter Reporting Bugs
7132 @cindex bugs in assembler
7133 @cindex reporting bugs in assembler
7135 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7137 Reporting a bug may help you by bringing a solution to your problem, or it may
7138 not. But in any case the principal function of a bug report is to help the
7139 entire community by making the next version of @command{@value{AS}} work better.
7140 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7142 In order for a bug report to serve its purpose, you must include the
7143 information that enables us to fix the bug.
7146 * Bug Criteria:: Have you found a bug?
7147 * Bug Reporting:: How to report bugs
7151 @section Have You Found a Bug?
7152 @cindex bug criteria
7154 If you are not sure whether you have found a bug, here are some guidelines:
7157 @cindex fatal signal
7158 @cindex assembler crash
7159 @cindex crash of assembler
7161 If the assembler gets a fatal signal, for any input whatever, that is a
7162 @command{@value{AS}} bug. Reliable assemblers never crash.
7164 @cindex error on valid input
7166 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7168 @cindex invalid input
7170 If @command{@value{AS}} does not produce an error message for invalid input, that
7171 is a bug. However, you should note that your idea of ``invalid input'' might
7172 be our idea of ``an extension'' or ``support for traditional practice''.
7175 If you are an experienced user of assemblers, your suggestions for improvement
7176 of @command{@value{AS}} are welcome in any case.
7180 @section How to Report Bugs
7182 @cindex assembler bugs, reporting
7184 A number of companies and individuals offer support for @sc{gnu} products. If
7185 you obtained @command{@value{AS}} from a support organization, we recommend you
7186 contact that organization first.
7188 You can find contact information for many support companies and
7189 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7193 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7197 The fundamental principle of reporting bugs usefully is this:
7198 @strong{report all the facts}. If you are not sure whether to state a
7199 fact or leave it out, state it!
7201 Often people omit facts because they think they know what causes the problem
7202 and assume that some details do not matter. Thus, you might assume that the
7203 name of a symbol you use in an example does not matter. Well, probably it does
7204 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7205 happens to fetch from the location where that name is stored in memory;
7206 perhaps, if the name were different, the contents of that location would fool
7207 the assembler into doing the right thing despite the bug. Play it safe and
7208 give a specific, complete example. That is the easiest thing for you to do,
7209 and the most helpful.
7211 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7212 it is new to us. Therefore, always write your bug reports on the assumption
7213 that the bug has not been reported previously.
7215 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7216 bell?'' This cannot help us fix a bug, so it is basically useless. We
7217 respond by asking for enough details to enable us to investigate.
7218 You might as well expedite matters by sending them to begin with.
7220 To enable us to fix the bug, you should include all these things:
7224 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
7225 it with the @samp{--version} argument.
7227 Without this, we will not know whether there is any point in looking for
7228 the bug in the current version of @command{@value{AS}}.
7231 Any patches you may have applied to the @command{@value{AS}} source.
7234 The type of machine you are using, and the operating system name and
7238 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
7242 The command arguments you gave the assembler to assemble your example and
7243 observe the bug. To guarantee you will not omit something important, list them
7244 all. A copy of the Makefile (or the output from make) is sufficient.
7246 If we were to try to guess the arguments, we would probably guess wrong
7247 and then we might not encounter the bug.
7250 A complete input file that will reproduce the bug. If the bug is observed when
7251 the assembler is invoked via a compiler, send the assembler source, not the
7252 high level language source. Most compilers will produce the assembler source
7253 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
7254 the options @samp{-v --save-temps}; this will save the assembler source in a
7255 file with an extension of @file{.s}, and also show you exactly how
7256 @command{@value{AS}} is being run.
7259 A description of what behavior you observe that you believe is
7260 incorrect. For example, ``It gets a fatal signal.''
7262 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
7263 will certainly notice it. But if the bug is incorrect output, we might not
7264 notice unless it is glaringly wrong. You might as well not give us a chance to
7267 Even if the problem you experience is a fatal signal, you should still say so
7268 explicitly. Suppose something strange is going on, such as, your copy of
7269 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
7270 library on your system. (This has happened!) Your copy might crash and ours
7271 would not. If you told us to expect a crash, then when ours fails to crash, we
7272 would know that the bug was not happening for us. If you had not told us to
7273 expect a crash, then we would not be able to draw any conclusion from our
7277 If you wish to suggest changes to the @command{@value{AS}} source, send us context
7278 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
7279 option. Always send diffs from the old file to the new file. If you even
7280 discuss something in the @command{@value{AS}} source, refer to it by context, not
7283 The line numbers in our development sources will not match those in your
7284 sources. Your line numbers would convey no useful information to us.
7287 Here are some things that are not necessary:
7291 A description of the envelope of the bug.
7293 Often people who encounter a bug spend a lot of time investigating
7294 which changes to the input file will make the bug go away and which
7295 changes will not affect it.
7297 This is often time consuming and not very useful, because the way we
7298 will find the bug is by running a single example under the debugger
7299 with breakpoints, not by pure deduction from a series of examples.
7300 We recommend that you save your time for something else.
7302 Of course, if you can find a simpler example to report @emph{instead}
7303 of the original one, that is a convenience for us. Errors in the
7304 output will be easier to spot, running under the debugger will take
7305 less time, and so on.
7307 However, simplification is not vital; if you do not want to do this,
7308 report the bug anyway and send us the entire test case you used.
7311 A patch for the bug.
7313 A patch for the bug does help us if it is a good one. But do not omit
7314 the necessary information, such as the test case, on the assumption that
7315 a patch is all we need. We might see problems with your patch and decide
7316 to fix the problem another way, or we might not understand it at all.
7318 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
7319 construct an example that will make the program follow a certain path through
7320 the code. If you do not send us the example, we will not be able to construct
7321 one, so we will not be able to verify that the bug is fixed.
7323 And if we cannot understand what bug you are trying to fix, or why your
7324 patch should be an improvement, we will not install it. A test case will
7325 help us to understand.
7328 A guess about what the bug is or what it depends on.
7330 Such guesses are usually wrong. Even we cannot guess right about such
7331 things without first using the debugger to find the facts.
7334 @node Acknowledgements
7335 @chapter Acknowledgements
7337 If you have contributed to GAS and your name isn't listed here,
7338 it is not meant as a slight. We just don't know about it. Send mail to the
7339 maintainer, and we'll correct the situation. Currently
7341 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
7343 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
7346 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
7347 information and the 68k series machines, most of the preprocessing pass, and
7348 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
7350 K. Richard Pixley maintained GAS for a while, adding various enhancements and
7351 many bug fixes, including merging support for several processors, breaking GAS
7352 up to handle multiple object file format back ends (including heavy rewrite,
7353 testing, an integration of the coff and b.out back ends), adding configuration
7354 including heavy testing and verification of cross assemblers and file splits
7355 and renaming, converted GAS to strictly ANSI C including full prototypes, added
7356 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
7357 port (including considerable amounts of reverse engineering), a SPARC opcode
7358 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
7359 assertions and made them work, much other reorganization, cleanup, and lint.
7361 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
7362 in format-specific I/O modules.
7364 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
7365 has done much work with it since.
7367 The Intel 80386 machine description was written by Eliot Dresselhaus.
7369 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
7371 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
7372 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
7374 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
7375 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
7376 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
7377 support a.out format.
7379 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
7380 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
7381 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
7382 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
7385 John Gilmore built the AMD 29000 support, added @code{.include} support, and
7386 simplified the configuration of which versions accept which directives. He
7387 updated the 68k machine description so that Motorola's opcodes always produced
7388 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
7389 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
7390 cross-compilation support, and one bug in relaxation that took a week and
7391 required the proverbial one-bit fix.
7393 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
7394 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
7395 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
7396 PowerPC assembler, and made a few other minor patches.
7398 Steve Chamberlain made GAS able to generate listings.
7400 Hewlett-Packard contributed support for the HP9000/300.
7402 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
7403 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
7404 formats). This work was supported by both the Center for Software Science at
7405 the University of Utah and Cygnus Support.
7407 Support for ELF format files has been worked on by Mark Eichin of Cygnus
7408 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
7409 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
7410 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
7411 and some initial 64-bit support).
7413 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
7415 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
7416 support for openVMS/Alpha.
7418 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
7421 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
7422 Inc.@: added support for Xtensa processors.
7424 Several engineers at Cygnus Support have also provided many small bug fixes and
7425 configuration enhancements.
7427 Jon Beniston added support for the Lattice Mico32 architecture.
7429 Many others have contributed large or small bugfixes and enhancements. If
7430 you have contributed significant work and are not mentioned on this list, and
7431 want to be, let us know. Some of the history has been lost; we are not
7432 intentionally leaving anyone out.
7434 @node GNU Free Documentation License
7435 @appendix GNU Free Documentation License
7439 @unnumbered AS Index
projects / external / binutils.git / blob