1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000,
3 @c 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009
4 @c Free Software Foundation, Inc.
5 @c UPDATE!! On future updates--
6 @c (1) check for new machine-dep cmdline options in
7 @c md_parse_option definitions in config/tc-*.c
8 @c (2) for platform-specific directives, examine md_pseudo_op
10 @c (3) for object-format specific directives, examine obj_pseudo_op
12 @c (4) portable directives in potable[] in read.c
16 @macro gcctabopt{body}
19 @c defaults, config file may override:
24 @include asconfig.texi
29 @c common OR combinations of conditions
52 @set abnormal-separator
56 @settitle Using @value{AS}
59 @settitle Using @value{AS} (@value{TARGET})
61 @setchapternewpage odd
66 @c WARE! Some of the machine-dependent sections contain tables of machine
67 @c instructions. Except in multi-column format, these tables look silly.
68 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
69 @c the multi-col format is faked within @example sections.
71 @c Again unfortunately, the natural size that fits on a page, for these tables,
72 @c is different depending on whether or not smallbook is turned on.
73 @c This matters, because of order: text flow switches columns at each page
76 @c The format faked in this source works reasonably well for smallbook,
77 @c not well for the default large-page format. This manual expects that if you
78 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
79 @c tables in question. You can turn on one without the other at your
80 @c discretion, of course.
83 @c the insn tables look just as silly in info files regardless of smallbook,
84 @c might as well show 'em anyways.
90 * As: (as). The GNU assembler.
91 * Gas: (as). The GNU assembler.
100 This file documents the GNU Assembler "@value{AS}".
102 @c man begin COPYRIGHT
103 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002,
104 2006, 2007, 2008, 2009 Free Software Foundation, Inc.
106 Permission is granted to copy, distribute and/or modify this document
107 under the terms of the GNU Free Documentation License, Version 1.3
108 or any later version published by the Free Software Foundation;
109 with no Invariant Sections, with no Front-Cover Texts, and with no
110 Back-Cover Texts. A copy of the license is included in the
111 section entitled ``GNU Free Documentation License''.
117 @title Using @value{AS}
118 @subtitle The @sc{gnu} Assembler
120 @subtitle for the @value{TARGET} family
122 @ifset VERSION_PACKAGE
124 @subtitle @value{VERSION_PACKAGE}
127 @subtitle Version @value{VERSION}
130 The Free Software Foundation Inc.@: thanks The Nice Computer
131 Company of Australia for loaning Dean Elsner to write the
132 first (Vax) version of @command{as} for Project @sc{gnu}.
133 The proprietors, management and staff of TNCCA thank FSF for
134 distracting the boss while they got some work
137 @author Dean Elsner, Jay Fenlason & friends
141 \hfill {\it Using {\tt @value{AS}}}\par
142 \hfill Edited by Cygnus Support\par
144 %"boxit" macro for figures:
145 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
146 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
147 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
148 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
149 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
152 @vskip 0pt plus 1filll
153 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002,
154 2006, 2007, 2008, 2009 Free Software Foundation, Inc.
156 Permission is granted to copy, distribute and/or modify this document
157 under the terms of the GNU Free Documentation License, Version 1.3
158 or any later version published by the Free Software Foundation;
159 with no Invariant Sections, with no Front-Cover Texts, and with no
160 Back-Cover Texts. A copy of the license is included in the
161 section entitled ``GNU Free Documentation License''.
168 @top Using @value{AS}
170 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}}
171 @ifset VERSION_PACKAGE
172 @value{VERSION_PACKAGE}
174 version @value{VERSION}.
176 This version of the file describes @command{@value{AS}} configured to generate
177 code for @value{TARGET} architectures.
180 This document is distributed under the terms of the GNU Free
181 Documentation License. A copy of the license is included in the
182 section entitled ``GNU Free Documentation License''.
185 * Overview:: Overview
186 * Invoking:: Command-Line Options
188 * Sections:: Sections and Relocation
190 * Expressions:: Expressions
191 * Pseudo Ops:: Assembler Directives
193 * Object Attributes:: Object Attributes
195 * Machine Dependencies:: Machine Dependent Features
196 * Reporting Bugs:: Reporting Bugs
197 * Acknowledgements:: Who Did What
198 * GNU Free Documentation License:: GNU Free Documentation License
199 * AS Index:: AS Index
206 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
208 This version of the manual describes @command{@value{AS}} configured to generate
209 code for @value{TARGET} architectures.
213 @cindex invocation summary
214 @cindex option summary
215 @cindex summary of options
216 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
217 see @ref{Invoking,,Command-Line Options}.
219 @c man title AS the portable GNU assembler.
223 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
227 @c We don't use deffn and friends for the following because they seem
228 @c to be limited to one line for the header.
230 @c man begin SYNOPSIS
231 @value{AS} [@b{-a}[@b{cdghlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
232 [@b{--debug-prefix-map} @var{old}=@var{new}]
233 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
234 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
235 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
236 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
237 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}] [@b{-o}
238 @var{objfile}] [@b{-R}] [@b{--reduce-memory-overheads}] [@b{--statistics}]
239 [@b{-v}] [@b{-version}] [@b{--version}] [@b{-W}] [@b{--warn}]
240 [@b{--fatal-warnings}] [@b{-w}] [@b{-x}] [@b{-Z}] [@b{@@@var{FILE}}]
241 [@b{--target-help}] [@var{target-options}]
242 [@b{--}|@var{files} @dots{}]
244 @c Target dependent options are listed below. Keep the list sorted.
245 @c Add an empty line for separation.
248 @emph{Target Alpha options:}
250 [@b{-mdebug} | @b{-no-mdebug}]
251 [@b{-replace} | @b{-noreplace}]
252 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
253 [@b{-F}] [@b{-32addr}]
257 @emph{Target ARC options:}
263 @emph{Target ARM options:}
264 @c Don't document the deprecated options
265 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
266 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
267 [@b{-mfpu}=@var{floating-point-format}]
268 [@b{-mfloat-abi}=@var{abi}]
269 [@b{-meabi}=@var{ver}]
272 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
273 @b{-mapcs-reentrant}]
274 [@b{-mthumb-interwork}] [@b{-k}]
278 @emph{Target CRIS options:}
279 [@b{--underscore} | @b{--no-underscore}]
281 [@b{--emulation=criself} | @b{--emulation=crisaout}]
282 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
283 @c Deprecated -- deliberately not documented.
288 @emph{Target D10V options:}
293 @emph{Target D30V options:}
294 [@b{-O}|@b{-n}|@b{-N}]
298 @emph{Target H8/300 options:}
302 @c HPPA has no machine-dependent assembler options (yet).
306 @emph{Target i386 options:}
307 [@b{--32}|@b{--64}] [@b{-n}]
308 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
312 @emph{Target i960 options:}
313 @c see md_parse_option in tc-i960.c
314 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
316 [@b{-b}] [@b{-no-relax}]
320 @emph{Target IA-64 options:}
321 [@b{-mconstant-gp}|@b{-mauto-pic}]
322 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
324 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
325 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
326 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
327 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
331 @emph{Target IP2K options:}
332 [@b{-mip2022}|@b{-mip2022ext}]
336 @emph{Target M32C options:}
337 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
341 @emph{Target M32R options:}
342 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
347 @emph{Target M680X0 options:}
348 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
352 @emph{Target M68HC11 options:}
353 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}]
354 [@b{-mshort}|@b{-mlong}]
355 [@b{-mshort-double}|@b{-mlong-double}]
356 [@b{--force-long-branches}] [@b{--short-branches}]
357 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
358 [@b{--print-opcodes}] [@b{--generate-example}]
362 @emph{Target MCORE options:}
363 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
364 [@b{-mcpu=[210|340]}]
367 @emph{Target MICROBLAZE options:}
368 @c MicroBlaze has no machine-dependent assembler options.
372 @emph{Target MIPS options:}
373 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
374 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
375 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
376 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
377 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
378 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
379 [@b{-mips64}] [@b{-mips64r2}]
380 [@b{-construct-floats}] [@b{-no-construct-floats}]
381 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
382 [@b{-mfix7000}] [@b{-mno-fix7000}]
383 [@b{-mips16}] [@b{-no-mips16}]
384 [@b{-msmartmips}] [@b{-mno-smartmips}]
385 [@b{-mips3d}] [@b{-no-mips3d}]
386 [@b{-mdmx}] [@b{-no-mdmx}]
387 [@b{-mdsp}] [@b{-mno-dsp}]
388 [@b{-mdspr2}] [@b{-mno-dspr2}]
389 [@b{-mmt}] [@b{-mno-mt}]
390 [@b{-mdebug}] [@b{-no-mdebug}]
391 [@b{-mpdr}] [@b{-mno-pdr}]
395 @emph{Target MMIX options:}
396 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
397 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
398 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
399 [@b{--linker-allocated-gregs}]
403 @emph{Target PDP11 options:}
404 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
405 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
406 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
410 @emph{Target picoJava options:}
415 @emph{Target PowerPC options:}
416 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|
417 @b{-m403}|@b{-m405}|@b{-mppc64}|@b{-m620}|@b{-mppc64bridge}|@b{-mbooke}]
418 [@b{-mcom}|@b{-many}|@b{-maltivec}|@b{-mvsx}] [@b{-memb}]
419 [@b{-mregnames}|@b{-mno-regnames}]
420 [@b{-mrelocatable}|@b{-mrelocatable-lib}]
421 [@b{-mlittle}|@b{-mlittle-endian}|@b{-mbig}|@b{-mbig-endian}]
422 [@b{-msolaris}|@b{-mno-solaris}]
426 @emph{Target s390 options:}
427 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
428 [@b{-mregnames}|@b{-mno-regnames}]
429 [@b{-mwarn-areg-zero}]
433 @emph{Target SCORE options:}
434 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
435 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
436 [@b{-march=score7}][@b{-march=score3}]
437 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
441 @emph{Target SPARC options:}
442 @c The order here is important. See c-sparc.texi.
443 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
444 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
445 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
450 @emph{Target TIC54X options:}
451 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
452 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
457 @emph{Target Z80 options:}
458 [@b{-z80}] [@b{-r800}]
459 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
460 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
461 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
462 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
463 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
464 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
468 @c Z8000 has no machine-dependent assembler options
472 @emph{Target Xtensa options:}
473 [@b{--[no-]text-section-literals}] [@b{--[no-]absolute-literals}]
474 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
475 [@b{--[no-]transform}]
476 [@b{--rename-section} @var{oldname}=@var{newname}]
484 @include at-file.texi
487 Turn on listings, in any of a variety of ways:
491 omit false conditionals
494 omit debugging directives
497 include general information, like @value{AS} version and options passed
500 include high-level source
506 include macro expansions
509 omit forms processing
515 set the name of the listing file
518 You may combine these options; for example, use @samp{-aln} for assembly
519 listing without forms processing. The @samp{=file} option, if used, must be
520 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
523 Begin in alternate macro mode.
525 @xref{Altmacro,,@code{.altmacro}}.
529 Ignored. This option is accepted for script compatibility with calls to
532 @item --debug-prefix-map @var{old}=@var{new}
533 When assembling files in directory @file{@var{old}}, record debugging
534 information describing them as in @file{@var{new}} instead.
536 @item --defsym @var{sym}=@var{value}
537 Define the symbol @var{sym} to be @var{value} before assembling the input file.
538 @var{value} must be an integer constant. As in C, a leading @samp{0x}
539 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
540 value. The value of the symbol can be overridden inside a source file via the
541 use of a @code{.set} pseudo-op.
544 ``fast''---skip whitespace and comment preprocessing (assume source is
549 Generate debugging information for each assembler source line using whichever
550 debug format is preferred by the target. This currently means either STABS,
554 Generate stabs debugging information for each assembler line. This
555 may help debugging assembler code, if the debugger can handle it.
558 Generate stabs debugging information for each assembler line, with GNU
559 extensions that probably only gdb can handle, and that could make other
560 debuggers crash or refuse to read your program. This
561 may help debugging assembler code. Currently the only GNU extension is
562 the location of the current working directory at assembling time.
565 Generate DWARF2 debugging information for each assembler line. This
566 may help debugging assembler code, if the debugger can handle it. Note---this
567 option is only supported by some targets, not all of them.
570 Print a summary of the command line options and exit.
573 Print a summary of all target specific options and exit.
576 Add directory @var{dir} to the search list for @code{.include} directives.
579 Don't warn about signed overflow.
582 @ifclear DIFF-TBL-KLUGE
583 This option is accepted but has no effect on the @value{TARGET} family.
585 @ifset DIFF-TBL-KLUGE
586 Issue warnings when difference tables altered for long displacements.
591 Keep (in the symbol table) local symbols. These symbols start with
592 system-specific local label prefixes, typically @samp{.L} for ELF systems
593 or @samp{L} for traditional a.out systems.
598 @item --listing-lhs-width=@var{number}
599 Set the maximum width, in words, of the output data column for an assembler
600 listing to @var{number}.
602 @item --listing-lhs-width2=@var{number}
603 Set the maximum width, in words, of the output data column for continuation
604 lines in an assembler listing to @var{number}.
606 @item --listing-rhs-width=@var{number}
607 Set the maximum width of an input source line, as displayed in a listing, to
610 @item --listing-cont-lines=@var{number}
611 Set the maximum number of lines printed in a listing for a single line of input
614 @item -o @var{objfile}
615 Name the object-file output from @command{@value{AS}} @var{objfile}.
618 Fold the data section into the text section.
620 @kindex --hash-size=@var{number}
621 Set the default size of GAS's hash tables to a prime number close to
622 @var{number}. Increasing this value can reduce the length of time it takes the
623 assembler to perform its tasks, at the expense of increasing the assembler's
624 memory requirements. Similarly reducing this value can reduce the memory
625 requirements at the expense of speed.
627 @item --reduce-memory-overheads
628 This option reduces GAS's memory requirements, at the expense of making the
629 assembly processes slower. Currently this switch is a synonym for
630 @samp{--hash-size=4051}, but in the future it may have other effects as well.
633 Print the maximum space (in bytes) and total time (in seconds) used by
636 @item --strip-local-absolute
637 Remove local absolute symbols from the outgoing symbol table.
641 Print the @command{as} version.
644 Print the @command{as} version and exit.
648 Suppress warning messages.
650 @item --fatal-warnings
651 Treat warnings as errors.
654 Don't suppress warning messages or treat them as errors.
663 Generate an object file even after errors.
665 @item -- | @var{files} @dots{}
666 Standard input, or source files to assemble.
671 The following options are available when @value{AS} is configured for
676 This option selects the core processor variant.
678 Select either big-endian (-EB) or little-endian (-EL) output.
683 The following options are available when @value{AS} is configured for the ARM
687 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
688 Specify which ARM processor variant is the target.
689 @item -march=@var{architecture}[+@var{extension}@dots{}]
690 Specify which ARM architecture variant is used by the target.
691 @item -mfpu=@var{floating-point-format}
692 Select which Floating Point architecture is the target.
693 @item -mfloat-abi=@var{abi}
694 Select which floating point ABI is in use.
696 Enable Thumb only instruction decoding.
697 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
698 Select which procedure calling convention is in use.
700 Select either big-endian (-EB) or little-endian (-EL) output.
701 @item -mthumb-interwork
702 Specify that the code has been generated with interworking between Thumb and
705 Specify that PIC code has been generated.
710 See the info pages for documentation of the CRIS-specific options.
714 The following options are available when @value{AS} is configured for
717 @cindex D10V optimization
718 @cindex optimization, D10V
720 Optimize output by parallelizing instructions.
725 The following options are available when @value{AS} is configured for a D30V
728 @cindex D30V optimization
729 @cindex optimization, D30V
731 Optimize output by parallelizing instructions.
735 Warn when nops are generated.
737 @cindex D30V nops after 32-bit multiply
739 Warn when a nop after a 32-bit multiply instruction is generated.
744 The following options are available when @value{AS} is configured for the
745 Intel 80960 processor.
748 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
749 Specify which variant of the 960 architecture is the target.
752 Add code to collect statistics about branches taken.
755 Do not alter compare-and-branch instructions for long displacements;
762 The following options are available when @value{AS} is configured for the
768 Specifies that the extended IP2022 instructions are allowed.
771 Restores the default behaviour, which restricts the permitted instructions to
772 just the basic IP2022 ones.
778 The following options are available when @value{AS} is configured for the
779 Renesas M32C and M16C processors.
784 Assemble M32C instructions.
787 Assemble M16C instructions (the default).
790 Enable support for link-time relaxations.
793 Support H'00 style hex constants in addition to 0x00 style.
799 The following options are available when @value{AS} is configured for the
800 Renesas M32R (formerly Mitsubishi M32R) series.
805 Specify which processor in the M32R family is the target. The default
806 is normally the M32R, but this option changes it to the M32RX.
808 @item --warn-explicit-parallel-conflicts or --Wp
809 Produce warning messages when questionable parallel constructs are
812 @item --no-warn-explicit-parallel-conflicts or --Wnp
813 Do not produce warning messages when questionable parallel constructs are
820 The following options are available when @value{AS} is configured for the
821 Motorola 68000 series.
826 Shorten references to undefined symbols, to one word instead of two.
828 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
829 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
830 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
831 Specify what processor in the 68000 family is the target. The default
832 is normally the 68020, but this can be changed at configuration time.
834 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
835 The target machine does (or does not) have a floating-point coprocessor.
836 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
837 the basic 68000 is not compatible with the 68881, a combination of the
838 two can be specified, since it's possible to do emulation of the
839 coprocessor instructions with the main processor.
841 @item -m68851 | -mno-68851
842 The target machine does (or does not) have a memory-management
843 unit coprocessor. The default is to assume an MMU for 68020 and up.
850 For details about the PDP-11 machine dependent features options,
851 see @ref{PDP-11-Options}.
854 @item -mpic | -mno-pic
855 Generate position-independent (or position-dependent) code. The
856 default is @option{-mpic}.
859 @itemx -mall-extensions
860 Enable all instruction set extensions. This is the default.
862 @item -mno-extensions
863 Disable all instruction set extensions.
865 @item -m@var{extension} | -mno-@var{extension}
866 Enable (or disable) a particular instruction set extension.
869 Enable the instruction set extensions supported by a particular CPU, and
870 disable all other extensions.
872 @item -m@var{machine}
873 Enable the instruction set extensions supported by a particular machine
874 model, and disable all other extensions.
880 The following options are available when @value{AS} is configured for
881 a picoJava processor.
885 @cindex PJ endianness
886 @cindex endianness, PJ
887 @cindex big endian output, PJ
889 Generate ``big endian'' format output.
891 @cindex little endian output, PJ
893 Generate ``little endian'' format output.
899 The following options are available when @value{AS} is configured for the
900 Motorola 68HC11 or 68HC12 series.
904 @item -m68hc11 | -m68hc12 | -m68hcs12
905 Specify what processor is the target. The default is
906 defined by the configuration option when building the assembler.
909 Specify to use the 16-bit integer ABI.
912 Specify to use the 32-bit integer ABI.
915 Specify to use the 32-bit double ABI.
918 Specify to use the 64-bit double ABI.
920 @item --force-long-branches
921 Relative branches are turned into absolute ones. This concerns
922 conditional branches, unconditional branches and branches to a
925 @item -S | --short-branches
926 Do not turn relative branches into absolute ones
927 when the offset is out of range.
929 @item --strict-direct-mode
930 Do not turn the direct addressing mode into extended addressing mode
931 when the instruction does not support direct addressing mode.
933 @item --print-insn-syntax
934 Print the syntax of instruction in case of error.
936 @item --print-opcodes
937 print the list of instructions with syntax and then exit.
939 @item --generate-example
940 print an example of instruction for each possible instruction and then exit.
941 This option is only useful for testing @command{@value{AS}}.
947 The following options are available when @command{@value{AS}} is configured
948 for the SPARC architecture:
951 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
952 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
953 Explicitly select a variant of the SPARC architecture.
955 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
956 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
958 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
959 UltraSPARC extensions.
961 @item -xarch=v8plus | -xarch=v8plusa
962 For compatibility with the Solaris v9 assembler. These options are
963 equivalent to -Av8plus and -Av8plusa, respectively.
966 Warn when the assembler switches to another architecture.
971 The following options are available when @value{AS} is configured for the 'c54x
976 Enable extended addressing mode. All addresses and relocations will assume
977 extended addressing (usually 23 bits).
978 @item -mcpu=@var{CPU_VERSION}
979 Sets the CPU version being compiled for.
980 @item -merrors-to-file @var{FILENAME}
981 Redirect error output to a file, for broken systems which don't support such
982 behaviour in the shell.
987 The following options are available when @value{AS} is configured for
988 a @sc{mips} processor.
992 This option sets the largest size of an object that can be referenced
993 implicitly with the @code{gp} register. It is only accepted for targets that
994 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
996 @cindex MIPS endianness
997 @cindex endianness, MIPS
998 @cindex big endian output, MIPS
1000 Generate ``big endian'' format output.
1002 @cindex little endian output, MIPS
1004 Generate ``little endian'' format output.
1016 Generate code for a particular @sc{mips} Instruction Set Architecture level.
1017 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1018 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1019 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1020 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
1022 correspond to generic
1023 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
1024 and @samp{MIPS64 Release 2}
1025 ISA processors, respectively.
1027 @item -march=@var{CPU}
1028 Generate code for a particular @sc{mips} cpu.
1030 @item -mtune=@var{cpu}
1031 Schedule and tune for a particular @sc{mips} cpu.
1035 Cause nops to be inserted if the read of the destination register
1036 of an mfhi or mflo instruction occurs in the following two instructions.
1040 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1041 section instead of the standard ELF .stabs sections.
1045 Control generation of @code{.pdr} sections.
1049 The register sizes are normally inferred from the ISA and ABI, but these
1050 flags force a certain group of registers to be treated as 32 bits wide at
1051 all times. @samp{-mgp32} controls the size of general-purpose registers
1052 and @samp{-mfp32} controls the size of floating-point registers.
1056 Generate code for the MIPS 16 processor. This is equivalent to putting
1057 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1058 turns off this option.
1061 @itemx -mno-smartmips
1062 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1063 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1064 @samp{-mno-smartmips} turns off this option.
1068 Generate code for the MIPS-3D Application Specific Extension.
1069 This tells the assembler to accept MIPS-3D instructions.
1070 @samp{-no-mips3d} turns off this option.
1074 Generate code for the MDMX Application Specific Extension.
1075 This tells the assembler to accept MDMX instructions.
1076 @samp{-no-mdmx} turns off this option.
1080 Generate code for the DSP Release 1 Application Specific Extension.
1081 This tells the assembler to accept DSP Release 1 instructions.
1082 @samp{-mno-dsp} turns off this option.
1086 Generate code for the DSP Release 2 Application Specific Extension.
1087 This option implies -mdsp.
1088 This tells the assembler to accept DSP Release 2 instructions.
1089 @samp{-mno-dspr2} turns off this option.
1093 Generate code for the MT Application Specific Extension.
1094 This tells the assembler to accept MT instructions.
1095 @samp{-mno-mt} turns off this option.
1097 @item --construct-floats
1098 @itemx --no-construct-floats
1099 The @samp{--no-construct-floats} option disables the construction of
1100 double width floating point constants by loading the two halves of the
1101 value into the two single width floating point registers that make up
1102 the double width register. By default @samp{--construct-floats} is
1103 selected, allowing construction of these floating point constants.
1106 @item --emulation=@var{name}
1107 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
1108 for some other target, in all respects, including output format (choosing
1109 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
1110 debugging information or store symbol table information, and default
1111 endianness. The available configuration names are: @samp{mipsecoff},
1112 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
1113 @samp{mipsbelf}. The first two do not alter the default endianness from that
1114 of the primary target for which the assembler was configured; the others change
1115 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
1116 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
1117 selection in any case.
1119 This option is currently supported only when the primary target
1120 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
1121 Furthermore, the primary target or others specified with
1122 @samp{--enable-targets=@dots{}} at configuration time must include support for
1123 the other format, if both are to be available. For example, the Irix 5
1124 configuration includes support for both.
1126 Eventually, this option will support more configurations, with more
1127 fine-grained control over the assembler's behavior, and will be supported for
1131 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1138 Control how to deal with multiplication overflow and division by zero.
1139 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1140 (and only work for Instruction Set Architecture level 2 and higher);
1141 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1145 When this option is used, @command{@value{AS}} will issue a warning every
1146 time it generates a nop instruction from a macro.
1151 The following options are available when @value{AS} is configured for
1157 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1158 The command line option @samp{-nojsri2bsr} can be used to disable it.
1162 Enable or disable the silicon filter behaviour. By default this is disabled.
1163 The default can be overridden by the @samp{-sifilter} command line option.
1166 Alter jump instructions for long displacements.
1168 @item -mcpu=[210|340]
1169 Select the cpu type on the target hardware. This controls which instructions
1173 Assemble for a big endian target.
1176 Assemble for a little endian target.
1182 See the info pages for documentation of the MMIX-specific options.
1186 The following options are available when @value{AS} is configured for the s390
1192 Select the word size, either 31/32 bits or 64 bits.
1195 Select the architecture mode, either the Enterprise System
1196 Architecture (esa) or the z/Architecture mode (zarch).
1197 @item -march=@var{processor}
1198 Specify which s390 processor variant is the target, @samp{g6}, @samp{g6},
1199 @samp{z900}, @samp{z990}, @samp{z9-109}, @samp{z9-ec}, or @samp{z10}.
1201 @itemx -mno-regnames
1202 Allow or disallow symbolic names for registers.
1203 @item -mwarn-areg-zero
1204 Warn whenever the operand for a base or index register has been specified
1205 but evaluates to zero.
1210 The following options are available when @value{AS} is configured for
1211 an Xtensa processor.
1214 @item --text-section-literals | --no-text-section-literals
1215 With @option{--text-@-section-@-literals}, literal pools are interspersed
1216 in the text section. The default is
1217 @option{--no-@-text-@-section-@-literals}, which places literals in a
1218 separate section in the output file. These options only affect literals
1219 referenced via PC-relative @code{L32R} instructions; literals for
1220 absolute mode @code{L32R} instructions are handled separately.
1222 @item --absolute-literals | --no-absolute-literals
1223 Indicate to the assembler whether @code{L32R} instructions use absolute
1224 or PC-relative addressing. The default is to assume absolute addressing
1225 if the Xtensa processor includes the absolute @code{L32R} addressing
1226 option. Otherwise, only the PC-relative @code{L32R} mode can be used.
1228 @item --target-align | --no-target-align
1229 Enable or disable automatic alignment to reduce branch penalties at the
1230 expense of some code density. The default is @option{--target-@-align}.
1232 @item --longcalls | --no-longcalls
1233 Enable or disable transformation of call instructions to allow calls
1234 across a greater range of addresses. The default is
1235 @option{--no-@-longcalls}.
1237 @item --transform | --no-transform
1238 Enable or disable all assembler transformations of Xtensa instructions.
1239 The default is @option{--transform};
1240 @option{--no-transform} should be used only in the rare cases when the
1241 instructions must be exactly as specified in the assembly source.
1243 @item --rename-section @var{oldname}=@var{newname}
1244 When generating output sections, rename the @var{oldname} section to
1250 The following options are available when @value{AS} is configured for
1251 a Z80 family processor.
1254 Assemble for Z80 processor.
1256 Assemble for R800 processor.
1257 @item -ignore-undocumented-instructions
1259 Assemble undocumented Z80 instructions that also work on R800 without warning.
1260 @item -ignore-unportable-instructions
1262 Assemble all undocumented Z80 instructions without warning.
1263 @item -warn-undocumented-instructions
1265 Issue a warning for undocumented Z80 instructions that also work on R800.
1266 @item -warn-unportable-instructions
1268 Issue a warning for undocumented Z80 instructions that do not work on R800.
1269 @item -forbid-undocumented-instructions
1271 Treat all undocumented instructions as errors.
1272 @item -forbid-unportable-instructions
1274 Treat undocumented Z80 instructions that do not work on R800 as errors.
1281 * Manual:: Structure of this Manual
1282 * GNU Assembler:: The GNU Assembler
1283 * Object Formats:: Object File Formats
1284 * Command Line:: Command Line
1285 * Input Files:: Input Files
1286 * Object:: Output (Object) File
1287 * Errors:: Error and Warning Messages
1291 @section Structure of this Manual
1293 @cindex manual, structure and purpose
1294 This manual is intended to describe what you need to know to use
1295 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1296 notation for symbols, constants, and expressions; the directives that
1297 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1300 We also cover special features in the @value{TARGET}
1301 configuration of @command{@value{AS}}, including assembler directives.
1304 This manual also describes some of the machine-dependent features of
1305 various flavors of the assembler.
1308 @cindex machine instructions (not covered)
1309 On the other hand, this manual is @emph{not} intended as an introduction
1310 to programming in assembly language---let alone programming in general!
1311 In a similar vein, we make no attempt to introduce the machine
1312 architecture; we do @emph{not} describe the instruction set, standard
1313 mnemonics, registers or addressing modes that are standard to a
1314 particular architecture.
1316 You may want to consult the manufacturer's
1317 machine architecture manual for this information.
1321 For information on the H8/300 machine instruction set, see @cite{H8/300
1322 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1323 Programming Manual} (Renesas).
1326 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1327 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1328 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1329 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1332 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1336 @c I think this is premature---doc@cygnus.com, 17jan1991
1338 Throughout this manual, we assume that you are running @dfn{GNU},
1339 the portable operating system from the @dfn{Free Software
1340 Foundation, Inc.}. This restricts our attention to certain kinds of
1341 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1342 once this assumption is granted examples and definitions need less
1345 @command{@value{AS}} is part of a team of programs that turn a high-level
1346 human-readable series of instructions into a low-level
1347 computer-readable series of instructions. Different versions of
1348 @command{@value{AS}} are used for different kinds of computer.
1351 @c There used to be a section "Terminology" here, which defined
1352 @c "contents", "byte", "word", and "long". Defining "word" to any
1353 @c particular size is confusing when the .word directive may generate 16
1354 @c bits on one machine and 32 bits on another; in general, for the user
1355 @c version of this manual, none of these terms seem essential to define.
1356 @c They were used very little even in the former draft of the manual;
1357 @c this draft makes an effort to avoid them (except in names of
1361 @section The GNU Assembler
1363 @c man begin DESCRIPTION
1365 @sc{gnu} @command{as} is really a family of assemblers.
1367 This manual describes @command{@value{AS}}, a member of that family which is
1368 configured for the @value{TARGET} architectures.
1370 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1371 should find a fairly similar environment when you use it on another
1372 architecture. Each version has much in common with the others,
1373 including object file formats, most assembler directives (often called
1374 @dfn{pseudo-ops}) and assembler syntax.@refill
1376 @cindex purpose of @sc{gnu} assembler
1377 @command{@value{AS}} is primarily intended to assemble the output of the
1378 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1379 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1380 assemble correctly everything that other assemblers for the same
1381 machine would assemble.
1383 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1386 @c This remark should appear in generic version of manual; assumption
1387 @c here is that generic version sets M680x0.
1388 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1389 assembler for the same architecture; for example, we know of several
1390 incompatible versions of 680x0 assembly language syntax.
1395 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1396 program in one pass of the source file. This has a subtle impact on the
1397 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1399 @node Object Formats
1400 @section Object File Formats
1402 @cindex object file format
1403 The @sc{gnu} assembler can be configured to produce several alternative
1404 object file formats. For the most part, this does not affect how you
1405 write assembly language programs; but directives for debugging symbols
1406 are typically different in different file formats. @xref{Symbol
1407 Attributes,,Symbol Attributes}.
1410 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1411 @value{OBJ-NAME} format object files.
1413 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1415 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1416 @code{b.out} or COFF format object files.
1419 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1420 SOM or ELF format object files.
1425 @section Command Line
1427 @cindex command line conventions
1429 After the program name @command{@value{AS}}, the command line may contain
1430 options and file names. Options may appear in any order, and may be
1431 before, after, or between file names. The order of file names is
1434 @cindex standard input, as input file
1436 @file{--} (two hyphens) by itself names the standard input file
1437 explicitly, as one of the files for @command{@value{AS}} to assemble.
1439 @cindex options, command line
1440 Except for @samp{--} any command line argument that begins with a
1441 hyphen (@samp{-}) is an option. Each option changes the behavior of
1442 @command{@value{AS}}. No option changes the way another option works. An
1443 option is a @samp{-} followed by one or more letters; the case of
1444 the letter is important. All options are optional.
1446 Some options expect exactly one file name to follow them. The file
1447 name may either immediately follow the option's letter (compatible
1448 with older assemblers) or it may be the next command argument (@sc{gnu}
1449 standard). These two command lines are equivalent:
1452 @value{AS} -o my-object-file.o mumble.s
1453 @value{AS} -omy-object-file.o mumble.s
1457 @section Input Files
1460 @cindex source program
1461 @cindex files, input
1462 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1463 describe the program input to one run of @command{@value{AS}}. The program may
1464 be in one or more files; how the source is partitioned into files
1465 doesn't change the meaning of the source.
1467 @c I added "con" prefix to "catenation" just to prove I can overcome my
1468 @c APL training... doc@cygnus.com
1469 The source program is a concatenation of the text in all the files, in the
1472 @c man begin DESCRIPTION
1473 Each time you run @command{@value{AS}} it assembles exactly one source
1474 program. The source program is made up of one or more files.
1475 (The standard input is also a file.)
1477 You give @command{@value{AS}} a command line that has zero or more input file
1478 names. The input files are read (from left file name to right). A
1479 command line argument (in any position) that has no special meaning
1480 is taken to be an input file name.
1482 If you give @command{@value{AS}} no file names it attempts to read one input file
1483 from the @command{@value{AS}} standard input, which is normally your terminal. You
1484 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1487 Use @samp{--} if you need to explicitly name the standard input file
1488 in your command line.
1490 If the source is empty, @command{@value{AS}} produces a small, empty object
1495 @subheading Filenames and Line-numbers
1497 @cindex input file linenumbers
1498 @cindex line numbers, in input files
1499 There are two ways of locating a line in the input file (or files) and
1500 either may be used in reporting error messages. One way refers to a line
1501 number in a physical file; the other refers to a line number in a
1502 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1504 @dfn{Physical files} are those files named in the command line given
1505 to @command{@value{AS}}.
1507 @dfn{Logical files} are simply names declared explicitly by assembler
1508 directives; they bear no relation to physical files. Logical file names help
1509 error messages reflect the original source file, when @command{@value{AS}} source
1510 is itself synthesized from other files. @command{@value{AS}} understands the
1511 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1512 @ref{File,,@code{.file}}.
1515 @section Output (Object) File
1521 Every time you run @command{@value{AS}} it produces an output file, which is
1522 your assembly language program translated into numbers. This file
1523 is the object file. Its default name is
1531 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1533 You can give it another name by using the @option{-o} option. Conventionally,
1534 object file names end with @file{.o}. The default name is used for historical
1535 reasons: older assemblers were capable of assembling self-contained programs
1536 directly into a runnable program. (For some formats, this isn't currently
1537 possible, but it can be done for the @code{a.out} format.)
1541 The object file is meant for input to the linker @code{@value{LD}}. It contains
1542 assembled program code, information to help @code{@value{LD}} integrate
1543 the assembled program into a runnable file, and (optionally) symbolic
1544 information for the debugger.
1546 @c link above to some info file(s) like the description of a.out.
1547 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1550 @section Error and Warning Messages
1552 @c man begin DESCRIPTION
1554 @cindex error messages
1555 @cindex warning messages
1556 @cindex messages from assembler
1557 @command{@value{AS}} may write warnings and error messages to the standard error
1558 file (usually your terminal). This should not happen when a compiler
1559 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1560 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1561 grave problem that stops the assembly.
1565 @cindex format of warning messages
1566 Warning messages have the format
1569 file_name:@b{NNN}:Warning Message Text
1573 @cindex line numbers, in warnings/errors
1574 (where @b{NNN} is a line number). If a logical file name has been given
1575 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1576 the current input file is used. If a logical line number was given
1578 (@pxref{Line,,@code{.line}})
1580 then it is used to calculate the number printed,
1581 otherwise the actual line in the current source file is printed. The
1582 message text is intended to be self explanatory (in the grand Unix
1585 @cindex format of error messages
1586 Error messages have the format
1588 file_name:@b{NNN}:FATAL:Error Message Text
1590 The file name and line number are derived as for warning
1591 messages. The actual message text may be rather less explanatory
1592 because many of them aren't supposed to happen.
1595 @chapter Command-Line Options
1597 @cindex options, all versions of assembler
1598 This chapter describes command-line options available in @emph{all}
1599 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
1600 for options specific
1602 to the @value{TARGET} target.
1605 to particular machine architectures.
1608 @c man begin DESCRIPTION
1610 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1611 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1612 The assembler arguments must be separated from each other (and the @samp{-Wa})
1613 by commas. For example:
1616 gcc -c -g -O -Wa,-alh,-L file.c
1620 This passes two options to the assembler: @samp{-alh} (emit a listing to
1621 standard output with high-level and assembly source) and @samp{-L} (retain
1622 local symbols in the symbol table).
1624 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1625 command-line options are automatically passed to the assembler by the compiler.
1626 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1627 precisely what options it passes to each compilation pass, including the
1633 * a:: -a[cdghlns] enable listings
1634 * alternate:: --alternate enable alternate macro syntax
1635 * D:: -D for compatibility
1636 * f:: -f to work faster
1637 * I:: -I for .include search path
1638 @ifclear DIFF-TBL-KLUGE
1639 * K:: -K for compatibility
1641 @ifset DIFF-TBL-KLUGE
1642 * K:: -K for difference tables
1645 * L:: -L to retain local symbols
1646 * listing:: --listing-XXX to configure listing output
1647 * M:: -M or --mri to assemble in MRI compatibility mode
1648 * MD:: --MD for dependency tracking
1649 * o:: -o to name the object file
1650 * R:: -R to join data and text sections
1651 * statistics:: --statistics to see statistics about assembly
1652 * traditional-format:: --traditional-format for compatible output
1653 * v:: -v to announce version
1654 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1655 * Z:: -Z to make object file even after errors
1659 @section Enable Listings: @option{-a[cdghlns]}
1669 @cindex listings, enabling
1670 @cindex assembly listings, enabling
1672 These options enable listing output from the assembler. By itself,
1673 @samp{-a} requests high-level, assembly, and symbols listing.
1674 You can use other letters to select specific options for the list:
1675 @samp{-ah} requests a high-level language listing,
1676 @samp{-al} requests an output-program assembly listing, and
1677 @samp{-as} requests a symbol table listing.
1678 High-level listings require that a compiler debugging option like
1679 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1682 Use the @samp{-ag} option to print a first section with general assembly
1683 information, like @value{AS} version, switches passed, or time stamp.
1685 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1686 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1687 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1688 omitted from the listing.
1690 Use the @samp{-ad} option to omit debugging directives from the
1693 Once you have specified one of these options, you can further control
1694 listing output and its appearance using the directives @code{.list},
1695 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1697 The @samp{-an} option turns off all forms processing.
1698 If you do not request listing output with one of the @samp{-a} options, the
1699 listing-control directives have no effect.
1701 The letters after @samp{-a} may be combined into one option,
1702 @emph{e.g.}, @samp{-aln}.
1704 Note if the assembler source is coming from the standard input (e.g.,
1706 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1707 is being used) then the listing will not contain any comments or preprocessor
1708 directives. This is because the listing code buffers input source lines from
1709 stdin only after they have been preprocessed by the assembler. This reduces
1710 memory usage and makes the code more efficient.
1713 @section @option{--alternate}
1716 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
1719 @section @option{-D}
1722 This option has no effect whatsoever, but it is accepted to make it more
1723 likely that scripts written for other assemblers also work with
1724 @command{@value{AS}}.
1727 @section Work Faster: @option{-f}
1730 @cindex trusted compiler
1731 @cindex faster processing (@option{-f})
1732 @samp{-f} should only be used when assembling programs written by a
1733 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1734 and comment preprocessing on
1735 the input file(s) before assembling them. @xref{Preprocessing,
1739 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1740 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1745 @section @code{.include} Search Path: @option{-I} @var{path}
1747 @kindex -I @var{path}
1748 @cindex paths for @code{.include}
1749 @cindex search path for @code{.include}
1750 @cindex @code{include} directive search path
1751 Use this option to add a @var{path} to the list of directories
1752 @command{@value{AS}} searches for files specified in @code{.include}
1753 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1754 many times as necessary to include a variety of paths. The current
1755 working directory is always searched first; after that, @command{@value{AS}}
1756 searches any @samp{-I} directories in the same order as they were
1757 specified (left to right) on the command line.
1760 @section Difference Tables: @option{-K}
1763 @ifclear DIFF-TBL-KLUGE
1764 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1765 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1766 where it can be used to warn when the assembler alters the machine code
1767 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1768 family does not have the addressing limitations that sometimes lead to this
1769 alteration on other platforms.
1772 @ifset DIFF-TBL-KLUGE
1773 @cindex difference tables, warning
1774 @cindex warning for altered difference tables
1775 @command{@value{AS}} sometimes alters the code emitted for directives of the
1776 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
1777 You can use the @samp{-K} option if you want a warning issued when this
1782 @section Include Local Symbols: @option{-L}
1785 @cindex local symbols, retaining in output
1786 Symbols beginning with system-specific local label prefixes, typically
1787 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
1788 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
1789 such symbols when debugging, because they are intended for the use of
1790 programs (like compilers) that compose assembler programs, not for your
1791 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
1792 such symbols, so you do not normally debug with them.
1794 This option tells @command{@value{AS}} to retain those local symbols
1795 in the object file. Usually if you do this you also tell the linker
1796 @code{@value{LD}} to preserve those symbols.
1799 @section Configuring listing output: @option{--listing}
1801 The listing feature of the assembler can be enabled via the command line switch
1802 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1803 hex dump of the corresponding locations in the output object file, and displays
1804 them as a listing file. The format of this listing can be controlled by
1805 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
1806 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
1807 @code{.psize} (@pxref{Psize}), and
1808 @code{.eject} (@pxref{Eject}) and also by the following switches:
1811 @item --listing-lhs-width=@samp{number}
1812 @kindex --listing-lhs-width
1813 @cindex Width of first line disassembly output
1814 Sets the maximum width, in words, of the first line of the hex byte dump. This
1815 dump appears on the left hand side of the listing output.
1817 @item --listing-lhs-width2=@samp{number}
1818 @kindex --listing-lhs-width2
1819 @cindex Width of continuation lines of disassembly output
1820 Sets the maximum width, in words, of any further lines of the hex byte dump for
1821 a given input source line. If this value is not specified, it defaults to being
1822 the same as the value specified for @samp{--listing-lhs-width}. If neither
1823 switch is used the default is to one.
1825 @item --listing-rhs-width=@samp{number}
1826 @kindex --listing-rhs-width
1827 @cindex Width of source line output
1828 Sets the maximum width, in characters, of the source line that is displayed
1829 alongside the hex dump. The default value for this parameter is 100. The
1830 source line is displayed on the right hand side of the listing output.
1832 @item --listing-cont-lines=@samp{number}
1833 @kindex --listing-cont-lines
1834 @cindex Maximum number of continuation lines
1835 Sets the maximum number of continuation lines of hex dump that will be
1836 displayed for a given single line of source input. The default value is 4.
1840 @section Assemble in MRI Compatibility Mode: @option{-M}
1843 @cindex MRI compatibility mode
1844 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
1845 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
1846 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1847 configured target) assembler from Microtec Research. The exact nature of the
1848 MRI syntax will not be documented here; see the MRI manuals for more
1849 information. Note in particular that the handling of macros and macro
1850 arguments is somewhat different. The purpose of this option is to permit
1851 assembling existing MRI assembler code using @command{@value{AS}}.
1853 The MRI compatibility is not complete. Certain operations of the MRI assembler
1854 depend upon its object file format, and can not be supported using other object
1855 file formats. Supporting these would require enhancing each object file format
1856 individually. These are:
1859 @item global symbols in common section
1861 The m68k MRI assembler supports common sections which are merged by the linker.
1862 Other object file formats do not support this. @command{@value{AS}} handles
1863 common sections by treating them as a single common symbol. It permits local
1864 symbols to be defined within a common section, but it can not support global
1865 symbols, since it has no way to describe them.
1867 @item complex relocations
1869 The MRI assemblers support relocations against a negated section address, and
1870 relocations which combine the start addresses of two or more sections. These
1871 are not support by other object file formats.
1873 @item @code{END} pseudo-op specifying start address
1875 The MRI @code{END} pseudo-op permits the specification of a start address.
1876 This is not supported by other object file formats. The start address may
1877 instead be specified using the @option{-e} option to the linker, or in a linker
1880 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1882 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1883 name to the output file. This is not supported by other object file formats.
1885 @item @code{ORG} pseudo-op
1887 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1888 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
1889 which changes the location within the current section. Absolute sections are
1890 not supported by other object file formats. The address of a section may be
1891 assigned within a linker script.
1894 There are some other features of the MRI assembler which are not supported by
1895 @command{@value{AS}}, typically either because they are difficult or because they
1896 seem of little consequence. Some of these may be supported in future releases.
1900 @item EBCDIC strings
1902 EBCDIC strings are not supported.
1904 @item packed binary coded decimal
1906 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1907 and @code{DCB.P} pseudo-ops are not supported.
1909 @item @code{FEQU} pseudo-op
1911 The m68k @code{FEQU} pseudo-op is not supported.
1913 @item @code{NOOBJ} pseudo-op
1915 The m68k @code{NOOBJ} pseudo-op is not supported.
1917 @item @code{OPT} branch control options
1919 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1920 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
1921 relaxes all branches, whether forward or backward, to an appropriate size, so
1922 these options serve no purpose.
1924 @item @code{OPT} list control options
1926 The following m68k @code{OPT} list control options are ignored: @code{C},
1927 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1928 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1930 @item other @code{OPT} options
1932 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1933 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1935 @item @code{OPT} @code{D} option is default
1937 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1938 @code{OPT NOD} may be used to turn it off.
1940 @item @code{XREF} pseudo-op.
1942 The m68k @code{XREF} pseudo-op is ignored.
1944 @item @code{.debug} pseudo-op
1946 The i960 @code{.debug} pseudo-op is not supported.
1948 @item @code{.extended} pseudo-op
1950 The i960 @code{.extended} pseudo-op is not supported.
1952 @item @code{.list} pseudo-op.
1954 The various options of the i960 @code{.list} pseudo-op are not supported.
1956 @item @code{.optimize} pseudo-op
1958 The i960 @code{.optimize} pseudo-op is not supported.
1960 @item @code{.output} pseudo-op
1962 The i960 @code{.output} pseudo-op is not supported.
1964 @item @code{.setreal} pseudo-op
1966 The i960 @code{.setreal} pseudo-op is not supported.
1971 @section Dependency Tracking: @option{--MD}
1974 @cindex dependency tracking
1977 @command{@value{AS}} can generate a dependency file for the file it creates. This
1978 file consists of a single rule suitable for @code{make} describing the
1979 dependencies of the main source file.
1981 The rule is written to the file named in its argument.
1983 This feature is used in the automatic updating of makefiles.
1986 @section Name the Object File: @option{-o}
1989 @cindex naming object file
1990 @cindex object file name
1991 There is always one object file output when you run @command{@value{AS}}. By
1992 default it has the name
1995 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2009 You use this option (which takes exactly one filename) to give the
2010 object file a different name.
2012 Whatever the object file is called, @command{@value{AS}} overwrites any
2013 existing file of the same name.
2016 @section Join Data and Text Sections: @option{-R}
2019 @cindex data and text sections, joining
2020 @cindex text and data sections, joining
2021 @cindex joining text and data sections
2022 @cindex merging text and data sections
2023 @option{-R} tells @command{@value{AS}} to write the object file as if all
2024 data-section data lives in the text section. This is only done at
2025 the very last moment: your binary data are the same, but data
2026 section parts are relocated differently. The data section part of
2027 your object file is zero bytes long because all its bytes are
2028 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2030 When you specify @option{-R} it would be possible to generate shorter
2031 address displacements (because we do not have to cross between text and
2032 data section). We refrain from doing this simply for compatibility with
2033 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2036 When @command{@value{AS}} is configured for COFF or ELF output,
2037 this option is only useful if you use sections named @samp{.text} and
2042 @option{-R} is not supported for any of the HPPA targets. Using
2043 @option{-R} generates a warning from @command{@value{AS}}.
2047 @section Display Assembly Statistics: @option{--statistics}
2049 @kindex --statistics
2050 @cindex statistics, about assembly
2051 @cindex time, total for assembly
2052 @cindex space used, maximum for assembly
2053 Use @samp{--statistics} to display two statistics about the resources used by
2054 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2055 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2058 @node traditional-format
2059 @section Compatible Output: @option{--traditional-format}
2061 @kindex --traditional-format
2062 For some targets, the output of @command{@value{AS}} is different in some ways
2063 from the output of some existing assembler. This switch requests
2064 @command{@value{AS}} to use the traditional format instead.
2066 For example, it disables the exception frame optimizations which
2067 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2070 @section Announce Version: @option{-v}
2074 @cindex assembler version
2075 @cindex version of assembler
2076 You can find out what version of as is running by including the
2077 option @samp{-v} (which you can also spell as @samp{-version}) on the
2081 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2083 @command{@value{AS}} should never give a warning or error message when
2084 assembling compiler output. But programs written by people often
2085 cause @command{@value{AS}} to give a warning that a particular assumption was
2086 made. All such warnings are directed to the standard error file.
2090 @cindex suppressing warnings
2091 @cindex warnings, suppressing
2092 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2093 This only affects the warning messages: it does not change any particular of
2094 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2097 @kindex --fatal-warnings
2098 @cindex errors, caused by warnings
2099 @cindex warnings, causing error
2100 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2101 files that generate warnings to be in error.
2104 @cindex warnings, switching on
2105 You can switch these options off again by specifying @option{--warn}, which
2106 causes warnings to be output as usual.
2109 @section Generate Object File in Spite of Errors: @option{-Z}
2110 @cindex object file, after errors
2111 @cindex errors, continuing after
2112 After an error message, @command{@value{AS}} normally produces no output. If for
2113 some reason you are interested in object file output even after
2114 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2115 option. If there are any errors, @command{@value{AS}} continues anyways, and
2116 writes an object file after a final warning message of the form @samp{@var{n}
2117 errors, @var{m} warnings, generating bad object file.}
2122 @cindex machine-independent syntax
2123 @cindex syntax, machine-independent
2124 This chapter describes the machine-independent syntax allowed in a
2125 source file. @command{@value{AS}} syntax is similar to what many other
2126 assemblers use; it is inspired by the BSD 4.2
2131 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2135 * Preprocessing:: Preprocessing
2136 * Whitespace:: Whitespace
2137 * Comments:: Comments
2138 * Symbol Intro:: Symbols
2139 * Statements:: Statements
2140 * Constants:: Constants
2144 @section Preprocessing
2146 @cindex preprocessing
2147 The @command{@value{AS}} internal preprocessor:
2149 @cindex whitespace, removed by preprocessor
2151 adjusts and removes extra whitespace. It leaves one space or tab before
2152 the keywords on a line, and turns any other whitespace on the line into
2155 @cindex comments, removed by preprocessor
2157 removes all comments, replacing them with a single space, or an
2158 appropriate number of newlines.
2160 @cindex constants, converted by preprocessor
2162 converts character constants into the appropriate numeric values.
2165 It does not do macro processing, include file handling, or
2166 anything else you may get from your C compiler's preprocessor. You can
2167 do include file processing with the @code{.include} directive
2168 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2169 to get other ``CPP'' style preprocessing by giving the input file a
2170 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2171 Output, gcc.info, Using GNU CC}.
2173 Excess whitespace, comments, and character constants
2174 cannot be used in the portions of the input text that are not
2177 @cindex turning preprocessing on and off
2178 @cindex preprocessing, turning on and off
2181 If the first line of an input file is @code{#NO_APP} or if you use the
2182 @samp{-f} option, whitespace and comments are not removed from the input file.
2183 Within an input file, you can ask for whitespace and comment removal in
2184 specific portions of the by putting a line that says @code{#APP} before the
2185 text that may contain whitespace or comments, and putting a line that says
2186 @code{#NO_APP} after this text. This feature is mainly intend to support
2187 @code{asm} statements in compilers whose output is otherwise free of comments
2194 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2195 Whitespace is used to separate symbols, and to make programs neater for
2196 people to read. Unless within character constants
2197 (@pxref{Characters,,Character Constants}), any whitespace means the same
2198 as exactly one space.
2204 There are two ways of rendering comments to @command{@value{AS}}. In both
2205 cases the comment is equivalent to one space.
2207 Anything from @samp{/*} through the next @samp{*/} is a comment.
2208 This means you may not nest these comments.
2212 The only way to include a newline ('\n') in a comment
2213 is to use this sort of comment.
2216 /* This sort of comment does not nest. */
2219 @cindex line comment character
2220 Anything from the @dfn{line comment} character to the next newline
2221 is considered a comment and is ignored. The line comment character is
2223 @samp{;} on the ARC;
2226 @samp{@@} on the ARM;
2229 @samp{;} for the H8/300 family;
2232 @samp{;} for the HPPA;
2235 @samp{#} on the i386 and x86-64;
2238 @samp{#} on the i960;
2241 @samp{;} for the PDP-11;
2244 @samp{;} for picoJava;
2247 @samp{#} for Motorola PowerPC;
2250 @samp{#} for IBM S/390;
2253 @samp{#} for the Sunplus SCORE;
2256 @samp{!} for the Renesas / SuperH SH;
2259 @samp{!} on the SPARC;
2262 @samp{#} on the ip2k;
2265 @samp{#} on the m32c;
2268 @samp{#} on the m32r;
2271 @samp{|} on the 680x0;
2274 @samp{#} on the 68HC11 and 68HC12;
2277 @samp{#} on the Vax;
2280 @samp{;} for the Z80;
2283 @samp{!} for the Z8000;
2286 @samp{#} on the V850;
2289 @samp{#} for Xtensa systems;
2291 see @ref{Machine Dependencies}. @refill
2292 @c FIXME What about i860?
2295 On some machines there are two different line comment characters. One
2296 character only begins a comment if it is the first non-whitespace character on
2297 a line, while the other always begins a comment.
2301 The V850 assembler also supports a double dash as starting a comment that
2302 extends to the end of the line.
2308 @cindex lines starting with @code{#}
2309 @cindex logical line numbers
2310 To be compatible with past assemblers, lines that begin with @samp{#} have a
2311 special interpretation. Following the @samp{#} should be an absolute
2312 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2313 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2314 new logical file name. The rest of the line, if any, should be whitespace.
2316 If the first non-whitespace characters on the line are not numeric,
2317 the line is ignored. (Just like a comment.)
2320 # This is an ordinary comment.
2321 # 42-6 "new_file_name" # New logical file name
2322 # This is logical line # 36.
2324 This feature is deprecated, and may disappear from future versions
2325 of @command{@value{AS}}.
2330 @cindex characters used in symbols
2331 @ifclear SPECIAL-SYMS
2332 A @dfn{symbol} is one or more characters chosen from the set of all
2333 letters (both upper and lower case), digits and the three characters
2339 A @dfn{symbol} is one or more characters chosen from the set of all
2340 letters (both upper and lower case), digits and the three characters
2341 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2347 On most machines, you can also use @code{$} in symbol names; exceptions
2348 are noted in @ref{Machine Dependencies}.
2350 No symbol may begin with a digit. Case is significant.
2351 There is no length limit: all characters are significant. Symbols are
2352 delimited by characters not in that set, or by the beginning of a file
2353 (since the source program must end with a newline, the end of a file is
2354 not a possible symbol delimiter). @xref{Symbols}.
2355 @cindex length of symbols
2360 @cindex statements, structure of
2361 @cindex line separator character
2362 @cindex statement separator character
2364 @ifclear abnormal-separator
2365 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
2366 semicolon (@samp{;}). The newline or semicolon is considered part of
2367 the preceding statement. Newlines and semicolons within character
2368 constants are an exception: they do not end statements.
2370 @ifset abnormal-separator
2372 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
2373 point (@samp{!}). The newline or exclamation point is considered part of the
2374 preceding statement. Newlines and exclamation points within character
2375 constants are an exception: they do not end statements.
2378 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
2379 H8/300) a dollar sign (@samp{$}); or (for the Renesas-SH) a semicolon
2380 (@samp{;}). The newline or separator character is considered part of
2381 the preceding statement. Newlines and separators within character
2382 constants are an exception: they do not end statements.
2387 A @dfn{statement} ends at a newline character (@samp{\n}) or line
2388 separator character. (The line separator is usually @samp{;}, unless this
2389 conflicts with the comment character; see @ref{Machine Dependencies}.) The
2390 newline or separator character is considered part of the preceding
2391 statement. Newlines and separators within character constants are an
2392 exception: they do not end statements.
2395 @cindex newline, required at file end
2396 @cindex EOF, newline must precede
2397 It is an error to end any statement with end-of-file: the last
2398 character of any input file should be a newline.@refill
2400 An empty statement is allowed, and may include whitespace. It is ignored.
2402 @cindex instructions and directives
2403 @cindex directives and instructions
2404 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2405 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2407 A statement begins with zero or more labels, optionally followed by a
2408 key symbol which determines what kind of statement it is. The key
2409 symbol determines the syntax of the rest of the statement. If the
2410 symbol begins with a dot @samp{.} then the statement is an assembler
2411 directive: typically valid for any computer. If the symbol begins with
2412 a letter the statement is an assembly language @dfn{instruction}: it
2413 assembles into a machine language instruction.
2415 Different versions of @command{@value{AS}} for different computers
2416 recognize different instructions. In fact, the same symbol may
2417 represent a different instruction in a different computer's assembly
2421 @cindex @code{:} (label)
2422 @cindex label (@code{:})
2423 A label is a symbol immediately followed by a colon (@code{:}).
2424 Whitespace before a label or after a colon is permitted, but you may not
2425 have whitespace between a label's symbol and its colon. @xref{Labels}.
2428 For HPPA targets, labels need not be immediately followed by a colon, but
2429 the definition of a label must begin in column zero. This also implies that
2430 only one label may be defined on each line.
2434 label: .directive followed by something
2435 another_label: # This is an empty statement.
2436 instruction operand_1, operand_2, @dots{}
2443 A constant is a number, written so that its value is known by
2444 inspection, without knowing any context. Like this:
2447 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2448 .ascii "Ring the bell\7" # A string constant.
2449 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2450 .float 0f-314159265358979323846264338327\
2451 95028841971.693993751E-40 # - pi, a flonum.
2456 * Characters:: Character Constants
2457 * Numbers:: Number Constants
2461 @subsection Character Constants
2463 @cindex character constants
2464 @cindex constants, character
2465 There are two kinds of character constants. A @dfn{character} stands
2466 for one character in one byte and its value may be used in
2467 numeric expressions. String constants (properly called string
2468 @emph{literals}) are potentially many bytes and their values may not be
2469 used in arithmetic expressions.
2473 * Chars:: Characters
2477 @subsubsection Strings
2479 @cindex string constants
2480 @cindex constants, string
2481 A @dfn{string} is written between double-quotes. It may contain
2482 double-quotes or null characters. The way to get special characters
2483 into a string is to @dfn{escape} these characters: precede them with
2484 a backslash @samp{\} character. For example @samp{\\} represents
2485 one backslash: the first @code{\} is an escape which tells
2486 @command{@value{AS}} to interpret the second character literally as a backslash
2487 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2488 escape character). The complete list of escapes follows.
2490 @cindex escape codes, character
2491 @cindex character escape codes
2494 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2496 @cindex @code{\b} (backspace character)
2497 @cindex backspace (@code{\b})
2499 Mnemonic for backspace; for ASCII this is octal code 010.
2502 @c Mnemonic for EOText; for ASCII this is octal code 004.
2504 @cindex @code{\f} (formfeed character)
2505 @cindex formfeed (@code{\f})
2507 Mnemonic for FormFeed; for ASCII this is octal code 014.
2509 @cindex @code{\n} (newline character)
2510 @cindex newline (@code{\n})
2512 Mnemonic for newline; for ASCII this is octal code 012.
2515 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2517 @cindex @code{\r} (carriage return character)
2518 @cindex carriage return (@code{\r})
2520 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2523 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2524 @c other assemblers.
2526 @cindex @code{\t} (tab)
2527 @cindex tab (@code{\t})
2529 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2532 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2533 @c @item \x @var{digit} @var{digit} @var{digit}
2534 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2536 @cindex @code{\@var{ddd}} (octal character code)
2537 @cindex octal character code (@code{\@var{ddd}})
2538 @item \ @var{digit} @var{digit} @var{digit}
2539 An octal character code. The numeric code is 3 octal digits.
2540 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2541 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2543 @cindex @code{\@var{xd...}} (hex character code)
2544 @cindex hex character code (@code{\@var{xd...}})
2545 @item \@code{x} @var{hex-digits...}
2546 A hex character code. All trailing hex digits are combined. Either upper or
2547 lower case @code{x} works.
2549 @cindex @code{\\} (@samp{\} character)
2550 @cindex backslash (@code{\\})
2552 Represents one @samp{\} character.
2555 @c Represents one @samp{'} (accent acute) character.
2556 @c This is needed in single character literals
2557 @c (@xref{Characters,,Character Constants}.) to represent
2560 @cindex @code{\"} (doublequote character)
2561 @cindex doublequote (@code{\"})
2563 Represents one @samp{"} character. Needed in strings to represent
2564 this character, because an unescaped @samp{"} would end the string.
2566 @item \ @var{anything-else}
2567 Any other character when escaped by @kbd{\} gives a warning, but
2568 assembles as if the @samp{\} was not present. The idea is that if
2569 you used an escape sequence you clearly didn't want the literal
2570 interpretation of the following character. However @command{@value{AS}} has no
2571 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2572 code and warns you of the fact.
2575 Which characters are escapable, and what those escapes represent,
2576 varies widely among assemblers. The current set is what we think
2577 the BSD 4.2 assembler recognizes, and is a subset of what most C
2578 compilers recognize. If you are in doubt, do not use an escape
2582 @subsubsection Characters
2584 @cindex single character constant
2585 @cindex character, single
2586 @cindex constant, single character
2587 A single character may be written as a single quote immediately
2588 followed by that character. The same escapes apply to characters as
2589 to strings. So if you want to write the character backslash, you
2590 must write @kbd{'\\} where the first @code{\} escapes the second
2591 @code{\}. As you can see, the quote is an acute accent, not a
2592 grave accent. A newline
2594 @ifclear abnormal-separator
2595 (or semicolon @samp{;})
2597 @ifset abnormal-separator
2599 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2604 immediately following an acute accent is taken as a literal character
2605 and does not count as the end of a statement. The value of a character
2606 constant in a numeric expression is the machine's byte-wide code for
2607 that character. @command{@value{AS}} assumes your character code is ASCII:
2608 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2611 @subsection Number Constants
2613 @cindex constants, number
2614 @cindex number constants
2615 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2616 are stored in the target machine. @emph{Integers} are numbers that
2617 would fit into an @code{int} in the C language. @emph{Bignums} are
2618 integers, but they are stored in more than 32 bits. @emph{Flonums}
2619 are floating point numbers, described below.
2622 * Integers:: Integers
2627 * Bit Fields:: Bit Fields
2633 @subsubsection Integers
2635 @cindex constants, integer
2637 @cindex binary integers
2638 @cindex integers, binary
2639 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2640 the binary digits @samp{01}.
2642 @cindex octal integers
2643 @cindex integers, octal
2644 An octal integer is @samp{0} followed by zero or more of the octal
2645 digits (@samp{01234567}).
2647 @cindex decimal integers
2648 @cindex integers, decimal
2649 A decimal integer starts with a non-zero digit followed by zero or
2650 more digits (@samp{0123456789}).
2652 @cindex hexadecimal integers
2653 @cindex integers, hexadecimal
2654 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2655 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2657 Integers have the usual values. To denote a negative integer, use
2658 the prefix operator @samp{-} discussed under expressions
2659 (@pxref{Prefix Ops,,Prefix Operators}).
2662 @subsubsection Bignums
2665 @cindex constants, bignum
2666 A @dfn{bignum} has the same syntax and semantics as an integer
2667 except that the number (or its negative) takes more than 32 bits to
2668 represent in binary. The distinction is made because in some places
2669 integers are permitted while bignums are not.
2672 @subsubsection Flonums
2674 @cindex floating point numbers
2675 @cindex constants, floating point
2677 @cindex precision, floating point
2678 A @dfn{flonum} represents a floating point number. The translation is
2679 indirect: a decimal floating point number from the text is converted by
2680 @command{@value{AS}} to a generic binary floating point number of more than
2681 sufficient precision. This generic floating point number is converted
2682 to a particular computer's floating point format (or formats) by a
2683 portion of @command{@value{AS}} specialized to that computer.
2685 A flonum is written by writing (in order)
2690 (@samp{0} is optional on the HPPA.)
2694 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2696 @kbd{e} is recommended. Case is not important.
2698 @c FIXME: verify if flonum syntax really this vague for most cases
2699 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2700 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2703 On the H8/300, Renesas / SuperH SH,
2704 and AMD 29K architectures, the letter must be
2705 one of the letters @samp{DFPRSX} (in upper or lower case).
2707 On the ARC, the letter must be one of the letters @samp{DFRS}
2708 (in upper or lower case).
2710 On the Intel 960 architecture, the letter must be
2711 one of the letters @samp{DFT} (in upper or lower case).
2713 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2717 One of the letters @samp{DFRS} (in upper or lower case).
2720 One of the letters @samp{DFPRSX} (in upper or lower case).
2723 The letter @samp{E} (upper case only).
2726 One of the letters @samp{DFT} (in upper or lower case).
2731 An optional sign: either @samp{+} or @samp{-}.
2734 An optional @dfn{integer part}: zero or more decimal digits.
2737 An optional @dfn{fractional part}: @samp{.} followed by zero
2738 or more decimal digits.
2741 An optional exponent, consisting of:
2745 An @samp{E} or @samp{e}.
2746 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2747 @c principle this can perfectly well be different on different targets.
2749 Optional sign: either @samp{+} or @samp{-}.
2751 One or more decimal digits.
2756 At least one of the integer part or the fractional part must be
2757 present. The floating point number has the usual base-10 value.
2759 @command{@value{AS}} does all processing using integers. Flonums are computed
2760 independently of any floating point hardware in the computer running
2761 @command{@value{AS}}.
2765 @c Bit fields are written as a general facility but are also controlled
2766 @c by a conditional-compilation flag---which is as of now (21mar91)
2767 @c turned on only by the i960 config of GAS.
2769 @subsubsection Bit Fields
2772 @cindex constants, bit field
2773 You can also define numeric constants as @dfn{bit fields}.
2774 Specify two numbers separated by a colon---
2776 @var{mask}:@var{value}
2779 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2782 The resulting number is then packed
2784 @c this conditional paren in case bit fields turned on elsewhere than 960
2785 (in host-dependent byte order)
2787 into a field whose width depends on which assembler directive has the
2788 bit-field as its argument. Overflow (a result from the bitwise and
2789 requiring more binary digits to represent) is not an error; instead,
2790 more constants are generated, of the specified width, beginning with the
2791 least significant digits.@refill
2793 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2794 @code{.short}, and @code{.word} accept bit-field arguments.
2799 @chapter Sections and Relocation
2804 * Secs Background:: Background
2805 * Ld Sections:: Linker Sections
2806 * As Sections:: Assembler Internal Sections
2807 * Sub-Sections:: Sub-Sections
2811 @node Secs Background
2814 Roughly, a section is a range of addresses, with no gaps; all data
2815 ``in'' those addresses is treated the same for some particular purpose.
2816 For example there may be a ``read only'' section.
2818 @cindex linker, and assembler
2819 @cindex assembler, and linker
2820 The linker @code{@value{LD}} reads many object files (partial programs) and
2821 combines their contents to form a runnable program. When @command{@value{AS}}
2822 emits an object file, the partial program is assumed to start at address 0.
2823 @code{@value{LD}} assigns the final addresses for the partial program, so that
2824 different partial programs do not overlap. This is actually an
2825 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2828 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2829 addresses. These blocks slide to their run-time addresses as rigid
2830 units; their length does not change and neither does the order of bytes
2831 within them. Such a rigid unit is called a @emph{section}. Assigning
2832 run-time addresses to sections is called @dfn{relocation}. It includes
2833 the task of adjusting mentions of object-file addresses so they refer to
2834 the proper run-time addresses.
2836 For the H8/300, and for the Renesas / SuperH SH,
2837 @command{@value{AS}} pads sections if needed to
2838 ensure they end on a word (sixteen bit) boundary.
2841 @cindex standard assembler sections
2842 An object file written by @command{@value{AS}} has at least three sections, any
2843 of which may be empty. These are named @dfn{text}, @dfn{data} and
2848 When it generates COFF or ELF output,
2850 @command{@value{AS}} can also generate whatever other named sections you specify
2851 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2852 If you do not use any directives that place output in the @samp{.text}
2853 or @samp{.data} sections, these sections still exist, but are empty.
2858 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2860 @command{@value{AS}} can also generate whatever other named sections you
2861 specify using the @samp{.space} and @samp{.subspace} directives. See
2862 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2863 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2864 assembler directives.
2867 Additionally, @command{@value{AS}} uses different names for the standard
2868 text, data, and bss sections when generating SOM output. Program text
2869 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2870 BSS into @samp{$BSS$}.
2874 Within the object file, the text section starts at address @code{0}, the
2875 data section follows, and the bss section follows the data section.
2878 When generating either SOM or ELF output files on the HPPA, the text
2879 section starts at address @code{0}, the data section at address
2880 @code{0x4000000}, and the bss section follows the data section.
2883 To let @code{@value{LD}} know which data changes when the sections are
2884 relocated, and how to change that data, @command{@value{AS}} also writes to the
2885 object file details of the relocation needed. To perform relocation
2886 @code{@value{LD}} must know, each time an address in the object
2890 Where in the object file is the beginning of this reference to
2893 How long (in bytes) is this reference?
2895 Which section does the address refer to? What is the numeric value of
2897 (@var{address}) @minus{} (@var{start-address of section})?
2900 Is the reference to an address ``Program-Counter relative''?
2903 @cindex addresses, format of
2904 @cindex section-relative addressing
2905 In fact, every address @command{@value{AS}} ever uses is expressed as
2907 (@var{section}) + (@var{offset into section})
2910 Further, most expressions @command{@value{AS}} computes have this section-relative
2913 (For some object formats, such as SOM for the HPPA, some expressions are
2914 symbol-relative instead.)
2917 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2918 @var{N} into section @var{secname}.''
2920 Apart from text, data and bss sections you need to know about the
2921 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2922 addresses in the absolute section remain unchanged. For example, address
2923 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2924 @code{@value{LD}}. Although the linker never arranges two partial programs'
2925 data sections with overlapping addresses after linking, @emph{by definition}
2926 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2927 part of a program is always the same address when the program is running as
2928 address @code{@{absolute@ 239@}} in any other part of the program.
2930 The idea of sections is extended to the @dfn{undefined} section. Any
2931 address whose section is unknown at assembly time is by definition
2932 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2933 Since numbers are always defined, the only way to generate an undefined
2934 address is to mention an undefined symbol. A reference to a named
2935 common block would be such a symbol: its value is unknown at assembly
2936 time so it has section @emph{undefined}.
2938 By analogy the word @emph{section} is used to describe groups of sections in
2939 the linked program. @code{@value{LD}} puts all partial programs' text
2940 sections in contiguous addresses in the linked program. It is
2941 customary to refer to the @emph{text section} of a program, meaning all
2942 the addresses of all partial programs' text sections. Likewise for
2943 data and bss sections.
2945 Some sections are manipulated by @code{@value{LD}}; others are invented for
2946 use of @command{@value{AS}} and have no meaning except during assembly.
2949 @section Linker Sections
2950 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2955 @cindex named sections
2956 @cindex sections, named
2957 @item named sections
2960 @cindex text section
2961 @cindex data section
2965 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
2966 separate but equal sections. Anything you can say of one section is
2969 When the program is running, however, it is
2970 customary for the text section to be unalterable. The
2971 text section is often shared among processes: it contains
2972 instructions, constants and the like. The data section of a running
2973 program is usually alterable: for example, C variables would be stored
2974 in the data section.
2979 This section contains zeroed bytes when your program begins running. It
2980 is used to hold uninitialized variables or common storage. The length of
2981 each partial program's bss section is important, but because it starts
2982 out containing zeroed bytes there is no need to store explicit zero
2983 bytes in the object file. The bss section was invented to eliminate
2984 those explicit zeros from object files.
2986 @cindex absolute section
2987 @item absolute section
2988 Address 0 of this section is always ``relocated'' to runtime address 0.
2989 This is useful if you want to refer to an address that @code{@value{LD}} must
2990 not change when relocating. In this sense we speak of absolute
2991 addresses being ``unrelocatable'': they do not change during relocation.
2993 @cindex undefined section
2994 @item undefined section
2995 This ``section'' is a catch-all for address references to objects not in
2996 the preceding sections.
2997 @c FIXME: ref to some other doc on obj-file formats could go here.
3000 @cindex relocation example
3001 An idealized example of three relocatable sections follows.
3003 The example uses the traditional section names @samp{.text} and @samp{.data}.
3005 Memory addresses are on the horizontal axis.
3009 @c END TEXI2ROFF-KILL
3012 partial program # 1: |ttttt|dddd|00|
3019 partial program # 2: |TTT|DDD|000|
3022 +--+---+-----+--+----+---+-----+~~
3023 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3024 +--+---+-----+--+----+---+-----+~~
3026 addresses: 0 @dots{}
3033 \line{\it Partial program \#1: \hfil}
3034 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3035 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3037 \line{\it Partial program \#2: \hfil}
3038 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3039 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3041 \line{\it linked program: \hfil}
3042 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3043 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3044 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3045 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3047 \line{\it addresses: \hfil}
3051 @c END TEXI2ROFF-KILL
3054 @section Assembler Internal Sections
3056 @cindex internal assembler sections
3057 @cindex sections in messages, internal
3058 These sections are meant only for the internal use of @command{@value{AS}}. They
3059 have no meaning at run-time. You do not really need to know about these
3060 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3061 warning messages, so it might be helpful to have an idea of their
3062 meanings to @command{@value{AS}}. These sections are used to permit the
3063 value of every expression in your assembly language program to be a
3064 section-relative address.
3067 @cindex assembler internal logic error
3068 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3069 An internal assembler logic error has been found. This means there is a
3070 bug in the assembler.
3072 @cindex expr (internal section)
3074 The assembler stores complex expression internally as combinations of
3075 symbols. When it needs to represent an expression as a symbol, it puts
3076 it in the expr section.
3078 @c FIXME item transfer[t] vector preload
3079 @c FIXME item transfer[t] vector postload
3080 @c FIXME item register
3084 @section Sub-Sections
3086 @cindex numbered subsections
3087 @cindex grouping data
3093 fall into two sections: text and data.
3095 You may have separate groups of
3097 data in named sections
3101 data in named sections
3107 that you want to end up near to each other in the object file, even though they
3108 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3109 use @dfn{subsections} for this purpose. Within each section, there can be
3110 numbered subsections with values from 0 to 8192. Objects assembled into the
3111 same subsection go into the object file together with other objects in the same
3112 subsection. For example, a compiler might want to store constants in the text
3113 section, but might not want to have them interspersed with the program being
3114 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3115 section of code being output, and a @samp{.text 1} before each group of
3116 constants being output.
3118 Subsections are optional. If you do not use subsections, everything
3119 goes in subsection number zero.
3122 Each subsection is zero-padded up to a multiple of four bytes.
3123 (Subsections may be padded a different amount on different flavors
3124 of @command{@value{AS}}.)
3128 On the H8/300 platform, each subsection is zero-padded to a word
3129 boundary (two bytes).
3130 The same is true on the Renesas SH.
3133 @c FIXME section padding (alignment)?
3134 @c Rich Pixley says padding here depends on target obj code format; that
3135 @c doesn't seem particularly useful to say without further elaboration,
3136 @c so for now I say nothing about it. If this is a generic BFD issue,
3137 @c these paragraphs might need to vanish from this manual, and be
3138 @c discussed in BFD chapter of binutils (or some such).
3142 Subsections appear in your object file in numeric order, lowest numbered
3143 to highest. (All this to be compatible with other people's assemblers.)
3144 The object file contains no representation of subsections; @code{@value{LD}} and
3145 other programs that manipulate object files see no trace of them.
3146 They just see all your text subsections as a text section, and all your
3147 data subsections as a data section.
3149 To specify which subsection you want subsequent statements assembled
3150 into, use a numeric argument to specify it, in a @samp{.text
3151 @var{expression}} or a @samp{.data @var{expression}} statement.
3154 When generating COFF output, you
3159 can also use an extra subsection
3160 argument with arbitrary named sections: @samp{.section @var{name},
3165 When generating ELF output, you
3170 can also use the @code{.subsection} directive (@pxref{SubSection})
3171 to specify a subsection: @samp{.subsection @var{expression}}.
3173 @var{Expression} should be an absolute expression
3174 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3175 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3176 begins in @code{text 0}. For instance:
3178 .text 0 # The default subsection is text 0 anyway.
3179 .ascii "This lives in the first text subsection. *"
3181 .ascii "But this lives in the second text subsection."
3183 .ascii "This lives in the data section,"
3184 .ascii "in the first data subsection."
3186 .ascii "This lives in the first text section,"
3187 .ascii "immediately following the asterisk (*)."
3190 Each section has a @dfn{location counter} incremented by one for every byte
3191 assembled into that section. Because subsections are merely a convenience
3192 restricted to @command{@value{AS}} there is no concept of a subsection location
3193 counter. There is no way to directly manipulate a location counter---but the
3194 @code{.align} directive changes it, and any label definition captures its
3195 current value. The location counter of the section where statements are being
3196 assembled is said to be the @dfn{active} location counter.
3199 @section bss Section
3202 @cindex common variable storage
3203 The bss section is used for local common variable storage.
3204 You may allocate address space in the bss section, but you may
3205 not dictate data to load into it before your program executes. When
3206 your program starts running, all the contents of the bss
3207 section are zeroed bytes.
3209 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3210 @ref{Lcomm,,@code{.lcomm}}.
3212 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3213 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3216 When assembling for a target which supports multiple sections, such as ELF or
3217 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3218 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3219 section. Typically the section will only contain symbol definitions and
3220 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3227 Symbols are a central concept: the programmer uses symbols to name
3228 things, the linker uses symbols to link, and the debugger uses symbols
3232 @cindex debuggers, and symbol order
3233 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3234 the same order they were declared. This may break some debuggers.
3239 * Setting Symbols:: Giving Symbols Other Values
3240 * Symbol Names:: Symbol Names
3241 * Dot:: The Special Dot Symbol
3242 * Symbol Attributes:: Symbol Attributes
3249 A @dfn{label} is written as a symbol immediately followed by a colon
3250 @samp{:}. The symbol then represents the current value of the
3251 active location counter, and is, for example, a suitable instruction
3252 operand. You are warned if you use the same symbol to represent two
3253 different locations: the first definition overrides any other
3257 On the HPPA, the usual form for a label need not be immediately followed by a
3258 colon, but instead must start in column zero. Only one label may be defined on
3259 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3260 provides a special directive @code{.label} for defining labels more flexibly.
3263 @node Setting Symbols
3264 @section Giving Symbols Other Values
3266 @cindex assigning values to symbols
3267 @cindex symbol values, assigning
3268 A symbol can be given an arbitrary value by writing a symbol, followed
3269 by an equals sign @samp{=}, followed by an expression
3270 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3271 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3272 equals sign @samp{=}@samp{=} here represents an equivalent of the
3273 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3276 Blackfin does not support symbol assignment with @samp{=}.
3280 @section Symbol Names
3282 @cindex symbol names
3283 @cindex names, symbol
3284 @ifclear SPECIAL-SYMS
3285 Symbol names begin with a letter or with one of @samp{._}. On most
3286 machines, you can also use @code{$} in symbol names; exceptions are
3287 noted in @ref{Machine Dependencies}. That character may be followed by any
3288 string of digits, letters, dollar signs (unless otherwise noted for a
3289 particular target machine), and underscores.
3293 Symbol names begin with a letter or with one of @samp{._}. On the
3294 Renesas SH you can also use @code{$} in symbol names. That
3295 character may be followed by any string of digits, letters, dollar signs (save
3296 on the H8/300), and underscores.
3300 Case of letters is significant: @code{foo} is a different symbol name
3303 Each symbol has exactly one name. Each name in an assembly language program
3304 refers to exactly one symbol. You may use that symbol name any number of times
3307 @subheading Local Symbol Names
3309 @cindex local symbol names
3310 @cindex symbol names, local
3311 A local symbol is any symbol beginning with certain local label prefixes.
3312 By default, the local label prefix is @samp{.L} for ELF systems or
3313 @samp{L} for traditional a.out systems, but each target may have its own
3314 set of local label prefixes.
3316 On the HPPA local symbols begin with @samp{L$}.
3319 Local symbols are defined and used within the assembler, but they are
3320 normally not saved in object files. Thus, they are not visible when debugging.
3321 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols:
3322 @option{-L}}) to retain the local symbols in the object files.
3324 @subheading Local Labels
3326 @cindex local labels
3327 @cindex temporary symbol names
3328 @cindex symbol names, temporary
3329 Local labels help compilers and programmers use names temporarily.
3330 They create symbols which are guaranteed to be unique over the entire scope of
3331 the input source code and which can be referred to by a simple notation.
3332 To define a local label, write a label of the form @samp{@b{N}:} (where @b{N}
3333 represents any positive integer). To refer to the most recent previous
3334 definition of that label write @samp{@b{N}b}, using the same number as when
3335 you defined the label. To refer to the next definition of a local label, write
3336 @samp{@b{N}f}---the @samp{b} stands for ``backwards'' and the @samp{f} stands
3339 There is no restriction on how you can use these labels, and you can reuse them
3340 too. So that it is possible to repeatedly define the same local label (using
3341 the same number @samp{@b{N}}), although you can only refer to the most recently
3342 defined local label of that number (for a backwards reference) or the next
3343 definition of a specific local label for a forward reference. It is also worth
3344 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3345 implemented in a slightly more efficient manner than the others.
3356 Which is the equivalent of:
3359 label_1: branch label_3
3360 label_2: branch label_1
3361 label_3: branch label_4
3362 label_4: branch label_3
3365 Local label names are only a notational device. They are immediately
3366 transformed into more conventional symbol names before the assembler uses them.
3367 The symbol names are stored in the symbol table, appear in error messages, and
3368 are optionally emitted to the object file. The names are constructed using
3372 @item @emph{local label prefix}
3373 All local symbols begin with the system-specific local label prefix.
3374 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3375 that start with the local label prefix. These labels are
3376 used for symbols you are never intended to see. If you use the
3377 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3378 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3379 you may use them in debugging.
3382 This is the number that was used in the local label definition. So if the
3383 label is written @samp{55:} then the number is @samp{55}.
3386 This unusual character is included so you do not accidentally invent a symbol
3387 of the same name. The character has ASCII value of @samp{\002} (control-B).
3389 @item @emph{ordinal number}
3390 This is a serial number to keep the labels distinct. The first definition of
3391 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3392 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3393 the number @samp{1} and its 15th definition gets @samp{15} as well.
3396 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3397 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3399 @subheading Dollar Local Labels
3400 @cindex dollar local symbols
3402 @code{@value{AS}} also supports an even more local form of local labels called
3403 dollar labels. These labels go out of scope (i.e., they become undefined) as
3404 soon as a non-local label is defined. Thus they remain valid for only a small
3405 region of the input source code. Normal local labels, by contrast, remain in
3406 scope for the entire file, or until they are redefined by another occurrence of
3407 the same local label.
3409 Dollar labels are defined in exactly the same way as ordinary local labels,
3410 except that they have a dollar sign suffix to their numeric value, e.g.,
3413 They can also be distinguished from ordinary local labels by their transformed
3414 names which use ASCII character @samp{\001} (control-A) as the magic character
3415 to distinguish them from ordinary labels. For example, the fifth definition of
3416 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3419 @section The Special Dot Symbol
3421 @cindex dot (symbol)
3422 @cindex @code{.} (symbol)
3423 @cindex current address
3424 @cindex location counter
3425 The special symbol @samp{.} refers to the current address that
3426 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3427 .long .} defines @code{melvin} to contain its own address.
3428 Assigning a value to @code{.} is treated the same as a @code{.org}
3430 @ifclear no-space-dir
3431 Thus, the expression @samp{.=.+4} is the same as saying
3435 @node Symbol Attributes
3436 @section Symbol Attributes
3438 @cindex symbol attributes
3439 @cindex attributes, symbol
3440 Every symbol has, as well as its name, the attributes ``Value'' and
3441 ``Type''. Depending on output format, symbols can also have auxiliary
3444 The detailed definitions are in @file{a.out.h}.
3447 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3448 all these attributes, and probably won't warn you. This makes the
3449 symbol an externally defined symbol, which is generally what you
3453 * Symbol Value:: Value
3454 * Symbol Type:: Type
3457 * a.out Symbols:: Symbol Attributes: @code{a.out}
3461 * a.out Symbols:: Symbol Attributes: @code{a.out}
3464 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3469 * COFF Symbols:: Symbol Attributes for COFF
3472 * SOM Symbols:: Symbol Attributes for SOM
3479 @cindex value of a symbol
3480 @cindex symbol value
3481 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3482 location in the text, data, bss or absolute sections the value is the
3483 number of addresses from the start of that section to the label.
3484 Naturally for text, data and bss sections the value of a symbol changes
3485 as @code{@value{LD}} changes section base addresses during linking. Absolute
3486 symbols' values do not change during linking: that is why they are
3489 The value of an undefined symbol is treated in a special way. If it is
3490 0 then the symbol is not defined in this assembler source file, and
3491 @code{@value{LD}} tries to determine its value from other files linked into the
3492 same program. You make this kind of symbol simply by mentioning a symbol
3493 name without defining it. A non-zero value represents a @code{.comm}
3494 common declaration. The value is how much common storage to reserve, in
3495 bytes (addresses). The symbol refers to the first address of the
3501 @cindex type of a symbol
3503 The type attribute of a symbol contains relocation (section)
3504 information, any flag settings indicating that a symbol is external, and
3505 (optionally), other information for linkers and debuggers. The exact
3506 format depends on the object-code output format in use.
3511 @c The following avoids a "widow" subsection title. @group would be
3512 @c better if it were available outside examples.
3515 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3517 @cindex @code{b.out} symbol attributes
3518 @cindex symbol attributes, @code{b.out}
3519 These symbol attributes appear only when @command{@value{AS}} is configured for
3520 one of the Berkeley-descended object output formats---@code{a.out} or
3526 @subsection Symbol Attributes: @code{a.out}
3528 @cindex @code{a.out} symbol attributes
3529 @cindex symbol attributes, @code{a.out}
3535 @subsection Symbol Attributes: @code{a.out}
3537 @cindex @code{a.out} symbol attributes
3538 @cindex symbol attributes, @code{a.out}
3542 * Symbol Desc:: Descriptor
3543 * Symbol Other:: Other
3547 @subsubsection Descriptor
3549 @cindex descriptor, of @code{a.out} symbol
3550 This is an arbitrary 16-bit value. You may establish a symbol's
3551 descriptor value by using a @code{.desc} statement
3552 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3553 @command{@value{AS}}.
3556 @subsubsection Other
3558 @cindex other attribute, of @code{a.out} symbol
3559 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3564 @subsection Symbol Attributes for COFF
3566 @cindex COFF symbol attributes
3567 @cindex symbol attributes, COFF
3569 The COFF format supports a multitude of auxiliary symbol attributes;
3570 like the primary symbol attributes, they are set between @code{.def} and
3571 @code{.endef} directives.
3573 @subsubsection Primary Attributes
3575 @cindex primary attributes, COFF symbols
3576 The symbol name is set with @code{.def}; the value and type,
3577 respectively, with @code{.val} and @code{.type}.
3579 @subsubsection Auxiliary Attributes
3581 @cindex auxiliary attributes, COFF symbols
3582 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3583 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3584 table information for COFF.
3589 @subsection Symbol Attributes for SOM
3591 @cindex SOM symbol attributes
3592 @cindex symbol attributes, SOM
3594 The SOM format for the HPPA supports a multitude of symbol attributes set with
3595 the @code{.EXPORT} and @code{.IMPORT} directives.
3597 The attributes are described in @cite{HP9000 Series 800 Assembly
3598 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3599 @code{EXPORT} assembler directive documentation.
3603 @chapter Expressions
3607 @cindex numeric values
3608 An @dfn{expression} specifies an address or numeric value.
3609 Whitespace may precede and/or follow an expression.
3611 The result of an expression must be an absolute number, or else an offset into
3612 a particular section. If an expression is not absolute, and there is not
3613 enough information when @command{@value{AS}} sees the expression to know its
3614 section, a second pass over the source program might be necessary to interpret
3615 the expression---but the second pass is currently not implemented.
3616 @command{@value{AS}} aborts with an error message in this situation.
3619 * Empty Exprs:: Empty Expressions
3620 * Integer Exprs:: Integer Expressions
3624 @section Empty Expressions
3626 @cindex empty expressions
3627 @cindex expressions, empty
3628 An empty expression has no value: it is just whitespace or null.
3629 Wherever an absolute expression is required, you may omit the
3630 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3631 is compatible with other assemblers.
3634 @section Integer Expressions
3636 @cindex integer expressions
3637 @cindex expressions, integer
3638 An @dfn{integer expression} is one or more @emph{arguments} delimited
3639 by @emph{operators}.
3642 * Arguments:: Arguments
3643 * Operators:: Operators
3644 * Prefix Ops:: Prefix Operators
3645 * Infix Ops:: Infix Operators
3649 @subsection Arguments
3651 @cindex expression arguments
3652 @cindex arguments in expressions
3653 @cindex operands in expressions
3654 @cindex arithmetic operands
3655 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3656 contexts arguments are sometimes called ``arithmetic operands''. In
3657 this manual, to avoid confusing them with the ``instruction operands'' of
3658 the machine language, we use the term ``argument'' to refer to parts of
3659 expressions only, reserving the word ``operand'' to refer only to machine
3660 instruction operands.
3662 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3663 @var{section} is one of text, data, bss, absolute,
3664 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3667 Numbers are usually integers.
3669 A number can be a flonum or bignum. In this case, you are warned
3670 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3671 these 32 bits are an integer. You may write integer-manipulating
3672 instructions that act on exotic constants, compatible with other
3675 @cindex subexpressions
3676 Subexpressions are a left parenthesis @samp{(} followed by an integer
3677 expression, followed by a right parenthesis @samp{)}; or a prefix
3678 operator followed by an argument.
3681 @subsection Operators
3683 @cindex operators, in expressions
3684 @cindex arithmetic functions
3685 @cindex functions, in expressions
3686 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3687 operators are followed by an argument. Infix operators appear
3688 between their arguments. Operators may be preceded and/or followed by
3692 @subsection Prefix Operator
3694 @cindex prefix operators
3695 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3696 one argument, which must be absolute.
3698 @c the tex/end tex stuff surrounding this small table is meant to make
3699 @c it align, on the printed page, with the similar table in the next
3700 @c section (which is inside an enumerate).
3702 \global\advance\leftskip by \itemindent
3707 @dfn{Negation}. Two's complement negation.
3709 @dfn{Complementation}. Bitwise not.
3713 \global\advance\leftskip by -\itemindent
3717 @subsection Infix Operators
3719 @cindex infix operators
3720 @cindex operators, permitted arguments
3721 @dfn{Infix operators} take two arguments, one on either side. Operators
3722 have precedence, but operations with equal precedence are performed left
3723 to right. Apart from @code{+} or @option{-}, both arguments must be
3724 absolute, and the result is absolute.
3727 @cindex operator precedence
3728 @cindex precedence of operators
3735 @dfn{Multiplication}.
3738 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3744 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3747 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3751 Intermediate precedence
3756 @dfn{Bitwise Inclusive Or}.
3762 @dfn{Bitwise Exclusive Or}.
3765 @dfn{Bitwise Or Not}.
3772 @cindex addition, permitted arguments
3773 @cindex plus, permitted arguments
3774 @cindex arguments for addition
3776 @dfn{Addition}. If either argument is absolute, the result has the section of
3777 the other argument. You may not add together arguments from different
3780 @cindex subtraction, permitted arguments
3781 @cindex minus, permitted arguments
3782 @cindex arguments for subtraction
3784 @dfn{Subtraction}. If the right argument is absolute, the
3785 result has the section of the left argument.
3786 If both arguments are in the same section, the result is absolute.
3787 You may not subtract arguments from different sections.
3788 @c FIXME is there still something useful to say about undefined - undefined ?
3790 @cindex comparison expressions
3791 @cindex expressions, comparison
3796 @dfn{Is Not Equal To}
3800 @dfn{Is Greater Than}
3802 @dfn{Is Greater Than Or Equal To}
3804 @dfn{Is Less Than Or Equal To}
3806 The comparison operators can be used as infix operators. A true results has a
3807 value of -1 whereas a false result has a value of 0. Note, these operators
3808 perform signed comparisons.
3811 @item Lowest Precedence
3820 These two logical operations can be used to combine the results of sub
3821 expressions. Note, unlike the comparison operators a true result returns a
3822 value of 1 but a false results does still return 0. Also note that the logical
3823 or operator has a slightly lower precedence than logical and.
3828 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3829 address; you can only have a defined section in one of the two arguments.
3832 @chapter Assembler Directives
3834 @cindex directives, machine independent
3835 @cindex pseudo-ops, machine independent
3836 @cindex machine independent directives
3837 All assembler directives have names that begin with a period (@samp{.}).
3838 The rest of the name is letters, usually in lower case.
3840 This chapter discusses directives that are available regardless of the
3841 target machine configuration for the @sc{gnu} assembler.
3843 Some machine configurations provide additional directives.
3844 @xref{Machine Dependencies}.
3847 @ifset machine-directives
3848 @xref{Machine Dependencies}, for additional directives.
3853 * Abort:: @code{.abort}
3855 * ABORT (COFF):: @code{.ABORT}
3858 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3859 * Altmacro:: @code{.altmacro}
3860 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3861 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3862 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3863 * Byte:: @code{.byte @var{expressions}}
3864 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
3865 * Comm:: @code{.comm @var{symbol} , @var{length} }
3866 * Data:: @code{.data @var{subsection}}
3868 * Def:: @code{.def @var{name}}
3871 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3877 * Double:: @code{.double @var{flonums}}
3878 * Eject:: @code{.eject}
3879 * Else:: @code{.else}
3880 * Elseif:: @code{.elseif}
3883 * Endef:: @code{.endef}
3886 * Endfunc:: @code{.endfunc}
3887 * Endif:: @code{.endif}
3888 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3889 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3890 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
3892 * Error:: @code{.error @var{string}}
3893 * Exitm:: @code{.exitm}
3894 * Extern:: @code{.extern}
3895 * Fail:: @code{.fail}
3896 * File:: @code{.file}
3897 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3898 * Float:: @code{.float @var{flonums}}
3899 * Func:: @code{.func}
3900 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3902 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
3903 * Hidden:: @code{.hidden @var{names}}
3906 * hword:: @code{.hword @var{expressions}}
3907 * Ident:: @code{.ident}
3908 * If:: @code{.if @var{absolute expression}}
3909 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
3910 * Include:: @code{.include "@var{file}"}
3911 * Int:: @code{.int @var{expressions}}
3913 * Internal:: @code{.internal @var{names}}
3916 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3917 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3918 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3919 * Lflags:: @code{.lflags}
3920 @ifclear no-line-dir
3921 * Line:: @code{.line @var{line-number}}
3924 * Linkonce:: @code{.linkonce [@var{type}]}
3925 * List:: @code{.list}
3926 * Ln:: @code{.ln @var{line-number}}
3927 * Loc:: @code{.loc @var{fileno} @var{lineno}}
3928 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
3930 * Local:: @code{.local @var{names}}
3933 * Long:: @code{.long @var{expressions}}
3935 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3938 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3939 * MRI:: @code{.mri @var{val}}
3940 * Noaltmacro:: @code{.noaltmacro}
3941 * Nolist:: @code{.nolist}
3942 * Octa:: @code{.octa @var{bignums}}
3943 * Org:: @code{.org @var{new-lc}, @var{fill}}
3944 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3946 * PopSection:: @code{.popsection}
3947 * Previous:: @code{.previous}
3950 * Print:: @code{.print @var{string}}
3952 * Protected:: @code{.protected @var{names}}
3955 * Psize:: @code{.psize @var{lines}, @var{columns}}
3956 * Purgem:: @code{.purgem @var{name}}
3958 * PushSection:: @code{.pushsection @var{name}}
3961 * Quad:: @code{.quad @var{bignums}}
3962 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
3963 * Rept:: @code{.rept @var{count}}
3964 * Sbttl:: @code{.sbttl "@var{subheading}"}
3966 * Scl:: @code{.scl @var{class}}
3969 * Section:: @code{.section @var{name}[, @var{flags}]}
3972 * Set:: @code{.set @var{symbol}, @var{expression}}
3973 * Short:: @code{.short @var{expressions}}
3974 * Single:: @code{.single @var{flonums}}
3976 * Size:: @code{.size [@var{name} , @var{expression}]}
3978 @ifclear no-space-dir
3979 * Skip:: @code{.skip @var{size} , @var{fill}}
3982 * Sleb128:: @code{.sleb128 @var{expressions}}
3983 @ifclear no-space-dir
3984 * Space:: @code{.space @var{size} , @var{fill}}
3987 * Stab:: @code{.stabd, .stabn, .stabs}
3990 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
3991 * Struct:: @code{.struct @var{expression}}
3993 * SubSection:: @code{.subsection}
3994 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
3998 * Tag:: @code{.tag @var{structname}}
4001 * Text:: @code{.text @var{subsection}}
4002 * Title:: @code{.title "@var{heading}"}
4004 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4007 * Uleb128:: @code{.uleb128 @var{expressions}}
4009 * Val:: @code{.val @var{addr}}
4013 * Version:: @code{.version "@var{string}"}
4014 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4015 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4018 * Warning:: @code{.warning @var{string}}
4019 * Weak:: @code{.weak @var{names}}
4020 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4021 * Word:: @code{.word @var{expressions}}
4022 * Deprecated:: Deprecated Directives
4026 @section @code{.abort}
4028 @cindex @code{abort} directive
4029 @cindex stopping the assembly
4030 This directive stops the assembly immediately. It is for
4031 compatibility with other assemblers. The original idea was that the
4032 assembly language source would be piped into the assembler. If the sender
4033 of the source quit, it could use this directive tells @command{@value{AS}} to
4034 quit also. One day @code{.abort} will not be supported.
4038 @section @code{.ABORT} (COFF)
4040 @cindex @code{ABORT} directive
4041 When producing COFF output, @command{@value{AS}} accepts this directive as a
4042 synonym for @samp{.abort}.
4045 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4051 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4053 @cindex padding the location counter
4054 @cindex @code{align} directive
4055 Pad the location counter (in the current subsection) to a particular storage
4056 boundary. The first expression (which must be absolute) is the alignment
4057 required, as described below.
4059 The second expression (also absolute) gives the fill value to be stored in the
4060 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4061 padding bytes are normally zero. However, on some systems, if the section is
4062 marked as containing code and the fill value is omitted, the space is filled
4063 with no-op instructions.
4065 The third expression is also absolute, and is also optional. If it is present,
4066 it is the maximum number of bytes that should be skipped by this alignment
4067 directive. If doing the alignment would require skipping more bytes than the
4068 specified maximum, then the alignment is not done at all. You can omit the
4069 fill value (the second argument) entirely by simply using two commas after the
4070 required alignment; this can be useful if you want the alignment to be filled
4071 with no-op instructions when appropriate.
4073 The way the required alignment is specified varies from system to system.
4074 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or32,
4075 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4076 alignment request in bytes. For example @samp{.align 8} advances
4077 the location counter until it is a multiple of 8. If the location counter
4078 is already a multiple of 8, no change is needed. For the tic54x, the
4079 first expression is the alignment request in words.
4081 For other systems, including ppc, i386 using a.out format, arm and
4082 strongarm, it is the
4083 number of low-order zero bits the location counter must have after
4084 advancement. For example @samp{.align 3} advances the location
4085 counter until it a multiple of 8. If the location counter is already a
4086 multiple of 8, no change is needed.
4088 This inconsistency is due to the different behaviors of the various
4089 native assemblers for these systems which GAS must emulate.
4090 GAS also provides @code{.balign} and @code{.p2align} directives,
4091 described later, which have a consistent behavior across all
4092 architectures (but are specific to GAS).
4095 @section @code{.altmacro}
4096 Enable alternate macro mode, enabling:
4099 @item LOCAL @var{name} [ , @dots{} ]
4100 One additional directive, @code{LOCAL}, is available. It is used to
4101 generate a string replacement for each of the @var{name} arguments, and
4102 replace any instances of @var{name} in each macro expansion. The
4103 replacement string is unique in the assembly, and different for each
4104 separate macro expansion. @code{LOCAL} allows you to write macros that
4105 define symbols, without fear of conflict between separate macro expansions.
4107 @item String delimiters
4108 You can write strings delimited in these other ways besides
4109 @code{"@var{string}"}:
4112 @item '@var{string}'
4113 You can delimit strings with single-quote characters.
4115 @item <@var{string}>
4116 You can delimit strings with matching angle brackets.
4119 @item single-character string escape
4120 To include any single character literally in a string (even if the
4121 character would otherwise have some special meaning), you can prefix the
4122 character with @samp{!} (an exclamation mark). For example, you can
4123 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4125 @item Expression results as strings
4126 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4127 and use the result as a string.
4131 @section @code{.ascii "@var{string}"}@dots{}
4133 @cindex @code{ascii} directive
4134 @cindex string literals
4135 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4136 separated by commas. It assembles each string (with no automatic
4137 trailing zero byte) into consecutive addresses.
4140 @section @code{.asciz "@var{string}"}@dots{}
4142 @cindex @code{asciz} directive
4143 @cindex zero-terminated strings
4144 @cindex null-terminated strings
4145 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4146 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4149 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4151 @cindex padding the location counter given number of bytes
4152 @cindex @code{balign} directive
4153 Pad the location counter (in the current subsection) to a particular
4154 storage boundary. The first expression (which must be absolute) is the
4155 alignment request in bytes. For example @samp{.balign 8} advances
4156 the location counter until it is a multiple of 8. If the location counter
4157 is already a multiple of 8, no change is needed.
4159 The second expression (also absolute) gives the fill value to be stored in the
4160 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4161 padding bytes are normally zero. However, on some systems, if the section is
4162 marked as containing code and the fill value is omitted, the space is filled
4163 with no-op instructions.
4165 The third expression is also absolute, and is also optional. If it is present,
4166 it is the maximum number of bytes that should be skipped by this alignment
4167 directive. If doing the alignment would require skipping more bytes than the
4168 specified maximum, then the alignment is not done at all. You can omit the
4169 fill value (the second argument) entirely by simply using two commas after the
4170 required alignment; this can be useful if you want the alignment to be filled
4171 with no-op instructions when appropriate.
4173 @cindex @code{balignw} directive
4174 @cindex @code{balignl} directive
4175 The @code{.balignw} and @code{.balignl} directives are variants of the
4176 @code{.balign} directive. The @code{.balignw} directive treats the fill
4177 pattern as a two byte word value. The @code{.balignl} directives treats the
4178 fill pattern as a four byte longword value. For example, @code{.balignw
4179 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4180 filled in with the value 0x368d (the exact placement of the bytes depends upon
4181 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4185 @section @code{.byte @var{expressions}}
4187 @cindex @code{byte} directive
4188 @cindex integers, one byte
4189 @code{.byte} expects zero or more expressions, separated by commas.
4190 Each expression is assembled into the next byte.
4192 @node CFI directives
4193 @section @code{.cfi_startproc [simple]}
4194 @cindex @code{cfi_startproc} directive
4195 @code{.cfi_startproc} is used at the beginning of each function that
4196 should have an entry in @code{.eh_frame}. It initializes some internal
4197 data structures. Don't forget to close the function by
4198 @code{.cfi_endproc}.
4200 @section @code{.cfi_sections @var{section_list}}
4201 @cindex @code{cfi_sections} directive
4202 @code{.cfi_sections} may be used to specify whether CFI directives
4203 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4204 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4205 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4206 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4207 directive is not used is @code{.cfi_sections .eh_frame}.
4209 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4210 it also emits some architecture dependent initial CFI instructions.
4212 @section @code{.cfi_endproc}
4213 @cindex @code{cfi_endproc} directive
4214 @code{.cfi_endproc} is used at the end of a function where it closes its
4215 unwind entry previously opened by
4216 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4218 @section @code{.cfi_personality @var{encoding} [, @var{exp}]}
4219 @code{.cfi_personality} defines personality routine and its encoding.
4220 @var{encoding} must be a constant determining how the personality
4221 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4222 argument is not present, otherwise second argument should be
4223 a constant or a symbol name. When using indirect encodings,
4224 the symbol provided should be the location where personality
4225 can be loaded from, not the personality routine itself.
4226 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4227 no personality routine.
4229 @section @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4230 @code{.cfi_lsda} defines LSDA and its encoding.
4231 @var{encoding} must be a constant determining how the LSDA
4232 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4233 argument is not present, otherwise second argument should be a constant
4234 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4237 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4238 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4239 address from @var{register} and add @var{offset} to it}.
4241 @section @code{.cfi_def_cfa_register @var{register}}
4242 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4243 now on @var{register} will be used instead of the old one. Offset
4246 @section @code{.cfi_def_cfa_offset @var{offset}}
4247 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4248 remains the same, but @var{offset} is new. Note that it is the
4249 absolute offset that will be added to a defined register to compute
4252 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4253 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4254 value that is added/substracted from the previous offset.
4256 @section @code{.cfi_offset @var{register}, @var{offset}}
4257 Previous value of @var{register} is saved at offset @var{offset} from
4260 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4261 Previous value of @var{register} is saved at offset @var{offset} from
4262 the current CFA register. This is transformed to @code{.cfi_offset}
4263 using the known displacement of the CFA register from the CFA.
4264 This is often easier to use, because the number will match the
4265 code it's annotating.
4267 @section @code{.cfi_register @var{register1}, @var{register2}}
4268 Previous value of @var{register1} is saved in register @var{register2}.
4270 @section @code{.cfi_restore @var{register}}
4271 @code{.cfi_restore} says that the rule for @var{register} is now the
4272 same as it was at the beginning of the function, after all initial
4273 instruction added by @code{.cfi_startproc} were executed.
4275 @section @code{.cfi_undefined @var{register}}
4276 From now on the previous value of @var{register} can't be restored anymore.
4278 @section @code{.cfi_same_value @var{register}}
4279 Current value of @var{register} is the same like in the previous frame,
4280 i.e. no restoration needed.
4282 @section @code{.cfi_remember_state},
4283 First save all current rules for all registers by @code{.cfi_remember_state},
4284 then totally screw them up by subsequent @code{.cfi_*} directives and when
4285 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4286 the previous saved state.
4288 @section @code{.cfi_return_column @var{register}}
4289 Change return column @var{register}, i.e. the return address is either
4290 directly in @var{register} or can be accessed by rules for @var{register}.
4292 @section @code{.cfi_signal_frame}
4293 Mark current function as signal trampoline.
4295 @section @code{.cfi_window_save}
4296 SPARC register window has been saved.
4298 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4299 Allows the user to add arbitrary bytes to the unwind info. One
4300 might use this to add OS-specific CFI opcodes, or generic CFI
4301 opcodes that GAS does not yet support.
4303 @section @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4304 The current value of @var{register} is @var{label}. The value of @var{label}
4305 will be encoded in the output file according to @var{encoding}; see the
4306 description of @code{.cfi_personality} for details on this encoding.
4308 The usefulness of equating a register to a fixed label is probably
4309 limited to the return address register. Here, it can be useful to
4310 mark a code segment that has only one return address which is reached
4311 by a direct branch and no copy of the return address exists in memory
4312 or another register.
4315 @section @code{.comm @var{symbol} , @var{length} }
4317 @cindex @code{comm} directive
4318 @cindex symbol, common
4319 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4320 common symbol in one object file may be merged with a defined or common symbol
4321 of the same name in another object file. If @code{@value{LD}} does not see a
4322 definition for the symbol--just one or more common symbols--then it will
4323 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4324 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4325 the same name, and they do not all have the same size, it will allocate space
4326 using the largest size.
4329 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4330 an optional third argument. This is the desired alignment of the symbol,
4331 specified for ELF as a byte boundary (for example, an alignment of 16 means
4332 that the least significant 4 bits of the address should be zero), and for PE
4333 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4334 boundary). The alignment must be an absolute expression, and it must be a
4335 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4336 common symbol, it will use the alignment when placing the symbol. If no
4337 alignment is specified, @command{@value{AS}} will set the alignment to the
4338 largest power of two less than or equal to the size of the symbol, up to a
4339 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4340 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4341 @samp{--section-alignment} option; image file sections in PE are aligned to
4342 multiples of 4096, which is far too large an alignment for ordinary variables.
4343 It is rather the default alignment for (non-debug) sections within object
4344 (@samp{*.o}) files, which are less strictly aligned.}.
4348 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4349 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4353 @section @code{.data @var{subsection}}
4355 @cindex @code{data} directive
4356 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4357 end of the data subsection numbered @var{subsection} (which is an
4358 absolute expression). If @var{subsection} is omitted, it defaults
4363 @section @code{.def @var{name}}
4365 @cindex @code{def} directive
4366 @cindex COFF symbols, debugging
4367 @cindex debugging COFF symbols
4368 Begin defining debugging information for a symbol @var{name}; the
4369 definition extends until the @code{.endef} directive is encountered.
4372 This directive is only observed when @command{@value{AS}} is configured for COFF
4373 format output; when producing @code{b.out}, @samp{.def} is recognized,
4380 @section @code{.desc @var{symbol}, @var{abs-expression}}
4382 @cindex @code{desc} directive
4383 @cindex COFF symbol descriptor
4384 @cindex symbol descriptor, COFF
4385 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4386 to the low 16 bits of an absolute expression.
4389 The @samp{.desc} directive is not available when @command{@value{AS}} is
4390 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4391 object format. For the sake of compatibility, @command{@value{AS}} accepts
4392 it, but produces no output, when configured for COFF.
4398 @section @code{.dim}
4400 @cindex @code{dim} directive
4401 @cindex COFF auxiliary symbol information
4402 @cindex auxiliary symbol information, COFF
4403 This directive is generated by compilers to include auxiliary debugging
4404 information in the symbol table. It is only permitted inside
4405 @code{.def}/@code{.endef} pairs.
4408 @samp{.dim} is only meaningful when generating COFF format output; when
4409 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4415 @section @code{.double @var{flonums}}
4417 @cindex @code{double} directive
4418 @cindex floating point numbers (double)
4419 @code{.double} expects zero or more flonums, separated by commas. It
4420 assembles floating point numbers.
4422 The exact kind of floating point numbers emitted depends on how
4423 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4427 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4428 in @sc{ieee} format.
4433 @section @code{.eject}
4435 @cindex @code{eject} directive
4436 @cindex new page, in listings
4437 @cindex page, in listings
4438 @cindex listing control: new page
4439 Force a page break at this point, when generating assembly listings.
4442 @section @code{.else}
4444 @cindex @code{else} directive
4445 @code{.else} is part of the @command{@value{AS}} support for conditional
4446 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4447 of code to be assembled if the condition for the preceding @code{.if}
4451 @section @code{.elseif}
4453 @cindex @code{elseif} directive
4454 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4455 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4456 @code{.if} block that would otherwise fill the entire @code{.else} section.
4459 @section @code{.end}
4461 @cindex @code{end} directive
4462 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4463 process anything in the file past the @code{.end} directive.
4467 @section @code{.endef}
4469 @cindex @code{endef} directive
4470 This directive flags the end of a symbol definition begun with
4474 @samp{.endef} is only meaningful when generating COFF format output; if
4475 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4476 directive but ignores it.
4481 @section @code{.endfunc}
4482 @cindex @code{endfunc} directive
4483 @code{.endfunc} marks the end of a function specified with @code{.func}.
4486 @section @code{.endif}
4488 @cindex @code{endif} directive
4489 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4490 it marks the end of a block of code that is only assembled
4491 conditionally. @xref{If,,@code{.if}}.
4494 @section @code{.equ @var{symbol}, @var{expression}}
4496 @cindex @code{equ} directive
4497 @cindex assigning values to symbols
4498 @cindex symbols, assigning values to
4499 This directive sets the value of @var{symbol} to @var{expression}.
4500 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4503 The syntax for @code{equ} on the HPPA is
4504 @samp{@var{symbol} .equ @var{expression}}.
4508 The syntax for @code{equ} on the Z80 is
4509 @samp{@var{symbol} equ @var{expression}}.
4510 On the Z80 it is an eror if @var{symbol} is already defined,
4511 but the symbol is not protected from later redefinition.
4512 Compare @ref{Equiv}.
4516 @section @code{.equiv @var{symbol}, @var{expression}}
4517 @cindex @code{equiv} directive
4518 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4519 the assembler will signal an error if @var{symbol} is already defined. Note a
4520 symbol which has been referenced but not actually defined is considered to be
4523 Except for the contents of the error message, this is roughly equivalent to
4530 plus it protects the symbol from later redefinition.
4533 @section @code{.eqv @var{symbol}, @var{expression}}
4534 @cindex @code{eqv} directive
4535 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
4536 evaluate the expression or any part of it immediately. Instead each time
4537 the resulting symbol is used in an expression, a snapshot of its current
4541 @section @code{.err}
4542 @cindex @code{err} directive
4543 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4544 message and, unless the @option{-Z} option was used, it will not generate an
4545 object file. This can be used to signal an error in conditionally compiled code.
4548 @section @code{.error "@var{string}"}
4549 @cindex error directive
4551 Similarly to @code{.err}, this directive emits an error, but you can specify a
4552 string that will be emitted as the error message. If you don't specify the
4553 message, it defaults to @code{".error directive invoked in source file"}.
4554 @xref{Errors, ,Error and Warning Messages}.
4557 .error "This code has not been assembled and tested."
4561 @section @code{.exitm}
4562 Exit early from the current macro definition. @xref{Macro}.
4565 @section @code{.extern}
4567 @cindex @code{extern} directive
4568 @code{.extern} is accepted in the source program---for compatibility
4569 with other assemblers---but it is ignored. @command{@value{AS}} treats
4570 all undefined symbols as external.
4573 @section @code{.fail @var{expression}}
4575 @cindex @code{fail} directive
4576 Generates an error or a warning. If the value of the @var{expression} is 500
4577 or more, @command{@value{AS}} will print a warning message. If the value is less
4578 than 500, @command{@value{AS}} will print an error message. The message will
4579 include the value of @var{expression}. This can occasionally be useful inside
4580 complex nested macros or conditional assembly.
4583 @section @code{.file}
4584 @cindex @code{file} directive
4586 @ifclear no-file-dir
4587 There are two different versions of the @code{.file} directive. Targets
4588 that support DWARF2 line number information use the DWARF2 version of
4589 @code{.file}. Other targets use the default version.
4591 @subheading Default Version
4593 @cindex logical file name
4594 @cindex file name, logical
4595 This version of the @code{.file} directive tells @command{@value{AS}} that we
4596 are about to start a new logical file. The syntax is:
4602 @var{string} is the new file name. In general, the filename is
4603 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4604 to specify an empty file name, you must give the quotes--@code{""}. This
4605 statement may go away in future: it is only recognized to be compatible with
4606 old @command{@value{AS}} programs.
4608 @subheading DWARF2 Version
4611 When emitting DWARF2 line number information, @code{.file} assigns filenames
4612 to the @code{.debug_line} file name table. The syntax is:
4615 .file @var{fileno} @var{filename}
4618 The @var{fileno} operand should be a unique positive integer to use as the
4619 index of the entry in the table. The @var{filename} operand is a C string
4622 The detail of filename indices is exposed to the user because the filename
4623 table is shared with the @code{.debug_info} section of the DWARF2 debugging
4624 information, and thus the user must know the exact indices that table
4628 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4630 @cindex @code{fill} directive
4631 @cindex writing patterns in memory
4632 @cindex patterns, writing in memory
4633 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4634 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4635 may be zero or more. @var{Size} may be zero or more, but if it is
4636 more than 8, then it is deemed to have the value 8, compatible with
4637 other people's assemblers. The contents of each @var{repeat} bytes
4638 is taken from an 8-byte number. The highest order 4 bytes are
4639 zero. The lowest order 4 bytes are @var{value} rendered in the
4640 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4641 Each @var{size} bytes in a repetition is taken from the lowest order
4642 @var{size} bytes of this number. Again, this bizarre behavior is
4643 compatible with other people's assemblers.
4645 @var{size} and @var{value} are optional.
4646 If the second comma and @var{value} are absent, @var{value} is
4647 assumed zero. If the first comma and following tokens are absent,
4648 @var{size} is assumed to be 1.
4651 @section @code{.float @var{flonums}}
4653 @cindex floating point numbers (single)
4654 @cindex @code{float} directive
4655 This directive assembles zero or more flonums, separated by commas. It
4656 has the same effect as @code{.single}.
4658 The exact kind of floating point numbers emitted depends on how
4659 @command{@value{AS}} is configured.
4660 @xref{Machine Dependencies}.
4664 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4665 in @sc{ieee} format.
4670 @section @code{.func @var{name}[,@var{label}]}
4671 @cindex @code{func} directive
4672 @code{.func} emits debugging information to denote function @var{name}, and
4673 is ignored unless the file is assembled with debugging enabled.
4674 Only @samp{--gstabs[+]} is currently supported.
4675 @var{label} is the entry point of the function and if omitted @var{name}
4676 prepended with the @samp{leading char} is used.
4677 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4678 All functions are currently defined to have @code{void} return type.
4679 The function must be terminated with @code{.endfunc}.
4682 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4684 @cindex @code{global} directive
4685 @cindex symbol, making visible to linker
4686 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4687 @var{symbol} in your partial program, its value is made available to
4688 other partial programs that are linked with it. Otherwise,
4689 @var{symbol} takes its attributes from a symbol of the same name
4690 from another file linked into the same program.
4692 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4693 compatibility with other assemblers.
4696 On the HPPA, @code{.global} is not always enough to make it accessible to other
4697 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4698 @xref{HPPA Directives, ,HPPA Assembler Directives}.
4703 @section @code{.gnu_attribute @var{tag},@var{value}}
4704 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
4707 @section @code{.hidden @var{names}}
4709 @cindex @code{hidden} directive
4711 This is one of the ELF visibility directives. The other two are
4712 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4713 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4715 This directive overrides the named symbols default visibility (which is set by
4716 their binding: local, global or weak). The directive sets the visibility to
4717 @code{hidden} which means that the symbols are not visible to other components.
4718 Such symbols are always considered to be @code{protected} as well.
4722 @section @code{.hword @var{expressions}}
4724 @cindex @code{hword} directive
4725 @cindex integers, 16-bit
4726 @cindex numbers, 16-bit
4727 @cindex sixteen bit integers
4728 This expects zero or more @var{expressions}, and emits
4729 a 16 bit number for each.
4732 This directive is a synonym for @samp{.short}; depending on the target
4733 architecture, it may also be a synonym for @samp{.word}.
4737 This directive is a synonym for @samp{.short}.
4740 This directive is a synonym for both @samp{.short} and @samp{.word}.
4745 @section @code{.ident}
4747 @cindex @code{ident} directive
4749 This directive is used by some assemblers to place tags in object files. The
4750 behavior of this directive varies depending on the target. When using the
4751 a.out object file format, @command{@value{AS}} simply accepts the directive for
4752 source-file compatibility with existing assemblers, but does not emit anything
4753 for it. When using COFF, comments are emitted to the @code{.comment} or
4754 @code{.rdata} section, depending on the target. When using ELF, comments are
4755 emitted to the @code{.comment} section.
4758 @section @code{.if @var{absolute expression}}
4760 @cindex conditional assembly
4761 @cindex @code{if} directive
4762 @code{.if} marks the beginning of a section of code which is only
4763 considered part of the source program being assembled if the argument
4764 (which must be an @var{absolute expression}) is non-zero. The end of
4765 the conditional section of code must be marked by @code{.endif}
4766 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4767 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4768 If you have several conditions to check, @code{.elseif} may be used to avoid
4769 nesting blocks if/else within each subsequent @code{.else} block.
4771 The following variants of @code{.if} are also supported:
4773 @cindex @code{ifdef} directive
4774 @item .ifdef @var{symbol}
4775 Assembles the following section of code if the specified @var{symbol}
4776 has been defined. Note a symbol which has been referenced but not yet defined
4777 is considered to be undefined.
4779 @cindex @code{ifb} directive
4780 @item .ifb @var{text}
4781 Assembles the following section of code if the operand is blank (empty).
4783 @cindex @code{ifc} directive
4784 @item .ifc @var{string1},@var{string2}
4785 Assembles the following section of code if the two strings are the same. The
4786 strings may be optionally quoted with single quotes. If they are not quoted,
4787 the first string stops at the first comma, and the second string stops at the
4788 end of the line. Strings which contain whitespace should be quoted. The
4789 string comparison is case sensitive.
4791 @cindex @code{ifeq} directive
4792 @item .ifeq @var{absolute expression}
4793 Assembles the following section of code if the argument is zero.
4795 @cindex @code{ifeqs} directive
4796 @item .ifeqs @var{string1},@var{string2}
4797 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4799 @cindex @code{ifge} directive
4800 @item .ifge @var{absolute expression}
4801 Assembles the following section of code if the argument is greater than or
4804 @cindex @code{ifgt} directive
4805 @item .ifgt @var{absolute expression}
4806 Assembles the following section of code if the argument is greater than zero.
4808 @cindex @code{ifle} directive
4809 @item .ifle @var{absolute expression}
4810 Assembles the following section of code if the argument is less than or equal
4813 @cindex @code{iflt} directive
4814 @item .iflt @var{absolute expression}
4815 Assembles the following section of code if the argument is less than zero.
4817 @cindex @code{ifnb} directive
4818 @item .ifnb @var{text}
4819 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
4820 following section of code if the operand is non-blank (non-empty).
4822 @cindex @code{ifnc} directive
4823 @item .ifnc @var{string1},@var{string2}.
4824 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4825 following section of code if the two strings are not the same.
4827 @cindex @code{ifndef} directive
4828 @cindex @code{ifnotdef} directive
4829 @item .ifndef @var{symbol}
4830 @itemx .ifnotdef @var{symbol}
4831 Assembles the following section of code if the specified @var{symbol}
4832 has not been defined. Both spelling variants are equivalent. Note a symbol
4833 which has been referenced but not yet defined is considered to be undefined.
4835 @cindex @code{ifne} directive
4836 @item .ifne @var{absolute expression}
4837 Assembles the following section of code if the argument is not equal to zero
4838 (in other words, this is equivalent to @code{.if}).
4840 @cindex @code{ifnes} directive
4841 @item .ifnes @var{string1},@var{string2}
4842 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4843 following section of code if the two strings are not the same.
4847 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4849 @cindex @code{incbin} directive
4850 @cindex binary files, including
4851 The @code{incbin} directive includes @var{file} verbatim at the current
4852 location. You can control the search paths used with the @samp{-I} command-line
4853 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4856 The @var{skip} argument skips a number of bytes from the start of the
4857 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4858 read. Note that the data is not aligned in any way, so it is the user's
4859 responsibility to make sure that proper alignment is provided both before and
4860 after the @code{incbin} directive.
4863 @section @code{.include "@var{file}"}
4865 @cindex @code{include} directive
4866 @cindex supporting files, including
4867 @cindex files, including
4868 This directive provides a way to include supporting files at specified
4869 points in your source program. The code from @var{file} is assembled as
4870 if it followed the point of the @code{.include}; when the end of the
4871 included file is reached, assembly of the original file continues. You
4872 can control the search paths used with the @samp{-I} command-line option
4873 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4877 @section @code{.int @var{expressions}}
4879 @cindex @code{int} directive
4880 @cindex integers, 32-bit
4881 Expect zero or more @var{expressions}, of any section, separated by commas.
4882 For each expression, emit a number that, at run time, is the value of that
4883 expression. The byte order and bit size of the number depends on what kind
4884 of target the assembly is for.
4888 On most forms of the H8/300, @code{.int} emits 16-bit
4889 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
4896 @section @code{.internal @var{names}}
4898 @cindex @code{internal} directive
4900 This is one of the ELF visibility directives. The other two are
4901 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
4902 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4904 This directive overrides the named symbols default visibility (which is set by
4905 their binding: local, global or weak). The directive sets the visibility to
4906 @code{internal} which means that the symbols are considered to be @code{hidden}
4907 (i.e., not visible to other components), and that some extra, processor specific
4908 processing must also be performed upon the symbols as well.
4912 @section @code{.irp @var{symbol},@var{values}}@dots{}
4914 @cindex @code{irp} directive
4915 Evaluate a sequence of statements assigning different values to @var{symbol}.
4916 The sequence of statements starts at the @code{.irp} directive, and is
4917 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
4918 set to @var{value}, and the sequence of statements is assembled. If no
4919 @var{value} is listed, the sequence of statements is assembled once, with
4920 @var{symbol} set to the null string. To refer to @var{symbol} within the
4921 sequence of statements, use @var{\symbol}.
4923 For example, assembling
4931 is equivalent to assembling
4939 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
4942 @section @code{.irpc @var{symbol},@var{values}}@dots{}
4944 @cindex @code{irpc} directive
4945 Evaluate a sequence of statements assigning different values to @var{symbol}.
4946 The sequence of statements starts at the @code{.irpc} directive, and is
4947 terminated by an @code{.endr} directive. For each character in @var{value},
4948 @var{symbol} is set to the character, and the sequence of statements is
4949 assembled. If no @var{value} is listed, the sequence of statements is
4950 assembled once, with @var{symbol} set to the null string. To refer to
4951 @var{symbol} within the sequence of statements, use @var{\symbol}.
4953 For example, assembling
4961 is equivalent to assembling
4969 For some caveats with the spelling of @var{symbol}, see also the discussion
4973 @section @code{.lcomm @var{symbol} , @var{length}}
4975 @cindex @code{lcomm} directive
4976 @cindex local common symbols
4977 @cindex symbols, local common
4978 Reserve @var{length} (an absolute expression) bytes for a local common
4979 denoted by @var{symbol}. The section and value of @var{symbol} are
4980 those of the new local common. The addresses are allocated in the bss
4981 section, so that at run-time the bytes start off zeroed. @var{Symbol}
4982 is not declared global (@pxref{Global,,@code{.global}}), so is normally
4983 not visible to @code{@value{LD}}.
4986 Some targets permit a third argument to be used with @code{.lcomm}. This
4987 argument specifies the desired alignment of the symbol in the bss section.
4991 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
4992 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
4996 @section @code{.lflags}
4998 @cindex @code{lflags} directive (ignored)
4999 @command{@value{AS}} accepts this directive, for compatibility with other
5000 assemblers, but ignores it.
5002 @ifclear no-line-dir
5004 @section @code{.line @var{line-number}}
5006 @cindex @code{line} directive
5007 @cindex logical line number
5009 Change the logical line number. @var{line-number} must be an absolute
5010 expression. The next line has that logical line number. Therefore any other
5011 statements on the current line (after a statement separator character) are
5012 reported as on logical line number @var{line-number} @minus{} 1. One day
5013 @command{@value{AS}} will no longer support this directive: it is recognized only
5014 for compatibility with existing assembler programs.
5017 Even though this is a directive associated with the @code{a.out} or
5018 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5019 when producing COFF output, and treats @samp{.line} as though it
5020 were the COFF @samp{.ln} @emph{if} it is found outside a
5021 @code{.def}/@code{.endef} pair.
5023 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5024 used by compilers to generate auxiliary symbol information for
5029 @section @code{.linkonce [@var{type}]}
5031 @cindex @code{linkonce} directive
5032 @cindex common sections
5033 Mark the current section so that the linker only includes a single copy of it.
5034 This may be used to include the same section in several different object files,
5035 but ensure that the linker will only include it once in the final output file.
5036 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5037 Duplicate sections are detected based on the section name, so it should be
5040 This directive is only supported by a few object file formats; as of this
5041 writing, the only object file format which supports it is the Portable
5042 Executable format used on Windows NT.
5044 The @var{type} argument is optional. If specified, it must be one of the
5045 following strings. For example:
5049 Not all types may be supported on all object file formats.
5053 Silently discard duplicate sections. This is the default.
5056 Warn if there are duplicate sections, but still keep only one copy.
5059 Warn if any of the duplicates have different sizes.
5062 Warn if any of the duplicates do not have exactly the same contents.
5066 @section @code{.list}
5068 @cindex @code{list} directive
5069 @cindex listing control, turning on
5070 Control (in conjunction with the @code{.nolist} directive) whether or
5071 not assembly listings are generated. These two directives maintain an
5072 internal counter (which is zero initially). @code{.list} increments the
5073 counter, and @code{.nolist} decrements it. Assembly listings are
5074 generated whenever the counter is greater than zero.
5076 By default, listings are disabled. When you enable them (with the
5077 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5078 the initial value of the listing counter is one.
5081 @section @code{.ln @var{line-number}}
5083 @cindex @code{ln} directive
5084 @ifclear no-line-dir
5085 @samp{.ln} is a synonym for @samp{.line}.
5088 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5089 must be an absolute expression. The next line has that logical
5090 line number, so any other statements on the current line (after a
5091 statement separator character @code{;}) are reported as on logical
5092 line number @var{line-number} @minus{} 1.
5095 This directive is accepted, but ignored, when @command{@value{AS}} is
5096 configured for @code{b.out}; its effect is only associated with COFF
5102 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5103 @cindex @code{loc} directive
5104 When emitting DWARF2 line number information,
5105 the @code{.loc} directive will add a row to the @code{.debug_line} line
5106 number matrix corresponding to the immediately following assembly
5107 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5108 arguments will be applied to the @code{.debug_line} state machine before
5111 The @var{options} are a sequence of the following tokens in any order:
5115 This option will set the @code{basic_block} register in the
5116 @code{.debug_line} state machine to @code{true}.
5119 This option will set the @code{prologue_end} register in the
5120 @code{.debug_line} state machine to @code{true}.
5122 @item epilogue_begin
5123 This option will set the @code{epilogue_begin} register in the
5124 @code{.debug_line} state machine to @code{true}.
5126 @item is_stmt @var{value}
5127 This option will set the @code{is_stmt} register in the
5128 @code{.debug_line} state machine to @code{value}, which must be
5131 @item isa @var{value}
5132 This directive will set the @code{isa} register in the @code{.debug_line}
5133 state machine to @var{value}, which must be an unsigned integer.
5135 @item discriminator @var{value}
5136 This directive will set the @code{discriminator} register in the @code{.debug_line}
5137 state machine to @var{value}, which must be an unsigned integer.
5141 @node Loc_mark_labels
5142 @section @code{.loc_mark_labels @var{enable}}
5143 @cindex @code{loc_mark_labels} directive
5144 When emitting DWARF2 line number information,
5145 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5146 to the @code{.debug_line} line number matrix with the @code{basic_block}
5147 register in the state machine set whenever a code label is seen.
5148 The @var{enable} argument should be either 1 or 0, to enable or disable
5149 this function respectively.
5153 @section @code{.local @var{names}}
5155 @cindex @code{local} directive
5156 This directive, which is available for ELF targets, marks each symbol in
5157 the comma-separated list of @code{names} as a local symbol so that it
5158 will not be externally visible. If the symbols do not already exist,
5159 they will be created.
5161 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5162 accept an alignment argument, which is the case for most ELF targets,
5163 the @code{.local} directive can be used in combination with @code{.comm}
5164 (@pxref{Comm}) to define aligned local common data.
5168 @section @code{.long @var{expressions}}
5170 @cindex @code{long} directive
5171 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5174 @c no one seems to know what this is for or whether this description is
5175 @c what it really ought to do
5177 @section @code{.lsym @var{symbol}, @var{expression}}
5179 @cindex @code{lsym} directive
5180 @cindex symbol, not referenced in assembly
5181 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5182 the hash table, ensuring it cannot be referenced by name during the
5183 rest of the assembly. This sets the attributes of the symbol to be
5184 the same as the expression value:
5186 @var{other} = @var{descriptor} = 0
5187 @var{type} = @r{(section of @var{expression})}
5188 @var{value} = @var{expression}
5191 The new symbol is not flagged as external.
5195 @section @code{.macro}
5198 The commands @code{.macro} and @code{.endm} allow you to define macros that
5199 generate assembly output. For example, this definition specifies a macro
5200 @code{sum} that puts a sequence of numbers into memory:
5203 .macro sum from=0, to=5
5212 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5224 @item .macro @var{macname}
5225 @itemx .macro @var{macname} @var{macargs} @dots{}
5226 @cindex @code{macro} directive
5227 Begin the definition of a macro called @var{macname}. If your macro
5228 definition requires arguments, specify their names after the macro name,
5229 separated by commas or spaces. You can qualify the macro argument to
5230 indicate whether all invocations must specify a non-blank value (through
5231 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5232 (through @samp{:@code{vararg}}). You can supply a default value for any
5233 macro argument by following the name with @samp{=@var{deflt}}. You
5234 cannot define two macros with the same @var{macname} unless it has been
5235 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5236 definitions. For example, these are all valid @code{.macro} statements:
5240 Begin the definition of a macro called @code{comm}, which takes no
5243 @item .macro plus1 p, p1
5244 @itemx .macro plus1 p p1
5245 Either statement begins the definition of a macro called @code{plus1},
5246 which takes two arguments; within the macro definition, write
5247 @samp{\p} or @samp{\p1} to evaluate the arguments.
5249 @item .macro reserve_str p1=0 p2
5250 Begin the definition of a macro called @code{reserve_str}, with two
5251 arguments. The first argument has a default value, but not the second.
5252 After the definition is complete, you can call the macro either as
5253 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5254 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5255 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5256 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5258 @item .macro m p1:req, p2=0, p3:vararg
5259 Begin the definition of a macro called @code{m}, with at least three
5260 arguments. The first argument must always have a value specified, but
5261 not the second, which instead has a default value. The third formal
5262 will get assigned all remaining arguments specified at invocation time.
5264 When you call a macro, you can specify the argument values either by
5265 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5266 @samp{sum to=17, from=9}.
5270 Note that since each of the @var{macargs} can be an identifier exactly
5271 as any other one permitted by the target architecture, there may be
5272 occasional problems if the target hand-crafts special meanings to certain
5273 characters when they occur in a special position. For example, if the colon
5274 (@code{:}) is generally permitted to be part of a symbol name, but the
5275 architecture specific code special-cases it when occurring as the final
5276 character of a symbol (to denote a label), then the macro parameter
5277 replacement code will have no way of knowing that and consider the whole
5278 construct (including the colon) an identifier, and check only this
5279 identifier for being the subject to parameter substitution. So for example
5280 this macro definition:
5288 might not work as expected. Invoking @samp{label foo} might not create a label
5289 called @samp{foo} but instead just insert the text @samp{\l:} into the
5290 assembler source, probably generating an error about an unrecognised
5293 Similarly problems might occur with the period character (@samp{.})
5294 which is often allowed inside opcode names (and hence identifier names). So
5295 for example constructing a macro to build an opcode from a base name and a
5296 length specifier like this:
5299 .macro opcode base length
5304 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5305 instruction but instead generate some kind of error as the assembler tries to
5306 interpret the text @samp{\base.\length}.
5308 There are several possible ways around this problem:
5311 @item Insert white space
5312 If it is possible to use white space characters then this is the simplest
5321 @item Use @samp{\()}
5322 The string @samp{\()} can be used to separate the end of a macro argument from
5323 the following text. eg:
5326 .macro opcode base length
5331 @item Use the alternate macro syntax mode
5332 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5333 used as a separator. eg:
5343 Note: this problem of correctly identifying string parameters to pseudo ops
5344 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5345 and @code{.irpc} (@pxref{Irpc}) as well.
5348 @cindex @code{endm} directive
5349 Mark the end of a macro definition.
5352 @cindex @code{exitm} directive
5353 Exit early from the current macro definition.
5355 @cindex number of macros executed
5356 @cindex macros, count executed
5358 @command{@value{AS}} maintains a counter of how many macros it has
5359 executed in this pseudo-variable; you can copy that number to your
5360 output with @samp{\@@}, but @emph{only within a macro definition}.
5362 @item LOCAL @var{name} [ , @dots{} ]
5363 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5364 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5365 @xref{Altmacro,,@code{.altmacro}}.
5369 @section @code{.mri @var{val}}
5371 @cindex @code{mri} directive
5372 @cindex MRI mode, temporarily
5373 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5374 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5375 affects code assembled until the next @code{.mri} directive, or until the end
5376 of the file. @xref{M, MRI mode, MRI mode}.
5379 @section @code{.noaltmacro}
5380 Disable alternate macro mode. @xref{Altmacro}.
5383 @section @code{.nolist}
5385 @cindex @code{nolist} directive
5386 @cindex listing control, turning off
5387 Control (in conjunction with the @code{.list} directive) whether or
5388 not assembly listings are generated. These two directives maintain an
5389 internal counter (which is zero initially). @code{.list} increments the
5390 counter, and @code{.nolist} decrements it. Assembly listings are
5391 generated whenever the counter is greater than zero.
5394 @section @code{.octa @var{bignums}}
5396 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5397 @cindex @code{octa} directive
5398 @cindex integer, 16-byte
5399 @cindex sixteen byte integer
5400 This directive expects zero or more bignums, separated by commas. For each
5401 bignum, it emits a 16-byte integer.
5403 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5404 hence @emph{octa}-word for 16 bytes.
5407 @section @code{.org @var{new-lc} , @var{fill}}
5409 @cindex @code{org} directive
5410 @cindex location counter, advancing
5411 @cindex advancing location counter
5412 @cindex current address, advancing
5413 Advance the location counter of the current section to
5414 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5415 expression with the same section as the current subsection. That is,
5416 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5417 wrong section, the @code{.org} directive is ignored. To be compatible
5418 with former assemblers, if the section of @var{new-lc} is absolute,
5419 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5420 is the same as the current subsection.
5422 @code{.org} may only increase the location counter, or leave it
5423 unchanged; you cannot use @code{.org} to move the location counter
5426 @c double negative used below "not undefined" because this is a specific
5427 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5428 @c section. doc@cygnus.com 18feb91
5429 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5430 may not be undefined. If you really detest this restriction we eagerly await
5431 a chance to share your improved assembler.
5433 Beware that the origin is relative to the start of the section, not
5434 to the start of the subsection. This is compatible with other
5435 people's assemblers.
5437 When the location counter (of the current subsection) is advanced, the
5438 intervening bytes are filled with @var{fill} which should be an
5439 absolute expression. If the comma and @var{fill} are omitted,
5440 @var{fill} defaults to zero.
5443 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5445 @cindex padding the location counter given a power of two
5446 @cindex @code{p2align} directive
5447 Pad the location counter (in the current subsection) to a particular
5448 storage boundary. The first expression (which must be absolute) is the
5449 number of low-order zero bits the location counter must have after
5450 advancement. For example @samp{.p2align 3} advances the location
5451 counter until it a multiple of 8. If the location counter is already a
5452 multiple of 8, no change is needed.
5454 The second expression (also absolute) gives the fill value to be stored in the
5455 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5456 padding bytes are normally zero. However, on some systems, if the section is
5457 marked as containing code and the fill value is omitted, the space is filled
5458 with no-op instructions.
5460 The third expression is also absolute, and is also optional. If it is present,
5461 it is the maximum number of bytes that should be skipped by this alignment
5462 directive. If doing the alignment would require skipping more bytes than the
5463 specified maximum, then the alignment is not done at all. You can omit the
5464 fill value (the second argument) entirely by simply using two commas after the
5465 required alignment; this can be useful if you want the alignment to be filled
5466 with no-op instructions when appropriate.
5468 @cindex @code{p2alignw} directive
5469 @cindex @code{p2alignl} directive
5470 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5471 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5472 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5473 fill pattern as a four byte longword value. For example, @code{.p2alignw
5474 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5475 filled in with the value 0x368d (the exact placement of the bytes depends upon
5476 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5481 @section @code{.popsection}
5483 @cindex @code{popsection} directive
5484 @cindex Section Stack
5485 This is one of the ELF section stack manipulation directives. The others are
5486 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5487 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5490 This directive replaces the current section (and subsection) with the top
5491 section (and subsection) on the section stack. This section is popped off the
5497 @section @code{.previous}
5499 @cindex @code{previous} directive
5500 @cindex Section Stack
5501 This is one of the ELF section stack manipulation directives. The others are
5502 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5503 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5504 (@pxref{PopSection}).
5506 This directive swaps the current section (and subsection) with most recently
5507 referenced section/subsection pair prior to this one. Multiple
5508 @code{.previous} directives in a row will flip between two sections (and their
5509 subsections). For example:
5521 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
5527 # Now in section A subsection 1
5531 # Now in section B subsection 0
5534 # Now in section B subsection 1
5537 # Now in section B subsection 0
5541 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
5542 section B and 0x9abc into subsection 1 of section B.
5544 In terms of the section stack, this directive swaps the current section with
5545 the top section on the section stack.
5549 @section @code{.print @var{string}}
5551 @cindex @code{print} directive
5552 @command{@value{AS}} will print @var{string} on the standard output during
5553 assembly. You must put @var{string} in double quotes.
5557 @section @code{.protected @var{names}}
5559 @cindex @code{protected} directive
5561 This is one of the ELF visibility directives. The other two are
5562 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5564 This directive overrides the named symbols default visibility (which is set by
5565 their binding: local, global or weak). The directive sets the visibility to
5566 @code{protected} which means that any references to the symbols from within the
5567 components that defines them must be resolved to the definition in that
5568 component, even if a definition in another component would normally preempt
5573 @section @code{.psize @var{lines} , @var{columns}}
5575 @cindex @code{psize} directive
5576 @cindex listing control: paper size
5577 @cindex paper size, for listings
5578 Use this directive to declare the number of lines---and, optionally, the
5579 number of columns---to use for each page, when generating listings.
5581 If you do not use @code{.psize}, listings use a default line-count
5582 of 60. You may omit the comma and @var{columns} specification; the
5583 default width is 200 columns.
5585 @command{@value{AS}} generates formfeeds whenever the specified number of
5586 lines is exceeded (or whenever you explicitly request one, using
5589 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5590 those explicitly specified with @code{.eject}.
5593 @section @code{.purgem @var{name}}
5595 @cindex @code{purgem} directive
5596 Undefine the macro @var{name}, so that later uses of the string will not be
5597 expanded. @xref{Macro}.
5601 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
5603 @cindex @code{pushsection} directive
5604 @cindex Section Stack
5605 This is one of the ELF section stack manipulation directives. The others are
5606 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5607 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5610 This directive pushes the current section (and subsection) onto the
5611 top of the section stack, and then replaces the current section and
5612 subsection with @code{name} and @code{subsection}. The optional
5613 @code{flags}, @code{type} and @code{arguments} are treated the same
5614 as in the @code{.section} (@pxref{Section}) directive.
5618 @section @code{.quad @var{bignums}}
5620 @cindex @code{quad} directive
5621 @code{.quad} expects zero or more bignums, separated by commas. For
5622 each bignum, it emits
5624 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5625 warning message; and just takes the lowest order 8 bytes of the bignum.
5626 @cindex eight-byte integer
5627 @cindex integer, 8-byte
5629 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5630 hence @emph{quad}-word for 8 bytes.
5633 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5634 warning message; and just takes the lowest order 16 bytes of the bignum.
5635 @cindex sixteen-byte integer
5636 @cindex integer, 16-byte
5640 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
5642 @cindex @code{reloc} directive
5643 Generate a relocation at @var{offset} of type @var{reloc_name} with value
5644 @var{expression}. If @var{offset} is a number, the relocation is generated in
5645 the current section. If @var{offset} is an expression that resolves to a
5646 symbol plus offset, the relocation is generated in the given symbol's section.
5647 @var{expression}, if present, must resolve to a symbol plus addend or to an
5648 absolute value, but note that not all targets support an addend. e.g. ELF REL
5649 targets such as i386 store an addend in the section contents rather than in the
5650 relocation. This low level interface does not support addends stored in the
5654 @section @code{.rept @var{count}}
5656 @cindex @code{rept} directive
5657 Repeat the sequence of lines between the @code{.rept} directive and the next
5658 @code{.endr} directive @var{count} times.
5660 For example, assembling
5668 is equivalent to assembling
5677 @section @code{.sbttl "@var{subheading}"}
5679 @cindex @code{sbttl} directive
5680 @cindex subtitles for listings
5681 @cindex listing control: subtitle
5682 Use @var{subheading} as the title (third line, immediately after the
5683 title line) when generating assembly listings.
5685 This directive affects subsequent pages, as well as the current page if
5686 it appears within ten lines of the top of a page.
5690 @section @code{.scl @var{class}}
5692 @cindex @code{scl} directive
5693 @cindex symbol storage class (COFF)
5694 @cindex COFF symbol storage class
5695 Set the storage-class value for a symbol. This directive may only be
5696 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5697 whether a symbol is static or external, or it may record further
5698 symbolic debugging information.
5701 The @samp{.scl} directive is primarily associated with COFF output; when
5702 configured to generate @code{b.out} output format, @command{@value{AS}}
5703 accepts this directive but ignores it.
5709 @section @code{.section @var{name}}
5711 @cindex named section
5712 Use the @code{.section} directive to assemble the following code into a section
5715 This directive is only supported for targets that actually support arbitrarily
5716 named sections; on @code{a.out} targets, for example, it is not accepted, even
5717 with a standard @code{a.out} section name.
5721 @c only print the extra heading if both COFF and ELF are set
5722 @subheading COFF Version
5725 @cindex @code{section} directive (COFF version)
5726 For COFF targets, the @code{.section} directive is used in one of the following
5730 .section @var{name}[, "@var{flags}"]
5731 .section @var{name}[, @var{subsection}]
5734 If the optional argument is quoted, it is taken as flags to use for the
5735 section. Each flag is a single character. The following flags are recognized:
5738 bss section (uninitialized data)
5740 section is not loaded
5750 shared section (meaningful for PE targets)
5752 ignored. (For compatibility with the ELF version)
5754 section is not readable (meaningful for PE targets)
5756 single-digit power-of-two section alignment (GNU extension)
5759 If no flags are specified, the default flags depend upon the section name. If
5760 the section name is not recognized, the default will be for the section to be
5761 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
5762 from the section, rather than adding them, so if they are used on their own it
5763 will be as if no flags had been specified at all.
5765 If the optional argument to the @code{.section} directive is not quoted, it is
5766 taken as a subsection number (@pxref{Sub-Sections}).
5771 @c only print the extra heading if both COFF and ELF are set
5772 @subheading ELF Version
5775 @cindex Section Stack
5776 This is one of the ELF section stack manipulation directives. The others are
5777 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
5778 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
5779 @code{.previous} (@pxref{Previous}).
5781 @cindex @code{section} directive (ELF version)
5782 For ELF targets, the @code{.section} directive is used like this:
5785 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
5788 The optional @var{flags} argument is a quoted string which may contain any
5789 combination of the following characters:
5792 section is allocatable
5796 section is executable
5798 section is mergeable
5800 section contains zero terminated strings
5802 section is a member of a section group
5804 section is used for thread-local-storage
5807 The optional @var{type} argument may contain one of the following constants:
5810 section contains data
5812 section does not contain data (i.e., section only occupies space)
5814 section contains data which is used by things other than the program
5816 section contains an array of pointers to init functions
5818 section contains an array of pointers to finish functions
5819 @item @@preinit_array
5820 section contains an array of pointers to pre-init functions
5823 Many targets only support the first three section types.
5825 Note on targets where the @code{@@} character is the start of a comment (eg
5826 ARM) then another character is used instead. For example the ARM port uses the
5829 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
5830 be specified as well as an extra argument---@var{entsize}---like this:
5833 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
5836 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
5837 constants, each @var{entsize} octets long. Sections with both @code{M} and
5838 @code{S} must contain zero terminated strings where each character is
5839 @var{entsize} bytes long. The linker may remove duplicates within sections with
5840 the same name, same entity size and same flags. @var{entsize} must be an
5841 absolute expression. For sections with both @code{M} and @code{S}, a string
5842 which is a suffix of a larger string is considered a duplicate. Thus
5843 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
5844 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
5846 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
5847 be present along with an additional field like this:
5850 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
5853 The @var{GroupName} field specifies the name of the section group to which this
5854 particular section belongs. The optional linkage field can contain:
5857 indicates that only one copy of this section should be retained
5862 Note: if both the @var{M} and @var{G} flags are present then the fields for
5863 the Merge flag should come first, like this:
5866 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
5869 If no flags are specified, the default flags depend upon the section name. If
5870 the section name is not recognized, the default will be for the section to have
5871 none of the above flags: it will not be allocated in memory, nor writable, nor
5872 executable. The section will contain data.
5874 For ELF targets, the assembler supports another type of @code{.section}
5875 directive for compatibility with the Solaris assembler:
5878 .section "@var{name}"[, @var{flags}...]
5881 Note that the section name is quoted. There may be a sequence of comma
5885 section is allocatable
5889 section is executable
5891 section is used for thread local storage
5894 This directive replaces the current section and subsection. See the
5895 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
5896 some examples of how this directive and the other section stack directives
5902 @section @code{.set @var{symbol}, @var{expression}}
5904 @cindex @code{set} directive
5905 @cindex symbol value, setting
5906 Set the value of @var{symbol} to @var{expression}. This
5907 changes @var{symbol}'s value and type to conform to
5908 @var{expression}. If @var{symbol} was flagged as external, it remains
5909 flagged (@pxref{Symbol Attributes}).
5911 You may @code{.set} a symbol many times in the same assembly.
5913 If you @code{.set} a global symbol, the value stored in the object
5914 file is the last value stored into it.
5917 The syntax for @code{set} on the HPPA is
5918 @samp{@var{symbol} .set @var{expression}}.
5922 On Z80 @code{set} is a real instruction, use
5923 @samp{@var{symbol} defl @var{expression}} instead.
5927 @section @code{.short @var{expressions}}
5929 @cindex @code{short} directive
5931 @code{.short} is normally the same as @samp{.word}.
5932 @xref{Word,,@code{.word}}.
5934 In some configurations, however, @code{.short} and @code{.word} generate
5935 numbers of different lengths. @xref{Machine Dependencies}.
5939 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
5942 This expects zero or more @var{expressions}, and emits
5943 a 16 bit number for each.
5948 @section @code{.single @var{flonums}}
5950 @cindex @code{single} directive
5951 @cindex floating point numbers (single)
5952 This directive assembles zero or more flonums, separated by commas. It
5953 has the same effect as @code{.float}.
5955 The exact kind of floating point numbers emitted depends on how
5956 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5960 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
5961 numbers in @sc{ieee} format.
5967 @section @code{.size}
5969 This directive is used to set the size associated with a symbol.
5973 @c only print the extra heading if both COFF and ELF are set
5974 @subheading COFF Version
5977 @cindex @code{size} directive (COFF version)
5978 For COFF targets, the @code{.size} directive is only permitted inside
5979 @code{.def}/@code{.endef} pairs. It is used like this:
5982 .size @var{expression}
5986 @samp{.size} is only meaningful when generating COFF format output; when
5987 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5994 @c only print the extra heading if both COFF and ELF are set
5995 @subheading ELF Version
5998 @cindex @code{size} directive (ELF version)
5999 For ELF targets, the @code{.size} directive is used like this:
6002 .size @var{name} , @var{expression}
6005 This directive sets the size associated with a symbol @var{name}.
6006 The size in bytes is computed from @var{expression} which can make use of label
6007 arithmetic. This directive is typically used to set the size of function
6012 @ifclear no-space-dir
6014 @section @code{.skip @var{size} , @var{fill}}
6016 @cindex @code{skip} directive
6017 @cindex filling memory
6018 This directive emits @var{size} bytes, each of value @var{fill}. Both
6019 @var{size} and @var{fill} are absolute expressions. If the comma and
6020 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6025 @section @code{.sleb128 @var{expressions}}
6027 @cindex @code{sleb128} directive
6028 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6029 compact, variable length representation of numbers used by the DWARF
6030 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6032 @ifclear no-space-dir
6034 @section @code{.space @var{size} , @var{fill}}
6036 @cindex @code{space} directive
6037 @cindex filling memory
6038 This directive emits @var{size} bytes, each of value @var{fill}. Both
6039 @var{size} and @var{fill} are absolute expressions. If the comma
6040 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6045 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6046 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6047 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6048 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6056 @section @code{.stabd, .stabn, .stabs}
6058 @cindex symbolic debuggers, information for
6059 @cindex @code{stab@var{x}} directives
6060 There are three directives that begin @samp{.stab}.
6061 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6062 The symbols are not entered in the @command{@value{AS}} hash table: they
6063 cannot be referenced elsewhere in the source file.
6064 Up to five fields are required:
6068 This is the symbol's name. It may contain any character except
6069 @samp{\000}, so is more general than ordinary symbol names. Some
6070 debuggers used to code arbitrarily complex structures into symbol names
6074 An absolute expression. The symbol's type is set to the low 8 bits of
6075 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6076 and debuggers choke on silly bit patterns.
6079 An absolute expression. The symbol's ``other'' attribute is set to the
6080 low 8 bits of this expression.
6083 An absolute expression. The symbol's descriptor is set to the low 16
6084 bits of this expression.
6087 An absolute expression which becomes the symbol's value.
6090 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6091 or @code{.stabs} statement, the symbol has probably already been created;
6092 you get a half-formed symbol in your object file. This is
6093 compatible with earlier assemblers!
6096 @cindex @code{stabd} directive
6097 @item .stabd @var{type} , @var{other} , @var{desc}
6099 The ``name'' of the symbol generated is not even an empty string.
6100 It is a null pointer, for compatibility. Older assemblers used a
6101 null pointer so they didn't waste space in object files with empty
6104 The symbol's value is set to the location counter,
6105 relocatably. When your program is linked, the value of this symbol
6106 is the address of the location counter when the @code{.stabd} was
6109 @cindex @code{stabn} directive
6110 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6111 The name of the symbol is set to the empty string @code{""}.
6113 @cindex @code{stabs} directive
6114 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6115 All five fields are specified.
6121 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6122 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6124 @cindex string, copying to object file
6125 @cindex string8, copying to object file
6126 @cindex string16, copying to object file
6127 @cindex string32, copying to object file
6128 @cindex string64, copying to object file
6129 @cindex @code{string} directive
6130 @cindex @code{string8} directive
6131 @cindex @code{string16} directive
6132 @cindex @code{string32} directive
6133 @cindex @code{string64} directive
6135 Copy the characters in @var{str} to the object file. You may specify more than
6136 one string to copy, separated by commas. Unless otherwise specified for a
6137 particular machine, the assembler marks the end of each string with a 0 byte.
6138 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6140 The variants @code{string16}, @code{string32} and @code{string64} differ from
6141 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6142 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6143 are stored in target endianness byte order.
6149 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6150 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6155 @section @code{.struct @var{expression}}
6157 @cindex @code{struct} directive
6158 Switch to the absolute section, and set the section offset to @var{expression},
6159 which must be an absolute expression. You might use this as follows:
6168 This would define the symbol @code{field1} to have the value 0, the symbol
6169 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6170 value 8. Assembly would be left in the absolute section, and you would need to
6171 use a @code{.section} directive of some sort to change to some other section
6172 before further assembly.
6176 @section @code{.subsection @var{name}}
6178 @cindex @code{subsection} directive
6179 @cindex Section Stack
6180 This is one of the ELF section stack manipulation directives. The others are
6181 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6182 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6185 This directive replaces the current subsection with @code{name}. The current
6186 section is not changed. The replaced subsection is put onto the section stack
6187 in place of the then current top of stack subsection.
6192 @section @code{.symver}
6193 @cindex @code{symver} directive
6194 @cindex symbol versioning
6195 @cindex versions of symbols
6196 Use the @code{.symver} directive to bind symbols to specific version nodes
6197 within a source file. This is only supported on ELF platforms, and is
6198 typically used when assembling files to be linked into a shared library.
6199 There are cases where it may make sense to use this in objects to be bound
6200 into an application itself so as to override a versioned symbol from a
6203 For ELF targets, the @code{.symver} directive can be used like this:
6205 .symver @var{name}, @var{name2@@nodename}
6207 If the symbol @var{name} is defined within the file
6208 being assembled, the @code{.symver} directive effectively creates a symbol
6209 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6210 just don't try and create a regular alias is that the @var{@@} character isn't
6211 permitted in symbol names. The @var{name2} part of the name is the actual name
6212 of the symbol by which it will be externally referenced. The name @var{name}
6213 itself is merely a name of convenience that is used so that it is possible to
6214 have definitions for multiple versions of a function within a single source
6215 file, and so that the compiler can unambiguously know which version of a
6216 function is being mentioned. The @var{nodename} portion of the alias should be
6217 the name of a node specified in the version script supplied to the linker when
6218 building a shared library. If you are attempting to override a versioned
6219 symbol from a shared library, then @var{nodename} should correspond to the
6220 nodename of the symbol you are trying to override.
6222 If the symbol @var{name} is not defined within the file being assembled, all
6223 references to @var{name} will be changed to @var{name2@@nodename}. If no
6224 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6227 Another usage of the @code{.symver} directive is:
6229 .symver @var{name}, @var{name2@@@@nodename}
6231 In this case, the symbol @var{name} must exist and be defined within
6232 the file being assembled. It is similar to @var{name2@@nodename}. The
6233 difference is @var{name2@@@@nodename} will also be used to resolve
6234 references to @var{name2} by the linker.
6236 The third usage of the @code{.symver} directive is:
6238 .symver @var{name}, @var{name2@@@@@@nodename}
6240 When @var{name} is not defined within the
6241 file being assembled, it is treated as @var{name2@@nodename}. When
6242 @var{name} is defined within the file being assembled, the symbol
6243 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6248 @section @code{.tag @var{structname}}
6250 @cindex COFF structure debugging
6251 @cindex structure debugging, COFF
6252 @cindex @code{tag} directive
6253 This directive is generated by compilers to include auxiliary debugging
6254 information in the symbol table. It is only permitted inside
6255 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6256 definitions in the symbol table with instances of those structures.
6259 @samp{.tag} is only used when generating COFF format output; when
6260 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6266 @section @code{.text @var{subsection}}
6268 @cindex @code{text} directive
6269 Tells @command{@value{AS}} to assemble the following statements onto the end of
6270 the text subsection numbered @var{subsection}, which is an absolute
6271 expression. If @var{subsection} is omitted, subsection number zero
6275 @section @code{.title "@var{heading}"}
6277 @cindex @code{title} directive
6278 @cindex listing control: title line
6279 Use @var{heading} as the title (second line, immediately after the
6280 source file name and pagenumber) when generating assembly listings.
6282 This directive affects subsequent pages, as well as the current page if
6283 it appears within ten lines of the top of a page.
6287 @section @code{.type}
6289 This directive is used to set the type of a symbol.
6293 @c only print the extra heading if both COFF and ELF are set
6294 @subheading COFF Version
6297 @cindex COFF symbol type
6298 @cindex symbol type, COFF
6299 @cindex @code{type} directive (COFF version)
6300 For COFF targets, this directive is permitted only within
6301 @code{.def}/@code{.endef} pairs. It is used like this:
6307 This records the integer @var{int} as the type attribute of a symbol table
6311 @samp{.type} is associated only with COFF format output; when
6312 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6313 directive but ignores it.
6319 @c only print the extra heading if both COFF and ELF are set
6320 @subheading ELF Version
6323 @cindex ELF symbol type
6324 @cindex symbol type, ELF
6325 @cindex @code{type} directive (ELF version)
6326 For ELF targets, the @code{.type} directive is used like this:
6329 .type @var{name} , @var{type description}
6332 This sets the type of symbol @var{name} to be either a
6333 function symbol or an object symbol. There are five different syntaxes
6334 supported for the @var{type description} field, in order to provide
6335 compatibility with various other assemblers.
6337 Because some of the characters used in these syntaxes (such as @samp{@@} and
6338 @samp{#}) are comment characters for some architectures, some of the syntaxes
6339 below do not work on all architectures. The first variant will be accepted by
6340 the GNU assembler on all architectures so that variant should be used for
6341 maximum portability, if you do not need to assemble your code with other
6344 The syntaxes supported are:
6347 .type <name> STT_<TYPE_IN_UPPER_CASE>
6348 .type <name>,#<type>
6349 .type <name>,@@<type>
6350 .type <name>,%<type>
6351 .type <name>,"<type>"
6354 The types supported are:
6359 Mark the symbol as being a function name.
6362 @itemx gnu_indirect_function
6363 Mark the symbol as an indirect function when evaluated during reloc
6364 processing. (This is only supported on Linux targeted assemblers).
6368 Mark the symbol as being a data object.
6372 Mark the symbol as being a thead-local data object.
6376 Mark the symbol as being a common data object.
6380 Does not mark the symbol in any way. It is supported just for completeness.
6382 @item gnu_unique_object
6383 Marks the symbol as being a globally unique data object. The dynamic linker
6384 will make sure that in the entire process there is just one symbol with this
6385 name and type in use. (This is only supported on Linux targeted assemblers).
6389 Note: Some targets support extra types in addition to those listed above.
6395 @section @code{.uleb128 @var{expressions}}
6397 @cindex @code{uleb128} directive
6398 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6399 compact, variable length representation of numbers used by the DWARF
6400 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6404 @section @code{.val @var{addr}}
6406 @cindex @code{val} directive
6407 @cindex COFF value attribute
6408 @cindex value attribute, COFF
6409 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6410 records the address @var{addr} as the value attribute of a symbol table
6414 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6415 configured for @code{b.out}, it accepts this directive but ignores it.
6421 @section @code{.version "@var{string}"}
6423 @cindex @code{version} directive
6424 This directive creates a @code{.note} section and places into it an ELF
6425 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6430 @section @code{.vtable_entry @var{table}, @var{offset}}
6432 @cindex @code{vtable_entry} directive
6433 This directive finds or creates a symbol @code{table} and creates a
6434 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6437 @section @code{.vtable_inherit @var{child}, @var{parent}}
6439 @cindex @code{vtable_inherit} directive
6440 This directive finds the symbol @code{child} and finds or creates the symbol
6441 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6442 parent whose addend is the value of the child symbol. As a special case the
6443 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6447 @section @code{.warning "@var{string}"}
6448 @cindex warning directive
6449 Similar to the directive @code{.error}
6450 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6453 @section @code{.weak @var{names}}
6455 @cindex @code{weak} directive
6456 This directive sets the weak attribute on the comma separated list of symbol
6457 @code{names}. If the symbols do not already exist, they will be created.
6459 On COFF targets other than PE, weak symbols are a GNU extension. This
6460 directive sets the weak attribute on the comma separated list of symbol
6461 @code{names}. If the symbols do not already exist, they will be created.
6463 On the PE target, weak symbols are supported natively as weak aliases.
6464 When a weak symbol is created that is not an alias, GAS creates an
6465 alternate symbol to hold the default value.
6468 @section @code{.weakref @var{alias}, @var{target}}
6470 @cindex @code{weakref} directive
6471 This directive creates an alias to the target symbol that enables the symbol to
6472 be referenced with weak-symbol semantics, but without actually making it weak.
6473 If direct references or definitions of the symbol are present, then the symbol
6474 will not be weak, but if all references to it are through weak references, the
6475 symbol will be marked as weak in the symbol table.
6477 The effect is equivalent to moving all references to the alias to a separate
6478 assembly source file, renaming the alias to the symbol in it, declaring the
6479 symbol as weak there, and running a reloadable link to merge the object files
6480 resulting from the assembly of the new source file and the old source file that
6481 had the references to the alias removed.
6483 The alias itself never makes to the symbol table, and is entirely handled
6484 within the assembler.
6487 @section @code{.word @var{expressions}}
6489 @cindex @code{word} directive
6490 This directive expects zero or more @var{expressions}, of any section,
6491 separated by commas.
6494 For each expression, @command{@value{AS}} emits a 32-bit number.
6497 For each expression, @command{@value{AS}} emits a 16-bit number.
6502 The size of the number emitted, and its byte order,
6503 depend on what target computer the assembly is for.
6506 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
6507 @c happen---32-bit addressability, period; no long/short jumps.
6508 @ifset DIFF-TBL-KLUGE
6509 @cindex difference tables altered
6510 @cindex altered difference tables
6512 @emph{Warning: Special Treatment to support Compilers}
6516 Machines with a 32-bit address space, but that do less than 32-bit
6517 addressing, require the following special treatment. If the machine of
6518 interest to you does 32-bit addressing (or doesn't require it;
6519 @pxref{Machine Dependencies}), you can ignore this issue.
6522 In order to assemble compiler output into something that works,
6523 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6524 Directives of the form @samp{.word sym1-sym2} are often emitted by
6525 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6526 directive of the form @samp{.word sym1-sym2}, and the difference between
6527 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6528 creates a @dfn{secondary jump table}, immediately before the next label.
6529 This secondary jump table is preceded by a short-jump to the
6530 first byte after the secondary table. This short-jump prevents the flow
6531 of control from accidentally falling into the new table. Inside the
6532 table is a long-jump to @code{sym2}. The original @samp{.word}
6533 contains @code{sym1} minus the address of the long-jump to
6536 If there were several occurrences of @samp{.word sym1-sym2} before the
6537 secondary jump table, all of them are adjusted. If there was a
6538 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6539 long-jump to @code{sym4} is included in the secondary jump table,
6540 and the @code{.word} directives are adjusted to contain @code{sym3}
6541 minus the address of the long-jump to @code{sym4}; and so on, for as many
6542 entries in the original jump table as necessary.
6545 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6546 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6547 assembly language programmers.
6550 @c end DIFF-TBL-KLUGE
6553 @section Deprecated Directives
6555 @cindex deprecated directives
6556 @cindex obsolescent directives
6557 One day these directives won't work.
6558 They are included for compatibility with older assemblers.
6565 @node Object Attributes
6566 @chapter Object Attributes
6567 @cindex object attributes
6569 @command{@value{AS}} assembles source files written for a specific architecture
6570 into object files for that architecture. But not all object files are alike.
6571 Many architectures support incompatible variations. For instance, floating
6572 point arguments might be passed in floating point registers if the object file
6573 requires hardware floating point support---or floating point arguments might be
6574 passed in integer registers if the object file supports processors with no
6575 hardware floating point unit. Or, if two objects are built for different
6576 generations of the same architecture, the combination may require the
6577 newer generation at run-time.
6579 This information is useful during and after linking. At link time,
6580 @command{@value{LD}} can warn about incompatible object files. After link
6581 time, tools like @command{gdb} can use it to process the linked file
6584 Compatibility information is recorded as a series of object attributes. Each
6585 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
6586 string, and indicates who sets the meaning of the tag. The tag is an integer,
6587 and indicates what property the attribute describes. The value may be a string
6588 or an integer, and indicates how the property affects this object. Missing
6589 attributes are the same as attributes with a zero value or empty string value.
6591 Object attributes were developed as part of the ABI for the ARM Architecture.
6592 The file format is documented in @cite{ELF for the ARM Architecture}.
6595 * GNU Object Attributes:: @sc{gnu} Object Attributes
6596 * Defining New Object Attributes:: Defining New Object Attributes
6599 @node GNU Object Attributes
6600 @section @sc{gnu} Object Attributes
6602 The @code{.gnu_attribute} directive records an object attribute
6603 with vendor @samp{gnu}.
6605 Except for @samp{Tag_compatibility}, which has both an integer and a string for
6606 its value, @sc{gnu} attributes have a string value if the tag number is odd and
6607 an integer value if the tag number is even. The second bit (@code{@var{tag} &
6608 2} is set for architecture-independent attributes and clear for
6609 architecture-dependent ones.
6611 @subsection Common @sc{gnu} attributes
6613 These attributes are valid on all architectures.
6616 @item Tag_compatibility (32)
6617 The compatibility attribute takes an integer flag value and a vendor name. If
6618 the flag value is 0, the file is compatible with other toolchains. If it is 1,
6619 then the file is only compatible with the named toolchain. If it is greater
6620 than 1, the file can only be processed by other toolchains under some private
6621 arrangement indicated by the flag value and the vendor name.
6624 @subsection MIPS Attributes
6627 @item Tag_GNU_MIPS_ABI_FP (4)
6628 The floating-point ABI used by this object file. The value will be:
6632 0 for files not affected by the floating-point ABI.
6634 1 for files using the hardware floating-point with a standard double-precision
6637 2 for files using the hardware floating-point ABI with a single-precision FPU.
6639 3 for files using the software floating-point ABI.
6641 4 for files using the hardware floating-point ABI with 64-bit wide
6642 double-precision floating-point registers and 32-bit wide general
6647 @subsection PowerPC Attributes
6650 @item Tag_GNU_Power_ABI_FP (4)
6651 The floating-point ABI used by this object file. The value will be:
6655 0 for files not affected by the floating-point ABI.
6657 1 for files using double-precision hardware floating-point ABI.
6659 2 for files using the software floating-point ABI.
6661 3 for files using single-precision hardware floating-point ABI.
6664 @item Tag_GNU_Power_ABI_Vector (8)
6665 The vector ABI used by this object file. The value will be:
6669 0 for files not affected by the vector ABI.
6671 1 for files using general purpose registers to pass vectors.
6673 2 for files using AltiVec registers to pass vectors.
6675 3 for files using SPE registers to pass vectors.
6679 @node Defining New Object Attributes
6680 @section Defining New Object Attributes
6682 If you want to define a new @sc{gnu} object attribute, here are the places you
6683 will need to modify. New attributes should be discussed on the @samp{binutils}
6688 This manual, which is the official register of attributes.
6690 The header for your architecture @file{include/elf}, to define the tag.
6692 The @file{bfd} support file for your architecture, to merge the attribute
6693 and issue any appropriate link warnings.
6695 Test cases in @file{ld/testsuite} for merging and link warnings.
6697 @file{binutils/readelf.c} to display your attribute.
6699 GCC, if you want the compiler to mark the attribute automatically.
6705 @node Machine Dependencies
6706 @chapter Machine Dependent Features
6708 @cindex machine dependencies
6709 The machine instruction sets are (almost by definition) different on
6710 each machine where @command{@value{AS}} runs. Floating point representations
6711 vary as well, and @command{@value{AS}} often supports a few additional
6712 directives or command-line options for compatibility with other
6713 assemblers on a particular platform. Finally, some versions of
6714 @command{@value{AS}} support special pseudo-instructions for branch
6717 This chapter discusses most of these differences, though it does not
6718 include details on any machine's instruction set. For details on that
6719 subject, see the hardware manufacturer's manual.
6723 * Alpha-Dependent:: Alpha Dependent Features
6726 * ARC-Dependent:: ARC Dependent Features
6729 * ARM-Dependent:: ARM Dependent Features
6732 * AVR-Dependent:: AVR Dependent Features
6735 * Blackfin-Dependent:: Blackfin Dependent Features
6738 * CR16-Dependent:: CR16 Dependent Features
6741 * CRIS-Dependent:: CRIS Dependent Features
6744 * D10V-Dependent:: D10V Dependent Features
6747 * D30V-Dependent:: D30V Dependent Features
6750 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6753 * HPPA-Dependent:: HPPA Dependent Features
6756 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
6759 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
6762 * i860-Dependent:: Intel 80860 Dependent Features
6765 * i960-Dependent:: Intel 80960 Dependent Features
6768 * IA-64-Dependent:: Intel IA-64 Dependent Features
6771 * IP2K-Dependent:: IP2K Dependent Features
6774 * LM32-Dependent:: LM32 Dependent Features
6777 * M32C-Dependent:: M32C Dependent Features
6780 * M32R-Dependent:: M32R Dependent Features
6783 * M68K-Dependent:: M680x0 Dependent Features
6786 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
6789 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
6792 * MIPS-Dependent:: MIPS Dependent Features
6795 * MMIX-Dependent:: MMIX Dependent Features
6798 * MSP430-Dependent:: MSP430 Dependent Features
6801 * SH-Dependent:: Renesas / SuperH SH Dependent Features
6802 * SH64-Dependent:: SuperH SH64 Dependent Features
6805 * PDP-11-Dependent:: PDP-11 Dependent Features
6808 * PJ-Dependent:: picoJava Dependent Features
6811 * PPC-Dependent:: PowerPC Dependent Features
6814 * S/390-Dependent:: IBM S/390 Dependent Features
6817 * SCORE-Dependent:: SCORE Dependent Features
6820 * Sparc-Dependent:: SPARC Dependent Features
6823 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
6826 * V850-Dependent:: V850 Dependent Features
6829 * Xtensa-Dependent:: Xtensa Dependent Features
6832 * Z80-Dependent:: Z80 Dependent Features
6835 * Z8000-Dependent:: Z8000 Dependent Features
6838 * Vax-Dependent:: VAX Dependent Features
6845 @c The following major nodes are *sections* in the GENERIC version, *chapters*
6846 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
6847 @c peculiarity: to preserve cross-references, there must be a node called
6848 @c "Machine Dependencies". Hence the conditional nodenames in each
6849 @c major node below. Node defaulting in makeinfo requires adjacency of
6850 @c node and sectioning commands; hence the repetition of @chapter BLAH
6851 @c in both conditional blocks.
6854 @include c-alpha.texi
6870 @include c-bfin.texi
6874 @include c-cr16.texi
6878 @include c-cris.texi
6883 @node Machine Dependencies
6884 @chapter Machine Dependent Features
6886 The machine instruction sets are different on each Renesas chip family,
6887 and there are also some syntax differences among the families. This
6888 chapter describes the specific @command{@value{AS}} features for each
6892 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6893 * SH-Dependent:: Renesas SH Dependent Features
6900 @include c-d10v.texi
6904 @include c-d30v.texi
6908 @include c-h8300.texi
6912 @include c-hppa.texi
6916 @include c-i370.texi
6920 @include c-i386.texi
6924 @include c-i860.texi
6928 @include c-i960.texi
6932 @include c-ia64.texi
6936 @include c-ip2k.texi
6940 @include c-lm32.texi
6944 @include c-m32c.texi
6948 @include c-m32r.texi
6952 @include c-m68k.texi
6956 @include c-m68hc11.texi
6960 @include c-microblaze.texi
6964 @include c-mips.texi
6968 @include c-mmix.texi
6972 @include c-msp430.texi
6976 @include c-ns32k.texi
6980 @include c-pdp11.texi
6992 @include c-s390.texi
6996 @include c-score.texi
7001 @include c-sh64.texi
7005 @include c-sparc.texi
7009 @include c-tic54x.texi
7025 @include c-v850.texi
7029 @include c-xtensa.texi
7033 @c reverse effect of @down at top of generic Machine-Dep chapter
7037 @node Reporting Bugs
7038 @chapter Reporting Bugs
7039 @cindex bugs in assembler
7040 @cindex reporting bugs in assembler
7042 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7044 Reporting a bug may help you by bringing a solution to your problem, or it may
7045 not. But in any case the principal function of a bug report is to help the
7046 entire community by making the next version of @command{@value{AS}} work better.
7047 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7049 In order for a bug report to serve its purpose, you must include the
7050 information that enables us to fix the bug.
7053 * Bug Criteria:: Have you found a bug?
7054 * Bug Reporting:: How to report bugs
7058 @section Have You Found a Bug?
7059 @cindex bug criteria
7061 If you are not sure whether you have found a bug, here are some guidelines:
7064 @cindex fatal signal
7065 @cindex assembler crash
7066 @cindex crash of assembler
7068 If the assembler gets a fatal signal, for any input whatever, that is a
7069 @command{@value{AS}} bug. Reliable assemblers never crash.
7071 @cindex error on valid input
7073 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7075 @cindex invalid input
7077 If @command{@value{AS}} does not produce an error message for invalid input, that
7078 is a bug. However, you should note that your idea of ``invalid input'' might
7079 be our idea of ``an extension'' or ``support for traditional practice''.
7082 If you are an experienced user of assemblers, your suggestions for improvement
7083 of @command{@value{AS}} are welcome in any case.
7087 @section How to Report Bugs
7089 @cindex assembler bugs, reporting
7091 A number of companies and individuals offer support for @sc{gnu} products. If
7092 you obtained @command{@value{AS}} from a support organization, we recommend you
7093 contact that organization first.
7095 You can find contact information for many support companies and
7096 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7100 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7104 The fundamental principle of reporting bugs usefully is this:
7105 @strong{report all the facts}. If you are not sure whether to state a
7106 fact or leave it out, state it!
7108 Often people omit facts because they think they know what causes the problem
7109 and assume that some details do not matter. Thus, you might assume that the
7110 name of a symbol you use in an example does not matter. Well, probably it does
7111 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7112 happens to fetch from the location where that name is stored in memory;
7113 perhaps, if the name were different, the contents of that location would fool
7114 the assembler into doing the right thing despite the bug. Play it safe and
7115 give a specific, complete example. That is the easiest thing for you to do,
7116 and the most helpful.
7118 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7119 it is new to us. Therefore, always write your bug reports on the assumption
7120 that the bug has not been reported previously.
7122 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7123 bell?'' This cannot help us fix a bug, so it is basically useless. We
7124 respond by asking for enough details to enable us to investigate.
7125 You might as well expedite matters by sending them to begin with.
7127 To enable us to fix the bug, you should include all these things:
7131 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
7132 it with the @samp{--version} argument.
7134 Without this, we will not know whether there is any point in looking for
7135 the bug in the current version of @command{@value{AS}}.
7138 Any patches you may have applied to the @command{@value{AS}} source.
7141 The type of machine you are using, and the operating system name and
7145 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
7149 The command arguments you gave the assembler to assemble your example and
7150 observe the bug. To guarantee you will not omit something important, list them
7151 all. A copy of the Makefile (or the output from make) is sufficient.
7153 If we were to try to guess the arguments, we would probably guess wrong
7154 and then we might not encounter the bug.
7157 A complete input file that will reproduce the bug. If the bug is observed when
7158 the assembler is invoked via a compiler, send the assembler source, not the
7159 high level language source. Most compilers will produce the assembler source
7160 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
7161 the options @samp{-v --save-temps}; this will save the assembler source in a
7162 file with an extension of @file{.s}, and also show you exactly how
7163 @command{@value{AS}} is being run.
7166 A description of what behavior you observe that you believe is
7167 incorrect. For example, ``It gets a fatal signal.''
7169 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
7170 will certainly notice it. But if the bug is incorrect output, we might not
7171 notice unless it is glaringly wrong. You might as well not give us a chance to
7174 Even if the problem you experience is a fatal signal, you should still say so
7175 explicitly. Suppose something strange is going on, such as, your copy of
7176 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
7177 library on your system. (This has happened!) Your copy might crash and ours
7178 would not. If you told us to expect a crash, then when ours fails to crash, we
7179 would know that the bug was not happening for us. If you had not told us to
7180 expect a crash, then we would not be able to draw any conclusion from our
7184 If you wish to suggest changes to the @command{@value{AS}} source, send us context
7185 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
7186 option. Always send diffs from the old file to the new file. If you even
7187 discuss something in the @command{@value{AS}} source, refer to it by context, not
7190 The line numbers in our development sources will not match those in your
7191 sources. Your line numbers would convey no useful information to us.
7194 Here are some things that are not necessary:
7198 A description of the envelope of the bug.
7200 Often people who encounter a bug spend a lot of time investigating
7201 which changes to the input file will make the bug go away and which
7202 changes will not affect it.
7204 This is often time consuming and not very useful, because the way we
7205 will find the bug is by running a single example under the debugger
7206 with breakpoints, not by pure deduction from a series of examples.
7207 We recommend that you save your time for something else.
7209 Of course, if you can find a simpler example to report @emph{instead}
7210 of the original one, that is a convenience for us. Errors in the
7211 output will be easier to spot, running under the debugger will take
7212 less time, and so on.
7214 However, simplification is not vital; if you do not want to do this,
7215 report the bug anyway and send us the entire test case you used.
7218 A patch for the bug.
7220 A patch for the bug does help us if it is a good one. But do not omit
7221 the necessary information, such as the test case, on the assumption that
7222 a patch is all we need. We might see problems with your patch and decide
7223 to fix the problem another way, or we might not understand it at all.
7225 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
7226 construct an example that will make the program follow a certain path through
7227 the code. If you do not send us the example, we will not be able to construct
7228 one, so we will not be able to verify that the bug is fixed.
7230 And if we cannot understand what bug you are trying to fix, or why your
7231 patch should be an improvement, we will not install it. A test case will
7232 help us to understand.
7235 A guess about what the bug is or what it depends on.
7237 Such guesses are usually wrong. Even we cannot guess right about such
7238 things without first using the debugger to find the facts.
7241 @node Acknowledgements
7242 @chapter Acknowledgements
7244 If you have contributed to GAS and your name isn't listed here,
7245 it is not meant as a slight. We just don't know about it. Send mail to the
7246 maintainer, and we'll correct the situation. Currently
7248 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
7250 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
7253 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
7254 information and the 68k series machines, most of the preprocessing pass, and
7255 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
7257 K. Richard Pixley maintained GAS for a while, adding various enhancements and
7258 many bug fixes, including merging support for several processors, breaking GAS
7259 up to handle multiple object file format back ends (including heavy rewrite,
7260 testing, an integration of the coff and b.out back ends), adding configuration
7261 including heavy testing and verification of cross assemblers and file splits
7262 and renaming, converted GAS to strictly ANSI C including full prototypes, added
7263 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
7264 port (including considerable amounts of reverse engineering), a SPARC opcode
7265 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
7266 assertions and made them work, much other reorganization, cleanup, and lint.
7268 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
7269 in format-specific I/O modules.
7271 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
7272 has done much work with it since.
7274 The Intel 80386 machine description was written by Eliot Dresselhaus.
7276 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
7278 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
7279 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
7281 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
7282 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
7283 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
7284 support a.out format.
7286 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
7287 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
7288 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
7289 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
7292 John Gilmore built the AMD 29000 support, added @code{.include} support, and
7293 simplified the configuration of which versions accept which directives. He
7294 updated the 68k machine description so that Motorola's opcodes always produced
7295 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
7296 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
7297 cross-compilation support, and one bug in relaxation that took a week and
7298 required the proverbial one-bit fix.
7300 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
7301 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
7302 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
7303 PowerPC assembler, and made a few other minor patches.
7305 Steve Chamberlain made GAS able to generate listings.
7307 Hewlett-Packard contributed support for the HP9000/300.
7309 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
7310 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
7311 formats). This work was supported by both the Center for Software Science at
7312 the University of Utah and Cygnus Support.
7314 Support for ELF format files has been worked on by Mark Eichin of Cygnus
7315 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
7316 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
7317 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
7318 and some initial 64-bit support).
7320 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
7322 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
7323 support for openVMS/Alpha.
7325 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
7328 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
7329 Inc.@: added support for Xtensa processors.
7331 Several engineers at Cygnus Support have also provided many small bug fixes and
7332 configuration enhancements.
7334 Jon Beniston added support for the Lattice Mico32 architecture.
7336 Many others have contributed large or small bugfixes and enhancements. If
7337 you have contributed significant work and are not mentioned on this list, and
7338 want to be, let us know. Some of the history has been lost; we are not
7339 intentionally leaving anyone out.
7341 @node GNU Free Documentation License
7342 @appendix GNU Free Documentation License
7346 @unnumbered AS Index
projects / dragonfly.git / blob