1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000,
3 @c 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
4 @c Free Software Foundation, Inc.
5 @c UPDATE!! On future updates--
6 @c (1) check for new machine-dep cmdline options in
7 @c md_parse_option definitions in config/tc-*.c
8 @c (2) for platform-specific directives, examine md_pseudo_op
10 @c (3) for object-format specific directives, examine obj_pseudo_op
12 @c (4) portable directives in potable[] in read.c
16 @macro gcctabopt{body}
19 @c defaults, config file may override:
24 @include asconfig.texi
29 @c common OR combinations of conditions
55 @set abnormal-separator
59 @settitle Using @value{AS}
62 @settitle Using @value{AS} (@value{TARGET})
64 @setchapternewpage odd
69 @c WARE! Some of the machine-dependent sections contain tables of machine
70 @c instructions. Except in multi-column format, these tables look silly.
71 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
72 @c the multi-col format is faked within @example sections.
74 @c Again unfortunately, the natural size that fits on a page, for these tables,
75 @c is different depending on whether or not smallbook is turned on.
76 @c This matters, because of order: text flow switches columns at each page
79 @c The format faked in this source works reasonably well for smallbook,
80 @c not well for the default large-page format. This manual expects that if you
81 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
82 @c tables in question. You can turn on one without the other at your
83 @c discretion, of course.
86 @c the insn tables look just as silly in info files regardless of smallbook,
87 @c might as well show 'em anyways.
93 * As: (as). The GNU assembler.
94 * Gas: (as). The GNU assembler.
103 This file documents the GNU Assembler "@value{AS}".
105 @c man begin COPYRIGHT
106 Copyright @copyright{} 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
107 2000, 2001, 2002, 2006, 2007, 2008, 2009, 2010 Free Software Foundation, Inc.
109 Permission is granted to copy, distribute and/or modify this document
110 under the terms of the GNU Free Documentation License, Version 1.3
111 or any later version published by the Free Software Foundation;
112 with no Invariant Sections, with no Front-Cover Texts, and with no
113 Back-Cover Texts. A copy of the license is included in the
114 section entitled ``GNU Free Documentation License''.
120 @title Using @value{AS}
121 @subtitle The @sc{gnu} Assembler
123 @subtitle for the @value{TARGET} family
125 @ifset VERSION_PACKAGE
127 @subtitle @value{VERSION_PACKAGE}
130 @subtitle Version @value{VERSION}
133 The Free Software Foundation Inc.@: thanks The Nice Computer
134 Company of Australia for loaning Dean Elsner to write the
135 first (Vax) version of @command{as} for Project @sc{gnu}.
136 The proprietors, management and staff of TNCCA thank FSF for
137 distracting the boss while they got some work
140 @author Dean Elsner, Jay Fenlason & friends
144 \hfill {\it Using {\tt @value{AS}}}\par
145 \hfill Edited by Cygnus Support\par
147 %"boxit" macro for figures:
148 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
149 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
150 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
151 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
152 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
155 @vskip 0pt plus 1filll
156 Copyright @copyright{} 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
157 2000, 2001, 2002, 2006, 2007, 2008, 2009, 2010 Free Software Foundation, Inc.
159 Permission is granted to copy, distribute and/or modify this document
160 under the terms of the GNU Free Documentation License, Version 1.3
161 or any later version published by the Free Software Foundation;
162 with no Invariant Sections, with no Front-Cover Texts, and with no
163 Back-Cover Texts. A copy of the license is included in the
164 section entitled ``GNU Free Documentation License''.
171 @top Using @value{AS}
173 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}}
174 @ifset VERSION_PACKAGE
175 @value{VERSION_PACKAGE}
177 version @value{VERSION}.
179 This version of the file describes @command{@value{AS}} configured to generate
180 code for @value{TARGET} architectures.
183 This document is distributed under the terms of the GNU Free
184 Documentation License. A copy of the license is included in the
185 section entitled ``GNU Free Documentation License''.
188 * Overview:: Overview
189 * Invoking:: Command-Line Options
191 * Sections:: Sections and Relocation
193 * Expressions:: Expressions
194 * Pseudo Ops:: Assembler Directives
196 * Object Attributes:: Object Attributes
198 * Machine Dependencies:: Machine Dependent Features
199 * Reporting Bugs:: Reporting Bugs
200 * Acknowledgements:: Who Did What
201 * GNU Free Documentation License:: GNU Free Documentation License
202 * AS Index:: AS Index
209 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
211 This version of the manual describes @command{@value{AS}} configured to generate
212 code for @value{TARGET} architectures.
216 @cindex invocation summary
217 @cindex option summary
218 @cindex summary of options
219 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
220 see @ref{Invoking,,Command-Line Options}.
222 @c man title AS the portable GNU assembler.
226 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
230 @c We don't use deffn and friends for the following because they seem
231 @c to be limited to one line for the header.
233 @c man begin SYNOPSIS
234 @value{AS} [@b{-a}[@b{cdghlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
235 [@b{--compress-debug-sections}] [@b{--nocompress-debug-sections}]
236 [@b{--debug-prefix-map} @var{old}=@var{new}]
237 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
238 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
239 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
240 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
241 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}] [@b{-o}
242 @var{objfile}] [@b{-R}] [@b{--reduce-memory-overheads}] [@b{--statistics}]
243 [@b{-v}] [@b{-version}] [@b{--version}] [@b{-W}] [@b{--warn}]
244 [@b{--fatal-warnings}] [@b{-w}] [@b{-x}] [@b{-Z}] [@b{@@@var{FILE}}]
245 [@b{--target-help}] [@var{target-options}]
246 [@b{--}|@var{files} @dots{}]
248 @c Target dependent options are listed below. Keep the list sorted.
249 @c Add an empty line for separation.
252 @emph{Target Alpha options:}
254 [@b{-mdebug} | @b{-no-mdebug}]
255 [@b{-replace} | @b{-noreplace}]
256 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
257 [@b{-F}] [@b{-32addr}]
261 @emph{Target ARC options:}
267 @emph{Target ARM options:}
268 @c Don't document the deprecated options
269 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
270 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
271 [@b{-mfpu}=@var{floating-point-format}]
272 [@b{-mfloat-abi}=@var{abi}]
273 [@b{-meabi}=@var{ver}]
276 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
277 @b{-mapcs-reentrant}]
278 [@b{-mthumb-interwork}] [@b{-k}]
282 @emph{Target Blackfin options:}
283 [@b{-mcpu}=@var{processor}[-@var{sirevision}]]
290 @emph{Target CRIS options:}
291 [@b{--underscore} | @b{--no-underscore}]
293 [@b{--emulation=criself} | @b{--emulation=crisaout}]
294 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
295 @c Deprecated -- deliberately not documented.
300 @emph{Target D10V options:}
305 @emph{Target D30V options:}
306 [@b{-O}|@b{-n}|@b{-N}]
310 @emph{Target H8/300 options:}
314 @c HPPA has no machine-dependent assembler options (yet).
318 @emph{Target i386 options:}
319 [@b{--32}|@b{--64}] [@b{-n}]
320 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
324 @emph{Target i960 options:}
325 @c see md_parse_option in tc-i960.c
326 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
328 [@b{-b}] [@b{-no-relax}]
332 @emph{Target IA-64 options:}
333 [@b{-mconstant-gp}|@b{-mauto-pic}]
334 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
336 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
337 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
338 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
339 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
343 @emph{Target IP2K options:}
344 [@b{-mip2022}|@b{-mip2022ext}]
348 @emph{Target M32C options:}
349 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
353 @emph{Target M32R options:}
354 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
359 @emph{Target M680X0 options:}
360 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
364 @emph{Target M68HC11 options:}
365 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}]
366 [@b{-mshort}|@b{-mlong}]
367 [@b{-mshort-double}|@b{-mlong-double}]
368 [@b{--force-long-branches}] [@b{--short-branches}]
369 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
370 [@b{--print-opcodes}] [@b{--generate-example}]
374 @emph{Target MCORE options:}
375 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
376 [@b{-mcpu=[210|340]}]
379 @emph{Target MICROBLAZE options:}
380 @c MicroBlaze has no machine-dependent assembler options.
384 @emph{Target MIPS options:}
385 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
386 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
387 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
388 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
389 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
390 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
391 [@b{-mips64}] [@b{-mips64r2}]
392 [@b{-construct-floats}] [@b{-no-construct-floats}]
393 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
394 [@b{-mips16}] [@b{-no-mips16}]
395 [@b{-msmartmips}] [@b{-mno-smartmips}]
396 [@b{-mips3d}] [@b{-no-mips3d}]
397 [@b{-mdmx}] [@b{-no-mdmx}]
398 [@b{-mdsp}] [@b{-mno-dsp}]
399 [@b{-mdspr2}] [@b{-mno-dspr2}]
400 [@b{-mmt}] [@b{-mno-mt}]
401 [@b{-mfix7000}] [@b{-mno-fix7000}]
402 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
403 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
404 [@b{-mdebug}] [@b{-no-mdebug}]
405 [@b{-mpdr}] [@b{-mno-pdr}]
409 @emph{Target MMIX options:}
410 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
411 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
412 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
413 [@b{--linker-allocated-gregs}]
417 @emph{Target PDP11 options:}
418 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
419 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
420 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
424 @emph{Target picoJava options:}
429 @emph{Target PowerPC options:}
430 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|
431 @b{-m403}|@b{-m405}|@b{-mppc64}|@b{-m620}|@b{-mppc64bridge}|@b{-mbooke}]
432 [@b{-mcom}|@b{-many}|@b{-maltivec}|@b{-mvsx}] [@b{-memb}]
433 [@b{-mregnames}|@b{-mno-regnames}]
434 [@b{-mrelocatable}|@b{-mrelocatable-lib}]
435 [@b{-mlittle}|@b{-mlittle-endian}|@b{-mbig}|@b{-mbig-endian}]
436 [@b{-msolaris}|@b{-mno-solaris}]
440 @emph{Target RX options:}
441 [@b{-mlittle-endian}|@b{-mbig-endian}]
442 [@b{-m32bit-ints}|@b{-m16bit-ints}]
443 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
447 @emph{Target s390 options:}
448 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
449 [@b{-mregnames}|@b{-mno-regnames}]
450 [@b{-mwarn-areg-zero}]
454 @emph{Target SCORE options:}
455 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
456 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
457 [@b{-march=score7}][@b{-march=score3}]
458 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
462 @emph{Target SPARC options:}
463 @c The order here is important. See c-sparc.texi.
464 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
465 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
466 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
471 @emph{Target TIC54X options:}
472 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
473 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
478 @emph{Target TIC6X options:}
479 [@b{-march=@var{arch}}] [@b{-matomic}|@b{-mno-atomic}]
480 [@b{-mbig-endian}|@b{-mlittle-endian}]
485 @emph{Target Z80 options:}
486 [@b{-z80}] [@b{-r800}]
487 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
488 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
489 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
490 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
491 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
492 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
496 @c Z8000 has no machine-dependent assembler options
500 @emph{Target Xtensa options:}
501 [@b{--[no-]text-section-literals}] [@b{--[no-]absolute-literals}]
502 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
503 [@b{--[no-]transform}]
504 [@b{--rename-section} @var{oldname}=@var{newname}]
512 @include at-file.texi
515 Turn on listings, in any of a variety of ways:
519 omit false conditionals
522 omit debugging directives
525 include general information, like @value{AS} version and options passed
528 include high-level source
534 include macro expansions
537 omit forms processing
543 set the name of the listing file
546 You may combine these options; for example, use @samp{-aln} for assembly
547 listing without forms processing. The @samp{=file} option, if used, must be
548 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
551 Begin in alternate macro mode.
553 @xref{Altmacro,,@code{.altmacro}}.
556 @item --compress-debug-sections
557 Compress DWARF debug sections using zlib. The debug sections are renamed
558 to begin with @samp{.zdebug}, and the resulting object file may not be
559 compatible with older linkers and object file utilities.
561 @item --nocompress-debug-sections
562 Do not compress DWARF debug sections. This is the default.
565 Ignored. This option is accepted for script compatibility with calls to
568 @item --debug-prefix-map @var{old}=@var{new}
569 When assembling files in directory @file{@var{old}}, record debugging
570 information describing them as in @file{@var{new}} instead.
572 @item --defsym @var{sym}=@var{value}
573 Define the symbol @var{sym} to be @var{value} before assembling the input file.
574 @var{value} must be an integer constant. As in C, a leading @samp{0x}
575 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
576 value. The value of the symbol can be overridden inside a source file via the
577 use of a @code{.set} pseudo-op.
580 ``fast''---skip whitespace and comment preprocessing (assume source is
585 Generate debugging information for each assembler source line using whichever
586 debug format is preferred by the target. This currently means either STABS,
590 Generate stabs debugging information for each assembler line. This
591 may help debugging assembler code, if the debugger can handle it.
594 Generate stabs debugging information for each assembler line, with GNU
595 extensions that probably only gdb can handle, and that could make other
596 debuggers crash or refuse to read your program. This
597 may help debugging assembler code. Currently the only GNU extension is
598 the location of the current working directory at assembling time.
601 Generate DWARF2 debugging information for each assembler line. This
602 may help debugging assembler code, if the debugger can handle it. Note---this
603 option is only supported by some targets, not all of them.
606 Print a summary of the command line options and exit.
609 Print a summary of all target specific options and exit.
612 Add directory @var{dir} to the search list for @code{.include} directives.
615 Don't warn about signed overflow.
618 @ifclear DIFF-TBL-KLUGE
619 This option is accepted but has no effect on the @value{TARGET} family.
621 @ifset DIFF-TBL-KLUGE
622 Issue warnings when difference tables altered for long displacements.
627 Keep (in the symbol table) local symbols. These symbols start with
628 system-specific local label prefixes, typically @samp{.L} for ELF systems
629 or @samp{L} for traditional a.out systems.
634 @item --listing-lhs-width=@var{number}
635 Set the maximum width, in words, of the output data column for an assembler
636 listing to @var{number}.
638 @item --listing-lhs-width2=@var{number}
639 Set the maximum width, in words, of the output data column for continuation
640 lines in an assembler listing to @var{number}.
642 @item --listing-rhs-width=@var{number}
643 Set the maximum width of an input source line, as displayed in a listing, to
646 @item --listing-cont-lines=@var{number}
647 Set the maximum number of lines printed in a listing for a single line of input
650 @item -o @var{objfile}
651 Name the object-file output from @command{@value{AS}} @var{objfile}.
654 Fold the data section into the text section.
656 @kindex --hash-size=@var{number}
657 Set the default size of GAS's hash tables to a prime number close to
658 @var{number}. Increasing this value can reduce the length of time it takes the
659 assembler to perform its tasks, at the expense of increasing the assembler's
660 memory requirements. Similarly reducing this value can reduce the memory
661 requirements at the expense of speed.
663 @item --reduce-memory-overheads
664 This option reduces GAS's memory requirements, at the expense of making the
665 assembly processes slower. Currently this switch is a synonym for
666 @samp{--hash-size=4051}, but in the future it may have other effects as well.
669 Print the maximum space (in bytes) and total time (in seconds) used by
672 @item --strip-local-absolute
673 Remove local absolute symbols from the outgoing symbol table.
677 Print the @command{as} version.
680 Print the @command{as} version and exit.
684 Suppress warning messages.
686 @item --fatal-warnings
687 Treat warnings as errors.
690 Don't suppress warning messages or treat them as errors.
699 Generate an object file even after errors.
701 @item -- | @var{files} @dots{}
702 Standard input, or source files to assemble.
707 The following options are available when @value{AS} is configured for
712 This option selects the core processor variant.
714 Select either big-endian (-EB) or little-endian (-EL) output.
719 The following options are available when @value{AS} is configured for the ARM
723 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
724 Specify which ARM processor variant is the target.
725 @item -march=@var{architecture}[+@var{extension}@dots{}]
726 Specify which ARM architecture variant is used by the target.
727 @item -mfpu=@var{floating-point-format}
728 Select which Floating Point architecture is the target.
729 @item -mfloat-abi=@var{abi}
730 Select which floating point ABI is in use.
732 Enable Thumb only instruction decoding.
733 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
734 Select which procedure calling convention is in use.
736 Select either big-endian (-EB) or little-endian (-EL) output.
737 @item -mthumb-interwork
738 Specify that the code has been generated with interworking between Thumb and
741 Specify that PIC code has been generated.
746 The following options are available when @value{AS} is configured for
747 the Blackfin processor family.
750 @item -mcpu=@var{processor}@r{[}-@var{sirevision}@r{]}
751 This option specifies the target processor. The optional @var{sirevision}
752 is not used in assembler.
754 Assemble for the FDPIC ABI.
762 See the info pages for documentation of the CRIS-specific options.
766 The following options are available when @value{AS} is configured for
769 @cindex D10V optimization
770 @cindex optimization, D10V
772 Optimize output by parallelizing instructions.
777 The following options are available when @value{AS} is configured for a D30V
780 @cindex D30V optimization
781 @cindex optimization, D30V
783 Optimize output by parallelizing instructions.
787 Warn when nops are generated.
789 @cindex D30V nops after 32-bit multiply
791 Warn when a nop after a 32-bit multiply instruction is generated.
796 The following options are available when @value{AS} is configured for the
797 Intel 80960 processor.
800 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
801 Specify which variant of the 960 architecture is the target.
804 Add code to collect statistics about branches taken.
807 Do not alter compare-and-branch instructions for long displacements;
814 The following options are available when @value{AS} is configured for the
820 Specifies that the extended IP2022 instructions are allowed.
823 Restores the default behaviour, which restricts the permitted instructions to
824 just the basic IP2022 ones.
830 The following options are available when @value{AS} is configured for the
831 Renesas M32C and M16C processors.
836 Assemble M32C instructions.
839 Assemble M16C instructions (the default).
842 Enable support for link-time relaxations.
845 Support H'00 style hex constants in addition to 0x00 style.
851 The following options are available when @value{AS} is configured for the
852 Renesas M32R (formerly Mitsubishi M32R) series.
857 Specify which processor in the M32R family is the target. The default
858 is normally the M32R, but this option changes it to the M32RX.
860 @item --warn-explicit-parallel-conflicts or --Wp
861 Produce warning messages when questionable parallel constructs are
864 @item --no-warn-explicit-parallel-conflicts or --Wnp
865 Do not produce warning messages when questionable parallel constructs are
872 The following options are available when @value{AS} is configured for the
873 Motorola 68000 series.
878 Shorten references to undefined symbols, to one word instead of two.
880 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
881 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
882 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
883 Specify what processor in the 68000 family is the target. The default
884 is normally the 68020, but this can be changed at configuration time.
886 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
887 The target machine does (or does not) have a floating-point coprocessor.
888 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
889 the basic 68000 is not compatible with the 68881, a combination of the
890 two can be specified, since it's possible to do emulation of the
891 coprocessor instructions with the main processor.
893 @item -m68851 | -mno-68851
894 The target machine does (or does not) have a memory-management
895 unit coprocessor. The default is to assume an MMU for 68020 and up.
902 For details about the PDP-11 machine dependent features options,
903 see @ref{PDP-11-Options}.
906 @item -mpic | -mno-pic
907 Generate position-independent (or position-dependent) code. The
908 default is @option{-mpic}.
911 @itemx -mall-extensions
912 Enable all instruction set extensions. This is the default.
914 @item -mno-extensions
915 Disable all instruction set extensions.
917 @item -m@var{extension} | -mno-@var{extension}
918 Enable (or disable) a particular instruction set extension.
921 Enable the instruction set extensions supported by a particular CPU, and
922 disable all other extensions.
924 @item -m@var{machine}
925 Enable the instruction set extensions supported by a particular machine
926 model, and disable all other extensions.
932 The following options are available when @value{AS} is configured for
933 a picoJava processor.
937 @cindex PJ endianness
938 @cindex endianness, PJ
939 @cindex big endian output, PJ
941 Generate ``big endian'' format output.
943 @cindex little endian output, PJ
945 Generate ``little endian'' format output.
951 The following options are available when @value{AS} is configured for the
952 Motorola 68HC11 or 68HC12 series.
956 @item -m68hc11 | -m68hc12 | -m68hcs12
957 Specify what processor is the target. The default is
958 defined by the configuration option when building the assembler.
961 Specify to use the 16-bit integer ABI.
964 Specify to use the 32-bit integer ABI.
967 Specify to use the 32-bit double ABI.
970 Specify to use the 64-bit double ABI.
972 @item --force-long-branches
973 Relative branches are turned into absolute ones. This concerns
974 conditional branches, unconditional branches and branches to a
977 @item -S | --short-branches
978 Do not turn relative branches into absolute ones
979 when the offset is out of range.
981 @item --strict-direct-mode
982 Do not turn the direct addressing mode into extended addressing mode
983 when the instruction does not support direct addressing mode.
985 @item --print-insn-syntax
986 Print the syntax of instruction in case of error.
988 @item --print-opcodes
989 print the list of instructions with syntax and then exit.
991 @item --generate-example
992 print an example of instruction for each possible instruction and then exit.
993 This option is only useful for testing @command{@value{AS}}.
999 The following options are available when @command{@value{AS}} is configured
1000 for the SPARC architecture:
1003 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1004 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1005 Explicitly select a variant of the SPARC architecture.
1007 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1008 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1010 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1011 UltraSPARC extensions.
1013 @item -xarch=v8plus | -xarch=v8plusa
1014 For compatibility with the Solaris v9 assembler. These options are
1015 equivalent to -Av8plus and -Av8plusa, respectively.
1018 Warn when the assembler switches to another architecture.
1023 The following options are available when @value{AS} is configured for the 'c54x
1028 Enable extended addressing mode. All addresses and relocations will assume
1029 extended addressing (usually 23 bits).
1030 @item -mcpu=@var{CPU_VERSION}
1031 Sets the CPU version being compiled for.
1032 @item -merrors-to-file @var{FILENAME}
1033 Redirect error output to a file, for broken systems which don't support such
1034 behaviour in the shell.
1039 The following options are available when @value{AS} is configured for
1040 a @sc{mips} processor.
1044 This option sets the largest size of an object that can be referenced
1045 implicitly with the @code{gp} register. It is only accepted for targets that
1046 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1048 @cindex MIPS endianness
1049 @cindex endianness, MIPS
1050 @cindex big endian output, MIPS
1052 Generate ``big endian'' format output.
1054 @cindex little endian output, MIPS
1056 Generate ``little endian'' format output.
1068 Generate code for a particular @sc{mips} Instruction Set Architecture level.
1069 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1070 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1071 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1072 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
1074 correspond to generic
1075 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
1076 and @samp{MIPS64 Release 2}
1077 ISA processors, respectively.
1079 @item -march=@var{CPU}
1080 Generate code for a particular @sc{mips} cpu.
1082 @item -mtune=@var{cpu}
1083 Schedule and tune for a particular @sc{mips} cpu.
1087 Cause nops to be inserted if the read of the destination register
1088 of an mfhi or mflo instruction occurs in the following two instructions.
1092 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1093 section instead of the standard ELF .stabs sections.
1097 Control generation of @code{.pdr} sections.
1101 The register sizes are normally inferred from the ISA and ABI, but these
1102 flags force a certain group of registers to be treated as 32 bits wide at
1103 all times. @samp{-mgp32} controls the size of general-purpose registers
1104 and @samp{-mfp32} controls the size of floating-point registers.
1108 Generate code for the MIPS 16 processor. This is equivalent to putting
1109 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1110 turns off this option.
1113 @itemx -mno-smartmips
1114 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1115 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1116 @samp{-mno-smartmips} turns off this option.
1120 Generate code for the MIPS-3D Application Specific Extension.
1121 This tells the assembler to accept MIPS-3D instructions.
1122 @samp{-no-mips3d} turns off this option.
1126 Generate code for the MDMX Application Specific Extension.
1127 This tells the assembler to accept MDMX instructions.
1128 @samp{-no-mdmx} turns off this option.
1132 Generate code for the DSP Release 1 Application Specific Extension.
1133 This tells the assembler to accept DSP Release 1 instructions.
1134 @samp{-mno-dsp} turns off this option.
1138 Generate code for the DSP Release 2 Application Specific Extension.
1139 This option implies -mdsp.
1140 This tells the assembler to accept DSP Release 2 instructions.
1141 @samp{-mno-dspr2} turns off this option.
1145 Generate code for the MT Application Specific Extension.
1146 This tells the assembler to accept MT instructions.
1147 @samp{-mno-mt} turns off this option.
1149 @item --construct-floats
1150 @itemx --no-construct-floats
1151 The @samp{--no-construct-floats} option disables the construction of
1152 double width floating point constants by loading the two halves of the
1153 value into the two single width floating point registers that make up
1154 the double width register. By default @samp{--construct-floats} is
1155 selected, allowing construction of these floating point constants.
1158 @item --emulation=@var{name}
1159 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
1160 for some other target, in all respects, including output format (choosing
1161 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
1162 debugging information or store symbol table information, and default
1163 endianness. The available configuration names are: @samp{mipsecoff},
1164 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
1165 @samp{mipsbelf}. The first two do not alter the default endianness from that
1166 of the primary target for which the assembler was configured; the others change
1167 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
1168 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
1169 selection in any case.
1171 This option is currently supported only when the primary target
1172 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
1173 Furthermore, the primary target or others specified with
1174 @samp{--enable-targets=@dots{}} at configuration time must include support for
1175 the other format, if both are to be available. For example, the Irix 5
1176 configuration includes support for both.
1178 Eventually, this option will support more configurations, with more
1179 fine-grained control over the assembler's behavior, and will be supported for
1183 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1190 Control how to deal with multiplication overflow and division by zero.
1191 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1192 (and only work for Instruction Set Architecture level 2 and higher);
1193 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1197 When this option is used, @command{@value{AS}} will issue a warning every
1198 time it generates a nop instruction from a macro.
1203 The following options are available when @value{AS} is configured for
1209 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1210 The command line option @samp{-nojsri2bsr} can be used to disable it.
1214 Enable or disable the silicon filter behaviour. By default this is disabled.
1215 The default can be overridden by the @samp{-sifilter} command line option.
1218 Alter jump instructions for long displacements.
1220 @item -mcpu=[210|340]
1221 Select the cpu type on the target hardware. This controls which instructions
1225 Assemble for a big endian target.
1228 Assemble for a little endian target.
1234 See the info pages for documentation of the MMIX-specific options.
1238 See the info pages for documentation of the RX-specific options.
1242 The following options are available when @value{AS} is configured for the s390
1248 Select the word size, either 31/32 bits or 64 bits.
1251 Select the architecture mode, either the Enterprise System
1252 Architecture (esa) or the z/Architecture mode (zarch).
1253 @item -march=@var{processor}
1254 Specify which s390 processor variant is the target, @samp{g6}, @samp{g6},
1255 @samp{z900}, @samp{z990}, @samp{z9-109}, @samp{z9-ec}, or @samp{z10}.
1257 @itemx -mno-regnames
1258 Allow or disallow symbolic names for registers.
1259 @item -mwarn-areg-zero
1260 Warn whenever the operand for a base or index register has been specified
1261 but evaluates to zero.
1266 The following options are available when @value{AS} is configured for a
1267 TMS320C6000 processor.
1270 @item -march=@var{arch}
1271 Enable (only) instructions from architecture @var{arch}. By default,
1272 all instructions are permitted.
1274 The following values of @var{arch} are accepted: @code{c62x},
1275 @code{c64x}, @code{c64x+}, @code{c67x}, @code{c67x+}, @code{c674x}.
1279 Enable or disable the optional C64x+ atomic operation instructions.
1280 By default, they are enabled if no @option{-march} option is given, or
1281 if an architecture is specified with @option{-march} that implies
1282 these instructions are present (currently, there are no such
1283 architectures); they are disabled if an architecture is specified with
1284 @option{-march} on which the instructions are optional or not
1285 present. This option overrides such a default from the architecture,
1286 independent of the order in which the @option{-march} or
1287 @option{-matomic} or @option{-mno-atomic} options are passed.
1290 @itemx -mlittle-endian
1291 Generate code for the specified endianness. The default is
1298 The following options are available when @value{AS} is configured for
1299 an Xtensa processor.
1302 @item --text-section-literals | --no-text-section-literals
1303 With @option{--text-@-section-@-literals}, literal pools are interspersed
1304 in the text section. The default is
1305 @option{--no-@-text-@-section-@-literals}, which places literals in a
1306 separate section in the output file. These options only affect literals
1307 referenced via PC-relative @code{L32R} instructions; literals for
1308 absolute mode @code{L32R} instructions are handled separately.
1310 @item --absolute-literals | --no-absolute-literals
1311 Indicate to the assembler whether @code{L32R} instructions use absolute
1312 or PC-relative addressing. The default is to assume absolute addressing
1313 if the Xtensa processor includes the absolute @code{L32R} addressing
1314 option. Otherwise, only the PC-relative @code{L32R} mode can be used.
1316 @item --target-align | --no-target-align
1317 Enable or disable automatic alignment to reduce branch penalties at the
1318 expense of some code density. The default is @option{--target-@-align}.
1320 @item --longcalls | --no-longcalls
1321 Enable or disable transformation of call instructions to allow calls
1322 across a greater range of addresses. The default is
1323 @option{--no-@-longcalls}.
1325 @item --transform | --no-transform
1326 Enable or disable all assembler transformations of Xtensa instructions.
1327 The default is @option{--transform};
1328 @option{--no-transform} should be used only in the rare cases when the
1329 instructions must be exactly as specified in the assembly source.
1331 @item --rename-section @var{oldname}=@var{newname}
1332 When generating output sections, rename the @var{oldname} section to
1338 The following options are available when @value{AS} is configured for
1339 a Z80 family processor.
1342 Assemble for Z80 processor.
1344 Assemble for R800 processor.
1345 @item -ignore-undocumented-instructions
1347 Assemble undocumented Z80 instructions that also work on R800 without warning.
1348 @item -ignore-unportable-instructions
1350 Assemble all undocumented Z80 instructions without warning.
1351 @item -warn-undocumented-instructions
1353 Issue a warning for undocumented Z80 instructions that also work on R800.
1354 @item -warn-unportable-instructions
1356 Issue a warning for undocumented Z80 instructions that do not work on R800.
1357 @item -forbid-undocumented-instructions
1359 Treat all undocumented instructions as errors.
1360 @item -forbid-unportable-instructions
1362 Treat undocumented Z80 instructions that do not work on R800 as errors.
1369 * Manual:: Structure of this Manual
1370 * GNU Assembler:: The GNU Assembler
1371 * Object Formats:: Object File Formats
1372 * Command Line:: Command Line
1373 * Input Files:: Input Files
1374 * Object:: Output (Object) File
1375 * Errors:: Error and Warning Messages
1379 @section Structure of this Manual
1381 @cindex manual, structure and purpose
1382 This manual is intended to describe what you need to know to use
1383 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1384 notation for symbols, constants, and expressions; the directives that
1385 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1388 We also cover special features in the @value{TARGET}
1389 configuration of @command{@value{AS}}, including assembler directives.
1392 This manual also describes some of the machine-dependent features of
1393 various flavors of the assembler.
1396 @cindex machine instructions (not covered)
1397 On the other hand, this manual is @emph{not} intended as an introduction
1398 to programming in assembly language---let alone programming in general!
1399 In a similar vein, we make no attempt to introduce the machine
1400 architecture; we do @emph{not} describe the instruction set, standard
1401 mnemonics, registers or addressing modes that are standard to a
1402 particular architecture.
1404 You may want to consult the manufacturer's
1405 machine architecture manual for this information.
1409 For information on the H8/300 machine instruction set, see @cite{H8/300
1410 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1411 Programming Manual} (Renesas).
1414 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1415 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1416 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1417 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1420 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1424 @c I think this is premature---doc@cygnus.com, 17jan1991
1426 Throughout this manual, we assume that you are running @dfn{GNU},
1427 the portable operating system from the @dfn{Free Software
1428 Foundation, Inc.}. This restricts our attention to certain kinds of
1429 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1430 once this assumption is granted examples and definitions need less
1433 @command{@value{AS}} is part of a team of programs that turn a high-level
1434 human-readable series of instructions into a low-level
1435 computer-readable series of instructions. Different versions of
1436 @command{@value{AS}} are used for different kinds of computer.
1439 @c There used to be a section "Terminology" here, which defined
1440 @c "contents", "byte", "word", and "long". Defining "word" to any
1441 @c particular size is confusing when the .word directive may generate 16
1442 @c bits on one machine and 32 bits on another; in general, for the user
1443 @c version of this manual, none of these terms seem essential to define.
1444 @c They were used very little even in the former draft of the manual;
1445 @c this draft makes an effort to avoid them (except in names of
1449 @section The GNU Assembler
1451 @c man begin DESCRIPTION
1453 @sc{gnu} @command{as} is really a family of assemblers.
1455 This manual describes @command{@value{AS}}, a member of that family which is
1456 configured for the @value{TARGET} architectures.
1458 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1459 should find a fairly similar environment when you use it on another
1460 architecture. Each version has much in common with the others,
1461 including object file formats, most assembler directives (often called
1462 @dfn{pseudo-ops}) and assembler syntax.@refill
1464 @cindex purpose of @sc{gnu} assembler
1465 @command{@value{AS}} is primarily intended to assemble the output of the
1466 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1467 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1468 assemble correctly everything that other assemblers for the same
1469 machine would assemble.
1471 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1474 @c This remark should appear in generic version of manual; assumption
1475 @c here is that generic version sets M680x0.
1476 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1477 assembler for the same architecture; for example, we know of several
1478 incompatible versions of 680x0 assembly language syntax.
1483 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1484 program in one pass of the source file. This has a subtle impact on the
1485 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1487 @node Object Formats
1488 @section Object File Formats
1490 @cindex object file format
1491 The @sc{gnu} assembler can be configured to produce several alternative
1492 object file formats. For the most part, this does not affect how you
1493 write assembly language programs; but directives for debugging symbols
1494 are typically different in different file formats. @xref{Symbol
1495 Attributes,,Symbol Attributes}.
1498 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1499 @value{OBJ-NAME} format object files.
1501 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1503 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1504 @code{b.out} or COFF format object files.
1507 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1508 SOM or ELF format object files.
1513 @section Command Line
1515 @cindex command line conventions
1517 After the program name @command{@value{AS}}, the command line may contain
1518 options and file names. Options may appear in any order, and may be
1519 before, after, or between file names. The order of file names is
1522 @cindex standard input, as input file
1524 @file{--} (two hyphens) by itself names the standard input file
1525 explicitly, as one of the files for @command{@value{AS}} to assemble.
1527 @cindex options, command line
1528 Except for @samp{--} any command line argument that begins with a
1529 hyphen (@samp{-}) is an option. Each option changes the behavior of
1530 @command{@value{AS}}. No option changes the way another option works. An
1531 option is a @samp{-} followed by one or more letters; the case of
1532 the letter is important. All options are optional.
1534 Some options expect exactly one file name to follow them. The file
1535 name may either immediately follow the option's letter (compatible
1536 with older assemblers) or it may be the next command argument (@sc{gnu}
1537 standard). These two command lines are equivalent:
1540 @value{AS} -o my-object-file.o mumble.s
1541 @value{AS} -omy-object-file.o mumble.s
1545 @section Input Files
1548 @cindex source program
1549 @cindex files, input
1550 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1551 describe the program input to one run of @command{@value{AS}}. The program may
1552 be in one or more files; how the source is partitioned into files
1553 doesn't change the meaning of the source.
1555 @c I added "con" prefix to "catenation" just to prove I can overcome my
1556 @c APL training... doc@cygnus.com
1557 The source program is a concatenation of the text in all the files, in the
1560 @c man begin DESCRIPTION
1561 Each time you run @command{@value{AS}} it assembles exactly one source
1562 program. The source program is made up of one or more files.
1563 (The standard input is also a file.)
1565 You give @command{@value{AS}} a command line that has zero or more input file
1566 names. The input files are read (from left file name to right). A
1567 command line argument (in any position) that has no special meaning
1568 is taken to be an input file name.
1570 If you give @command{@value{AS}} no file names it attempts to read one input file
1571 from the @command{@value{AS}} standard input, which is normally your terminal. You
1572 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1575 Use @samp{--} if you need to explicitly name the standard input file
1576 in your command line.
1578 If the source is empty, @command{@value{AS}} produces a small, empty object
1583 @subheading Filenames and Line-numbers
1585 @cindex input file linenumbers
1586 @cindex line numbers, in input files
1587 There are two ways of locating a line in the input file (or files) and
1588 either may be used in reporting error messages. One way refers to a line
1589 number in a physical file; the other refers to a line number in a
1590 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1592 @dfn{Physical files} are those files named in the command line given
1593 to @command{@value{AS}}.
1595 @dfn{Logical files} are simply names declared explicitly by assembler
1596 directives; they bear no relation to physical files. Logical file names help
1597 error messages reflect the original source file, when @command{@value{AS}} source
1598 is itself synthesized from other files. @command{@value{AS}} understands the
1599 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1600 @ref{File,,@code{.file}}.
1603 @section Output (Object) File
1609 Every time you run @command{@value{AS}} it produces an output file, which is
1610 your assembly language program translated into numbers. This file
1611 is the object file. Its default name is
1619 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1621 You can give it another name by using the @option{-o} option. Conventionally,
1622 object file names end with @file{.o}. The default name is used for historical
1623 reasons: older assemblers were capable of assembling self-contained programs
1624 directly into a runnable program. (For some formats, this isn't currently
1625 possible, but it can be done for the @code{a.out} format.)
1629 The object file is meant for input to the linker @code{@value{LD}}. It contains
1630 assembled program code, information to help @code{@value{LD}} integrate
1631 the assembled program into a runnable file, and (optionally) symbolic
1632 information for the debugger.
1634 @c link above to some info file(s) like the description of a.out.
1635 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1638 @section Error and Warning Messages
1640 @c man begin DESCRIPTION
1642 @cindex error messages
1643 @cindex warning messages
1644 @cindex messages from assembler
1645 @command{@value{AS}} may write warnings and error messages to the standard error
1646 file (usually your terminal). This should not happen when a compiler
1647 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1648 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1649 grave problem that stops the assembly.
1653 @cindex format of warning messages
1654 Warning messages have the format
1657 file_name:@b{NNN}:Warning Message Text
1661 @cindex line numbers, in warnings/errors
1662 (where @b{NNN} is a line number). If a logical file name has been given
1663 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1664 the current input file is used. If a logical line number was given
1666 (@pxref{Line,,@code{.line}})
1668 then it is used to calculate the number printed,
1669 otherwise the actual line in the current source file is printed. The
1670 message text is intended to be self explanatory (in the grand Unix
1673 @cindex format of error messages
1674 Error messages have the format
1676 file_name:@b{NNN}:FATAL:Error Message Text
1678 The file name and line number are derived as for warning
1679 messages. The actual message text may be rather less explanatory
1680 because many of them aren't supposed to happen.
1683 @chapter Command-Line Options
1685 @cindex options, all versions of assembler
1686 This chapter describes command-line options available in @emph{all}
1687 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
1688 for options specific
1690 to the @value{TARGET} target.
1693 to particular machine architectures.
1696 @c man begin DESCRIPTION
1698 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1699 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1700 The assembler arguments must be separated from each other (and the @samp{-Wa})
1701 by commas. For example:
1704 gcc -c -g -O -Wa,-alh,-L file.c
1708 This passes two options to the assembler: @samp{-alh} (emit a listing to
1709 standard output with high-level and assembly source) and @samp{-L} (retain
1710 local symbols in the symbol table).
1712 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1713 command-line options are automatically passed to the assembler by the compiler.
1714 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1715 precisely what options it passes to each compilation pass, including the
1721 * a:: -a[cdghlns] enable listings
1722 * alternate:: --alternate enable alternate macro syntax
1723 * D:: -D for compatibility
1724 * f:: -f to work faster
1725 * I:: -I for .include search path
1726 @ifclear DIFF-TBL-KLUGE
1727 * K:: -K for compatibility
1729 @ifset DIFF-TBL-KLUGE
1730 * K:: -K for difference tables
1733 * L:: -L to retain local symbols
1734 * listing:: --listing-XXX to configure listing output
1735 * M:: -M or --mri to assemble in MRI compatibility mode
1736 * MD:: --MD for dependency tracking
1737 * o:: -o to name the object file
1738 * R:: -R to join data and text sections
1739 * statistics:: --statistics to see statistics about assembly
1740 * traditional-format:: --traditional-format for compatible output
1741 * v:: -v to announce version
1742 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1743 * Z:: -Z to make object file even after errors
1747 @section Enable Listings: @option{-a[cdghlns]}
1757 @cindex listings, enabling
1758 @cindex assembly listings, enabling
1760 These options enable listing output from the assembler. By itself,
1761 @samp{-a} requests high-level, assembly, and symbols listing.
1762 You can use other letters to select specific options for the list:
1763 @samp{-ah} requests a high-level language listing,
1764 @samp{-al} requests an output-program assembly listing, and
1765 @samp{-as} requests a symbol table listing.
1766 High-level listings require that a compiler debugging option like
1767 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1770 Use the @samp{-ag} option to print a first section with general assembly
1771 information, like @value{AS} version, switches passed, or time stamp.
1773 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1774 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1775 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1776 omitted from the listing.
1778 Use the @samp{-ad} option to omit debugging directives from the
1781 Once you have specified one of these options, you can further control
1782 listing output and its appearance using the directives @code{.list},
1783 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1785 The @samp{-an} option turns off all forms processing.
1786 If you do not request listing output with one of the @samp{-a} options, the
1787 listing-control directives have no effect.
1789 The letters after @samp{-a} may be combined into one option,
1790 @emph{e.g.}, @samp{-aln}.
1792 Note if the assembler source is coming from the standard input (e.g.,
1794 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1795 is being used) then the listing will not contain any comments or preprocessor
1796 directives. This is because the listing code buffers input source lines from
1797 stdin only after they have been preprocessed by the assembler. This reduces
1798 memory usage and makes the code more efficient.
1801 @section @option{--alternate}
1804 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
1807 @section @option{-D}
1810 This option has no effect whatsoever, but it is accepted to make it more
1811 likely that scripts written for other assemblers also work with
1812 @command{@value{AS}}.
1815 @section Work Faster: @option{-f}
1818 @cindex trusted compiler
1819 @cindex faster processing (@option{-f})
1820 @samp{-f} should only be used when assembling programs written by a
1821 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1822 and comment preprocessing on
1823 the input file(s) before assembling them. @xref{Preprocessing,
1827 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1828 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1833 @section @code{.include} Search Path: @option{-I} @var{path}
1835 @kindex -I @var{path}
1836 @cindex paths for @code{.include}
1837 @cindex search path for @code{.include}
1838 @cindex @code{include} directive search path
1839 Use this option to add a @var{path} to the list of directories
1840 @command{@value{AS}} searches for files specified in @code{.include}
1841 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1842 many times as necessary to include a variety of paths. The current
1843 working directory is always searched first; after that, @command{@value{AS}}
1844 searches any @samp{-I} directories in the same order as they were
1845 specified (left to right) on the command line.
1848 @section Difference Tables: @option{-K}
1851 @ifclear DIFF-TBL-KLUGE
1852 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1853 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1854 where it can be used to warn when the assembler alters the machine code
1855 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1856 family does not have the addressing limitations that sometimes lead to this
1857 alteration on other platforms.
1860 @ifset DIFF-TBL-KLUGE
1861 @cindex difference tables, warning
1862 @cindex warning for altered difference tables
1863 @command{@value{AS}} sometimes alters the code emitted for directives of the
1864 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
1865 You can use the @samp{-K} option if you want a warning issued when this
1870 @section Include Local Symbols: @option{-L}
1873 @cindex local symbols, retaining in output
1874 Symbols beginning with system-specific local label prefixes, typically
1875 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
1876 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
1877 such symbols when debugging, because they are intended for the use of
1878 programs (like compilers) that compose assembler programs, not for your
1879 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
1880 such symbols, so you do not normally debug with them.
1882 This option tells @command{@value{AS}} to retain those local symbols
1883 in the object file. Usually if you do this you also tell the linker
1884 @code{@value{LD}} to preserve those symbols.
1887 @section Configuring listing output: @option{--listing}
1889 The listing feature of the assembler can be enabled via the command line switch
1890 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1891 hex dump of the corresponding locations in the output object file, and displays
1892 them as a listing file. The format of this listing can be controlled by
1893 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
1894 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
1895 @code{.psize} (@pxref{Psize}), and
1896 @code{.eject} (@pxref{Eject}) and also by the following switches:
1899 @item --listing-lhs-width=@samp{number}
1900 @kindex --listing-lhs-width
1901 @cindex Width of first line disassembly output
1902 Sets the maximum width, in words, of the first line of the hex byte dump. This
1903 dump appears on the left hand side of the listing output.
1905 @item --listing-lhs-width2=@samp{number}
1906 @kindex --listing-lhs-width2
1907 @cindex Width of continuation lines of disassembly output
1908 Sets the maximum width, in words, of any further lines of the hex byte dump for
1909 a given input source line. If this value is not specified, it defaults to being
1910 the same as the value specified for @samp{--listing-lhs-width}. If neither
1911 switch is used the default is to one.
1913 @item --listing-rhs-width=@samp{number}
1914 @kindex --listing-rhs-width
1915 @cindex Width of source line output
1916 Sets the maximum width, in characters, of the source line that is displayed
1917 alongside the hex dump. The default value for this parameter is 100. The
1918 source line is displayed on the right hand side of the listing output.
1920 @item --listing-cont-lines=@samp{number}
1921 @kindex --listing-cont-lines
1922 @cindex Maximum number of continuation lines
1923 Sets the maximum number of continuation lines of hex dump that will be
1924 displayed for a given single line of source input. The default value is 4.
1928 @section Assemble in MRI Compatibility Mode: @option{-M}
1931 @cindex MRI compatibility mode
1932 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
1933 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
1934 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1935 configured target) assembler from Microtec Research. The exact nature of the
1936 MRI syntax will not be documented here; see the MRI manuals for more
1937 information. Note in particular that the handling of macros and macro
1938 arguments is somewhat different. The purpose of this option is to permit
1939 assembling existing MRI assembler code using @command{@value{AS}}.
1941 The MRI compatibility is not complete. Certain operations of the MRI assembler
1942 depend upon its object file format, and can not be supported using other object
1943 file formats. Supporting these would require enhancing each object file format
1944 individually. These are:
1947 @item global symbols in common section
1949 The m68k MRI assembler supports common sections which are merged by the linker.
1950 Other object file formats do not support this. @command{@value{AS}} handles
1951 common sections by treating them as a single common symbol. It permits local
1952 symbols to be defined within a common section, but it can not support global
1953 symbols, since it has no way to describe them.
1955 @item complex relocations
1957 The MRI assemblers support relocations against a negated section address, and
1958 relocations which combine the start addresses of two or more sections. These
1959 are not support by other object file formats.
1961 @item @code{END} pseudo-op specifying start address
1963 The MRI @code{END} pseudo-op permits the specification of a start address.
1964 This is not supported by other object file formats. The start address may
1965 instead be specified using the @option{-e} option to the linker, or in a linker
1968 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1970 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1971 name to the output file. This is not supported by other object file formats.
1973 @item @code{ORG} pseudo-op
1975 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1976 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
1977 which changes the location within the current section. Absolute sections are
1978 not supported by other object file formats. The address of a section may be
1979 assigned within a linker script.
1982 There are some other features of the MRI assembler which are not supported by
1983 @command{@value{AS}}, typically either because they are difficult or because they
1984 seem of little consequence. Some of these may be supported in future releases.
1988 @item EBCDIC strings
1990 EBCDIC strings are not supported.
1992 @item packed binary coded decimal
1994 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1995 and @code{DCB.P} pseudo-ops are not supported.
1997 @item @code{FEQU} pseudo-op
1999 The m68k @code{FEQU} pseudo-op is not supported.
2001 @item @code{NOOBJ} pseudo-op
2003 The m68k @code{NOOBJ} pseudo-op is not supported.
2005 @item @code{OPT} branch control options
2007 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2008 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2009 relaxes all branches, whether forward or backward, to an appropriate size, so
2010 these options serve no purpose.
2012 @item @code{OPT} list control options
2014 The following m68k @code{OPT} list control options are ignored: @code{C},
2015 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2016 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2018 @item other @code{OPT} options
2020 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2021 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2023 @item @code{OPT} @code{D} option is default
2025 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2026 @code{OPT NOD} may be used to turn it off.
2028 @item @code{XREF} pseudo-op.
2030 The m68k @code{XREF} pseudo-op is ignored.
2032 @item @code{.debug} pseudo-op
2034 The i960 @code{.debug} pseudo-op is not supported.
2036 @item @code{.extended} pseudo-op
2038 The i960 @code{.extended} pseudo-op is not supported.
2040 @item @code{.list} pseudo-op.
2042 The various options of the i960 @code{.list} pseudo-op are not supported.
2044 @item @code{.optimize} pseudo-op
2046 The i960 @code{.optimize} pseudo-op is not supported.
2048 @item @code{.output} pseudo-op
2050 The i960 @code{.output} pseudo-op is not supported.
2052 @item @code{.setreal} pseudo-op
2054 The i960 @code{.setreal} pseudo-op is not supported.
2059 @section Dependency Tracking: @option{--MD}
2062 @cindex dependency tracking
2065 @command{@value{AS}} can generate a dependency file for the file it creates. This
2066 file consists of a single rule suitable for @code{make} describing the
2067 dependencies of the main source file.
2069 The rule is written to the file named in its argument.
2071 This feature is used in the automatic updating of makefiles.
2074 @section Name the Object File: @option{-o}
2077 @cindex naming object file
2078 @cindex object file name
2079 There is always one object file output when you run @command{@value{AS}}. By
2080 default it has the name
2083 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2097 You use this option (which takes exactly one filename) to give the
2098 object file a different name.
2100 Whatever the object file is called, @command{@value{AS}} overwrites any
2101 existing file of the same name.
2104 @section Join Data and Text Sections: @option{-R}
2107 @cindex data and text sections, joining
2108 @cindex text and data sections, joining
2109 @cindex joining text and data sections
2110 @cindex merging text and data sections
2111 @option{-R} tells @command{@value{AS}} to write the object file as if all
2112 data-section data lives in the text section. This is only done at
2113 the very last moment: your binary data are the same, but data
2114 section parts are relocated differently. The data section part of
2115 your object file is zero bytes long because all its bytes are
2116 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2118 When you specify @option{-R} it would be possible to generate shorter
2119 address displacements (because we do not have to cross between text and
2120 data section). We refrain from doing this simply for compatibility with
2121 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2124 When @command{@value{AS}} is configured for COFF or ELF output,
2125 this option is only useful if you use sections named @samp{.text} and
2130 @option{-R} is not supported for any of the HPPA targets. Using
2131 @option{-R} generates a warning from @command{@value{AS}}.
2135 @section Display Assembly Statistics: @option{--statistics}
2137 @kindex --statistics
2138 @cindex statistics, about assembly
2139 @cindex time, total for assembly
2140 @cindex space used, maximum for assembly
2141 Use @samp{--statistics} to display two statistics about the resources used by
2142 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2143 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2146 @node traditional-format
2147 @section Compatible Output: @option{--traditional-format}
2149 @kindex --traditional-format
2150 For some targets, the output of @command{@value{AS}} is different in some ways
2151 from the output of some existing assembler. This switch requests
2152 @command{@value{AS}} to use the traditional format instead.
2154 For example, it disables the exception frame optimizations which
2155 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2158 @section Announce Version: @option{-v}
2162 @cindex assembler version
2163 @cindex version of assembler
2164 You can find out what version of as is running by including the
2165 option @samp{-v} (which you can also spell as @samp{-version}) on the
2169 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2171 @command{@value{AS}} should never give a warning or error message when
2172 assembling compiler output. But programs written by people often
2173 cause @command{@value{AS}} to give a warning that a particular assumption was
2174 made. All such warnings are directed to the standard error file.
2178 @cindex suppressing warnings
2179 @cindex warnings, suppressing
2180 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2181 This only affects the warning messages: it does not change any particular of
2182 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2185 @kindex --fatal-warnings
2186 @cindex errors, caused by warnings
2187 @cindex warnings, causing error
2188 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2189 files that generate warnings to be in error.
2192 @cindex warnings, switching on
2193 You can switch these options off again by specifying @option{--warn}, which
2194 causes warnings to be output as usual.
2197 @section Generate Object File in Spite of Errors: @option{-Z}
2198 @cindex object file, after errors
2199 @cindex errors, continuing after
2200 After an error message, @command{@value{AS}} normally produces no output. If for
2201 some reason you are interested in object file output even after
2202 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2203 option. If there are any errors, @command{@value{AS}} continues anyways, and
2204 writes an object file after a final warning message of the form @samp{@var{n}
2205 errors, @var{m} warnings, generating bad object file.}
2210 @cindex machine-independent syntax
2211 @cindex syntax, machine-independent
2212 This chapter describes the machine-independent syntax allowed in a
2213 source file. @command{@value{AS}} syntax is similar to what many other
2214 assemblers use; it is inspired by the BSD 4.2
2219 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2223 * Preprocessing:: Preprocessing
2224 * Whitespace:: Whitespace
2225 * Comments:: Comments
2226 * Symbol Intro:: Symbols
2227 * Statements:: Statements
2228 * Constants:: Constants
2232 @section Preprocessing
2234 @cindex preprocessing
2235 The @command{@value{AS}} internal preprocessor:
2237 @cindex whitespace, removed by preprocessor
2239 adjusts and removes extra whitespace. It leaves one space or tab before
2240 the keywords on a line, and turns any other whitespace on the line into
2243 @cindex comments, removed by preprocessor
2245 removes all comments, replacing them with a single space, or an
2246 appropriate number of newlines.
2248 @cindex constants, converted by preprocessor
2250 converts character constants into the appropriate numeric values.
2253 It does not do macro processing, include file handling, or
2254 anything else you may get from your C compiler's preprocessor. You can
2255 do include file processing with the @code{.include} directive
2256 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2257 to get other ``CPP'' style preprocessing by giving the input file a
2258 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2259 Output, gcc.info, Using GNU CC}.
2261 Excess whitespace, comments, and character constants
2262 cannot be used in the portions of the input text that are not
2265 @cindex turning preprocessing on and off
2266 @cindex preprocessing, turning on and off
2269 If the first line of an input file is @code{#NO_APP} or if you use the
2270 @samp{-f} option, whitespace and comments are not removed from the input file.
2271 Within an input file, you can ask for whitespace and comment removal in
2272 specific portions of the by putting a line that says @code{#APP} before the
2273 text that may contain whitespace or comments, and putting a line that says
2274 @code{#NO_APP} after this text. This feature is mainly intend to support
2275 @code{asm} statements in compilers whose output is otherwise free of comments
2282 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2283 Whitespace is used to separate symbols, and to make programs neater for
2284 people to read. Unless within character constants
2285 (@pxref{Characters,,Character Constants}), any whitespace means the same
2286 as exactly one space.
2292 There are two ways of rendering comments to @command{@value{AS}}. In both
2293 cases the comment is equivalent to one space.
2295 Anything from @samp{/*} through the next @samp{*/} is a comment.
2296 This means you may not nest these comments.
2300 The only way to include a newline ('\n') in a comment
2301 is to use this sort of comment.
2304 /* This sort of comment does not nest. */
2307 @cindex line comment character
2308 Anything from the @dfn{line comment} character to the next newline
2309 is considered a comment and is ignored. The line comment character is
2311 @samp{;} on the ARC;
2314 @samp{@@} on the ARM;
2317 @samp{;} for the H8/300 family;
2320 @samp{;} for the HPPA;
2323 @samp{#} on the i386 and x86-64;
2326 @samp{#} on the i960;
2329 @samp{;} for the PDP-11;
2332 @samp{;} for picoJava;
2335 @samp{#} for Motorola PowerPC;
2338 @samp{#} for IBM S/390;
2341 @samp{#} for the Sunplus SCORE;
2344 @samp{!} for the Renesas / SuperH SH;
2347 @samp{!} on the SPARC;
2350 @samp{#} on the ip2k;
2353 @samp{#} on the m32c;
2356 @samp{#} on the m32r;
2359 @samp{|} on the 680x0;
2362 @samp{#} on the 68HC11 and 68HC12;
2368 @samp{;} on the TMS320C6X;
2371 @samp{#} on the Vax;
2374 @samp{;} for the Z80;
2377 @samp{!} for the Z8000;
2380 @samp{#} on the V850;
2383 @samp{#} for Xtensa systems;
2385 see @ref{Machine Dependencies}. @refill
2386 @c FIXME What about i860?
2389 On some machines there are two different line comment characters. One
2390 character only begins a comment if it is the first non-whitespace character on
2391 a line, while the other always begins a comment.
2395 The V850 assembler also supports a double dash as starting a comment that
2396 extends to the end of the line.
2402 @cindex lines starting with @code{#}
2403 @cindex logical line numbers
2404 To be compatible with past assemblers, lines that begin with @samp{#} have a
2405 special interpretation. Following the @samp{#} should be an absolute
2406 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2407 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2408 new logical file name. The rest of the line, if any, should be whitespace.
2410 If the first non-whitespace characters on the line are not numeric,
2411 the line is ignored. (Just like a comment.)
2414 # This is an ordinary comment.
2415 # 42-6 "new_file_name" # New logical file name
2416 # This is logical line # 36.
2418 This feature is deprecated, and may disappear from future versions
2419 of @command{@value{AS}}.
2424 @cindex characters used in symbols
2425 @ifclear SPECIAL-SYMS
2426 A @dfn{symbol} is one or more characters chosen from the set of all
2427 letters (both upper and lower case), digits and the three characters
2433 A @dfn{symbol} is one or more characters chosen from the set of all
2434 letters (both upper and lower case), digits and the three characters
2435 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2441 On most machines, you can also use @code{$} in symbol names; exceptions
2442 are noted in @ref{Machine Dependencies}.
2444 No symbol may begin with a digit. Case is significant.
2445 There is no length limit: all characters are significant. Symbols are
2446 delimited by characters not in that set, or by the beginning of a file
2447 (since the source program must end with a newline, the end of a file is
2448 not a possible symbol delimiter). @xref{Symbols}.
2449 @cindex length of symbols
2454 @cindex statements, structure of
2455 @cindex line separator character
2456 @cindex statement separator character
2458 @ifclear abnormal-separator
2459 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
2460 semicolon (@samp{;}). The newline or semicolon is considered part of
2461 the preceding statement. Newlines and semicolons within character
2462 constants are an exception: they do not end statements.
2464 @ifset abnormal-separator
2466 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
2467 point (@samp{!}). The newline or exclamation point is considered part of the
2468 preceding statement. Newlines and exclamation points within character
2469 constants are an exception: they do not end statements.
2472 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
2473 H8/300) a dollar sign (@samp{$}); or (for the Renesas-SH) a semicolon
2474 (@samp{;}). The newline or separator character is considered part of
2475 the preceding statement. Newlines and separators within character
2476 constants are an exception: they do not end statements.
2481 A @dfn{statement} ends at a newline character (@samp{\n}) or line
2482 separator character. (The line separator is usually @samp{;}, unless this
2483 conflicts with the comment character; see @ref{Machine Dependencies}.) The
2484 newline or separator character is considered part of the preceding
2485 statement. Newlines and separators within character constants are an
2486 exception: they do not end statements.
2489 @cindex newline, required at file end
2490 @cindex EOF, newline must precede
2491 It is an error to end any statement with end-of-file: the last
2492 character of any input file should be a newline.@refill
2494 An empty statement is allowed, and may include whitespace. It is ignored.
2496 @cindex instructions and directives
2497 @cindex directives and instructions
2498 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2499 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2501 A statement begins with zero or more labels, optionally followed by a
2502 key symbol which determines what kind of statement it is. The key
2503 symbol determines the syntax of the rest of the statement. If the
2504 symbol begins with a dot @samp{.} then the statement is an assembler
2505 directive: typically valid for any computer. If the symbol begins with
2506 a letter the statement is an assembly language @dfn{instruction}: it
2507 assembles into a machine language instruction.
2509 Different versions of @command{@value{AS}} for different computers
2510 recognize different instructions. In fact, the same symbol may
2511 represent a different instruction in a different computer's assembly
2515 @cindex @code{:} (label)
2516 @cindex label (@code{:})
2517 A label is a symbol immediately followed by a colon (@code{:}).
2518 Whitespace before a label or after a colon is permitted, but you may not
2519 have whitespace between a label's symbol and its colon. @xref{Labels}.
2522 For HPPA targets, labels need not be immediately followed by a colon, but
2523 the definition of a label must begin in column zero. This also implies that
2524 only one label may be defined on each line.
2528 label: .directive followed by something
2529 another_label: # This is an empty statement.
2530 instruction operand_1, operand_2, @dots{}
2537 A constant is a number, written so that its value is known by
2538 inspection, without knowing any context. Like this:
2541 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2542 .ascii "Ring the bell\7" # A string constant.
2543 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2544 .float 0f-314159265358979323846264338327\
2545 95028841971.693993751E-40 # - pi, a flonum.
2550 * Characters:: Character Constants
2551 * Numbers:: Number Constants
2555 @subsection Character Constants
2557 @cindex character constants
2558 @cindex constants, character
2559 There are two kinds of character constants. A @dfn{character} stands
2560 for one character in one byte and its value may be used in
2561 numeric expressions. String constants (properly called string
2562 @emph{literals}) are potentially many bytes and their values may not be
2563 used in arithmetic expressions.
2567 * Chars:: Characters
2571 @subsubsection Strings
2573 @cindex string constants
2574 @cindex constants, string
2575 A @dfn{string} is written between double-quotes. It may contain
2576 double-quotes or null characters. The way to get special characters
2577 into a string is to @dfn{escape} these characters: precede them with
2578 a backslash @samp{\} character. For example @samp{\\} represents
2579 one backslash: the first @code{\} is an escape which tells
2580 @command{@value{AS}} to interpret the second character literally as a backslash
2581 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2582 escape character). The complete list of escapes follows.
2584 @cindex escape codes, character
2585 @cindex character escape codes
2588 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2590 @cindex @code{\b} (backspace character)
2591 @cindex backspace (@code{\b})
2593 Mnemonic for backspace; for ASCII this is octal code 010.
2596 @c Mnemonic for EOText; for ASCII this is octal code 004.
2598 @cindex @code{\f} (formfeed character)
2599 @cindex formfeed (@code{\f})
2601 Mnemonic for FormFeed; for ASCII this is octal code 014.
2603 @cindex @code{\n} (newline character)
2604 @cindex newline (@code{\n})
2606 Mnemonic for newline; for ASCII this is octal code 012.
2609 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2611 @cindex @code{\r} (carriage return character)
2612 @cindex carriage return (@code{\r})
2614 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2617 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2618 @c other assemblers.
2620 @cindex @code{\t} (tab)
2621 @cindex tab (@code{\t})
2623 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2626 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2627 @c @item \x @var{digit} @var{digit} @var{digit}
2628 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2630 @cindex @code{\@var{ddd}} (octal character code)
2631 @cindex octal character code (@code{\@var{ddd}})
2632 @item \ @var{digit} @var{digit} @var{digit}
2633 An octal character code. The numeric code is 3 octal digits.
2634 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2635 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2637 @cindex @code{\@var{xd...}} (hex character code)
2638 @cindex hex character code (@code{\@var{xd...}})
2639 @item \@code{x} @var{hex-digits...}
2640 A hex character code. All trailing hex digits are combined. Either upper or
2641 lower case @code{x} works.
2643 @cindex @code{\\} (@samp{\} character)
2644 @cindex backslash (@code{\\})
2646 Represents one @samp{\} character.
2649 @c Represents one @samp{'} (accent acute) character.
2650 @c This is needed in single character literals
2651 @c (@xref{Characters,,Character Constants}.) to represent
2654 @cindex @code{\"} (doublequote character)
2655 @cindex doublequote (@code{\"})
2657 Represents one @samp{"} character. Needed in strings to represent
2658 this character, because an unescaped @samp{"} would end the string.
2660 @item \ @var{anything-else}
2661 Any other character when escaped by @kbd{\} gives a warning, but
2662 assembles as if the @samp{\} was not present. The idea is that if
2663 you used an escape sequence you clearly didn't want the literal
2664 interpretation of the following character. However @command{@value{AS}} has no
2665 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2666 code and warns you of the fact.
2669 Which characters are escapable, and what those escapes represent,
2670 varies widely among assemblers. The current set is what we think
2671 the BSD 4.2 assembler recognizes, and is a subset of what most C
2672 compilers recognize. If you are in doubt, do not use an escape
2676 @subsubsection Characters
2678 @cindex single character constant
2679 @cindex character, single
2680 @cindex constant, single character
2681 A single character may be written as a single quote immediately
2682 followed by that character. The same escapes apply to characters as
2683 to strings. So if you want to write the character backslash, you
2684 must write @kbd{'\\} where the first @code{\} escapes the second
2685 @code{\}. As you can see, the quote is an acute accent, not a
2686 grave accent. A newline
2688 @ifclear abnormal-separator
2689 (or semicolon @samp{;})
2691 @ifset abnormal-separator
2693 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2698 immediately following an acute accent is taken as a literal character
2699 and does not count as the end of a statement. The value of a character
2700 constant in a numeric expression is the machine's byte-wide code for
2701 that character. @command{@value{AS}} assumes your character code is ASCII:
2702 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2705 @subsection Number Constants
2707 @cindex constants, number
2708 @cindex number constants
2709 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2710 are stored in the target machine. @emph{Integers} are numbers that
2711 would fit into an @code{int} in the C language. @emph{Bignums} are
2712 integers, but they are stored in more than 32 bits. @emph{Flonums}
2713 are floating point numbers, described below.
2716 * Integers:: Integers
2721 * Bit Fields:: Bit Fields
2727 @subsubsection Integers
2729 @cindex constants, integer
2731 @cindex binary integers
2732 @cindex integers, binary
2733 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2734 the binary digits @samp{01}.
2736 @cindex octal integers
2737 @cindex integers, octal
2738 An octal integer is @samp{0} followed by zero or more of the octal
2739 digits (@samp{01234567}).
2741 @cindex decimal integers
2742 @cindex integers, decimal
2743 A decimal integer starts with a non-zero digit followed by zero or
2744 more digits (@samp{0123456789}).
2746 @cindex hexadecimal integers
2747 @cindex integers, hexadecimal
2748 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2749 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2751 Integers have the usual values. To denote a negative integer, use
2752 the prefix operator @samp{-} discussed under expressions
2753 (@pxref{Prefix Ops,,Prefix Operators}).
2756 @subsubsection Bignums
2759 @cindex constants, bignum
2760 A @dfn{bignum} has the same syntax and semantics as an integer
2761 except that the number (or its negative) takes more than 32 bits to
2762 represent in binary. The distinction is made because in some places
2763 integers are permitted while bignums are not.
2766 @subsubsection Flonums
2768 @cindex floating point numbers
2769 @cindex constants, floating point
2771 @cindex precision, floating point
2772 A @dfn{flonum} represents a floating point number. The translation is
2773 indirect: a decimal floating point number from the text is converted by
2774 @command{@value{AS}} to a generic binary floating point number of more than
2775 sufficient precision. This generic floating point number is converted
2776 to a particular computer's floating point format (or formats) by a
2777 portion of @command{@value{AS}} specialized to that computer.
2779 A flonum is written by writing (in order)
2784 (@samp{0} is optional on the HPPA.)
2788 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2790 @kbd{e} is recommended. Case is not important.
2792 @c FIXME: verify if flonum syntax really this vague for most cases
2793 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2794 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2797 On the H8/300, Renesas / SuperH SH,
2798 and AMD 29K architectures, the letter must be
2799 one of the letters @samp{DFPRSX} (in upper or lower case).
2801 On the ARC, the letter must be one of the letters @samp{DFRS}
2802 (in upper or lower case).
2804 On the Intel 960 architecture, the letter must be
2805 one of the letters @samp{DFT} (in upper or lower case).
2807 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2811 One of the letters @samp{DFRS} (in upper or lower case).
2814 One of the letters @samp{DFPRSX} (in upper or lower case).
2817 The letter @samp{E} (upper case only).
2820 One of the letters @samp{DFT} (in upper or lower case).
2825 An optional sign: either @samp{+} or @samp{-}.
2828 An optional @dfn{integer part}: zero or more decimal digits.
2831 An optional @dfn{fractional part}: @samp{.} followed by zero
2832 or more decimal digits.
2835 An optional exponent, consisting of:
2839 An @samp{E} or @samp{e}.
2840 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2841 @c principle this can perfectly well be different on different targets.
2843 Optional sign: either @samp{+} or @samp{-}.
2845 One or more decimal digits.
2850 At least one of the integer part or the fractional part must be
2851 present. The floating point number has the usual base-10 value.
2853 @command{@value{AS}} does all processing using integers. Flonums are computed
2854 independently of any floating point hardware in the computer running
2855 @command{@value{AS}}.
2859 @c Bit fields are written as a general facility but are also controlled
2860 @c by a conditional-compilation flag---which is as of now (21mar91)
2861 @c turned on only by the i960 config of GAS.
2863 @subsubsection Bit Fields
2866 @cindex constants, bit field
2867 You can also define numeric constants as @dfn{bit fields}.
2868 Specify two numbers separated by a colon---
2870 @var{mask}:@var{value}
2873 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2876 The resulting number is then packed
2878 @c this conditional paren in case bit fields turned on elsewhere than 960
2879 (in host-dependent byte order)
2881 into a field whose width depends on which assembler directive has the
2882 bit-field as its argument. Overflow (a result from the bitwise and
2883 requiring more binary digits to represent) is not an error; instead,
2884 more constants are generated, of the specified width, beginning with the
2885 least significant digits.@refill
2887 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2888 @code{.short}, and @code{.word} accept bit-field arguments.
2893 @chapter Sections and Relocation
2898 * Secs Background:: Background
2899 * Ld Sections:: Linker Sections
2900 * As Sections:: Assembler Internal Sections
2901 * Sub-Sections:: Sub-Sections
2905 @node Secs Background
2908 Roughly, a section is a range of addresses, with no gaps; all data
2909 ``in'' those addresses is treated the same for some particular purpose.
2910 For example there may be a ``read only'' section.
2912 @cindex linker, and assembler
2913 @cindex assembler, and linker
2914 The linker @code{@value{LD}} reads many object files (partial programs) and
2915 combines their contents to form a runnable program. When @command{@value{AS}}
2916 emits an object file, the partial program is assumed to start at address 0.
2917 @code{@value{LD}} assigns the final addresses for the partial program, so that
2918 different partial programs do not overlap. This is actually an
2919 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2922 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2923 addresses. These blocks slide to their run-time addresses as rigid
2924 units; their length does not change and neither does the order of bytes
2925 within them. Such a rigid unit is called a @emph{section}. Assigning
2926 run-time addresses to sections is called @dfn{relocation}. It includes
2927 the task of adjusting mentions of object-file addresses so they refer to
2928 the proper run-time addresses.
2930 For the H8/300, and for the Renesas / SuperH SH,
2931 @command{@value{AS}} pads sections if needed to
2932 ensure they end on a word (sixteen bit) boundary.
2935 @cindex standard assembler sections
2936 An object file written by @command{@value{AS}} has at least three sections, any
2937 of which may be empty. These are named @dfn{text}, @dfn{data} and
2942 When it generates COFF or ELF output,
2944 @command{@value{AS}} can also generate whatever other named sections you specify
2945 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2946 If you do not use any directives that place output in the @samp{.text}
2947 or @samp{.data} sections, these sections still exist, but are empty.
2952 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2954 @command{@value{AS}} can also generate whatever other named sections you
2955 specify using the @samp{.space} and @samp{.subspace} directives. See
2956 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2957 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2958 assembler directives.
2961 Additionally, @command{@value{AS}} uses different names for the standard
2962 text, data, and bss sections when generating SOM output. Program text
2963 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2964 BSS into @samp{$BSS$}.
2968 Within the object file, the text section starts at address @code{0}, the
2969 data section follows, and the bss section follows the data section.
2972 When generating either SOM or ELF output files on the HPPA, the text
2973 section starts at address @code{0}, the data section at address
2974 @code{0x4000000}, and the bss section follows the data section.
2977 To let @code{@value{LD}} know which data changes when the sections are
2978 relocated, and how to change that data, @command{@value{AS}} also writes to the
2979 object file details of the relocation needed. To perform relocation
2980 @code{@value{LD}} must know, each time an address in the object
2984 Where in the object file is the beginning of this reference to
2987 How long (in bytes) is this reference?
2989 Which section does the address refer to? What is the numeric value of
2991 (@var{address}) @minus{} (@var{start-address of section})?
2994 Is the reference to an address ``Program-Counter relative''?
2997 @cindex addresses, format of
2998 @cindex section-relative addressing
2999 In fact, every address @command{@value{AS}} ever uses is expressed as
3001 (@var{section}) + (@var{offset into section})
3004 Further, most expressions @command{@value{AS}} computes have this section-relative
3007 (For some object formats, such as SOM for the HPPA, some expressions are
3008 symbol-relative instead.)
3011 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3012 @var{N} into section @var{secname}.''
3014 Apart from text, data and bss sections you need to know about the
3015 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3016 addresses in the absolute section remain unchanged. For example, address
3017 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3018 @code{@value{LD}}. Although the linker never arranges two partial programs'
3019 data sections with overlapping addresses after linking, @emph{by definition}
3020 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3021 part of a program is always the same address when the program is running as
3022 address @code{@{absolute@ 239@}} in any other part of the program.
3024 The idea of sections is extended to the @dfn{undefined} section. Any
3025 address whose section is unknown at assembly time is by definition
3026 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3027 Since numbers are always defined, the only way to generate an undefined
3028 address is to mention an undefined symbol. A reference to a named
3029 common block would be such a symbol: its value is unknown at assembly
3030 time so it has section @emph{undefined}.
3032 By analogy the word @emph{section} is used to describe groups of sections in
3033 the linked program. @code{@value{LD}} puts all partial programs' text
3034 sections in contiguous addresses in the linked program. It is
3035 customary to refer to the @emph{text section} of a program, meaning all
3036 the addresses of all partial programs' text sections. Likewise for
3037 data and bss sections.
3039 Some sections are manipulated by @code{@value{LD}}; others are invented for
3040 use of @command{@value{AS}} and have no meaning except during assembly.
3043 @section Linker Sections
3044 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3049 @cindex named sections
3050 @cindex sections, named
3051 @item named sections
3054 @cindex text section
3055 @cindex data section
3059 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3060 separate but equal sections. Anything you can say of one section is
3063 When the program is running, however, it is
3064 customary for the text section to be unalterable. The
3065 text section is often shared among processes: it contains
3066 instructions, constants and the like. The data section of a running
3067 program is usually alterable: for example, C variables would be stored
3068 in the data section.
3073 This section contains zeroed bytes when your program begins running. It
3074 is used to hold uninitialized variables or common storage. The length of
3075 each partial program's bss section is important, but because it starts
3076 out containing zeroed bytes there is no need to store explicit zero
3077 bytes in the object file. The bss section was invented to eliminate
3078 those explicit zeros from object files.
3080 @cindex absolute section
3081 @item absolute section
3082 Address 0 of this section is always ``relocated'' to runtime address 0.
3083 This is useful if you want to refer to an address that @code{@value{LD}} must
3084 not change when relocating. In this sense we speak of absolute
3085 addresses being ``unrelocatable'': they do not change during relocation.
3087 @cindex undefined section
3088 @item undefined section
3089 This ``section'' is a catch-all for address references to objects not in
3090 the preceding sections.
3091 @c FIXME: ref to some other doc on obj-file formats could go here.
3094 @cindex relocation example
3095 An idealized example of three relocatable sections follows.
3097 The example uses the traditional section names @samp{.text} and @samp{.data}.
3099 Memory addresses are on the horizontal axis.
3103 @c END TEXI2ROFF-KILL
3106 partial program # 1: |ttttt|dddd|00|
3113 partial program # 2: |TTT|DDD|000|
3116 +--+---+-----+--+----+---+-----+~~
3117 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3118 +--+---+-----+--+----+---+-----+~~
3120 addresses: 0 @dots{}
3127 \line{\it Partial program \#1: \hfil}
3128 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3129 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3131 \line{\it Partial program \#2: \hfil}
3132 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3133 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3135 \line{\it linked program: \hfil}
3136 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3137 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3138 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3139 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3141 \line{\it addresses: \hfil}
3145 @c END TEXI2ROFF-KILL
3148 @section Assembler Internal Sections
3150 @cindex internal assembler sections
3151 @cindex sections in messages, internal
3152 These sections are meant only for the internal use of @command{@value{AS}}. They
3153 have no meaning at run-time. You do not really need to know about these
3154 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3155 warning messages, so it might be helpful to have an idea of their
3156 meanings to @command{@value{AS}}. These sections are used to permit the
3157 value of every expression in your assembly language program to be a
3158 section-relative address.
3161 @cindex assembler internal logic error
3162 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3163 An internal assembler logic error has been found. This means there is a
3164 bug in the assembler.
3166 @cindex expr (internal section)
3168 The assembler stores complex expression internally as combinations of
3169 symbols. When it needs to represent an expression as a symbol, it puts
3170 it in the expr section.
3172 @c FIXME item transfer[t] vector preload
3173 @c FIXME item transfer[t] vector postload
3174 @c FIXME item register
3178 @section Sub-Sections
3180 @cindex numbered subsections
3181 @cindex grouping data
3187 fall into two sections: text and data.
3189 You may have separate groups of
3191 data in named sections
3195 data in named sections
3201 that you want to end up near to each other in the object file, even though they
3202 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3203 use @dfn{subsections} for this purpose. Within each section, there can be
3204 numbered subsections with values from 0 to 8192. Objects assembled into the
3205 same subsection go into the object file together with other objects in the same
3206 subsection. For example, a compiler might want to store constants in the text
3207 section, but might not want to have them interspersed with the program being
3208 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3209 section of code being output, and a @samp{.text 1} before each group of
3210 constants being output.
3212 Subsections are optional. If you do not use subsections, everything
3213 goes in subsection number zero.
3216 Each subsection is zero-padded up to a multiple of four bytes.
3217 (Subsections may be padded a different amount on different flavors
3218 of @command{@value{AS}}.)
3222 On the H8/300 platform, each subsection is zero-padded to a word
3223 boundary (two bytes).
3224 The same is true on the Renesas SH.
3227 @c FIXME section padding (alignment)?
3228 @c Rich Pixley says padding here depends on target obj code format; that
3229 @c doesn't seem particularly useful to say without further elaboration,
3230 @c so for now I say nothing about it. If this is a generic BFD issue,
3231 @c these paragraphs might need to vanish from this manual, and be
3232 @c discussed in BFD chapter of binutils (or some such).
3236 Subsections appear in your object file in numeric order, lowest numbered
3237 to highest. (All this to be compatible with other people's assemblers.)
3238 The object file contains no representation of subsections; @code{@value{LD}} and
3239 other programs that manipulate object files see no trace of them.
3240 They just see all your text subsections as a text section, and all your
3241 data subsections as a data section.
3243 To specify which subsection you want subsequent statements assembled
3244 into, use a numeric argument to specify it, in a @samp{.text
3245 @var{expression}} or a @samp{.data @var{expression}} statement.
3248 When generating COFF output, you
3253 can also use an extra subsection
3254 argument with arbitrary named sections: @samp{.section @var{name},
3259 When generating ELF output, you
3264 can also use the @code{.subsection} directive (@pxref{SubSection})
3265 to specify a subsection: @samp{.subsection @var{expression}}.
3267 @var{Expression} should be an absolute expression
3268 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3269 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3270 begins in @code{text 0}. For instance:
3272 .text 0 # The default subsection is text 0 anyway.
3273 .ascii "This lives in the first text subsection. *"
3275 .ascii "But this lives in the second text subsection."
3277 .ascii "This lives in the data section,"
3278 .ascii "in the first data subsection."
3280 .ascii "This lives in the first text section,"
3281 .ascii "immediately following the asterisk (*)."
3284 Each section has a @dfn{location counter} incremented by one for every byte
3285 assembled into that section. Because subsections are merely a convenience
3286 restricted to @command{@value{AS}} there is no concept of a subsection location
3287 counter. There is no way to directly manipulate a location counter---but the
3288 @code{.align} directive changes it, and any label definition captures its
3289 current value. The location counter of the section where statements are being
3290 assembled is said to be the @dfn{active} location counter.
3293 @section bss Section
3296 @cindex common variable storage
3297 The bss section is used for local common variable storage.
3298 You may allocate address space in the bss section, but you may
3299 not dictate data to load into it before your program executes. When
3300 your program starts running, all the contents of the bss
3301 section are zeroed bytes.
3303 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3304 @ref{Lcomm,,@code{.lcomm}}.
3306 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3307 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3310 When assembling for a target which supports multiple sections, such as ELF or
3311 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3312 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3313 section. Typically the section will only contain symbol definitions and
3314 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3321 Symbols are a central concept: the programmer uses symbols to name
3322 things, the linker uses symbols to link, and the debugger uses symbols
3326 @cindex debuggers, and symbol order
3327 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3328 the same order they were declared. This may break some debuggers.
3333 * Setting Symbols:: Giving Symbols Other Values
3334 * Symbol Names:: Symbol Names
3335 * Dot:: The Special Dot Symbol
3336 * Symbol Attributes:: Symbol Attributes
3343 A @dfn{label} is written as a symbol immediately followed by a colon
3344 @samp{:}. The symbol then represents the current value of the
3345 active location counter, and is, for example, a suitable instruction
3346 operand. You are warned if you use the same symbol to represent two
3347 different locations: the first definition overrides any other
3351 On the HPPA, the usual form for a label need not be immediately followed by a
3352 colon, but instead must start in column zero. Only one label may be defined on
3353 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3354 provides a special directive @code{.label} for defining labels more flexibly.
3357 @node Setting Symbols
3358 @section Giving Symbols Other Values
3360 @cindex assigning values to symbols
3361 @cindex symbol values, assigning
3362 A symbol can be given an arbitrary value by writing a symbol, followed
3363 by an equals sign @samp{=}, followed by an expression
3364 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3365 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3366 equals sign @samp{=}@samp{=} here represents an equivalent of the
3367 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3370 Blackfin does not support symbol assignment with @samp{=}.
3374 @section Symbol Names
3376 @cindex symbol names
3377 @cindex names, symbol
3378 @ifclear SPECIAL-SYMS
3379 Symbol names begin with a letter or with one of @samp{._}. On most
3380 machines, you can also use @code{$} in symbol names; exceptions are
3381 noted in @ref{Machine Dependencies}. That character may be followed by any
3382 string of digits, letters, dollar signs (unless otherwise noted for a
3383 particular target machine), and underscores.
3387 Symbol names begin with a letter or with one of @samp{._}. On the
3388 Renesas SH you can also use @code{$} in symbol names. That
3389 character may be followed by any string of digits, letters, dollar signs (save
3390 on the H8/300), and underscores.
3394 Case of letters is significant: @code{foo} is a different symbol name
3397 Each symbol has exactly one name. Each name in an assembly language program
3398 refers to exactly one symbol. You may use that symbol name any number of times
3401 @subheading Local Symbol Names
3403 @cindex local symbol names
3404 @cindex symbol names, local
3405 A local symbol is any symbol beginning with certain local label prefixes.
3406 By default, the local label prefix is @samp{.L} for ELF systems or
3407 @samp{L} for traditional a.out systems, but each target may have its own
3408 set of local label prefixes.
3410 On the HPPA local symbols begin with @samp{L$}.
3413 Local symbols are defined and used within the assembler, but they are
3414 normally not saved in object files. Thus, they are not visible when debugging.
3415 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols:
3416 @option{-L}}) to retain the local symbols in the object files.
3418 @subheading Local Labels
3420 @cindex local labels
3421 @cindex temporary symbol names
3422 @cindex symbol names, temporary
3423 Local labels help compilers and programmers use names temporarily.
3424 They create symbols which are guaranteed to be unique over the entire scope of
3425 the input source code and which can be referred to by a simple notation.
3426 To define a local label, write a label of the form @samp{@b{N}:} (where @b{N}
3427 represents any positive integer). To refer to the most recent previous
3428 definition of that label write @samp{@b{N}b}, using the same number as when
3429 you defined the label. To refer to the next definition of a local label, write
3430 @samp{@b{N}f}---the @samp{b} stands for ``backwards'' and the @samp{f} stands
3433 There is no restriction on how you can use these labels, and you can reuse them
3434 too. So that it is possible to repeatedly define the same local label (using
3435 the same number @samp{@b{N}}), although you can only refer to the most recently
3436 defined local label of that number (for a backwards reference) or the next
3437 definition of a specific local label for a forward reference. It is also worth
3438 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3439 implemented in a slightly more efficient manner than the others.
3450 Which is the equivalent of:
3453 label_1: branch label_3
3454 label_2: branch label_1
3455 label_3: branch label_4
3456 label_4: branch label_3
3459 Local label names are only a notational device. They are immediately
3460 transformed into more conventional symbol names before the assembler uses them.
3461 The symbol names are stored in the symbol table, appear in error messages, and
3462 are optionally emitted to the object file. The names are constructed using
3466 @item @emph{local label prefix}
3467 All local symbols begin with the system-specific local label prefix.
3468 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3469 that start with the local label prefix. These labels are
3470 used for symbols you are never intended to see. If you use the
3471 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3472 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3473 you may use them in debugging.
3476 This is the number that was used in the local label definition. So if the
3477 label is written @samp{55:} then the number is @samp{55}.
3480 This unusual character is included so you do not accidentally invent a symbol
3481 of the same name. The character has ASCII value of @samp{\002} (control-B).
3483 @item @emph{ordinal number}
3484 This is a serial number to keep the labels distinct. The first definition of
3485 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3486 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3487 the number @samp{1} and its 15th definition gets @samp{15} as well.
3490 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3491 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3493 @subheading Dollar Local Labels
3494 @cindex dollar local symbols
3496 @code{@value{AS}} also supports an even more local form of local labels called
3497 dollar labels. These labels go out of scope (i.e., they become undefined) as
3498 soon as a non-local label is defined. Thus they remain valid for only a small
3499 region of the input source code. Normal local labels, by contrast, remain in
3500 scope for the entire file, or until they are redefined by another occurrence of
3501 the same local label.
3503 Dollar labels are defined in exactly the same way as ordinary local labels,
3504 except that they have a dollar sign suffix to their numeric value, e.g.,
3507 They can also be distinguished from ordinary local labels by their transformed
3508 names which use ASCII character @samp{\001} (control-A) as the magic character
3509 to distinguish them from ordinary labels. For example, the fifth definition of
3510 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3513 @section The Special Dot Symbol
3515 @cindex dot (symbol)
3516 @cindex @code{.} (symbol)
3517 @cindex current address
3518 @cindex location counter
3519 The special symbol @samp{.} refers to the current address that
3520 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3521 .long .} defines @code{melvin} to contain its own address.
3522 Assigning a value to @code{.} is treated the same as a @code{.org}
3524 @ifclear no-space-dir
3525 Thus, the expression @samp{.=.+4} is the same as saying
3529 @node Symbol Attributes
3530 @section Symbol Attributes
3532 @cindex symbol attributes
3533 @cindex attributes, symbol
3534 Every symbol has, as well as its name, the attributes ``Value'' and
3535 ``Type''. Depending on output format, symbols can also have auxiliary
3538 The detailed definitions are in @file{a.out.h}.
3541 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3542 all these attributes, and probably won't warn you. This makes the
3543 symbol an externally defined symbol, which is generally what you
3547 * Symbol Value:: Value
3548 * Symbol Type:: Type
3551 * a.out Symbols:: Symbol Attributes: @code{a.out}
3555 * a.out Symbols:: Symbol Attributes: @code{a.out}
3558 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3563 * COFF Symbols:: Symbol Attributes for COFF
3566 * SOM Symbols:: Symbol Attributes for SOM
3573 @cindex value of a symbol
3574 @cindex symbol value
3575 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3576 location in the text, data, bss or absolute sections the value is the
3577 number of addresses from the start of that section to the label.
3578 Naturally for text, data and bss sections the value of a symbol changes
3579 as @code{@value{LD}} changes section base addresses during linking. Absolute
3580 symbols' values do not change during linking: that is why they are
3583 The value of an undefined symbol is treated in a special way. If it is
3584 0 then the symbol is not defined in this assembler source file, and
3585 @code{@value{LD}} tries to determine its value from other files linked into the
3586 same program. You make this kind of symbol simply by mentioning a symbol
3587 name without defining it. A non-zero value represents a @code{.comm}
3588 common declaration. The value is how much common storage to reserve, in
3589 bytes (addresses). The symbol refers to the first address of the
3595 @cindex type of a symbol
3597 The type attribute of a symbol contains relocation (section)
3598 information, any flag settings indicating that a symbol is external, and
3599 (optionally), other information for linkers and debuggers. The exact
3600 format depends on the object-code output format in use.
3605 @c The following avoids a "widow" subsection title. @group would be
3606 @c better if it were available outside examples.
3609 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3611 @cindex @code{b.out} symbol attributes
3612 @cindex symbol attributes, @code{b.out}
3613 These symbol attributes appear only when @command{@value{AS}} is configured for
3614 one of the Berkeley-descended object output formats---@code{a.out} or
3620 @subsection Symbol Attributes: @code{a.out}
3622 @cindex @code{a.out} symbol attributes
3623 @cindex symbol attributes, @code{a.out}
3629 @subsection Symbol Attributes: @code{a.out}
3631 @cindex @code{a.out} symbol attributes
3632 @cindex symbol attributes, @code{a.out}
3636 * Symbol Desc:: Descriptor
3637 * Symbol Other:: Other
3641 @subsubsection Descriptor
3643 @cindex descriptor, of @code{a.out} symbol
3644 This is an arbitrary 16-bit value. You may establish a symbol's
3645 descriptor value by using a @code{.desc} statement
3646 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3647 @command{@value{AS}}.
3650 @subsubsection Other
3652 @cindex other attribute, of @code{a.out} symbol
3653 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3658 @subsection Symbol Attributes for COFF
3660 @cindex COFF symbol attributes
3661 @cindex symbol attributes, COFF
3663 The COFF format supports a multitude of auxiliary symbol attributes;
3664 like the primary symbol attributes, they are set between @code{.def} and
3665 @code{.endef} directives.
3667 @subsubsection Primary Attributes
3669 @cindex primary attributes, COFF symbols
3670 The symbol name is set with @code{.def}; the value and type,
3671 respectively, with @code{.val} and @code{.type}.
3673 @subsubsection Auxiliary Attributes
3675 @cindex auxiliary attributes, COFF symbols
3676 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3677 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3678 table information for COFF.
3683 @subsection Symbol Attributes for SOM
3685 @cindex SOM symbol attributes
3686 @cindex symbol attributes, SOM
3688 The SOM format for the HPPA supports a multitude of symbol attributes set with
3689 the @code{.EXPORT} and @code{.IMPORT} directives.
3691 The attributes are described in @cite{HP9000 Series 800 Assembly
3692 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3693 @code{EXPORT} assembler directive documentation.
3697 @chapter Expressions
3701 @cindex numeric values
3702 An @dfn{expression} specifies an address or numeric value.
3703 Whitespace may precede and/or follow an expression.
3705 The result of an expression must be an absolute number, or else an offset into
3706 a particular section. If an expression is not absolute, and there is not
3707 enough information when @command{@value{AS}} sees the expression to know its
3708 section, a second pass over the source program might be necessary to interpret
3709 the expression---but the second pass is currently not implemented.
3710 @command{@value{AS}} aborts with an error message in this situation.
3713 * Empty Exprs:: Empty Expressions
3714 * Integer Exprs:: Integer Expressions
3718 @section Empty Expressions
3720 @cindex empty expressions
3721 @cindex expressions, empty
3722 An empty expression has no value: it is just whitespace or null.
3723 Wherever an absolute expression is required, you may omit the
3724 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3725 is compatible with other assemblers.
3728 @section Integer Expressions
3730 @cindex integer expressions
3731 @cindex expressions, integer
3732 An @dfn{integer expression} is one or more @emph{arguments} delimited
3733 by @emph{operators}.
3736 * Arguments:: Arguments
3737 * Operators:: Operators
3738 * Prefix Ops:: Prefix Operators
3739 * Infix Ops:: Infix Operators
3743 @subsection Arguments
3745 @cindex expression arguments
3746 @cindex arguments in expressions
3747 @cindex operands in expressions
3748 @cindex arithmetic operands
3749 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3750 contexts arguments are sometimes called ``arithmetic operands''. In
3751 this manual, to avoid confusing them with the ``instruction operands'' of
3752 the machine language, we use the term ``argument'' to refer to parts of
3753 expressions only, reserving the word ``operand'' to refer only to machine
3754 instruction operands.
3756 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3757 @var{section} is one of text, data, bss, absolute,
3758 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3761 Numbers are usually integers.
3763 A number can be a flonum or bignum. In this case, you are warned
3764 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3765 these 32 bits are an integer. You may write integer-manipulating
3766 instructions that act on exotic constants, compatible with other
3769 @cindex subexpressions
3770 Subexpressions are a left parenthesis @samp{(} followed by an integer
3771 expression, followed by a right parenthesis @samp{)}; or a prefix
3772 operator followed by an argument.
3775 @subsection Operators
3777 @cindex operators, in expressions
3778 @cindex arithmetic functions
3779 @cindex functions, in expressions
3780 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3781 operators are followed by an argument. Infix operators appear
3782 between their arguments. Operators may be preceded and/or followed by
3786 @subsection Prefix Operator
3788 @cindex prefix operators
3789 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3790 one argument, which must be absolute.
3792 @c the tex/end tex stuff surrounding this small table is meant to make
3793 @c it align, on the printed page, with the similar table in the next
3794 @c section (which is inside an enumerate).
3796 \global\advance\leftskip by \itemindent
3801 @dfn{Negation}. Two's complement negation.
3803 @dfn{Complementation}. Bitwise not.
3807 \global\advance\leftskip by -\itemindent
3811 @subsection Infix Operators
3813 @cindex infix operators
3814 @cindex operators, permitted arguments
3815 @dfn{Infix operators} take two arguments, one on either side. Operators
3816 have precedence, but operations with equal precedence are performed left
3817 to right. Apart from @code{+} or @option{-}, both arguments must be
3818 absolute, and the result is absolute.
3821 @cindex operator precedence
3822 @cindex precedence of operators
3829 @dfn{Multiplication}.
3832 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3838 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3841 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3845 Intermediate precedence
3850 @dfn{Bitwise Inclusive Or}.
3856 @dfn{Bitwise Exclusive Or}.
3859 @dfn{Bitwise Or Not}.
3866 @cindex addition, permitted arguments
3867 @cindex plus, permitted arguments
3868 @cindex arguments for addition
3870 @dfn{Addition}. If either argument is absolute, the result has the section of
3871 the other argument. You may not add together arguments from different
3874 @cindex subtraction, permitted arguments
3875 @cindex minus, permitted arguments
3876 @cindex arguments for subtraction
3878 @dfn{Subtraction}. If the right argument is absolute, the
3879 result has the section of the left argument.
3880 If both arguments are in the same section, the result is absolute.
3881 You may not subtract arguments from different sections.
3882 @c FIXME is there still something useful to say about undefined - undefined ?
3884 @cindex comparison expressions
3885 @cindex expressions, comparison
3890 @dfn{Is Not Equal To}
3894 @dfn{Is Greater Than}
3896 @dfn{Is Greater Than Or Equal To}
3898 @dfn{Is Less Than Or Equal To}
3900 The comparison operators can be used as infix operators. A true results has a
3901 value of -1 whereas a false result has a value of 0. Note, these operators
3902 perform signed comparisons.
3905 @item Lowest Precedence
3914 These two logical operations can be used to combine the results of sub
3915 expressions. Note, unlike the comparison operators a true result returns a
3916 value of 1 but a false results does still return 0. Also note that the logical
3917 or operator has a slightly lower precedence than logical and.
3922 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3923 address; you can only have a defined section in one of the two arguments.
3926 @chapter Assembler Directives
3928 @cindex directives, machine independent
3929 @cindex pseudo-ops, machine independent
3930 @cindex machine independent directives
3931 All assembler directives have names that begin with a period (@samp{.}).
3932 The rest of the name is letters, usually in lower case.
3934 This chapter discusses directives that are available regardless of the
3935 target machine configuration for the @sc{gnu} assembler.
3937 Some machine configurations provide additional directives.
3938 @xref{Machine Dependencies}.
3941 @ifset machine-directives
3942 @xref{Machine Dependencies}, for additional directives.
3947 * Abort:: @code{.abort}
3949 * ABORT (COFF):: @code{.ABORT}
3952 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3953 * Altmacro:: @code{.altmacro}
3954 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3955 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3956 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3957 * Byte:: @code{.byte @var{expressions}}
3958 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
3959 * Comm:: @code{.comm @var{symbol} , @var{length} }
3960 * Data:: @code{.data @var{subsection}}
3962 * Def:: @code{.def @var{name}}
3965 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3971 * Double:: @code{.double @var{flonums}}
3972 * Eject:: @code{.eject}
3973 * Else:: @code{.else}
3974 * Elseif:: @code{.elseif}
3977 * Endef:: @code{.endef}
3980 * Endfunc:: @code{.endfunc}
3981 * Endif:: @code{.endif}
3982 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3983 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3984 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
3986 * Error:: @code{.error @var{string}}
3987 * Exitm:: @code{.exitm}
3988 * Extern:: @code{.extern}
3989 * Fail:: @code{.fail}
3990 * File:: @code{.file}
3991 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3992 * Float:: @code{.float @var{flonums}}
3993 * Func:: @code{.func}
3994 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3996 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
3997 * Hidden:: @code{.hidden @var{names}}
4000 * hword:: @code{.hword @var{expressions}}
4001 * Ident:: @code{.ident}
4002 * If:: @code{.if @var{absolute expression}}
4003 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4004 * Include:: @code{.include "@var{file}"}
4005 * Int:: @code{.int @var{expressions}}
4007 * Internal:: @code{.internal @var{names}}
4010 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4011 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4012 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4013 * Lflags:: @code{.lflags}
4014 @ifclear no-line-dir
4015 * Line:: @code{.line @var{line-number}}
4018 * Linkonce:: @code{.linkonce [@var{type}]}
4019 * List:: @code{.list}
4020 * Ln:: @code{.ln @var{line-number}}
4021 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4022 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4024 * Local:: @code{.local @var{names}}
4027 * Long:: @code{.long @var{expressions}}
4029 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4032 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4033 * MRI:: @code{.mri @var{val}}
4034 * Noaltmacro:: @code{.noaltmacro}
4035 * Nolist:: @code{.nolist}
4036 * Octa:: @code{.octa @var{bignums}}
4037 * Org:: @code{.org @var{new-lc}, @var{fill}}
4038 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4040 * PopSection:: @code{.popsection}
4041 * Previous:: @code{.previous}
4044 * Print:: @code{.print @var{string}}
4046 * Protected:: @code{.protected @var{names}}
4049 * Psize:: @code{.psize @var{lines}, @var{columns}}
4050 * Purgem:: @code{.purgem @var{name}}
4052 * PushSection:: @code{.pushsection @var{name}}
4055 * Quad:: @code{.quad @var{bignums}}
4056 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4057 * Rept:: @code{.rept @var{count}}
4058 * Sbttl:: @code{.sbttl "@var{subheading}"}
4060 * Scl:: @code{.scl @var{class}}
4063 * Section:: @code{.section @var{name}[, @var{flags}]}
4066 * Set:: @code{.set @var{symbol}, @var{expression}}
4067 * Short:: @code{.short @var{expressions}}
4068 * Single:: @code{.single @var{flonums}}
4070 * Size:: @code{.size [@var{name} , @var{expression}]}
4072 @ifclear no-space-dir
4073 * Skip:: @code{.skip @var{size} , @var{fill}}
4076 * Sleb128:: @code{.sleb128 @var{expressions}}
4077 @ifclear no-space-dir
4078 * Space:: @code{.space @var{size} , @var{fill}}
4081 * Stab:: @code{.stabd, .stabn, .stabs}
4084 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4085 * Struct:: @code{.struct @var{expression}}
4087 * SubSection:: @code{.subsection}
4088 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4092 * Tag:: @code{.tag @var{structname}}
4095 * Text:: @code{.text @var{subsection}}
4096 * Title:: @code{.title "@var{heading}"}
4098 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4101 * Uleb128:: @code{.uleb128 @var{expressions}}
4103 * Val:: @code{.val @var{addr}}
4107 * Version:: @code{.version "@var{string}"}
4108 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4109 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4112 * Warning:: @code{.warning @var{string}}
4113 * Weak:: @code{.weak @var{names}}
4114 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4115 * Word:: @code{.word @var{expressions}}
4116 * Deprecated:: Deprecated Directives
4120 @section @code{.abort}
4122 @cindex @code{abort} directive
4123 @cindex stopping the assembly
4124 This directive stops the assembly immediately. It is for
4125 compatibility with other assemblers. The original idea was that the
4126 assembly language source would be piped into the assembler. If the sender
4127 of the source quit, it could use this directive tells @command{@value{AS}} to
4128 quit also. One day @code{.abort} will not be supported.
4132 @section @code{.ABORT} (COFF)
4134 @cindex @code{ABORT} directive
4135 When producing COFF output, @command{@value{AS}} accepts this directive as a
4136 synonym for @samp{.abort}.
4139 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4145 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4147 @cindex padding the location counter
4148 @cindex @code{align} directive
4149 Pad the location counter (in the current subsection) to a particular storage
4150 boundary. The first expression (which must be absolute) is the alignment
4151 required, as described below.
4153 The second expression (also absolute) gives the fill value to be stored in the
4154 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4155 padding bytes are normally zero. However, on some systems, if the section is
4156 marked as containing code and the fill value is omitted, the space is filled
4157 with no-op instructions.
4159 The third expression is also absolute, and is also optional. If it is present,
4160 it is the maximum number of bytes that should be skipped by this alignment
4161 directive. If doing the alignment would require skipping more bytes than the
4162 specified maximum, then the alignment is not done at all. You can omit the
4163 fill value (the second argument) entirely by simply using two commas after the
4164 required alignment; this can be useful if you want the alignment to be filled
4165 with no-op instructions when appropriate.
4167 The way the required alignment is specified varies from system to system.
4168 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or32,
4169 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4170 alignment request in bytes. For example @samp{.align 8} advances
4171 the location counter until it is a multiple of 8. If the location counter
4172 is already a multiple of 8, no change is needed. For the tic54x, the
4173 first expression is the alignment request in words.
4175 For other systems, including ppc, i386 using a.out format, arm and
4176 strongarm, it is the
4177 number of low-order zero bits the location counter must have after
4178 advancement. For example @samp{.align 3} advances the location
4179 counter until it a multiple of 8. If the location counter is already a
4180 multiple of 8, no change is needed.
4182 This inconsistency is due to the different behaviors of the various
4183 native assemblers for these systems which GAS must emulate.
4184 GAS also provides @code{.balign} and @code{.p2align} directives,
4185 described later, which have a consistent behavior across all
4186 architectures (but are specific to GAS).
4189 @section @code{.altmacro}
4190 Enable alternate macro mode, enabling:
4193 @item LOCAL @var{name} [ , @dots{} ]
4194 One additional directive, @code{LOCAL}, is available. It is used to
4195 generate a string replacement for each of the @var{name} arguments, and
4196 replace any instances of @var{name} in each macro expansion. The
4197 replacement string is unique in the assembly, and different for each
4198 separate macro expansion. @code{LOCAL} allows you to write macros that
4199 define symbols, without fear of conflict between separate macro expansions.
4201 @item String delimiters
4202 You can write strings delimited in these other ways besides
4203 @code{"@var{string}"}:
4206 @item '@var{string}'
4207 You can delimit strings with single-quote characters.
4209 @item <@var{string}>
4210 You can delimit strings with matching angle brackets.
4213 @item single-character string escape
4214 To include any single character literally in a string (even if the
4215 character would otherwise have some special meaning), you can prefix the
4216 character with @samp{!} (an exclamation mark). For example, you can
4217 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4219 @item Expression results as strings
4220 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4221 and use the result as a string.
4225 @section @code{.ascii "@var{string}"}@dots{}
4227 @cindex @code{ascii} directive
4228 @cindex string literals
4229 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4230 separated by commas. It assembles each string (with no automatic
4231 trailing zero byte) into consecutive addresses.
4234 @section @code{.asciz "@var{string}"}@dots{}
4236 @cindex @code{asciz} directive
4237 @cindex zero-terminated strings
4238 @cindex null-terminated strings
4239 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4240 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4243 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4245 @cindex padding the location counter given number of bytes
4246 @cindex @code{balign} directive
4247 Pad the location counter (in the current subsection) to a particular
4248 storage boundary. The first expression (which must be absolute) is the
4249 alignment request in bytes. For example @samp{.balign 8} advances
4250 the location counter until it is a multiple of 8. If the location counter
4251 is already a multiple of 8, no change is needed.
4253 The second expression (also absolute) gives the fill value to be stored in the
4254 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4255 padding bytes are normally zero. However, on some systems, if the section is
4256 marked as containing code and the fill value is omitted, the space is filled
4257 with no-op instructions.
4259 The third expression is also absolute, and is also optional. If it is present,
4260 it is the maximum number of bytes that should be skipped by this alignment
4261 directive. If doing the alignment would require skipping more bytes than the
4262 specified maximum, then the alignment is not done at all. You can omit the
4263 fill value (the second argument) entirely by simply using two commas after the
4264 required alignment; this can be useful if you want the alignment to be filled
4265 with no-op instructions when appropriate.
4267 @cindex @code{balignw} directive
4268 @cindex @code{balignl} directive
4269 The @code{.balignw} and @code{.balignl} directives are variants of the
4270 @code{.balign} directive. The @code{.balignw} directive treats the fill
4271 pattern as a two byte word value. The @code{.balignl} directives treats the
4272 fill pattern as a four byte longword value. For example, @code{.balignw
4273 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4274 filled in with the value 0x368d (the exact placement of the bytes depends upon
4275 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4279 @section @code{.byte @var{expressions}}
4281 @cindex @code{byte} directive
4282 @cindex integers, one byte
4283 @code{.byte} expects zero or more expressions, separated by commas.
4284 Each expression is assembled into the next byte.
4286 @node CFI directives
4287 @section @code{.cfi_sections @var{section_list}}
4288 @cindex @code{cfi_sections} directive
4289 @code{.cfi_sections} may be used to specify whether CFI directives
4290 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4291 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4292 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4293 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4294 directive is not used is @code{.cfi_sections .eh_frame}.
4296 @section @code{.cfi_startproc [simple]}
4297 @cindex @code{cfi_startproc} directive
4298 @code{.cfi_startproc} is used at the beginning of each function that
4299 should have an entry in @code{.eh_frame}. It initializes some internal
4300 data structures. Don't forget to close the function by
4301 @code{.cfi_endproc}.
4303 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4304 it also emits some architecture dependent initial CFI instructions.
4306 @section @code{.cfi_endproc}
4307 @cindex @code{cfi_endproc} directive
4308 @code{.cfi_endproc} is used at the end of a function where it closes its
4309 unwind entry previously opened by
4310 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4312 @section @code{.cfi_personality @var{encoding} [, @var{exp}]}
4313 @code{.cfi_personality} defines personality routine and its encoding.
4314 @var{encoding} must be a constant determining how the personality
4315 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4316 argument is not present, otherwise second argument should be
4317 a constant or a symbol name. When using indirect encodings,
4318 the symbol provided should be the location where personality
4319 can be loaded from, not the personality routine itself.
4320 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4321 no personality routine.
4323 @section @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4324 @code{.cfi_lsda} defines LSDA and its encoding.
4325 @var{encoding} must be a constant determining how the LSDA
4326 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4327 argument is not present, otherwise second argument should be a constant
4328 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4331 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4332 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4333 address from @var{register} and add @var{offset} to it}.
4335 @section @code{.cfi_def_cfa_register @var{register}}
4336 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4337 now on @var{register} will be used instead of the old one. Offset
4340 @section @code{.cfi_def_cfa_offset @var{offset}}
4341 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4342 remains the same, but @var{offset} is new. Note that it is the
4343 absolute offset that will be added to a defined register to compute
4346 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4347 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4348 value that is added/substracted from the previous offset.
4350 @section @code{.cfi_offset @var{register}, @var{offset}}
4351 Previous value of @var{register} is saved at offset @var{offset} from
4354 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4355 Previous value of @var{register} is saved at offset @var{offset} from
4356 the current CFA register. This is transformed to @code{.cfi_offset}
4357 using the known displacement of the CFA register from the CFA.
4358 This is often easier to use, because the number will match the
4359 code it's annotating.
4361 @section @code{.cfi_register @var{register1}, @var{register2}}
4362 Previous value of @var{register1} is saved in register @var{register2}.
4364 @section @code{.cfi_restore @var{register}}
4365 @code{.cfi_restore} says that the rule for @var{register} is now the
4366 same as it was at the beginning of the function, after all initial
4367 instruction added by @code{.cfi_startproc} were executed.
4369 @section @code{.cfi_undefined @var{register}}
4370 From now on the previous value of @var{register} can't be restored anymore.
4372 @section @code{.cfi_same_value @var{register}}
4373 Current value of @var{register} is the same like in the previous frame,
4374 i.e. no restoration needed.
4376 @section @code{.cfi_remember_state},
4377 First save all current rules for all registers by @code{.cfi_remember_state},
4378 then totally screw them up by subsequent @code{.cfi_*} directives and when
4379 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4380 the previous saved state.
4382 @section @code{.cfi_return_column @var{register}}
4383 Change return column @var{register}, i.e. the return address is either
4384 directly in @var{register} or can be accessed by rules for @var{register}.
4386 @section @code{.cfi_signal_frame}
4387 Mark current function as signal trampoline.
4389 @section @code{.cfi_window_save}
4390 SPARC register window has been saved.
4392 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4393 Allows the user to add arbitrary bytes to the unwind info. One
4394 might use this to add OS-specific CFI opcodes, or generic CFI
4395 opcodes that GAS does not yet support.
4397 @section @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4398 The current value of @var{register} is @var{label}. The value of @var{label}
4399 will be encoded in the output file according to @var{encoding}; see the
4400 description of @code{.cfi_personality} for details on this encoding.
4402 The usefulness of equating a register to a fixed label is probably
4403 limited to the return address register. Here, it can be useful to
4404 mark a code segment that has only one return address which is reached
4405 by a direct branch and no copy of the return address exists in memory
4406 or another register.
4409 @section @code{.comm @var{symbol} , @var{length} }
4411 @cindex @code{comm} directive
4412 @cindex symbol, common
4413 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4414 common symbol in one object file may be merged with a defined or common symbol
4415 of the same name in another object file. If @code{@value{LD}} does not see a
4416 definition for the symbol--just one or more common symbols--then it will
4417 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4418 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4419 the same name, and they do not all have the same size, it will allocate space
4420 using the largest size.
4423 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4424 an optional third argument. This is the desired alignment of the symbol,
4425 specified for ELF as a byte boundary (for example, an alignment of 16 means
4426 that the least significant 4 bits of the address should be zero), and for PE
4427 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4428 boundary). The alignment must be an absolute expression, and it must be a
4429 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4430 common symbol, it will use the alignment when placing the symbol. If no
4431 alignment is specified, @command{@value{AS}} will set the alignment to the
4432 largest power of two less than or equal to the size of the symbol, up to a
4433 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4434 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4435 @samp{--section-alignment} option; image file sections in PE are aligned to
4436 multiples of 4096, which is far too large an alignment for ordinary variables.
4437 It is rather the default alignment for (non-debug) sections within object
4438 (@samp{*.o}) files, which are less strictly aligned.}.
4442 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4443 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4447 @section @code{.data @var{subsection}}
4449 @cindex @code{data} directive
4450 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4451 end of the data subsection numbered @var{subsection} (which is an
4452 absolute expression). If @var{subsection} is omitted, it defaults
4457 @section @code{.def @var{name}}
4459 @cindex @code{def} directive
4460 @cindex COFF symbols, debugging
4461 @cindex debugging COFF symbols
4462 Begin defining debugging information for a symbol @var{name}; the
4463 definition extends until the @code{.endef} directive is encountered.
4466 This directive is only observed when @command{@value{AS}} is configured for COFF
4467 format output; when producing @code{b.out}, @samp{.def} is recognized,
4474 @section @code{.desc @var{symbol}, @var{abs-expression}}
4476 @cindex @code{desc} directive
4477 @cindex COFF symbol descriptor
4478 @cindex symbol descriptor, COFF
4479 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4480 to the low 16 bits of an absolute expression.
4483 The @samp{.desc} directive is not available when @command{@value{AS}} is
4484 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4485 object format. For the sake of compatibility, @command{@value{AS}} accepts
4486 it, but produces no output, when configured for COFF.
4492 @section @code{.dim}
4494 @cindex @code{dim} directive
4495 @cindex COFF auxiliary symbol information
4496 @cindex auxiliary symbol information, COFF
4497 This directive is generated by compilers to include auxiliary debugging
4498 information in the symbol table. It is only permitted inside
4499 @code{.def}/@code{.endef} pairs.
4502 @samp{.dim} is only meaningful when generating COFF format output; when
4503 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4509 @section @code{.double @var{flonums}}
4511 @cindex @code{double} directive
4512 @cindex floating point numbers (double)
4513 @code{.double} expects zero or more flonums, separated by commas. It
4514 assembles floating point numbers.
4516 The exact kind of floating point numbers emitted depends on how
4517 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4521 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4522 in @sc{ieee} format.
4527 @section @code{.eject}
4529 @cindex @code{eject} directive
4530 @cindex new page, in listings
4531 @cindex page, in listings
4532 @cindex listing control: new page
4533 Force a page break at this point, when generating assembly listings.
4536 @section @code{.else}
4538 @cindex @code{else} directive
4539 @code{.else} is part of the @command{@value{AS}} support for conditional
4540 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4541 of code to be assembled if the condition for the preceding @code{.if}
4545 @section @code{.elseif}
4547 @cindex @code{elseif} directive
4548 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4549 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4550 @code{.if} block that would otherwise fill the entire @code{.else} section.
4553 @section @code{.end}
4555 @cindex @code{end} directive
4556 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4557 process anything in the file past the @code{.end} directive.
4561 @section @code{.endef}
4563 @cindex @code{endef} directive
4564 This directive flags the end of a symbol definition begun with
4568 @samp{.endef} is only meaningful when generating COFF format output; if
4569 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4570 directive but ignores it.
4575 @section @code{.endfunc}
4576 @cindex @code{endfunc} directive
4577 @code{.endfunc} marks the end of a function specified with @code{.func}.
4580 @section @code{.endif}
4582 @cindex @code{endif} directive
4583 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4584 it marks the end of a block of code that is only assembled
4585 conditionally. @xref{If,,@code{.if}}.
4588 @section @code{.equ @var{symbol}, @var{expression}}
4590 @cindex @code{equ} directive
4591 @cindex assigning values to symbols
4592 @cindex symbols, assigning values to
4593 This directive sets the value of @var{symbol} to @var{expression}.
4594 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4597 The syntax for @code{equ} on the HPPA is
4598 @samp{@var{symbol} .equ @var{expression}}.
4602 The syntax for @code{equ} on the Z80 is
4603 @samp{@var{symbol} equ @var{expression}}.
4604 On the Z80 it is an eror if @var{symbol} is already defined,
4605 but the symbol is not protected from later redefinition.
4606 Compare @ref{Equiv}.
4610 @section @code{.equiv @var{symbol}, @var{expression}}
4611 @cindex @code{equiv} directive
4612 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4613 the assembler will signal an error if @var{symbol} is already defined. Note a
4614 symbol which has been referenced but not actually defined is considered to be
4617 Except for the contents of the error message, this is roughly equivalent to
4624 plus it protects the symbol from later redefinition.
4627 @section @code{.eqv @var{symbol}, @var{expression}}
4628 @cindex @code{eqv} directive
4629 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
4630 evaluate the expression or any part of it immediately. Instead each time
4631 the resulting symbol is used in an expression, a snapshot of its current
4635 @section @code{.err}
4636 @cindex @code{err} directive
4637 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4638 message and, unless the @option{-Z} option was used, it will not generate an
4639 object file. This can be used to signal an error in conditionally compiled code.
4642 @section @code{.error "@var{string}"}
4643 @cindex error directive
4645 Similarly to @code{.err}, this directive emits an error, but you can specify a
4646 string that will be emitted as the error message. If you don't specify the
4647 message, it defaults to @code{".error directive invoked in source file"}.
4648 @xref{Errors, ,Error and Warning Messages}.
4651 .error "This code has not been assembled and tested."
4655 @section @code{.exitm}
4656 Exit early from the current macro definition. @xref{Macro}.
4659 @section @code{.extern}
4661 @cindex @code{extern} directive
4662 @code{.extern} is accepted in the source program---for compatibility
4663 with other assemblers---but it is ignored. @command{@value{AS}} treats
4664 all undefined symbols as external.
4667 @section @code{.fail @var{expression}}
4669 @cindex @code{fail} directive
4670 Generates an error or a warning. If the value of the @var{expression} is 500
4671 or more, @command{@value{AS}} will print a warning message. If the value is less
4672 than 500, @command{@value{AS}} will print an error message. The message will
4673 include the value of @var{expression}. This can occasionally be useful inside
4674 complex nested macros or conditional assembly.
4677 @section @code{.file}
4678 @cindex @code{file} directive
4680 @ifclear no-file-dir
4681 There are two different versions of the @code{.file} directive. Targets
4682 that support DWARF2 line number information use the DWARF2 version of
4683 @code{.file}. Other targets use the default version.
4685 @subheading Default Version
4687 @cindex logical file name
4688 @cindex file name, logical
4689 This version of the @code{.file} directive tells @command{@value{AS}} that we
4690 are about to start a new logical file. The syntax is:
4696 @var{string} is the new file name. In general, the filename is
4697 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4698 to specify an empty file name, you must give the quotes--@code{""}. This
4699 statement may go away in future: it is only recognized to be compatible with
4700 old @command{@value{AS}} programs.
4702 @subheading DWARF2 Version
4705 When emitting DWARF2 line number information, @code{.file} assigns filenames
4706 to the @code{.debug_line} file name table. The syntax is:
4709 .file @var{fileno} @var{filename}
4712 The @var{fileno} operand should be a unique positive integer to use as the
4713 index of the entry in the table. The @var{filename} operand is a C string
4716 The detail of filename indices is exposed to the user because the filename
4717 table is shared with the @code{.debug_info} section of the DWARF2 debugging
4718 information, and thus the user must know the exact indices that table
4722 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4724 @cindex @code{fill} directive
4725 @cindex writing patterns in memory
4726 @cindex patterns, writing in memory
4727 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4728 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4729 may be zero or more. @var{Size} may be zero or more, but if it is
4730 more than 8, then it is deemed to have the value 8, compatible with
4731 other people's assemblers. The contents of each @var{repeat} bytes
4732 is taken from an 8-byte number. The highest order 4 bytes are
4733 zero. The lowest order 4 bytes are @var{value} rendered in the
4734 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4735 Each @var{size} bytes in a repetition is taken from the lowest order
4736 @var{size} bytes of this number. Again, this bizarre behavior is
4737 compatible with other people's assemblers.
4739 @var{size} and @var{value} are optional.
4740 If the second comma and @var{value} are absent, @var{value} is
4741 assumed zero. If the first comma and following tokens are absent,
4742 @var{size} is assumed to be 1.
4745 @section @code{.float @var{flonums}}
4747 @cindex floating point numbers (single)
4748 @cindex @code{float} directive
4749 This directive assembles zero or more flonums, separated by commas. It
4750 has the same effect as @code{.single}.
4752 The exact kind of floating point numbers emitted depends on how
4753 @command{@value{AS}} is configured.
4754 @xref{Machine Dependencies}.
4758 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4759 in @sc{ieee} format.
4764 @section @code{.func @var{name}[,@var{label}]}
4765 @cindex @code{func} directive
4766 @code{.func} emits debugging information to denote function @var{name}, and
4767 is ignored unless the file is assembled with debugging enabled.
4768 Only @samp{--gstabs[+]} is currently supported.
4769 @var{label} is the entry point of the function and if omitted @var{name}
4770 prepended with the @samp{leading char} is used.
4771 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4772 All functions are currently defined to have @code{void} return type.
4773 The function must be terminated with @code{.endfunc}.
4776 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4778 @cindex @code{global} directive
4779 @cindex symbol, making visible to linker
4780 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4781 @var{symbol} in your partial program, its value is made available to
4782 other partial programs that are linked with it. Otherwise,
4783 @var{symbol} takes its attributes from a symbol of the same name
4784 from another file linked into the same program.
4786 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4787 compatibility with other assemblers.
4790 On the HPPA, @code{.global} is not always enough to make it accessible to other
4791 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4792 @xref{HPPA Directives, ,HPPA Assembler Directives}.
4797 @section @code{.gnu_attribute @var{tag},@var{value}}
4798 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
4801 @section @code{.hidden @var{names}}
4803 @cindex @code{hidden} directive
4805 This is one of the ELF visibility directives. The other two are
4806 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4807 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4809 This directive overrides the named symbols default visibility (which is set by
4810 their binding: local, global or weak). The directive sets the visibility to
4811 @code{hidden} which means that the symbols are not visible to other components.
4812 Such symbols are always considered to be @code{protected} as well.
4816 @section @code{.hword @var{expressions}}
4818 @cindex @code{hword} directive
4819 @cindex integers, 16-bit
4820 @cindex numbers, 16-bit
4821 @cindex sixteen bit integers
4822 This expects zero or more @var{expressions}, and emits
4823 a 16 bit number for each.
4826 This directive is a synonym for @samp{.short}; depending on the target
4827 architecture, it may also be a synonym for @samp{.word}.
4831 This directive is a synonym for @samp{.short}.
4834 This directive is a synonym for both @samp{.short} and @samp{.word}.
4839 @section @code{.ident}
4841 @cindex @code{ident} directive
4843 This directive is used by some assemblers to place tags in object files. The
4844 behavior of this directive varies depending on the target. When using the
4845 a.out object file format, @command{@value{AS}} simply accepts the directive for
4846 source-file compatibility with existing assemblers, but does not emit anything
4847 for it. When using COFF, comments are emitted to the @code{.comment} or
4848 @code{.rdata} section, depending on the target. When using ELF, comments are
4849 emitted to the @code{.comment} section.
4852 @section @code{.if @var{absolute expression}}
4854 @cindex conditional assembly
4855 @cindex @code{if} directive
4856 @code{.if} marks the beginning of a section of code which is only
4857 considered part of the source program being assembled if the argument
4858 (which must be an @var{absolute expression}) is non-zero. The end of
4859 the conditional section of code must be marked by @code{.endif}
4860 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4861 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4862 If you have several conditions to check, @code{.elseif} may be used to avoid
4863 nesting blocks if/else within each subsequent @code{.else} block.
4865 The following variants of @code{.if} are also supported:
4867 @cindex @code{ifdef} directive
4868 @item .ifdef @var{symbol}
4869 Assembles the following section of code if the specified @var{symbol}
4870 has been defined. Note a symbol which has been referenced but not yet defined
4871 is considered to be undefined.
4873 @cindex @code{ifb} directive
4874 @item .ifb @var{text}
4875 Assembles the following section of code if the operand is blank (empty).
4877 @cindex @code{ifc} directive
4878 @item .ifc @var{string1},@var{string2}
4879 Assembles the following section of code if the two strings are the same. The
4880 strings may be optionally quoted with single quotes. If they are not quoted,
4881 the first string stops at the first comma, and the second string stops at the
4882 end of the line. Strings which contain whitespace should be quoted. The
4883 string comparison is case sensitive.
4885 @cindex @code{ifeq} directive
4886 @item .ifeq @var{absolute expression}
4887 Assembles the following section of code if the argument is zero.
4889 @cindex @code{ifeqs} directive
4890 @item .ifeqs @var{string1},@var{string2}
4891 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4893 @cindex @code{ifge} directive
4894 @item .ifge @var{absolute expression}
4895 Assembles the following section of code if the argument is greater than or
4898 @cindex @code{ifgt} directive
4899 @item .ifgt @var{absolute expression}
4900 Assembles the following section of code if the argument is greater than zero.
4902 @cindex @code{ifle} directive
4903 @item .ifle @var{absolute expression}
4904 Assembles the following section of code if the argument is less than or equal
4907 @cindex @code{iflt} directive
4908 @item .iflt @var{absolute expression}
4909 Assembles the following section of code if the argument is less than zero.
4911 @cindex @code{ifnb} directive
4912 @item .ifnb @var{text}
4913 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
4914 following section of code if the operand is non-blank (non-empty).
4916 @cindex @code{ifnc} directive
4917 @item .ifnc @var{string1},@var{string2}.
4918 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4919 following section of code if the two strings are not the same.
4921 @cindex @code{ifndef} directive
4922 @cindex @code{ifnotdef} directive
4923 @item .ifndef @var{symbol}
4924 @itemx .ifnotdef @var{symbol}
4925 Assembles the following section of code if the specified @var{symbol}
4926 has not been defined. Both spelling variants are equivalent. Note a symbol
4927 which has been referenced but not yet defined is considered to be undefined.
4929 @cindex @code{ifne} directive
4930 @item .ifne @var{absolute expression}
4931 Assembles the following section of code if the argument is not equal to zero
4932 (in other words, this is equivalent to @code{.if}).
4934 @cindex @code{ifnes} directive
4935 @item .ifnes @var{string1},@var{string2}
4936 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4937 following section of code if the two strings are not the same.
4941 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4943 @cindex @code{incbin} directive
4944 @cindex binary files, including
4945 The @code{incbin} directive includes @var{file} verbatim at the current
4946 location. You can control the search paths used with the @samp{-I} command-line
4947 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4950 The @var{skip} argument skips a number of bytes from the start of the
4951 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4952 read. Note that the data is not aligned in any way, so it is the user's
4953 responsibility to make sure that proper alignment is provided both before and
4954 after the @code{incbin} directive.
4957 @section @code{.include "@var{file}"}
4959 @cindex @code{include} directive
4960 @cindex supporting files, including
4961 @cindex files, including
4962 This directive provides a way to include supporting files at specified
4963 points in your source program. The code from @var{file} is assembled as
4964 if it followed the point of the @code{.include}; when the end of the
4965 included file is reached, assembly of the original file continues. You
4966 can control the search paths used with the @samp{-I} command-line option
4967 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4971 @section @code{.int @var{expressions}}
4973 @cindex @code{int} directive
4974 @cindex integers, 32-bit
4975 Expect zero or more @var{expressions}, of any section, separated by commas.
4976 For each expression, emit a number that, at run time, is the value of that
4977 expression. The byte order and bit size of the number depends on what kind
4978 of target the assembly is for.
4982 On most forms of the H8/300, @code{.int} emits 16-bit
4983 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
4990 @section @code{.internal @var{names}}
4992 @cindex @code{internal} directive
4994 This is one of the ELF visibility directives. The other two are
4995 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
4996 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4998 This directive overrides the named symbols default visibility (which is set by
4999 their binding: local, global or weak). The directive sets the visibility to
5000 @code{internal} which means that the symbols are considered to be @code{hidden}
5001 (i.e., not visible to other components), and that some extra, processor specific
5002 processing must also be performed upon the symbols as well.
5006 @section @code{.irp @var{symbol},@var{values}}@dots{}
5008 @cindex @code{irp} directive
5009 Evaluate a sequence of statements assigning different values to @var{symbol}.
5010 The sequence of statements starts at the @code{.irp} directive, and is
5011 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5012 set to @var{value}, and the sequence of statements is assembled. If no
5013 @var{value} is listed, the sequence of statements is assembled once, with
5014 @var{symbol} set to the null string. To refer to @var{symbol} within the
5015 sequence of statements, use @var{\symbol}.
5017 For example, assembling
5025 is equivalent to assembling
5033 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5036 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5038 @cindex @code{irpc} directive
5039 Evaluate a sequence of statements assigning different values to @var{symbol}.
5040 The sequence of statements starts at the @code{.irpc} directive, and is
5041 terminated by an @code{.endr} directive. For each character in @var{value},
5042 @var{symbol} is set to the character, and the sequence of statements is
5043 assembled. If no @var{value} is listed, the sequence of statements is
5044 assembled once, with @var{symbol} set to the null string. To refer to
5045 @var{symbol} within the sequence of statements, use @var{\symbol}.
5047 For example, assembling
5055 is equivalent to assembling
5063 For some caveats with the spelling of @var{symbol}, see also the discussion
5067 @section @code{.lcomm @var{symbol} , @var{length}}
5069 @cindex @code{lcomm} directive
5070 @cindex local common symbols
5071 @cindex symbols, local common
5072 Reserve @var{length} (an absolute expression) bytes for a local common
5073 denoted by @var{symbol}. The section and value of @var{symbol} are
5074 those of the new local common. The addresses are allocated in the bss
5075 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5076 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5077 not visible to @code{@value{LD}}.
5080 Some targets permit a third argument to be used with @code{.lcomm}. This
5081 argument specifies the desired alignment of the symbol in the bss section.
5085 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5086 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5090 @section @code{.lflags}
5092 @cindex @code{lflags} directive (ignored)
5093 @command{@value{AS}} accepts this directive, for compatibility with other
5094 assemblers, but ignores it.
5096 @ifclear no-line-dir
5098 @section @code{.line @var{line-number}}
5100 @cindex @code{line} directive
5101 @cindex logical line number
5103 Change the logical line number. @var{line-number} must be an absolute
5104 expression. The next line has that logical line number. Therefore any other
5105 statements on the current line (after a statement separator character) are
5106 reported as on logical line number @var{line-number} @minus{} 1. One day
5107 @command{@value{AS}} will no longer support this directive: it is recognized only
5108 for compatibility with existing assembler programs.
5111 Even though this is a directive associated with the @code{a.out} or
5112 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5113 when producing COFF output, and treats @samp{.line} as though it
5114 were the COFF @samp{.ln} @emph{if} it is found outside a
5115 @code{.def}/@code{.endef} pair.
5117 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5118 used by compilers to generate auxiliary symbol information for
5123 @section @code{.linkonce [@var{type}]}
5125 @cindex @code{linkonce} directive
5126 @cindex common sections
5127 Mark the current section so that the linker only includes a single copy of it.
5128 This may be used to include the same section in several different object files,
5129 but ensure that the linker will only include it once in the final output file.
5130 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5131 Duplicate sections are detected based on the section name, so it should be
5134 This directive is only supported by a few object file formats; as of this
5135 writing, the only object file format which supports it is the Portable
5136 Executable format used on Windows NT.
5138 The @var{type} argument is optional. If specified, it must be one of the
5139 following strings. For example:
5143 Not all types may be supported on all object file formats.
5147 Silently discard duplicate sections. This is the default.
5150 Warn if there are duplicate sections, but still keep only one copy.
5153 Warn if any of the duplicates have different sizes.
5156 Warn if any of the duplicates do not have exactly the same contents.
5160 @section @code{.list}
5162 @cindex @code{list} directive
5163 @cindex listing control, turning on
5164 Control (in conjunction with the @code{.nolist} directive) whether or
5165 not assembly listings are generated. These two directives maintain an
5166 internal counter (which is zero initially). @code{.list} increments the
5167 counter, and @code{.nolist} decrements it. Assembly listings are
5168 generated whenever the counter is greater than zero.
5170 By default, listings are disabled. When you enable them (with the
5171 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5172 the initial value of the listing counter is one.
5175 @section @code{.ln @var{line-number}}
5177 @cindex @code{ln} directive
5178 @ifclear no-line-dir
5179 @samp{.ln} is a synonym for @samp{.line}.
5182 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5183 must be an absolute expression. The next line has that logical
5184 line number, so any other statements on the current line (after a
5185 statement separator character @code{;}) are reported as on logical
5186 line number @var{line-number} @minus{} 1.
5189 This directive is accepted, but ignored, when @command{@value{AS}} is
5190 configured for @code{b.out}; its effect is only associated with COFF
5196 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5197 @cindex @code{loc} directive
5198 When emitting DWARF2 line number information,
5199 the @code{.loc} directive will add a row to the @code{.debug_line} line
5200 number matrix corresponding to the immediately following assembly
5201 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5202 arguments will be applied to the @code{.debug_line} state machine before
5205 The @var{options} are a sequence of the following tokens in any order:
5209 This option will set the @code{basic_block} register in the
5210 @code{.debug_line} state machine to @code{true}.
5213 This option will set the @code{prologue_end} register in the
5214 @code{.debug_line} state machine to @code{true}.
5216 @item epilogue_begin
5217 This option will set the @code{epilogue_begin} register in the
5218 @code{.debug_line} state machine to @code{true}.
5220 @item is_stmt @var{value}
5221 This option will set the @code{is_stmt} register in the
5222 @code{.debug_line} state machine to @code{value}, which must be
5225 @item isa @var{value}
5226 This directive will set the @code{isa} register in the @code{.debug_line}
5227 state machine to @var{value}, which must be an unsigned integer.
5229 @item discriminator @var{value}
5230 This directive will set the @code{discriminator} register in the @code{.debug_line}
5231 state machine to @var{value}, which must be an unsigned integer.
5235 @node Loc_mark_labels
5236 @section @code{.loc_mark_labels @var{enable}}
5237 @cindex @code{loc_mark_labels} directive
5238 When emitting DWARF2 line number information,
5239 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5240 to the @code{.debug_line} line number matrix with the @code{basic_block}
5241 register in the state machine set whenever a code label is seen.
5242 The @var{enable} argument should be either 1 or 0, to enable or disable
5243 this function respectively.
5247 @section @code{.local @var{names}}
5249 @cindex @code{local} directive
5250 This directive, which is available for ELF targets, marks each symbol in
5251 the comma-separated list of @code{names} as a local symbol so that it
5252 will not be externally visible. If the symbols do not already exist,
5253 they will be created.
5255 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5256 accept an alignment argument, which is the case for most ELF targets,
5257 the @code{.local} directive can be used in combination with @code{.comm}
5258 (@pxref{Comm}) to define aligned local common data.
5262 @section @code{.long @var{expressions}}
5264 @cindex @code{long} directive
5265 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5268 @c no one seems to know what this is for or whether this description is
5269 @c what it really ought to do
5271 @section @code{.lsym @var{symbol}, @var{expression}}
5273 @cindex @code{lsym} directive
5274 @cindex symbol, not referenced in assembly
5275 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5276 the hash table, ensuring it cannot be referenced by name during the
5277 rest of the assembly. This sets the attributes of the symbol to be
5278 the same as the expression value:
5280 @var{other} = @var{descriptor} = 0
5281 @var{type} = @r{(section of @var{expression})}
5282 @var{value} = @var{expression}
5285 The new symbol is not flagged as external.
5289 @section @code{.macro}
5292 The commands @code{.macro} and @code{.endm} allow you to define macros that
5293 generate assembly output. For example, this definition specifies a macro
5294 @code{sum} that puts a sequence of numbers into memory:
5297 .macro sum from=0, to=5
5306 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5318 @item .macro @var{macname}
5319 @itemx .macro @var{macname} @var{macargs} @dots{}
5320 @cindex @code{macro} directive
5321 Begin the definition of a macro called @var{macname}. If your macro
5322 definition requires arguments, specify their names after the macro name,
5323 separated by commas or spaces. You can qualify the macro argument to
5324 indicate whether all invocations must specify a non-blank value (through
5325 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5326 (through @samp{:@code{vararg}}). You can supply a default value for any
5327 macro argument by following the name with @samp{=@var{deflt}}. You
5328 cannot define two macros with the same @var{macname} unless it has been
5329 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5330 definitions. For example, these are all valid @code{.macro} statements:
5334 Begin the definition of a macro called @code{comm}, which takes no
5337 @item .macro plus1 p, p1
5338 @itemx .macro plus1 p p1
5339 Either statement begins the definition of a macro called @code{plus1},
5340 which takes two arguments; within the macro definition, write
5341 @samp{\p} or @samp{\p1} to evaluate the arguments.
5343 @item .macro reserve_str p1=0 p2
5344 Begin the definition of a macro called @code{reserve_str}, with two
5345 arguments. The first argument has a default value, but not the second.
5346 After the definition is complete, you can call the macro either as
5347 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5348 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5349 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5350 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5352 @item .macro m p1:req, p2=0, p3:vararg
5353 Begin the definition of a macro called @code{m}, with at least three
5354 arguments. The first argument must always have a value specified, but
5355 not the second, which instead has a default value. The third formal
5356 will get assigned all remaining arguments specified at invocation time.
5358 When you call a macro, you can specify the argument values either by
5359 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5360 @samp{sum to=17, from=9}.
5364 Note that since each of the @var{macargs} can be an identifier exactly
5365 as any other one permitted by the target architecture, there may be
5366 occasional problems if the target hand-crafts special meanings to certain
5367 characters when they occur in a special position. For example, if the colon
5368 (@code{:}) is generally permitted to be part of a symbol name, but the
5369 architecture specific code special-cases it when occurring as the final
5370 character of a symbol (to denote a label), then the macro parameter
5371 replacement code will have no way of knowing that and consider the whole
5372 construct (including the colon) an identifier, and check only this
5373 identifier for being the subject to parameter substitution. So for example
5374 this macro definition:
5382 might not work as expected. Invoking @samp{label foo} might not create a label
5383 called @samp{foo} but instead just insert the text @samp{\l:} into the
5384 assembler source, probably generating an error about an unrecognised
5387 Similarly problems might occur with the period character (@samp{.})
5388 which is often allowed inside opcode names (and hence identifier names). So
5389 for example constructing a macro to build an opcode from a base name and a
5390 length specifier like this:
5393 .macro opcode base length
5398 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5399 instruction but instead generate some kind of error as the assembler tries to
5400 interpret the text @samp{\base.\length}.
5402 There are several possible ways around this problem:
5405 @item Insert white space
5406 If it is possible to use white space characters then this is the simplest
5415 @item Use @samp{\()}
5416 The string @samp{\()} can be used to separate the end of a macro argument from
5417 the following text. eg:
5420 .macro opcode base length
5425 @item Use the alternate macro syntax mode
5426 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5427 used as a separator. eg:
5437 Note: this problem of correctly identifying string parameters to pseudo ops
5438 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5439 and @code{.irpc} (@pxref{Irpc}) as well.
5442 @cindex @code{endm} directive
5443 Mark the end of a macro definition.
5446 @cindex @code{exitm} directive
5447 Exit early from the current macro definition.
5449 @cindex number of macros executed
5450 @cindex macros, count executed
5452 @command{@value{AS}} maintains a counter of how many macros it has
5453 executed in this pseudo-variable; you can copy that number to your
5454 output with @samp{\@@}, but @emph{only within a macro definition}.
5456 @item LOCAL @var{name} [ , @dots{} ]
5457 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5458 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5459 @xref{Altmacro,,@code{.altmacro}}.
5463 @section @code{.mri @var{val}}
5465 @cindex @code{mri} directive
5466 @cindex MRI mode, temporarily
5467 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5468 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5469 affects code assembled until the next @code{.mri} directive, or until the end
5470 of the file. @xref{M, MRI mode, MRI mode}.
5473 @section @code{.noaltmacro}
5474 Disable alternate macro mode. @xref{Altmacro}.
5477 @section @code{.nolist}
5479 @cindex @code{nolist} directive
5480 @cindex listing control, turning off
5481 Control (in conjunction with the @code{.list} directive) whether or
5482 not assembly listings are generated. These two directives maintain an
5483 internal counter (which is zero initially). @code{.list} increments the
5484 counter, and @code{.nolist} decrements it. Assembly listings are
5485 generated whenever the counter is greater than zero.
5488 @section @code{.octa @var{bignums}}
5490 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5491 @cindex @code{octa} directive
5492 @cindex integer, 16-byte
5493 @cindex sixteen byte integer
5494 This directive expects zero or more bignums, separated by commas. For each
5495 bignum, it emits a 16-byte integer.
5497 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5498 hence @emph{octa}-word for 16 bytes.
5501 @section @code{.org @var{new-lc} , @var{fill}}
5503 @cindex @code{org} directive
5504 @cindex location counter, advancing
5505 @cindex advancing location counter
5506 @cindex current address, advancing
5507 Advance the location counter of the current section to
5508 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5509 expression with the same section as the current subsection. That is,
5510 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5511 wrong section, the @code{.org} directive is ignored. To be compatible
5512 with former assemblers, if the section of @var{new-lc} is absolute,
5513 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5514 is the same as the current subsection.
5516 @code{.org} may only increase the location counter, or leave it
5517 unchanged; you cannot use @code{.org} to move the location counter
5520 @c double negative used below "not undefined" because this is a specific
5521 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5522 @c section. doc@cygnus.com 18feb91
5523 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5524 may not be undefined. If you really detest this restriction we eagerly await
5525 a chance to share your improved assembler.
5527 Beware that the origin is relative to the start of the section, not
5528 to the start of the subsection. This is compatible with other
5529 people's assemblers.
5531 When the location counter (of the current subsection) is advanced, the
5532 intervening bytes are filled with @var{fill} which should be an
5533 absolute expression. If the comma and @var{fill} are omitted,
5534 @var{fill} defaults to zero.
5537 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5539 @cindex padding the location counter given a power of two
5540 @cindex @code{p2align} directive
5541 Pad the location counter (in the current subsection) to a particular
5542 storage boundary. The first expression (which must be absolute) is the
5543 number of low-order zero bits the location counter must have after
5544 advancement. For example @samp{.p2align 3} advances the location
5545 counter until it a multiple of 8. If the location counter is already a
5546 multiple of 8, no change is needed.
5548 The second expression (also absolute) gives the fill value to be stored in the
5549 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5550 padding bytes are normally zero. However, on some systems, if the section is
5551 marked as containing code and the fill value is omitted, the space is filled
5552 with no-op instructions.
5554 The third expression is also absolute, and is also optional. If it is present,
5555 it is the maximum number of bytes that should be skipped by this alignment
5556 directive. If doing the alignment would require skipping more bytes than the
5557 specified maximum, then the alignment is not done at all. You can omit the
5558 fill value (the second argument) entirely by simply using two commas after the
5559 required alignment; this can be useful if you want the alignment to be filled
5560 with no-op instructions when appropriate.
5562 @cindex @code{p2alignw} directive
5563 @cindex @code{p2alignl} directive
5564 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5565 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5566 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5567 fill pattern as a four byte longword value. For example, @code{.p2alignw
5568 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5569 filled in with the value 0x368d (the exact placement of the bytes depends upon
5570 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5575 @section @code{.popsection}
5577 @cindex @code{popsection} directive
5578 @cindex Section Stack
5579 This is one of the ELF section stack manipulation directives. The others are
5580 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5581 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5584 This directive replaces the current section (and subsection) with the top
5585 section (and subsection) on the section stack. This section is popped off the
5591 @section @code{.previous}
5593 @cindex @code{previous} directive
5594 @cindex Section Stack
5595 This is one of the ELF section stack manipulation directives. The others are
5596 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5597 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5598 (@pxref{PopSection}).
5600 This directive swaps the current section (and subsection) with most recently
5601 referenced section/subsection pair prior to this one. Multiple
5602 @code{.previous} directives in a row will flip between two sections (and their
5603 subsections). For example:
5615 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
5621 # Now in section A subsection 1
5625 # Now in section B subsection 0
5628 # Now in section B subsection 1
5631 # Now in section B subsection 0
5635 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
5636 section B and 0x9abc into subsection 1 of section B.
5638 In terms of the section stack, this directive swaps the current section with
5639 the top section on the section stack.
5643 @section @code{.print @var{string}}
5645 @cindex @code{print} directive
5646 @command{@value{AS}} will print @var{string} on the standard output during
5647 assembly. You must put @var{string} in double quotes.
5651 @section @code{.protected @var{names}}
5653 @cindex @code{protected} directive
5655 This is one of the ELF visibility directives. The other two are
5656 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5658 This directive overrides the named symbols default visibility (which is set by
5659 their binding: local, global or weak). The directive sets the visibility to
5660 @code{protected} which means that any references to the symbols from within the
5661 components that defines them must be resolved to the definition in that
5662 component, even if a definition in another component would normally preempt
5667 @section @code{.psize @var{lines} , @var{columns}}
5669 @cindex @code{psize} directive
5670 @cindex listing control: paper size
5671 @cindex paper size, for listings
5672 Use this directive to declare the number of lines---and, optionally, the
5673 number of columns---to use for each page, when generating listings.
5675 If you do not use @code{.psize}, listings use a default line-count
5676 of 60. You may omit the comma and @var{columns} specification; the
5677 default width is 200 columns.
5679 @command{@value{AS}} generates formfeeds whenever the specified number of
5680 lines is exceeded (or whenever you explicitly request one, using
5683 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5684 those explicitly specified with @code{.eject}.
5687 @section @code{.purgem @var{name}}
5689 @cindex @code{purgem} directive
5690 Undefine the macro @var{name}, so that later uses of the string will not be
5691 expanded. @xref{Macro}.
5695 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
5697 @cindex @code{pushsection} directive
5698 @cindex Section Stack
5699 This is one of the ELF section stack manipulation directives. The others are
5700 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5701 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5704 This directive pushes the current section (and subsection) onto the
5705 top of the section stack, and then replaces the current section and
5706 subsection with @code{name} and @code{subsection}. The optional
5707 @code{flags}, @code{type} and @code{arguments} are treated the same
5708 as in the @code{.section} (@pxref{Section}) directive.
5712 @section @code{.quad @var{bignums}}
5714 @cindex @code{quad} directive
5715 @code{.quad} expects zero or more bignums, separated by commas. For
5716 each bignum, it emits
5718 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5719 warning message; and just takes the lowest order 8 bytes of the bignum.
5720 @cindex eight-byte integer
5721 @cindex integer, 8-byte
5723 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5724 hence @emph{quad}-word for 8 bytes.
5727 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5728 warning message; and just takes the lowest order 16 bytes of the bignum.
5729 @cindex sixteen-byte integer
5730 @cindex integer, 16-byte
5734 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
5736 @cindex @code{reloc} directive
5737 Generate a relocation at @var{offset} of type @var{reloc_name} with value
5738 @var{expression}. If @var{offset} is a number, the relocation is generated in
5739 the current section. If @var{offset} is an expression that resolves to a
5740 symbol plus offset, the relocation is generated in the given symbol's section.
5741 @var{expression}, if present, must resolve to a symbol plus addend or to an
5742 absolute value, but note that not all targets support an addend. e.g. ELF REL
5743 targets such as i386 store an addend in the section contents rather than in the
5744 relocation. This low level interface does not support addends stored in the
5748 @section @code{.rept @var{count}}
5750 @cindex @code{rept} directive
5751 Repeat the sequence of lines between the @code{.rept} directive and the next
5752 @code{.endr} directive @var{count} times.
5754 For example, assembling
5762 is equivalent to assembling
5771 @section @code{.sbttl "@var{subheading}"}
5773 @cindex @code{sbttl} directive
5774 @cindex subtitles for listings
5775 @cindex listing control: subtitle
5776 Use @var{subheading} as the title (third line, immediately after the
5777 title line) when generating assembly listings.
5779 This directive affects subsequent pages, as well as the current page if
5780 it appears within ten lines of the top of a page.
5784 @section @code{.scl @var{class}}
5786 @cindex @code{scl} directive
5787 @cindex symbol storage class (COFF)
5788 @cindex COFF symbol storage class
5789 Set the storage-class value for a symbol. This directive may only be
5790 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5791 whether a symbol is static or external, or it may record further
5792 symbolic debugging information.
5795 The @samp{.scl} directive is primarily associated with COFF output; when
5796 configured to generate @code{b.out} output format, @command{@value{AS}}
5797 accepts this directive but ignores it.
5803 @section @code{.section @var{name}}
5805 @cindex named section
5806 Use the @code{.section} directive to assemble the following code into a section
5809 This directive is only supported for targets that actually support arbitrarily
5810 named sections; on @code{a.out} targets, for example, it is not accepted, even
5811 with a standard @code{a.out} section name.
5815 @c only print the extra heading if both COFF and ELF are set
5816 @subheading COFF Version
5819 @cindex @code{section} directive (COFF version)
5820 For COFF targets, the @code{.section} directive is used in one of the following
5824 .section @var{name}[, "@var{flags}"]
5825 .section @var{name}[, @var{subsection}]
5828 If the optional argument is quoted, it is taken as flags to use for the
5829 section. Each flag is a single character. The following flags are recognized:
5832 bss section (uninitialized data)
5834 section is not loaded
5844 shared section (meaningful for PE targets)
5846 ignored. (For compatibility with the ELF version)
5848 section is not readable (meaningful for PE targets)
5850 single-digit power-of-two section alignment (GNU extension)
5853 If no flags are specified, the default flags depend upon the section name. If
5854 the section name is not recognized, the default will be for the section to be
5855 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
5856 from the section, rather than adding them, so if they are used on their own it
5857 will be as if no flags had been specified at all.
5859 If the optional argument to the @code{.section} directive is not quoted, it is
5860 taken as a subsection number (@pxref{Sub-Sections}).
5865 @c only print the extra heading if both COFF and ELF are set
5866 @subheading ELF Version
5869 @cindex Section Stack
5870 This is one of the ELF section stack manipulation directives. The others are
5871 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
5872 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
5873 @code{.previous} (@pxref{Previous}).
5875 @cindex @code{section} directive (ELF version)
5876 For ELF targets, the @code{.section} directive is used like this:
5879 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
5882 The optional @var{flags} argument is a quoted string which may contain any
5883 combination of the following characters:
5886 section is allocatable
5888 section is excluded from executable and shared library.
5892 section is executable
5894 section is mergeable
5896 section contains zero terminated strings
5898 section is a member of a section group
5900 section is used for thread-local-storage
5902 section is a member of the previously-current section's group, if any
5905 The optional @var{type} argument may contain one of the following constants:
5908 section contains data
5910 section does not contain data (i.e., section only occupies space)
5912 section contains data which is used by things other than the program
5914 section contains an array of pointers to init functions
5916 section contains an array of pointers to finish functions
5917 @item @@preinit_array
5918 section contains an array of pointers to pre-init functions
5921 Many targets only support the first three section types.
5923 Note on targets where the @code{@@} character is the start of a comment (eg
5924 ARM) then another character is used instead. For example the ARM port uses the
5927 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
5928 be specified as well as an extra argument---@var{entsize}---like this:
5931 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
5934 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
5935 constants, each @var{entsize} octets long. Sections with both @code{M} and
5936 @code{S} must contain zero terminated strings where each character is
5937 @var{entsize} bytes long. The linker may remove duplicates within sections with
5938 the same name, same entity size and same flags. @var{entsize} must be an
5939 absolute expression. For sections with both @code{M} and @code{S}, a string
5940 which is a suffix of a larger string is considered a duplicate. Thus
5941 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
5942 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
5944 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
5945 be present along with an additional field like this:
5948 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
5951 The @var{GroupName} field specifies the name of the section group to which this
5952 particular section belongs. The optional linkage field can contain:
5955 indicates that only one copy of this section should be retained
5960 Note: if both the @var{M} and @var{G} flags are present then the fields for
5961 the Merge flag should come first, like this:
5964 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
5967 If @var{flags} contains the @code{?} symbol then it may not also contain the
5968 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
5969 present. Instead, @code{?} says to consider the section that's current before
5970 this directive. If that section used @code{G}, then the new section will use
5971 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
5972 If not, then the @code{?} symbol has no effect.
5974 If no flags are specified, the default flags depend upon the section name. If
5975 the section name is not recognized, the default will be for the section to have
5976 none of the above flags: it will not be allocated in memory, nor writable, nor
5977 executable. The section will contain data.
5979 For ELF targets, the assembler supports another type of @code{.section}
5980 directive for compatibility with the Solaris assembler:
5983 .section "@var{name}"[, @var{flags}...]
5986 Note that the section name is quoted. There may be a sequence of comma
5990 section is allocatable
5994 section is executable
5996 section is excluded from executable and shared library.
5998 section is used for thread local storage
6001 This directive replaces the current section and subsection. See the
6002 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6003 some examples of how this directive and the other section stack directives
6009 @section @code{.set @var{symbol}, @var{expression}}
6011 @cindex @code{set} directive
6012 @cindex symbol value, setting
6013 Set the value of @var{symbol} to @var{expression}. This
6014 changes @var{symbol}'s value and type to conform to
6015 @var{expression}. If @var{symbol} was flagged as external, it remains
6016 flagged (@pxref{Symbol Attributes}).
6018 You may @code{.set} a symbol many times in the same assembly.
6020 If you @code{.set} a global symbol, the value stored in the object
6021 file is the last value stored into it.
6024 On Z80 @code{set} is a real instruction, use
6025 @samp{@var{symbol} defl @var{expression}} instead.
6029 @section @code{.short @var{expressions}}
6031 @cindex @code{short} directive
6033 @code{.short} is normally the same as @samp{.word}.
6034 @xref{Word,,@code{.word}}.
6036 In some configurations, however, @code{.short} and @code{.word} generate
6037 numbers of different lengths. @xref{Machine Dependencies}.
6041 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6044 This expects zero or more @var{expressions}, and emits
6045 a 16 bit number for each.
6050 @section @code{.single @var{flonums}}
6052 @cindex @code{single} directive
6053 @cindex floating point numbers (single)
6054 This directive assembles zero or more flonums, separated by commas. It
6055 has the same effect as @code{.float}.
6057 The exact kind of floating point numbers emitted depends on how
6058 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6062 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6063 numbers in @sc{ieee} format.
6069 @section @code{.size}
6071 This directive is used to set the size associated with a symbol.
6075 @c only print the extra heading if both COFF and ELF are set
6076 @subheading COFF Version
6079 @cindex @code{size} directive (COFF version)
6080 For COFF targets, the @code{.size} directive is only permitted inside
6081 @code{.def}/@code{.endef} pairs. It is used like this:
6084 .size @var{expression}
6088 @samp{.size} is only meaningful when generating COFF format output; when
6089 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6096 @c only print the extra heading if both COFF and ELF are set
6097 @subheading ELF Version
6100 @cindex @code{size} directive (ELF version)
6101 For ELF targets, the @code{.size} directive is used like this:
6104 .size @var{name} , @var{expression}
6107 This directive sets the size associated with a symbol @var{name}.
6108 The size in bytes is computed from @var{expression} which can make use of label
6109 arithmetic. This directive is typically used to set the size of function
6114 @ifclear no-space-dir
6116 @section @code{.skip @var{size} , @var{fill}}
6118 @cindex @code{skip} directive
6119 @cindex filling memory
6120 This directive emits @var{size} bytes, each of value @var{fill}. Both
6121 @var{size} and @var{fill} are absolute expressions. If the comma and
6122 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6127 @section @code{.sleb128 @var{expressions}}
6129 @cindex @code{sleb128} directive
6130 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6131 compact, variable length representation of numbers used by the DWARF
6132 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6134 @ifclear no-space-dir
6136 @section @code{.space @var{size} , @var{fill}}
6138 @cindex @code{space} directive
6139 @cindex filling memory
6140 This directive emits @var{size} bytes, each of value @var{fill}. Both
6141 @var{size} and @var{fill} are absolute expressions. If the comma
6142 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6147 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6148 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6149 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6150 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6158 @section @code{.stabd, .stabn, .stabs}
6160 @cindex symbolic debuggers, information for
6161 @cindex @code{stab@var{x}} directives
6162 There are three directives that begin @samp{.stab}.
6163 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6164 The symbols are not entered in the @command{@value{AS}} hash table: they
6165 cannot be referenced elsewhere in the source file.
6166 Up to five fields are required:
6170 This is the symbol's name. It may contain any character except
6171 @samp{\000}, so is more general than ordinary symbol names. Some
6172 debuggers used to code arbitrarily complex structures into symbol names
6176 An absolute expression. The symbol's type is set to the low 8 bits of
6177 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6178 and debuggers choke on silly bit patterns.
6181 An absolute expression. The symbol's ``other'' attribute is set to the
6182 low 8 bits of this expression.
6185 An absolute expression. The symbol's descriptor is set to the low 16
6186 bits of this expression.
6189 An absolute expression which becomes the symbol's value.
6192 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6193 or @code{.stabs} statement, the symbol has probably already been created;
6194 you get a half-formed symbol in your object file. This is
6195 compatible with earlier assemblers!
6198 @cindex @code{stabd} directive
6199 @item .stabd @var{type} , @var{other} , @var{desc}
6201 The ``name'' of the symbol generated is not even an empty string.
6202 It is a null pointer, for compatibility. Older assemblers used a
6203 null pointer so they didn't waste space in object files with empty
6206 The symbol's value is set to the location counter,
6207 relocatably. When your program is linked, the value of this symbol
6208 is the address of the location counter when the @code{.stabd} was
6211 @cindex @code{stabn} directive
6212 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6213 The name of the symbol is set to the empty string @code{""}.
6215 @cindex @code{stabs} directive
6216 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6217 All five fields are specified.
6223 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6224 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6226 @cindex string, copying to object file
6227 @cindex string8, copying to object file
6228 @cindex string16, copying to object file
6229 @cindex string32, copying to object file
6230 @cindex string64, copying to object file
6231 @cindex @code{string} directive
6232 @cindex @code{string8} directive
6233 @cindex @code{string16} directive
6234 @cindex @code{string32} directive
6235 @cindex @code{string64} directive
6237 Copy the characters in @var{str} to the object file. You may specify more than
6238 one string to copy, separated by commas. Unless otherwise specified for a
6239 particular machine, the assembler marks the end of each string with a 0 byte.
6240 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6242 The variants @code{string16}, @code{string32} and @code{string64} differ from
6243 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6244 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6245 are stored in target endianness byte order.
6251 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6252 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6257 @section @code{.struct @var{expression}}
6259 @cindex @code{struct} directive
6260 Switch to the absolute section, and set the section offset to @var{expression},
6261 which must be an absolute expression. You might use this as follows:
6270 This would define the symbol @code{field1} to have the value 0, the symbol
6271 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6272 value 8. Assembly would be left in the absolute section, and you would need to
6273 use a @code{.section} directive of some sort to change to some other section
6274 before further assembly.
6278 @section @code{.subsection @var{name}}
6280 @cindex @code{subsection} directive
6281 @cindex Section Stack
6282 This is one of the ELF section stack manipulation directives. The others are
6283 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6284 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6287 This directive replaces the current subsection with @code{name}. The current
6288 section is not changed. The replaced subsection is put onto the section stack
6289 in place of the then current top of stack subsection.
6294 @section @code{.symver}
6295 @cindex @code{symver} directive
6296 @cindex symbol versioning
6297 @cindex versions of symbols
6298 Use the @code{.symver} directive to bind symbols to specific version nodes
6299 within a source file. This is only supported on ELF platforms, and is
6300 typically used when assembling files to be linked into a shared library.
6301 There are cases where it may make sense to use this in objects to be bound
6302 into an application itself so as to override a versioned symbol from a
6305 For ELF targets, the @code{.symver} directive can be used like this:
6307 .symver @var{name}, @var{name2@@nodename}
6309 If the symbol @var{name} is defined within the file
6310 being assembled, the @code{.symver} directive effectively creates a symbol
6311 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6312 just don't try and create a regular alias is that the @var{@@} character isn't
6313 permitted in symbol names. The @var{name2} part of the name is the actual name
6314 of the symbol by which it will be externally referenced. The name @var{name}
6315 itself is merely a name of convenience that is used so that it is possible to
6316 have definitions for multiple versions of a function within a single source
6317 file, and so that the compiler can unambiguously know which version of a
6318 function is being mentioned. The @var{nodename} portion of the alias should be
6319 the name of a node specified in the version script supplied to the linker when
6320 building a shared library. If you are attempting to override a versioned
6321 symbol from a shared library, then @var{nodename} should correspond to the
6322 nodename of the symbol you are trying to override.
6324 If the symbol @var{name} is not defined within the file being assembled, all
6325 references to @var{name} will be changed to @var{name2@@nodename}. If no
6326 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6329 Another usage of the @code{.symver} directive is:
6331 .symver @var{name}, @var{name2@@@@nodename}
6333 In this case, the symbol @var{name} must exist and be defined within
6334 the file being assembled. It is similar to @var{name2@@nodename}. The
6335 difference is @var{name2@@@@nodename} will also be used to resolve
6336 references to @var{name2} by the linker.
6338 The third usage of the @code{.symver} directive is:
6340 .symver @var{name}, @var{name2@@@@@@nodename}
6342 When @var{name} is not defined within the
6343 file being assembled, it is treated as @var{name2@@nodename}. When
6344 @var{name} is defined within the file being assembled, the symbol
6345 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6350 @section @code{.tag @var{structname}}
6352 @cindex COFF structure debugging
6353 @cindex structure debugging, COFF
6354 @cindex @code{tag} directive
6355 This directive is generated by compilers to include auxiliary debugging
6356 information in the symbol table. It is only permitted inside
6357 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6358 definitions in the symbol table with instances of those structures.
6361 @samp{.tag} is only used when generating COFF format output; when
6362 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6368 @section @code{.text @var{subsection}}
6370 @cindex @code{text} directive
6371 Tells @command{@value{AS}} to assemble the following statements onto the end of
6372 the text subsection numbered @var{subsection}, which is an absolute
6373 expression. If @var{subsection} is omitted, subsection number zero
6377 @section @code{.title "@var{heading}"}
6379 @cindex @code{title} directive
6380 @cindex listing control: title line
6381 Use @var{heading} as the title (second line, immediately after the
6382 source file name and pagenumber) when generating assembly listings.
6384 This directive affects subsequent pages, as well as the current page if
6385 it appears within ten lines of the top of a page.
6389 @section @code{.type}
6391 This directive is used to set the type of a symbol.
6395 @c only print the extra heading if both COFF and ELF are set
6396 @subheading COFF Version
6399 @cindex COFF symbol type
6400 @cindex symbol type, COFF
6401 @cindex @code{type} directive (COFF version)
6402 For COFF targets, this directive is permitted only within
6403 @code{.def}/@code{.endef} pairs. It is used like this:
6409 This records the integer @var{int} as the type attribute of a symbol table
6413 @samp{.type} is associated only with COFF format output; when
6414 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6415 directive but ignores it.
6421 @c only print the extra heading if both COFF and ELF are set
6422 @subheading ELF Version
6425 @cindex ELF symbol type
6426 @cindex symbol type, ELF
6427 @cindex @code{type} directive (ELF version)
6428 For ELF targets, the @code{.type} directive is used like this:
6431 .type @var{name} , @var{type description}
6434 This sets the type of symbol @var{name} to be either a
6435 function symbol or an object symbol. There are five different syntaxes
6436 supported for the @var{type description} field, in order to provide
6437 compatibility with various other assemblers.
6439 Because some of the characters used in these syntaxes (such as @samp{@@} and
6440 @samp{#}) are comment characters for some architectures, some of the syntaxes
6441 below do not work on all architectures. The first variant will be accepted by
6442 the GNU assembler on all architectures so that variant should be used for
6443 maximum portability, if you do not need to assemble your code with other
6446 The syntaxes supported are:
6449 .type <name> STT_<TYPE_IN_UPPER_CASE>
6450 .type <name>,#<type>
6451 .type <name>,@@<type>
6452 .type <name>,%<type>
6453 .type <name>,"<type>"
6456 The types supported are:
6461 Mark the symbol as being a function name.
6464 @itemx gnu_indirect_function
6465 Mark the symbol as an indirect function when evaluated during reloc
6466 processing. (This is only supported on Linux targeted assemblers).
6470 Mark the symbol as being a data object.
6474 Mark the symbol as being a thead-local data object.
6478 Mark the symbol as being a common data object.
6482 Does not mark the symbol in any way. It is supported just for completeness.
6484 @item gnu_unique_object
6485 Marks the symbol as being a globally unique data object. The dynamic linker
6486 will make sure that in the entire process there is just one symbol with this
6487 name and type in use. (This is only supported on Linux targeted assemblers).
6491 Note: Some targets support extra types in addition to those listed above.
6497 @section @code{.uleb128 @var{expressions}}
6499 @cindex @code{uleb128} directive
6500 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6501 compact, variable length representation of numbers used by the DWARF
6502 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6506 @section @code{.val @var{addr}}
6508 @cindex @code{val} directive
6509 @cindex COFF value attribute
6510 @cindex value attribute, COFF
6511 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6512 records the address @var{addr} as the value attribute of a symbol table
6516 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6517 configured for @code{b.out}, it accepts this directive but ignores it.
6523 @section @code{.version "@var{string}"}
6525 @cindex @code{version} directive
6526 This directive creates a @code{.note} section and places into it an ELF
6527 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6532 @section @code{.vtable_entry @var{table}, @var{offset}}
6534 @cindex @code{vtable_entry} directive
6535 This directive finds or creates a symbol @code{table} and creates a
6536 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6539 @section @code{.vtable_inherit @var{child}, @var{parent}}
6541 @cindex @code{vtable_inherit} directive
6542 This directive finds the symbol @code{child} and finds or creates the symbol
6543 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6544 parent whose addend is the value of the child symbol. As a special case the
6545 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6549 @section @code{.warning "@var{string}"}
6550 @cindex warning directive
6551 Similar to the directive @code{.error}
6552 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6555 @section @code{.weak @var{names}}
6557 @cindex @code{weak} directive
6558 This directive sets the weak attribute on the comma separated list of symbol
6559 @code{names}. If the symbols do not already exist, they will be created.
6561 On COFF targets other than PE, weak symbols are a GNU extension. This
6562 directive sets the weak attribute on the comma separated list of symbol
6563 @code{names}. If the symbols do not already exist, they will be created.
6565 On the PE target, weak symbols are supported natively as weak aliases.
6566 When a weak symbol is created that is not an alias, GAS creates an
6567 alternate symbol to hold the default value.
6570 @section @code{.weakref @var{alias}, @var{target}}
6572 @cindex @code{weakref} directive
6573 This directive creates an alias to the target symbol that enables the symbol to
6574 be referenced with weak-symbol semantics, but without actually making it weak.
6575 If direct references or definitions of the symbol are present, then the symbol
6576 will not be weak, but if all references to it are through weak references, the
6577 symbol will be marked as weak in the symbol table.
6579 The effect is equivalent to moving all references to the alias to a separate
6580 assembly source file, renaming the alias to the symbol in it, declaring the
6581 symbol as weak there, and running a reloadable link to merge the object files
6582 resulting from the assembly of the new source file and the old source file that
6583 had the references to the alias removed.
6585 The alias itself never makes to the symbol table, and is entirely handled
6586 within the assembler.
6589 @section @code{.word @var{expressions}}
6591 @cindex @code{word} directive
6592 This directive expects zero or more @var{expressions}, of any section,
6593 separated by commas.
6596 For each expression, @command{@value{AS}} emits a 32-bit number.
6599 For each expression, @command{@value{AS}} emits a 16-bit number.
6604 The size of the number emitted, and its byte order,
6605 depend on what target computer the assembly is for.
6608 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
6609 @c happen---32-bit addressability, period; no long/short jumps.
6610 @ifset DIFF-TBL-KLUGE
6611 @cindex difference tables altered
6612 @cindex altered difference tables
6614 @emph{Warning: Special Treatment to support Compilers}
6618 Machines with a 32-bit address space, but that do less than 32-bit
6619 addressing, require the following special treatment. If the machine of
6620 interest to you does 32-bit addressing (or doesn't require it;
6621 @pxref{Machine Dependencies}), you can ignore this issue.
6624 In order to assemble compiler output into something that works,
6625 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6626 Directives of the form @samp{.word sym1-sym2} are often emitted by
6627 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6628 directive of the form @samp{.word sym1-sym2}, and the difference between
6629 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6630 creates a @dfn{secondary jump table}, immediately before the next label.
6631 This secondary jump table is preceded by a short-jump to the
6632 first byte after the secondary table. This short-jump prevents the flow
6633 of control from accidentally falling into the new table. Inside the
6634 table is a long-jump to @code{sym2}. The original @samp{.word}
6635 contains @code{sym1} minus the address of the long-jump to
6638 If there were several occurrences of @samp{.word sym1-sym2} before the
6639 secondary jump table, all of them are adjusted. If there was a
6640 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6641 long-jump to @code{sym4} is included in the secondary jump table,
6642 and the @code{.word} directives are adjusted to contain @code{sym3}
6643 minus the address of the long-jump to @code{sym4}; and so on, for as many
6644 entries in the original jump table as necessary.
6647 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6648 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6649 assembly language programmers.
6652 @c end DIFF-TBL-KLUGE
6655 @section Deprecated Directives
6657 @cindex deprecated directives
6658 @cindex obsolescent directives
6659 One day these directives won't work.
6660 They are included for compatibility with older assemblers.
6667 @node Object Attributes
6668 @chapter Object Attributes
6669 @cindex object attributes
6671 @command{@value{AS}} assembles source files written for a specific architecture
6672 into object files for that architecture. But not all object files are alike.
6673 Many architectures support incompatible variations. For instance, floating
6674 point arguments might be passed in floating point registers if the object file
6675 requires hardware floating point support---or floating point arguments might be
6676 passed in integer registers if the object file supports processors with no
6677 hardware floating point unit. Or, if two objects are built for different
6678 generations of the same architecture, the combination may require the
6679 newer generation at run-time.
6681 This information is useful during and after linking. At link time,
6682 @command{@value{LD}} can warn about incompatible object files. After link
6683 time, tools like @command{gdb} can use it to process the linked file
6686 Compatibility information is recorded as a series of object attributes. Each
6687 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
6688 string, and indicates who sets the meaning of the tag. The tag is an integer,
6689 and indicates what property the attribute describes. The value may be a string
6690 or an integer, and indicates how the property affects this object. Missing
6691 attributes are the same as attributes with a zero value or empty string value.
6693 Object attributes were developed as part of the ABI for the ARM Architecture.
6694 The file format is documented in @cite{ELF for the ARM Architecture}.
6697 * GNU Object Attributes:: @sc{gnu} Object Attributes
6698 * Defining New Object Attributes:: Defining New Object Attributes
6701 @node GNU Object Attributes
6702 @section @sc{gnu} Object Attributes
6704 The @code{.gnu_attribute} directive records an object attribute
6705 with vendor @samp{gnu}.
6707 Except for @samp{Tag_compatibility}, which has both an integer and a string for
6708 its value, @sc{gnu} attributes have a string value if the tag number is odd and
6709 an integer value if the tag number is even. The second bit (@code{@var{tag} &
6710 2} is set for architecture-independent attributes and clear for
6711 architecture-dependent ones.
6713 @subsection Common @sc{gnu} attributes
6715 These attributes are valid on all architectures.
6718 @item Tag_compatibility (32)
6719 The compatibility attribute takes an integer flag value and a vendor name. If
6720 the flag value is 0, the file is compatible with other toolchains. If it is 1,
6721 then the file is only compatible with the named toolchain. If it is greater
6722 than 1, the file can only be processed by other toolchains under some private
6723 arrangement indicated by the flag value and the vendor name.
6726 @subsection MIPS Attributes
6729 @item Tag_GNU_MIPS_ABI_FP (4)
6730 The floating-point ABI used by this object file. The value will be:
6734 0 for files not affected by the floating-point ABI.
6736 1 for files using the hardware floating-point with a standard double-precision
6739 2 for files using the hardware floating-point ABI with a single-precision FPU.
6741 3 for files using the software floating-point ABI.
6743 4 for files using the hardware floating-point ABI with 64-bit wide
6744 double-precision floating-point registers and 32-bit wide general
6749 @subsection PowerPC Attributes
6752 @item Tag_GNU_Power_ABI_FP (4)
6753 The floating-point ABI used by this object file. The value will be:
6757 0 for files not affected by the floating-point ABI.
6759 1 for files using double-precision hardware floating-point ABI.
6761 2 for files using the software floating-point ABI.
6763 3 for files using single-precision hardware floating-point ABI.
6766 @item Tag_GNU_Power_ABI_Vector (8)
6767 The vector ABI used by this object file. The value will be:
6771 0 for files not affected by the vector ABI.
6773 1 for files using general purpose registers to pass vectors.
6775 2 for files using AltiVec registers to pass vectors.
6777 3 for files using SPE registers to pass vectors.
6781 @node Defining New Object Attributes
6782 @section Defining New Object Attributes
6784 If you want to define a new @sc{gnu} object attribute, here are the places you
6785 will need to modify. New attributes should be discussed on the @samp{binutils}
6790 This manual, which is the official register of attributes.
6792 The header for your architecture @file{include/elf}, to define the tag.
6794 The @file{bfd} support file for your architecture, to merge the attribute
6795 and issue any appropriate link warnings.
6797 Test cases in @file{ld/testsuite} for merging and link warnings.
6799 @file{binutils/readelf.c} to display your attribute.
6801 GCC, if you want the compiler to mark the attribute automatically.
6807 @node Machine Dependencies
6808 @chapter Machine Dependent Features
6810 @cindex machine dependencies
6811 The machine instruction sets are (almost by definition) different on
6812 each machine where @command{@value{AS}} runs. Floating point representations
6813 vary as well, and @command{@value{AS}} often supports a few additional
6814 directives or command-line options for compatibility with other
6815 assemblers on a particular platform. Finally, some versions of
6816 @command{@value{AS}} support special pseudo-instructions for branch
6819 This chapter discusses most of these differences, though it does not
6820 include details on any machine's instruction set. For details on that
6821 subject, see the hardware manufacturer's manual.
6825 * Alpha-Dependent:: Alpha Dependent Features
6828 * ARC-Dependent:: ARC Dependent Features
6831 * ARM-Dependent:: ARM Dependent Features
6834 * AVR-Dependent:: AVR Dependent Features
6837 * Blackfin-Dependent:: Blackfin Dependent Features
6840 * CR16-Dependent:: CR16 Dependent Features
6843 * CRIS-Dependent:: CRIS Dependent Features
6846 * D10V-Dependent:: D10V Dependent Features
6849 * D30V-Dependent:: D30V Dependent Features
6852 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6855 * HPPA-Dependent:: HPPA Dependent Features
6858 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
6861 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
6864 * i860-Dependent:: Intel 80860 Dependent Features
6867 * i960-Dependent:: Intel 80960 Dependent Features
6870 * IA-64-Dependent:: Intel IA-64 Dependent Features
6873 * IP2K-Dependent:: IP2K Dependent Features
6876 * LM32-Dependent:: LM32 Dependent Features
6879 * M32C-Dependent:: M32C Dependent Features
6882 * M32R-Dependent:: M32R Dependent Features
6885 * M68K-Dependent:: M680x0 Dependent Features
6888 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
6891 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
6894 * MIPS-Dependent:: MIPS Dependent Features
6897 * MMIX-Dependent:: MMIX Dependent Features
6900 * MSP430-Dependent:: MSP430 Dependent Features
6903 * SH-Dependent:: Renesas / SuperH SH Dependent Features
6904 * SH64-Dependent:: SuperH SH64 Dependent Features
6907 * PDP-11-Dependent:: PDP-11 Dependent Features
6910 * PJ-Dependent:: picoJava Dependent Features
6913 * PPC-Dependent:: PowerPC Dependent Features
6916 * RX-Dependent:: RX Dependent Features
6919 * S/390-Dependent:: IBM S/390 Dependent Features
6922 * SCORE-Dependent:: SCORE Dependent Features
6925 * Sparc-Dependent:: SPARC Dependent Features
6928 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
6931 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
6934 * V850-Dependent:: V850 Dependent Features
6937 * Xtensa-Dependent:: Xtensa Dependent Features
6940 * Z80-Dependent:: Z80 Dependent Features
6943 * Z8000-Dependent:: Z8000 Dependent Features
6946 * Vax-Dependent:: VAX Dependent Features
6953 @c The following major nodes are *sections* in the GENERIC version, *chapters*
6954 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
6955 @c peculiarity: to preserve cross-references, there must be a node called
6956 @c "Machine Dependencies". Hence the conditional nodenames in each
6957 @c major node below. Node defaulting in makeinfo requires adjacency of
6958 @c node and sectioning commands; hence the repetition of @chapter BLAH
6959 @c in both conditional blocks.
6962 @include c-alpha.texi
6978 @include c-bfin.texi
6982 @include c-cr16.texi
6986 @include c-cris.texi
6991 @node Machine Dependencies
6992 @chapter Machine Dependent Features
6994 The machine instruction sets are different on each Renesas chip family,
6995 and there are also some syntax differences among the families. This
6996 chapter describes the specific @command{@value{AS}} features for each
7000 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7001 * SH-Dependent:: Renesas SH Dependent Features
7008 @include c-d10v.texi
7012 @include c-d30v.texi
7016 @include c-h8300.texi
7020 @include c-hppa.texi
7024 @include c-i370.texi
7028 @include c-i386.texi
7032 @include c-i860.texi
7036 @include c-i960.texi
7040 @include c-ia64.texi
7044 @include c-ip2k.texi
7048 @include c-lm32.texi
7052 @include c-m32c.texi
7056 @include c-m32r.texi
7060 @include c-m68k.texi
7064 @include c-m68hc11.texi
7068 @include c-microblaze.texi
7072 @include c-mips.texi
7076 @include c-mmix.texi
7080 @include c-msp430.texi
7084 @include c-ns32k.texi
7088 @include c-pdp11.texi
7104 @include c-s390.texi
7108 @include c-score.texi
7113 @include c-sh64.texi
7117 @include c-sparc.texi
7121 @include c-tic54x.texi
7125 @include c-tic6x.texi
7141 @include c-v850.texi
7145 @include c-xtensa.texi
7149 @c reverse effect of @down at top of generic Machine-Dep chapter
7153 @node Reporting Bugs
7154 @chapter Reporting Bugs
7155 @cindex bugs in assembler
7156 @cindex reporting bugs in assembler
7158 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7160 Reporting a bug may help you by bringing a solution to your problem, or it may
7161 not. But in any case the principal function of a bug report is to help the
7162 entire community by making the next version of @command{@value{AS}} work better.
7163 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7165 In order for a bug report to serve its purpose, you must include the
7166 information that enables us to fix the bug.
7169 * Bug Criteria:: Have you found a bug?
7170 * Bug Reporting:: How to report bugs
7174 @section Have You Found a Bug?
7175 @cindex bug criteria
7177 If you are not sure whether you have found a bug, here are some guidelines:
7180 @cindex fatal signal
7181 @cindex assembler crash
7182 @cindex crash of assembler
7184 If the assembler gets a fatal signal, for any input whatever, that is a
7185 @command{@value{AS}} bug. Reliable assemblers never crash.
7187 @cindex error on valid input
7189 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7191 @cindex invalid input
7193 If @command{@value{AS}} does not produce an error message for invalid input, that
7194 is a bug. However, you should note that your idea of ``invalid input'' might
7195 be our idea of ``an extension'' or ``support for traditional practice''.
7198 If you are an experienced user of assemblers, your suggestions for improvement
7199 of @command{@value{AS}} are welcome in any case.
7203 @section How to Report Bugs
7205 @cindex assembler bugs, reporting
7207 A number of companies and individuals offer support for @sc{gnu} products. If
7208 you obtained @command{@value{AS}} from a support organization, we recommend you
7209 contact that organization first.
7211 You can find contact information for many support companies and
7212 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7216 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7220 The fundamental principle of reporting bugs usefully is this:
7221 @strong{report all the facts}. If you are not sure whether to state a
7222 fact or leave it out, state it!
7224 Often people omit facts because they think they know what causes the problem
7225 and assume that some details do not matter. Thus, you might assume that the
7226 name of a symbol you use in an example does not matter. Well, probably it does
7227 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7228 happens to fetch from the location where that name is stored in memory;
7229 perhaps, if the name were different, the contents of that location would fool
7230 the assembler into doing the right thing despite the bug. Play it safe and
7231 give a specific, complete example. That is the easiest thing for you to do,
7232 and the most helpful.
7234 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7235 it is new to us. Therefore, always write your bug reports on the assumption
7236 that the bug has not been reported previously.
7238 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7239 bell?'' This cannot help us fix a bug, so it is basically useless. We
7240 respond by asking for enough details to enable us to investigate.
7241 You might as well expedite matters by sending them to begin with.
7243 To enable us to fix the bug, you should include all these things:
7247 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
7248 it with the @samp{--version} argument.
7250 Without this, we will not know whether there is any point in looking for
7251 the bug in the current version of @command{@value{AS}}.
7254 Any patches you may have applied to the @command{@value{AS}} source.
7257 The type of machine you are using, and the operating system name and
7261 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
7265 The command arguments you gave the assembler to assemble your example and
7266 observe the bug. To guarantee you will not omit something important, list them
7267 all. A copy of the Makefile (or the output from make) is sufficient.
7269 If we were to try to guess the arguments, we would probably guess wrong
7270 and then we might not encounter the bug.
7273 A complete input file that will reproduce the bug. If the bug is observed when
7274 the assembler is invoked via a compiler, send the assembler source, not the
7275 high level language source. Most compilers will produce the assembler source
7276 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
7277 the options @samp{-v --save-temps}; this will save the assembler source in a
7278 file with an extension of @file{.s}, and also show you exactly how
7279 @command{@value{AS}} is being run.
7282 A description of what behavior you observe that you believe is
7283 incorrect. For example, ``It gets a fatal signal.''
7285 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
7286 will certainly notice it. But if the bug is incorrect output, we might not
7287 notice unless it is glaringly wrong. You might as well not give us a chance to
7290 Even if the problem you experience is a fatal signal, you should still say so
7291 explicitly. Suppose something strange is going on, such as, your copy of
7292 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
7293 library on your system. (This has happened!) Your copy might crash and ours
7294 would not. If you told us to expect a crash, then when ours fails to crash, we
7295 would know that the bug was not happening for us. If you had not told us to
7296 expect a crash, then we would not be able to draw any conclusion from our
7300 If you wish to suggest changes to the @command{@value{AS}} source, send us context
7301 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
7302 option. Always send diffs from the old file to the new file. If you even
7303 discuss something in the @command{@value{AS}} source, refer to it by context, not
7306 The line numbers in our development sources will not match those in your
7307 sources. Your line numbers would convey no useful information to us.
7310 Here are some things that are not necessary:
7314 A description of the envelope of the bug.
7316 Often people who encounter a bug spend a lot of time investigating
7317 which changes to the input file will make the bug go away and which
7318 changes will not affect it.
7320 This is often time consuming and not very useful, because the way we
7321 will find the bug is by running a single example under the debugger
7322 with breakpoints, not by pure deduction from a series of examples.
7323 We recommend that you save your time for something else.
7325 Of course, if you can find a simpler example to report @emph{instead}
7326 of the original one, that is a convenience for us. Errors in the
7327 output will be easier to spot, running under the debugger will take
7328 less time, and so on.
7330 However, simplification is not vital; if you do not want to do this,
7331 report the bug anyway and send us the entire test case you used.
7334 A patch for the bug.
7336 A patch for the bug does help us if it is a good one. But do not omit
7337 the necessary information, such as the test case, on the assumption that
7338 a patch is all we need. We might see problems with your patch and decide
7339 to fix the problem another way, or we might not understand it at all.
7341 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
7342 construct an example that will make the program follow a certain path through
7343 the code. If you do not send us the example, we will not be able to construct
7344 one, so we will not be able to verify that the bug is fixed.
7346 And if we cannot understand what bug you are trying to fix, or why your
7347 patch should be an improvement, we will not install it. A test case will
7348 help us to understand.
7351 A guess about what the bug is or what it depends on.
7353 Such guesses are usually wrong. Even we cannot guess right about such
7354 things without first using the debugger to find the facts.
7357 @node Acknowledgements
7358 @chapter Acknowledgements
7360 If you have contributed to GAS and your name isn't listed here,
7361 it is not meant as a slight. We just don't know about it. Send mail to the
7362 maintainer, and we'll correct the situation. Currently
7364 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
7366 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
7369 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
7370 information and the 68k series machines, most of the preprocessing pass, and
7371 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
7373 K. Richard Pixley maintained GAS for a while, adding various enhancements and
7374 many bug fixes, including merging support for several processors, breaking GAS
7375 up to handle multiple object file format back ends (including heavy rewrite,
7376 testing, an integration of the coff and b.out back ends), adding configuration
7377 including heavy testing and verification of cross assemblers and file splits
7378 and renaming, converted GAS to strictly ANSI C including full prototypes, added
7379 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
7380 port (including considerable amounts of reverse engineering), a SPARC opcode
7381 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
7382 assertions and made them work, much other reorganization, cleanup, and lint.
7384 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
7385 in format-specific I/O modules.
7387 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
7388 has done much work with it since.
7390 The Intel 80386 machine description was written by Eliot Dresselhaus.
7392 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
7394 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
7395 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
7397 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
7398 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
7399 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
7400 support a.out format.
7402 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
7403 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
7404 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
7405 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
7408 John Gilmore built the AMD 29000 support, added @code{.include} support, and
7409 simplified the configuration of which versions accept which directives. He
7410 updated the 68k machine description so that Motorola's opcodes always produced
7411 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
7412 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
7413 cross-compilation support, and one bug in relaxation that took a week and
7414 required the proverbial one-bit fix.
7416 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
7417 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
7418 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
7419 PowerPC assembler, and made a few other minor patches.
7421 Steve Chamberlain made GAS able to generate listings.
7423 Hewlett-Packard contributed support for the HP9000/300.
7425 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
7426 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
7427 formats). This work was supported by both the Center for Software Science at
7428 the University of Utah and Cygnus Support.
7430 Support for ELF format files has been worked on by Mark Eichin of Cygnus
7431 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
7432 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
7433 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
7434 and some initial 64-bit support).
7436 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
7438 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
7439 support for openVMS/Alpha.
7441 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
7444 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
7445 Inc.@: added support for Xtensa processors.
7447 Several engineers at Cygnus Support have also provided many small bug fixes and
7448 configuration enhancements.
7450 Jon Beniston added support for the Lattice Mico32 architecture.
7452 Many others have contributed large or small bugfixes and enhancements. If
7453 you have contributed significant work and are not mentioned on this list, and
7454 want to be, let us know. Some of the history has been lost; we are not
7455 intentionally leaving anyone out.
7457 @node GNU Free Documentation License
7458 @appendix GNU Free Documentation License
7462 @unnumbered AS Index
projects / deliverable / binutils-gdb.git / blob