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, 2011
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.
91 @dircategory Software development
93 * As: (as). The GNU assembler.
94 * Gas: (as). The GNU assembler.
102 This file documents the GNU Assembler "@value{AS}".
104 @c man begin COPYRIGHT
105 Copyright @copyright{} 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
106 2000, 2001, 2002, 2006, 2007, 2008, 2009, 2010, 2011 Free Software Foundation,
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, 2011 Free Software Foundation,
160 Permission is granted to copy, distribute and/or modify this document
161 under the terms of the GNU Free Documentation License, Version 1.3
162 or any later version published by the Free Software Foundation;
163 with no Invariant Sections, with no Front-Cover Texts, and with no
164 Back-Cover Texts. A copy of the license is included in the
165 section entitled ``GNU Free Documentation License''.
172 @top Using @value{AS}
174 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}}
175 @ifset VERSION_PACKAGE
176 @value{VERSION_PACKAGE}
178 version @value{VERSION}.
180 This version of the file describes @command{@value{AS}} configured to generate
181 code for @value{TARGET} architectures.
184 This document is distributed under the terms of the GNU Free
185 Documentation License. A copy of the license is included in the
186 section entitled ``GNU Free Documentation License''.
189 * Overview:: Overview
190 * Invoking:: Command-Line Options
192 * Sections:: Sections and Relocation
194 * Expressions:: Expressions
195 * Pseudo Ops:: Assembler Directives
197 * Object Attributes:: Object Attributes
199 * Machine Dependencies:: Machine Dependent Features
200 * Reporting Bugs:: Reporting Bugs
201 * Acknowledgements:: Who Did What
202 * GNU Free Documentation License:: GNU Free Documentation License
203 * AS Index:: AS Index
210 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
212 This version of the manual describes @command{@value{AS}} configured to generate
213 code for @value{TARGET} architectures.
217 @cindex invocation summary
218 @cindex option summary
219 @cindex summary of options
220 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
221 see @ref{Invoking,,Command-Line Options}.
223 @c man title AS the portable GNU assembler.
227 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
231 @c We don't use deffn and friends for the following because they seem
232 @c to be limited to one line for the header.
234 @c man begin SYNOPSIS
235 @value{AS} [@b{-a}[@b{cdghlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
236 [@b{--compress-debug-sections}] [@b{--nocompress-debug-sections}]
237 [@b{--debug-prefix-map} @var{old}=@var{new}]
238 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
239 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
240 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
241 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
242 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}] [@b{-o}
243 @var{objfile}] [@b{-R}] [@b{--reduce-memory-overheads}] [@b{--statistics}]
244 [@b{-v}] [@b{-version}] [@b{--version}] [@b{-W}] [@b{--warn}]
245 [@b{--fatal-warnings}] [@b{-w}] [@b{-x}] [@b{-Z}] [@b{@@@var{FILE}}]
246 [@b{--size-check=[error|warning]}]
247 [@b{--target-help}] [@var{target-options}]
248 [@b{--}|@var{files} @dots{}]
250 @c Target dependent options are listed below. Keep the list sorted.
251 @c Add an empty line for separation.
254 @emph{Target Alpha options:}
256 [@b{-mdebug} | @b{-no-mdebug}]
257 [@b{-replace} | @b{-noreplace}]
258 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
259 [@b{-F}] [@b{-32addr}]
263 @emph{Target ARC options:}
269 @emph{Target ARM options:}
270 @c Don't document the deprecated options
271 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
272 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
273 [@b{-mfpu}=@var{floating-point-format}]
274 [@b{-mfloat-abi}=@var{abi}]
275 [@b{-meabi}=@var{ver}]
278 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
279 @b{-mapcs-reentrant}]
280 [@b{-mthumb-interwork}] [@b{-k}]
284 @emph{Target Blackfin options:}
285 [@b{-mcpu}=@var{processor}[-@var{sirevision}]]
292 @emph{Target CRIS options:}
293 [@b{--underscore} | @b{--no-underscore}]
295 [@b{--emulation=criself} | @b{--emulation=crisaout}]
296 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
297 @c Deprecated -- deliberately not documented.
302 @emph{Target D10V options:}
307 @emph{Target D30V options:}
308 [@b{-O}|@b{-n}|@b{-N}]
312 @emph{Target H8/300 options:}
316 @c HPPA has no machine-dependent assembler options (yet).
320 @emph{Target i386 options:}
321 [@b{--32}|@b{--n32}|@b{--64}] [@b{-n}]
322 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
326 @emph{Target i960 options:}
327 @c see md_parse_option in tc-i960.c
328 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
330 [@b{-b}] [@b{-no-relax}]
334 @emph{Target IA-64 options:}
335 [@b{-mconstant-gp}|@b{-mauto-pic}]
336 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
338 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
339 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
340 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
341 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
345 @emph{Target IP2K options:}
346 [@b{-mip2022}|@b{-mip2022ext}]
350 @emph{Target M32C options:}
351 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
355 @emph{Target M32R options:}
356 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
361 @emph{Target M680X0 options:}
362 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
366 @emph{Target M68HC11 options:}
367 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}]
368 [@b{-mshort}|@b{-mlong}]
369 [@b{-mshort-double}|@b{-mlong-double}]
370 [@b{--force-long-branches}] [@b{--short-branches}]
371 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
372 [@b{--print-opcodes}] [@b{--generate-example}]
376 @emph{Target MCORE options:}
377 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
378 [@b{-mcpu=[210|340]}]
381 @emph{Target MICROBLAZE options:}
382 @c MicroBlaze has no machine-dependent assembler options.
386 @emph{Target MIPS options:}
387 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
388 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
389 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
390 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
391 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
392 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
393 [@b{-mips64}] [@b{-mips64r2}]
394 [@b{-construct-floats}] [@b{-no-construct-floats}]
395 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
396 [@b{-mips16}] [@b{-no-mips16}]
397 [@b{-msmartmips}] [@b{-mno-smartmips}]
398 [@b{-mips3d}] [@b{-no-mips3d}]
399 [@b{-mdmx}] [@b{-no-mdmx}]
400 [@b{-mdsp}] [@b{-mno-dsp}]
401 [@b{-mdspr2}] [@b{-mno-dspr2}]
402 [@b{-mmt}] [@b{-mno-mt}]
403 [@b{-mfix7000}] [@b{-mno-fix7000}]
404 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
405 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
406 [@b{-mdebug}] [@b{-no-mdebug}]
407 [@b{-mpdr}] [@b{-mno-pdr}]
411 @emph{Target MMIX options:}
412 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
413 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
414 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
415 [@b{--linker-allocated-gregs}]
419 @emph{Target PDP11 options:}
420 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
421 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
422 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
426 @emph{Target picoJava options:}
431 @emph{Target PowerPC options:}
433 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|@b{-m403}|@b{-m405}|
434 @b{-m440}|@b{-m464}|@b{-m476}|@b{-m7400}|@b{-m7410}|@b{-m7450}|@b{-m7455}|@b{-m750cl}|@b{-mppc64}|
435 @b{-m620}|@b{-me500}|@b{-e500x2}|@b{-me500mc}|@b{-me500mc64}|@b{-mppc64bridge}|@b{-mbooke}|
436 @b{-mpower4}|@b{-mpr4}|@b{-mpower5}|@b{-mpwr5}|@b{-mpwr5x}|@b{-mpower6}|@b{-mpwr6}|
437 @b{-mpower7}|@b{-mpw7}|@b{-ma2}|@b{-mcell}|@b{-mspe}|@b{-mtitan}|@b{-me300}|@b{-mcom}]
438 [@b{-many}] [@b{-maltivec}|@b{-mvsx}]
439 [@b{-mregnames}|@b{-mno-regnames}]
440 [@b{-mrelocatable}|@b{-mrelocatable-lib}|@b{-K PIC}] [@b{-memb}]
441 [@b{-mlittle}|@b{-mlittle-endian}|@b{-le}|@b{-mbig}|@b{-mbig-endian}|@b{-be}]
442 [@b{-msolaris}|@b{-mno-solaris}]
443 [@b{-nops=@var{count}}]
447 @emph{Target RX options:}
448 [@b{-mlittle-endian}|@b{-mbig-endian}]
449 [@b{-m32bit-ints}|@b{-m16bit-ints}]
450 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
454 @emph{Target s390 options:}
455 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
456 [@b{-mregnames}|@b{-mno-regnames}]
457 [@b{-mwarn-areg-zero}]
461 @emph{Target SCORE options:}
462 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
463 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
464 [@b{-march=score7}][@b{-march=score3}]
465 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
469 @emph{Target SPARC options:}
470 @c The order here is important. See c-sparc.texi.
471 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
472 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
473 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
478 @emph{Target TIC54X options:}
479 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
480 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
485 @emph{Target TIC6X options:}
486 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
487 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
488 [@b{-mpic}|@b{-mno-pic}]
492 @emph{Target TILE-Gx options:}
496 @c TILEPro has no machine-dependent assembler options
501 @emph{Target Xtensa options:}
502 [@b{--[no-]text-section-literals}] [@b{--[no-]absolute-literals}]
503 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
504 [@b{--[no-]transform}]
505 [@b{--rename-section} @var{oldname}=@var{newname}]
510 @emph{Target Z80 options:}
511 [@b{-z80}] [@b{-r800}]
512 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
513 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
514 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
515 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
516 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
517 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
521 @c Z8000 has no machine-dependent assembler options
530 @include at-file.texi
533 Turn on listings, in any of a variety of ways:
537 omit false conditionals
540 omit debugging directives
543 include general information, like @value{AS} version and options passed
546 include high-level source
552 include macro expansions
555 omit forms processing
561 set the name of the listing file
564 You may combine these options; for example, use @samp{-aln} for assembly
565 listing without forms processing. The @samp{=file} option, if used, must be
566 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
569 Begin in alternate macro mode.
571 @xref{Altmacro,,@code{.altmacro}}.
574 @item --compress-debug-sections
575 Compress DWARF debug sections using zlib. The debug sections are renamed
576 to begin with @samp{.zdebug}, and the resulting object file may not be
577 compatible with older linkers and object file utilities.
579 @item --nocompress-debug-sections
580 Do not compress DWARF debug sections. This is the default.
583 Ignored. This option is accepted for script compatibility with calls to
586 @item --debug-prefix-map @var{old}=@var{new}
587 When assembling files in directory @file{@var{old}}, record debugging
588 information describing them as in @file{@var{new}} instead.
590 @item --defsym @var{sym}=@var{value}
591 Define the symbol @var{sym} to be @var{value} before assembling the input file.
592 @var{value} must be an integer constant. As in C, a leading @samp{0x}
593 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
594 value. The value of the symbol can be overridden inside a source file via the
595 use of a @code{.set} pseudo-op.
598 ``fast''---skip whitespace and comment preprocessing (assume source is
603 Generate debugging information for each assembler source line using whichever
604 debug format is preferred by the target. This currently means either STABS,
608 Generate stabs debugging information for each assembler line. This
609 may help debugging assembler code, if the debugger can handle it.
612 Generate stabs debugging information for each assembler line, with GNU
613 extensions that probably only gdb can handle, and that could make other
614 debuggers crash or refuse to read your program. This
615 may help debugging assembler code. Currently the only GNU extension is
616 the location of the current working directory at assembling time.
619 Generate DWARF2 debugging information for each assembler line. This
620 may help debugging assembler code, if the debugger can handle it. Note---this
621 option is only supported by some targets, not all of them.
623 @item --size-check=error
624 @itemx --size-check=warning
625 Issue an error or warning for invalid ELF .size directive.
628 Print a summary of the command line options and exit.
631 Print a summary of all target specific options and exit.
634 Add directory @var{dir} to the search list for @code{.include} directives.
637 Don't warn about signed overflow.
640 @ifclear DIFF-TBL-KLUGE
641 This option is accepted but has no effect on the @value{TARGET} family.
643 @ifset DIFF-TBL-KLUGE
644 Issue warnings when difference tables altered for long displacements.
649 Keep (in the symbol table) local symbols. These symbols start with
650 system-specific local label prefixes, typically @samp{.L} for ELF systems
651 or @samp{L} for traditional a.out systems.
656 @item --listing-lhs-width=@var{number}
657 Set the maximum width, in words, of the output data column for an assembler
658 listing to @var{number}.
660 @item --listing-lhs-width2=@var{number}
661 Set the maximum width, in words, of the output data column for continuation
662 lines in an assembler listing to @var{number}.
664 @item --listing-rhs-width=@var{number}
665 Set the maximum width of an input source line, as displayed in a listing, to
668 @item --listing-cont-lines=@var{number}
669 Set the maximum number of lines printed in a listing for a single line of input
672 @item -o @var{objfile}
673 Name the object-file output from @command{@value{AS}} @var{objfile}.
676 Fold the data section into the text section.
678 @kindex --hash-size=@var{number}
679 Set the default size of GAS's hash tables to a prime number close to
680 @var{number}. Increasing this value can reduce the length of time it takes the
681 assembler to perform its tasks, at the expense of increasing the assembler's
682 memory requirements. Similarly reducing this value can reduce the memory
683 requirements at the expense of speed.
685 @item --reduce-memory-overheads
686 This option reduces GAS's memory requirements, at the expense of making the
687 assembly processes slower. Currently this switch is a synonym for
688 @samp{--hash-size=4051}, but in the future it may have other effects as well.
691 Print the maximum space (in bytes) and total time (in seconds) used by
694 @item --strip-local-absolute
695 Remove local absolute symbols from the outgoing symbol table.
699 Print the @command{as} version.
702 Print the @command{as} version and exit.
706 Suppress warning messages.
708 @item --fatal-warnings
709 Treat warnings as errors.
712 Don't suppress warning messages or treat them as errors.
721 Generate an object file even after errors.
723 @item -- | @var{files} @dots{}
724 Standard input, or source files to assemble.
732 @xref{Alpha Options}, for the options available when @value{AS} is configured
733 for an Alpha processor.
738 The following options are available when @value{AS} is configured for an Alpha
742 @include c-alpha.texi
743 @c ended inside the included file
750 The following options are available when @value{AS} is configured for
755 This option selects the core processor variant.
757 Select either big-endian (-EB) or little-endian (-EL) output.
762 The following options are available when @value{AS} is configured for the ARM
766 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
767 Specify which ARM processor variant is the target.
768 @item -march=@var{architecture}[+@var{extension}@dots{}]
769 Specify which ARM architecture variant is used by the target.
770 @item -mfpu=@var{floating-point-format}
771 Select which Floating Point architecture is the target.
772 @item -mfloat-abi=@var{abi}
773 Select which floating point ABI is in use.
775 Enable Thumb only instruction decoding.
776 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
777 Select which procedure calling convention is in use.
779 Select either big-endian (-EB) or little-endian (-EL) output.
780 @item -mthumb-interwork
781 Specify that the code has been generated with interworking between Thumb and
784 Specify that PIC code has been generated.
792 @xref{Blackfin Options}, for the options available when @value{AS} is
793 configured for the Blackfin processor family.
798 The following options are available when @value{AS} is configured for
799 the Blackfin processor family.
803 @c ended inside the included file
810 See the info pages for documentation of the CRIS-specific options.
814 The following options are available when @value{AS} is configured for
817 @cindex D10V optimization
818 @cindex optimization, D10V
820 Optimize output by parallelizing instructions.
825 The following options are available when @value{AS} is configured for a D30V
828 @cindex D30V optimization
829 @cindex optimization, D30V
831 Optimize output by parallelizing instructions.
835 Warn when nops are generated.
837 @cindex D30V nops after 32-bit multiply
839 Warn when a nop after a 32-bit multiply instruction is generated.
847 @xref{i386-Options}, for the options available when @value{AS} is
848 configured for an i386 processor.
853 The following options are available when @value{AS} is configured for
858 @c ended inside the included file
865 The following options are available when @value{AS} is configured for the
866 Intel 80960 processor.
869 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
870 Specify which variant of the 960 architecture is the target.
873 Add code to collect statistics about branches taken.
876 Do not alter compare-and-branch instructions for long displacements;
883 The following options are available when @value{AS} is configured for the
889 Specifies that the extended IP2022 instructions are allowed.
892 Restores the default behaviour, which restricts the permitted instructions to
893 just the basic IP2022 ones.
899 The following options are available when @value{AS} is configured for the
900 Renesas M32C and M16C processors.
905 Assemble M32C instructions.
908 Assemble M16C instructions (the default).
911 Enable support for link-time relaxations.
914 Support H'00 style hex constants in addition to 0x00 style.
920 The following options are available when @value{AS} is configured for the
921 Renesas M32R (formerly Mitsubishi M32R) series.
926 Specify which processor in the M32R family is the target. The default
927 is normally the M32R, but this option changes it to the M32RX.
929 @item --warn-explicit-parallel-conflicts or --Wp
930 Produce warning messages when questionable parallel constructs are
933 @item --no-warn-explicit-parallel-conflicts or --Wnp
934 Do not produce warning messages when questionable parallel constructs are
941 The following options are available when @value{AS} is configured for the
942 Motorola 68000 series.
947 Shorten references to undefined symbols, to one word instead of two.
949 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
950 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
951 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
952 Specify what processor in the 68000 family is the target. The default
953 is normally the 68020, but this can be changed at configuration time.
955 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
956 The target machine does (or does not) have a floating-point coprocessor.
957 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
958 the basic 68000 is not compatible with the 68881, a combination of the
959 two can be specified, since it's possible to do emulation of the
960 coprocessor instructions with the main processor.
962 @item -m68851 | -mno-68851
963 The target machine does (or does not) have a memory-management
964 unit coprocessor. The default is to assume an MMU for 68020 and up.
971 For details about the PDP-11 machine dependent features options,
972 see @ref{PDP-11-Options}.
975 @item -mpic | -mno-pic
976 Generate position-independent (or position-dependent) code. The
977 default is @option{-mpic}.
980 @itemx -mall-extensions
981 Enable all instruction set extensions. This is the default.
983 @item -mno-extensions
984 Disable all instruction set extensions.
986 @item -m@var{extension} | -mno-@var{extension}
987 Enable (or disable) a particular instruction set extension.
990 Enable the instruction set extensions supported by a particular CPU, and
991 disable all other extensions.
993 @item -m@var{machine}
994 Enable the instruction set extensions supported by a particular machine
995 model, and disable all other extensions.
1001 The following options are available when @value{AS} is configured for
1002 a picoJava processor.
1006 @cindex PJ endianness
1007 @cindex endianness, PJ
1008 @cindex big endian output, PJ
1010 Generate ``big endian'' format output.
1012 @cindex little endian output, PJ
1014 Generate ``little endian'' format output.
1020 The following options are available when @value{AS} is configured for the
1021 Motorola 68HC11 or 68HC12 series.
1025 @item -m68hc11 | -m68hc12 | -m68hcs12
1026 Specify what processor is the target. The default is
1027 defined by the configuration option when building the assembler.
1030 Specify to use the 16-bit integer ABI.
1033 Specify to use the 32-bit integer ABI.
1035 @item -mshort-double
1036 Specify to use the 32-bit double ABI.
1039 Specify to use the 64-bit double ABI.
1041 @item --force-long-branches
1042 Relative branches are turned into absolute ones. This concerns
1043 conditional branches, unconditional branches and branches to a
1046 @item -S | --short-branches
1047 Do not turn relative branches into absolute ones
1048 when the offset is out of range.
1050 @item --strict-direct-mode
1051 Do not turn the direct addressing mode into extended addressing mode
1052 when the instruction does not support direct addressing mode.
1054 @item --print-insn-syntax
1055 Print the syntax of instruction in case of error.
1057 @item --print-opcodes
1058 print the list of instructions with syntax and then exit.
1060 @item --generate-example
1061 print an example of instruction for each possible instruction and then exit.
1062 This option is only useful for testing @command{@value{AS}}.
1068 The following options are available when @command{@value{AS}} is configured
1069 for the SPARC architecture:
1072 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1073 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1074 Explicitly select a variant of the SPARC architecture.
1076 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1077 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1079 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1080 UltraSPARC extensions.
1082 @item -xarch=v8plus | -xarch=v8plusa
1083 For compatibility with the Solaris v9 assembler. These options are
1084 equivalent to -Av8plus and -Av8plusa, respectively.
1087 Warn when the assembler switches to another architecture.
1092 The following options are available when @value{AS} is configured for the 'c54x
1097 Enable extended addressing mode. All addresses and relocations will assume
1098 extended addressing (usually 23 bits).
1099 @item -mcpu=@var{CPU_VERSION}
1100 Sets the CPU version being compiled for.
1101 @item -merrors-to-file @var{FILENAME}
1102 Redirect error output to a file, for broken systems which don't support such
1103 behaviour in the shell.
1108 The following options are available when @value{AS} is configured for
1109 a @sc{mips} processor.
1113 This option sets the largest size of an object that can be referenced
1114 implicitly with the @code{gp} register. It is only accepted for targets that
1115 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1117 @cindex MIPS endianness
1118 @cindex endianness, MIPS
1119 @cindex big endian output, MIPS
1121 Generate ``big endian'' format output.
1123 @cindex little endian output, MIPS
1125 Generate ``little endian'' format output.
1137 Generate code for a particular @sc{mips} Instruction Set Architecture level.
1138 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1139 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1140 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1141 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
1143 correspond to generic
1144 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
1145 and @samp{MIPS64 Release 2}
1146 ISA processors, respectively.
1148 @item -march=@var{CPU}
1149 Generate code for a particular @sc{mips} cpu.
1151 @item -mtune=@var{cpu}
1152 Schedule and tune for a particular @sc{mips} cpu.
1156 Cause nops to be inserted if the read of the destination register
1157 of an mfhi or mflo instruction occurs in the following two instructions.
1161 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1162 section instead of the standard ELF .stabs sections.
1166 Control generation of @code{.pdr} sections.
1170 The register sizes are normally inferred from the ISA and ABI, but these
1171 flags force a certain group of registers to be treated as 32 bits wide at
1172 all times. @samp{-mgp32} controls the size of general-purpose registers
1173 and @samp{-mfp32} controls the size of floating-point registers.
1177 Generate code for the MIPS 16 processor. This is equivalent to putting
1178 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1179 turns off this option.
1182 @itemx -mno-smartmips
1183 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1184 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1185 @samp{-mno-smartmips} turns off this option.
1189 Generate code for the MIPS-3D Application Specific Extension.
1190 This tells the assembler to accept MIPS-3D instructions.
1191 @samp{-no-mips3d} turns off this option.
1195 Generate code for the MDMX Application Specific Extension.
1196 This tells the assembler to accept MDMX instructions.
1197 @samp{-no-mdmx} turns off this option.
1201 Generate code for the DSP Release 1 Application Specific Extension.
1202 This tells the assembler to accept DSP Release 1 instructions.
1203 @samp{-mno-dsp} turns off this option.
1207 Generate code for the DSP Release 2 Application Specific Extension.
1208 This option implies -mdsp.
1209 This tells the assembler to accept DSP Release 2 instructions.
1210 @samp{-mno-dspr2} turns off this option.
1214 Generate code for the MT Application Specific Extension.
1215 This tells the assembler to accept MT instructions.
1216 @samp{-mno-mt} turns off this option.
1218 @item --construct-floats
1219 @itemx --no-construct-floats
1220 The @samp{--no-construct-floats} option disables the construction of
1221 double width floating point constants by loading the two halves of the
1222 value into the two single width floating point registers that make up
1223 the double width register. By default @samp{--construct-floats} is
1224 selected, allowing construction of these floating point constants.
1227 @item --emulation=@var{name}
1228 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
1229 for some other target, in all respects, including output format (choosing
1230 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
1231 debugging information or store symbol table information, and default
1232 endianness. The available configuration names are: @samp{mipsecoff},
1233 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
1234 @samp{mipsbelf}. The first two do not alter the default endianness from that
1235 of the primary target for which the assembler was configured; the others change
1236 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
1237 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
1238 selection in any case.
1240 This option is currently supported only when the primary target
1241 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
1242 Furthermore, the primary target or others specified with
1243 @samp{--enable-targets=@dots{}} at configuration time must include support for
1244 the other format, if both are to be available. For example, the Irix 5
1245 configuration includes support for both.
1247 Eventually, this option will support more configurations, with more
1248 fine-grained control over the assembler's behavior, and will be supported for
1252 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1259 Control how to deal with multiplication overflow and division by zero.
1260 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1261 (and only work for Instruction Set Architecture level 2 and higher);
1262 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1266 When this option is used, @command{@value{AS}} will issue a warning every
1267 time it generates a nop instruction from a macro.
1272 The following options are available when @value{AS} is configured for
1278 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1279 The command line option @samp{-nojsri2bsr} can be used to disable it.
1283 Enable or disable the silicon filter behaviour. By default this is disabled.
1284 The default can be overridden by the @samp{-sifilter} command line option.
1287 Alter jump instructions for long displacements.
1289 @item -mcpu=[210|340]
1290 Select the cpu type on the target hardware. This controls which instructions
1294 Assemble for a big endian target.
1297 Assemble for a little endian target.
1303 See the info pages for documentation of the MMIX-specific options.
1310 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1311 for a PowerPC processor.
1315 @c man begin OPTIONS
1316 The following options are available when @value{AS} is configured for a
1319 @c man begin INCLUDE
1321 @c ended inside the included file
1326 @c man begin OPTIONS
1328 See the info pages for documentation of the RX-specific options.
1332 The following options are available when @value{AS} is configured for the s390
1338 Select the word size, either 31/32 bits or 64 bits.
1341 Select the architecture mode, either the Enterprise System
1342 Architecture (esa) or the z/Architecture mode (zarch).
1343 @item -march=@var{processor}
1344 Specify which s390 processor variant is the target, @samp{g6}, @samp{g6},
1345 @samp{z900}, @samp{z990}, @samp{z9-109}, @samp{z9-ec}, or @samp{z10}.
1347 @itemx -mno-regnames
1348 Allow or disallow symbolic names for registers.
1349 @item -mwarn-areg-zero
1350 Warn whenever the operand for a base or index register has been specified
1351 but evaluates to zero.
1359 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1360 for a TMS320C6000 processor.
1364 @c man begin OPTIONS
1365 The following options are available when @value{AS} is configured for a
1366 TMS320C6000 processor.
1368 @c man begin INCLUDE
1369 @include c-tic6x.texi
1370 @c ended inside the included file
1378 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1379 for a TILE-Gx processor.
1383 @c man begin OPTIONS
1384 The following options are available when @value{AS} is configured for a TILE-Gx
1387 @c man begin INCLUDE
1388 @include c-tilegx.texi
1389 @c ended inside the included file
1397 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1398 for an Xtensa processor.
1402 @c man begin OPTIONS
1403 The following options are available when @value{AS} is configured for an
1406 @c man begin INCLUDE
1407 @include c-xtensa.texi
1408 @c ended inside the included file
1413 @c man begin OPTIONS
1416 The following options are available when @value{AS} is configured for
1417 a Z80 family processor.
1420 Assemble for Z80 processor.
1422 Assemble for R800 processor.
1423 @item -ignore-undocumented-instructions
1425 Assemble undocumented Z80 instructions that also work on R800 without warning.
1426 @item -ignore-unportable-instructions
1428 Assemble all undocumented Z80 instructions without warning.
1429 @item -warn-undocumented-instructions
1431 Issue a warning for undocumented Z80 instructions that also work on R800.
1432 @item -warn-unportable-instructions
1434 Issue a warning for undocumented Z80 instructions that do not work on R800.
1435 @item -forbid-undocumented-instructions
1437 Treat all undocumented instructions as errors.
1438 @item -forbid-unportable-instructions
1440 Treat undocumented Z80 instructions that do not work on R800 as errors.
1447 * Manual:: Structure of this Manual
1448 * GNU Assembler:: The GNU Assembler
1449 * Object Formats:: Object File Formats
1450 * Command Line:: Command Line
1451 * Input Files:: Input Files
1452 * Object:: Output (Object) File
1453 * Errors:: Error and Warning Messages
1457 @section Structure of this Manual
1459 @cindex manual, structure and purpose
1460 This manual is intended to describe what you need to know to use
1461 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1462 notation for symbols, constants, and expressions; the directives that
1463 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1466 We also cover special features in the @value{TARGET}
1467 configuration of @command{@value{AS}}, including assembler directives.
1470 This manual also describes some of the machine-dependent features of
1471 various flavors of the assembler.
1474 @cindex machine instructions (not covered)
1475 On the other hand, this manual is @emph{not} intended as an introduction
1476 to programming in assembly language---let alone programming in general!
1477 In a similar vein, we make no attempt to introduce the machine
1478 architecture; we do @emph{not} describe the instruction set, standard
1479 mnemonics, registers or addressing modes that are standard to a
1480 particular architecture.
1482 You may want to consult the manufacturer's
1483 machine architecture manual for this information.
1487 For information on the H8/300 machine instruction set, see @cite{H8/300
1488 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1489 Programming Manual} (Renesas).
1492 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1493 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1494 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1495 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1498 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1502 @c I think this is premature---doc@cygnus.com, 17jan1991
1504 Throughout this manual, we assume that you are running @dfn{GNU},
1505 the portable operating system from the @dfn{Free Software
1506 Foundation, Inc.}. This restricts our attention to certain kinds of
1507 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1508 once this assumption is granted examples and definitions need less
1511 @command{@value{AS}} is part of a team of programs that turn a high-level
1512 human-readable series of instructions into a low-level
1513 computer-readable series of instructions. Different versions of
1514 @command{@value{AS}} are used for different kinds of computer.
1517 @c There used to be a section "Terminology" here, which defined
1518 @c "contents", "byte", "word", and "long". Defining "word" to any
1519 @c particular size is confusing when the .word directive may generate 16
1520 @c bits on one machine and 32 bits on another; in general, for the user
1521 @c version of this manual, none of these terms seem essential to define.
1522 @c They were used very little even in the former draft of the manual;
1523 @c this draft makes an effort to avoid them (except in names of
1527 @section The GNU Assembler
1529 @c man begin DESCRIPTION
1531 @sc{gnu} @command{as} is really a family of assemblers.
1533 This manual describes @command{@value{AS}}, a member of that family which is
1534 configured for the @value{TARGET} architectures.
1536 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1537 should find a fairly similar environment when you use it on another
1538 architecture. Each version has much in common with the others,
1539 including object file formats, most assembler directives (often called
1540 @dfn{pseudo-ops}) and assembler syntax.@refill
1542 @cindex purpose of @sc{gnu} assembler
1543 @command{@value{AS}} is primarily intended to assemble the output of the
1544 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1545 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1546 assemble correctly everything that other assemblers for the same
1547 machine would assemble.
1549 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1552 @c This remark should appear in generic version of manual; assumption
1553 @c here is that generic version sets M680x0.
1554 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1555 assembler for the same architecture; for example, we know of several
1556 incompatible versions of 680x0 assembly language syntax.
1561 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1562 program in one pass of the source file. This has a subtle impact on the
1563 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1565 @node Object Formats
1566 @section Object File Formats
1568 @cindex object file format
1569 The @sc{gnu} assembler can be configured to produce several alternative
1570 object file formats. For the most part, this does not affect how you
1571 write assembly language programs; but directives for debugging symbols
1572 are typically different in different file formats. @xref{Symbol
1573 Attributes,,Symbol Attributes}.
1576 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1577 @value{OBJ-NAME} format object files.
1579 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1581 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1582 @code{b.out} or COFF format object files.
1585 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1586 SOM or ELF format object files.
1591 @section Command Line
1593 @cindex command line conventions
1595 After the program name @command{@value{AS}}, the command line may contain
1596 options and file names. Options may appear in any order, and may be
1597 before, after, or between file names. The order of file names is
1600 @cindex standard input, as input file
1602 @file{--} (two hyphens) by itself names the standard input file
1603 explicitly, as one of the files for @command{@value{AS}} to assemble.
1605 @cindex options, command line
1606 Except for @samp{--} any command line argument that begins with a
1607 hyphen (@samp{-}) is an option. Each option changes the behavior of
1608 @command{@value{AS}}. No option changes the way another option works. An
1609 option is a @samp{-} followed by one or more letters; the case of
1610 the letter is important. All options are optional.
1612 Some options expect exactly one file name to follow them. The file
1613 name may either immediately follow the option's letter (compatible
1614 with older assemblers) or it may be the next command argument (@sc{gnu}
1615 standard). These two command lines are equivalent:
1618 @value{AS} -o my-object-file.o mumble.s
1619 @value{AS} -omy-object-file.o mumble.s
1623 @section Input Files
1626 @cindex source program
1627 @cindex files, input
1628 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1629 describe the program input to one run of @command{@value{AS}}. The program may
1630 be in one or more files; how the source is partitioned into files
1631 doesn't change the meaning of the source.
1633 @c I added "con" prefix to "catenation" just to prove I can overcome my
1634 @c APL training... doc@cygnus.com
1635 The source program is a concatenation of the text in all the files, in the
1638 @c man begin DESCRIPTION
1639 Each time you run @command{@value{AS}} it assembles exactly one source
1640 program. The source program is made up of one or more files.
1641 (The standard input is also a file.)
1643 You give @command{@value{AS}} a command line that has zero or more input file
1644 names. The input files are read (from left file name to right). A
1645 command line argument (in any position) that has no special meaning
1646 is taken to be an input file name.
1648 If you give @command{@value{AS}} no file names it attempts to read one input file
1649 from the @command{@value{AS}} standard input, which is normally your terminal. You
1650 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1653 Use @samp{--} if you need to explicitly name the standard input file
1654 in your command line.
1656 If the source is empty, @command{@value{AS}} produces a small, empty object
1661 @subheading Filenames and Line-numbers
1663 @cindex input file linenumbers
1664 @cindex line numbers, in input files
1665 There are two ways of locating a line in the input file (or files) and
1666 either may be used in reporting error messages. One way refers to a line
1667 number in a physical file; the other refers to a line number in a
1668 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1670 @dfn{Physical files} are those files named in the command line given
1671 to @command{@value{AS}}.
1673 @dfn{Logical files} are simply names declared explicitly by assembler
1674 directives; they bear no relation to physical files. Logical file names help
1675 error messages reflect the original source file, when @command{@value{AS}} source
1676 is itself synthesized from other files. @command{@value{AS}} understands the
1677 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1678 @ref{File,,@code{.file}}.
1681 @section Output (Object) File
1687 Every time you run @command{@value{AS}} it produces an output file, which is
1688 your assembly language program translated into numbers. This file
1689 is the object file. Its default name is
1697 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1699 You can give it another name by using the @option{-o} option. Conventionally,
1700 object file names end with @file{.o}. The default name is used for historical
1701 reasons: older assemblers were capable of assembling self-contained programs
1702 directly into a runnable program. (For some formats, this isn't currently
1703 possible, but it can be done for the @code{a.out} format.)
1707 The object file is meant for input to the linker @code{@value{LD}}. It contains
1708 assembled program code, information to help @code{@value{LD}} integrate
1709 the assembled program into a runnable file, and (optionally) symbolic
1710 information for the debugger.
1712 @c link above to some info file(s) like the description of a.out.
1713 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1716 @section Error and Warning Messages
1718 @c man begin DESCRIPTION
1720 @cindex error messages
1721 @cindex warning messages
1722 @cindex messages from assembler
1723 @command{@value{AS}} may write warnings and error messages to the standard error
1724 file (usually your terminal). This should not happen when a compiler
1725 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1726 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1727 grave problem that stops the assembly.
1731 @cindex format of warning messages
1732 Warning messages have the format
1735 file_name:@b{NNN}:Warning Message Text
1739 @cindex line numbers, in warnings/errors
1740 (where @b{NNN} is a line number). If a logical file name has been given
1741 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1742 the current input file is used. If a logical line number was given
1744 (@pxref{Line,,@code{.line}})
1746 then it is used to calculate the number printed,
1747 otherwise the actual line in the current source file is printed. The
1748 message text is intended to be self explanatory (in the grand Unix
1751 @cindex format of error messages
1752 Error messages have the format
1754 file_name:@b{NNN}:FATAL:Error Message Text
1756 The file name and line number are derived as for warning
1757 messages. The actual message text may be rather less explanatory
1758 because many of them aren't supposed to happen.
1761 @chapter Command-Line Options
1763 @cindex options, all versions of assembler
1764 This chapter describes command-line options available in @emph{all}
1765 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
1766 for options specific
1768 to the @value{TARGET} target.
1771 to particular machine architectures.
1774 @c man begin DESCRIPTION
1776 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1777 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1778 The assembler arguments must be separated from each other (and the @samp{-Wa})
1779 by commas. For example:
1782 gcc -c -g -O -Wa,-alh,-L file.c
1786 This passes two options to the assembler: @samp{-alh} (emit a listing to
1787 standard output with high-level and assembly source) and @samp{-L} (retain
1788 local symbols in the symbol table).
1790 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1791 command-line options are automatically passed to the assembler by the compiler.
1792 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1793 precisely what options it passes to each compilation pass, including the
1799 * a:: -a[cdghlns] enable listings
1800 * alternate:: --alternate enable alternate macro syntax
1801 * D:: -D for compatibility
1802 * f:: -f to work faster
1803 * I:: -I for .include search path
1804 @ifclear DIFF-TBL-KLUGE
1805 * K:: -K for compatibility
1807 @ifset DIFF-TBL-KLUGE
1808 * K:: -K for difference tables
1811 * L:: -L to retain local symbols
1812 * listing:: --listing-XXX to configure listing output
1813 * M:: -M or --mri to assemble in MRI compatibility mode
1814 * MD:: --MD for dependency tracking
1815 * o:: -o to name the object file
1816 * R:: -R to join data and text sections
1817 * statistics:: --statistics to see statistics about assembly
1818 * traditional-format:: --traditional-format for compatible output
1819 * v:: -v to announce version
1820 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1821 * Z:: -Z to make object file even after errors
1825 @section Enable Listings: @option{-a[cdghlns]}
1835 @cindex listings, enabling
1836 @cindex assembly listings, enabling
1838 These options enable listing output from the assembler. By itself,
1839 @samp{-a} requests high-level, assembly, and symbols listing.
1840 You can use other letters to select specific options for the list:
1841 @samp{-ah} requests a high-level language listing,
1842 @samp{-al} requests an output-program assembly listing, and
1843 @samp{-as} requests a symbol table listing.
1844 High-level listings require that a compiler debugging option like
1845 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1848 Use the @samp{-ag} option to print a first section with general assembly
1849 information, like @value{AS} version, switches passed, or time stamp.
1851 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1852 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1853 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1854 omitted from the listing.
1856 Use the @samp{-ad} option to omit debugging directives from the
1859 Once you have specified one of these options, you can further control
1860 listing output and its appearance using the directives @code{.list},
1861 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1863 The @samp{-an} option turns off all forms processing.
1864 If you do not request listing output with one of the @samp{-a} options, the
1865 listing-control directives have no effect.
1867 The letters after @samp{-a} may be combined into one option,
1868 @emph{e.g.}, @samp{-aln}.
1870 Note if the assembler source is coming from the standard input (e.g.,
1872 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1873 is being used) then the listing will not contain any comments or preprocessor
1874 directives. This is because the listing code buffers input source lines from
1875 stdin only after they have been preprocessed by the assembler. This reduces
1876 memory usage and makes the code more efficient.
1879 @section @option{--alternate}
1882 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
1885 @section @option{-D}
1888 This option has no effect whatsoever, but it is accepted to make it more
1889 likely that scripts written for other assemblers also work with
1890 @command{@value{AS}}.
1893 @section Work Faster: @option{-f}
1896 @cindex trusted compiler
1897 @cindex faster processing (@option{-f})
1898 @samp{-f} should only be used when assembling programs written by a
1899 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1900 and comment preprocessing on
1901 the input file(s) before assembling them. @xref{Preprocessing,
1905 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1906 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1911 @section @code{.include} Search Path: @option{-I} @var{path}
1913 @kindex -I @var{path}
1914 @cindex paths for @code{.include}
1915 @cindex search path for @code{.include}
1916 @cindex @code{include} directive search path
1917 Use this option to add a @var{path} to the list of directories
1918 @command{@value{AS}} searches for files specified in @code{.include}
1919 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1920 many times as necessary to include a variety of paths. The current
1921 working directory is always searched first; after that, @command{@value{AS}}
1922 searches any @samp{-I} directories in the same order as they were
1923 specified (left to right) on the command line.
1926 @section Difference Tables: @option{-K}
1929 @ifclear DIFF-TBL-KLUGE
1930 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1931 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1932 where it can be used to warn when the assembler alters the machine code
1933 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1934 family does not have the addressing limitations that sometimes lead to this
1935 alteration on other platforms.
1938 @ifset DIFF-TBL-KLUGE
1939 @cindex difference tables, warning
1940 @cindex warning for altered difference tables
1941 @command{@value{AS}} sometimes alters the code emitted for directives of the
1942 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
1943 You can use the @samp{-K} option if you want a warning issued when this
1948 @section Include Local Symbols: @option{-L}
1951 @cindex local symbols, retaining in output
1952 Symbols beginning with system-specific local label prefixes, typically
1953 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
1954 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
1955 such symbols when debugging, because they are intended for the use of
1956 programs (like compilers) that compose assembler programs, not for your
1957 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
1958 such symbols, so you do not normally debug with them.
1960 This option tells @command{@value{AS}} to retain those local symbols
1961 in the object file. Usually if you do this you also tell the linker
1962 @code{@value{LD}} to preserve those symbols.
1965 @section Configuring listing output: @option{--listing}
1967 The listing feature of the assembler can be enabled via the command line switch
1968 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1969 hex dump of the corresponding locations in the output object file, and displays
1970 them as a listing file. The format of this listing can be controlled by
1971 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
1972 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
1973 @code{.psize} (@pxref{Psize}), and
1974 @code{.eject} (@pxref{Eject}) and also by the following switches:
1977 @item --listing-lhs-width=@samp{number}
1978 @kindex --listing-lhs-width
1979 @cindex Width of first line disassembly output
1980 Sets the maximum width, in words, of the first line of the hex byte dump. This
1981 dump appears on the left hand side of the listing output.
1983 @item --listing-lhs-width2=@samp{number}
1984 @kindex --listing-lhs-width2
1985 @cindex Width of continuation lines of disassembly output
1986 Sets the maximum width, in words, of any further lines of the hex byte dump for
1987 a given input source line. If this value is not specified, it defaults to being
1988 the same as the value specified for @samp{--listing-lhs-width}. If neither
1989 switch is used the default is to one.
1991 @item --listing-rhs-width=@samp{number}
1992 @kindex --listing-rhs-width
1993 @cindex Width of source line output
1994 Sets the maximum width, in characters, of the source line that is displayed
1995 alongside the hex dump. The default value for this parameter is 100. The
1996 source line is displayed on the right hand side of the listing output.
1998 @item --listing-cont-lines=@samp{number}
1999 @kindex --listing-cont-lines
2000 @cindex Maximum number of continuation lines
2001 Sets the maximum number of continuation lines of hex dump that will be
2002 displayed for a given single line of source input. The default value is 4.
2006 @section Assemble in MRI Compatibility Mode: @option{-M}
2009 @cindex MRI compatibility mode
2010 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2011 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2012 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
2013 configured target) assembler from Microtec Research. The exact nature of the
2014 MRI syntax will not be documented here; see the MRI manuals for more
2015 information. Note in particular that the handling of macros and macro
2016 arguments is somewhat different. The purpose of this option is to permit
2017 assembling existing MRI assembler code using @command{@value{AS}}.
2019 The MRI compatibility is not complete. Certain operations of the MRI assembler
2020 depend upon its object file format, and can not be supported using other object
2021 file formats. Supporting these would require enhancing each object file format
2022 individually. These are:
2025 @item global symbols in common section
2027 The m68k MRI assembler supports common sections which are merged by the linker.
2028 Other object file formats do not support this. @command{@value{AS}} handles
2029 common sections by treating them as a single common symbol. It permits local
2030 symbols to be defined within a common section, but it can not support global
2031 symbols, since it has no way to describe them.
2033 @item complex relocations
2035 The MRI assemblers support relocations against a negated section address, and
2036 relocations which combine the start addresses of two or more sections. These
2037 are not support by other object file formats.
2039 @item @code{END} pseudo-op specifying start address
2041 The MRI @code{END} pseudo-op permits the specification of a start address.
2042 This is not supported by other object file formats. The start address may
2043 instead be specified using the @option{-e} option to the linker, or in a linker
2046 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2048 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2049 name to the output file. This is not supported by other object file formats.
2051 @item @code{ORG} pseudo-op
2053 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2054 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2055 which changes the location within the current section. Absolute sections are
2056 not supported by other object file formats. The address of a section may be
2057 assigned within a linker script.
2060 There are some other features of the MRI assembler which are not supported by
2061 @command{@value{AS}}, typically either because they are difficult or because they
2062 seem of little consequence. Some of these may be supported in future releases.
2066 @item EBCDIC strings
2068 EBCDIC strings are not supported.
2070 @item packed binary coded decimal
2072 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2073 and @code{DCB.P} pseudo-ops are not supported.
2075 @item @code{FEQU} pseudo-op
2077 The m68k @code{FEQU} pseudo-op is not supported.
2079 @item @code{NOOBJ} pseudo-op
2081 The m68k @code{NOOBJ} pseudo-op is not supported.
2083 @item @code{OPT} branch control options
2085 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2086 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2087 relaxes all branches, whether forward or backward, to an appropriate size, so
2088 these options serve no purpose.
2090 @item @code{OPT} list control options
2092 The following m68k @code{OPT} list control options are ignored: @code{C},
2093 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2094 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2096 @item other @code{OPT} options
2098 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2099 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2101 @item @code{OPT} @code{D} option is default
2103 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2104 @code{OPT NOD} may be used to turn it off.
2106 @item @code{XREF} pseudo-op.
2108 The m68k @code{XREF} pseudo-op is ignored.
2110 @item @code{.debug} pseudo-op
2112 The i960 @code{.debug} pseudo-op is not supported.
2114 @item @code{.extended} pseudo-op
2116 The i960 @code{.extended} pseudo-op is not supported.
2118 @item @code{.list} pseudo-op.
2120 The various options of the i960 @code{.list} pseudo-op are not supported.
2122 @item @code{.optimize} pseudo-op
2124 The i960 @code{.optimize} pseudo-op is not supported.
2126 @item @code{.output} pseudo-op
2128 The i960 @code{.output} pseudo-op is not supported.
2130 @item @code{.setreal} pseudo-op
2132 The i960 @code{.setreal} pseudo-op is not supported.
2137 @section Dependency Tracking: @option{--MD}
2140 @cindex dependency tracking
2143 @command{@value{AS}} can generate a dependency file for the file it creates. This
2144 file consists of a single rule suitable for @code{make} describing the
2145 dependencies of the main source file.
2147 The rule is written to the file named in its argument.
2149 This feature is used in the automatic updating of makefiles.
2152 @section Name the Object File: @option{-o}
2155 @cindex naming object file
2156 @cindex object file name
2157 There is always one object file output when you run @command{@value{AS}}. By
2158 default it has the name
2161 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2175 You use this option (which takes exactly one filename) to give the
2176 object file a different name.
2178 Whatever the object file is called, @command{@value{AS}} overwrites any
2179 existing file of the same name.
2182 @section Join Data and Text Sections: @option{-R}
2185 @cindex data and text sections, joining
2186 @cindex text and data sections, joining
2187 @cindex joining text and data sections
2188 @cindex merging text and data sections
2189 @option{-R} tells @command{@value{AS}} to write the object file as if all
2190 data-section data lives in the text section. This is only done at
2191 the very last moment: your binary data are the same, but data
2192 section parts are relocated differently. The data section part of
2193 your object file is zero bytes long because all its bytes are
2194 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2196 When you specify @option{-R} it would be possible to generate shorter
2197 address displacements (because we do not have to cross between text and
2198 data section). We refrain from doing this simply for compatibility with
2199 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2202 When @command{@value{AS}} is configured for COFF or ELF output,
2203 this option is only useful if you use sections named @samp{.text} and
2208 @option{-R} is not supported for any of the HPPA targets. Using
2209 @option{-R} generates a warning from @command{@value{AS}}.
2213 @section Display Assembly Statistics: @option{--statistics}
2215 @kindex --statistics
2216 @cindex statistics, about assembly
2217 @cindex time, total for assembly
2218 @cindex space used, maximum for assembly
2219 Use @samp{--statistics} to display two statistics about the resources used by
2220 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2221 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2224 @node traditional-format
2225 @section Compatible Output: @option{--traditional-format}
2227 @kindex --traditional-format
2228 For some targets, the output of @command{@value{AS}} is different in some ways
2229 from the output of some existing assembler. This switch requests
2230 @command{@value{AS}} to use the traditional format instead.
2232 For example, it disables the exception frame optimizations which
2233 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2236 @section Announce Version: @option{-v}
2240 @cindex assembler version
2241 @cindex version of assembler
2242 You can find out what version of as is running by including the
2243 option @samp{-v} (which you can also spell as @samp{-version}) on the
2247 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2249 @command{@value{AS}} should never give a warning or error message when
2250 assembling compiler output. But programs written by people often
2251 cause @command{@value{AS}} to give a warning that a particular assumption was
2252 made. All such warnings are directed to the standard error file.
2256 @cindex suppressing warnings
2257 @cindex warnings, suppressing
2258 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2259 This only affects the warning messages: it does not change any particular of
2260 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2263 @kindex --fatal-warnings
2264 @cindex errors, caused by warnings
2265 @cindex warnings, causing error
2266 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2267 files that generate warnings to be in error.
2270 @cindex warnings, switching on
2271 You can switch these options off again by specifying @option{--warn}, which
2272 causes warnings to be output as usual.
2275 @section Generate Object File in Spite of Errors: @option{-Z}
2276 @cindex object file, after errors
2277 @cindex errors, continuing after
2278 After an error message, @command{@value{AS}} normally produces no output. If for
2279 some reason you are interested in object file output even after
2280 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2281 option. If there are any errors, @command{@value{AS}} continues anyways, and
2282 writes an object file after a final warning message of the form @samp{@var{n}
2283 errors, @var{m} warnings, generating bad object file.}
2288 @cindex machine-independent syntax
2289 @cindex syntax, machine-independent
2290 This chapter describes the machine-independent syntax allowed in a
2291 source file. @command{@value{AS}} syntax is similar to what many other
2292 assemblers use; it is inspired by the BSD 4.2
2297 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2301 * Preprocessing:: Preprocessing
2302 * Whitespace:: Whitespace
2303 * Comments:: Comments
2304 * Symbol Intro:: Symbols
2305 * Statements:: Statements
2306 * Constants:: Constants
2310 @section Preprocessing
2312 @cindex preprocessing
2313 The @command{@value{AS}} internal preprocessor:
2315 @cindex whitespace, removed by preprocessor
2317 adjusts and removes extra whitespace. It leaves one space or tab before
2318 the keywords on a line, and turns any other whitespace on the line into
2321 @cindex comments, removed by preprocessor
2323 removes all comments, replacing them with a single space, or an
2324 appropriate number of newlines.
2326 @cindex constants, converted by preprocessor
2328 converts character constants into the appropriate numeric values.
2331 It does not do macro processing, include file handling, or
2332 anything else you may get from your C compiler's preprocessor. You can
2333 do include file processing with the @code{.include} directive
2334 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2335 to get other ``CPP'' style preprocessing by giving the input file a
2336 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2337 Output, gcc.info, Using GNU CC}.
2339 Excess whitespace, comments, and character constants
2340 cannot be used in the portions of the input text that are not
2343 @cindex turning preprocessing on and off
2344 @cindex preprocessing, turning on and off
2347 If the first line of an input file is @code{#NO_APP} or if you use the
2348 @samp{-f} option, whitespace and comments are not removed from the input file.
2349 Within an input file, you can ask for whitespace and comment removal in
2350 specific portions of the by putting a line that says @code{#APP} before the
2351 text that may contain whitespace or comments, and putting a line that says
2352 @code{#NO_APP} after this text. This feature is mainly intend to support
2353 @code{asm} statements in compilers whose output is otherwise free of comments
2360 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2361 Whitespace is used to separate symbols, and to make programs neater for
2362 people to read. Unless within character constants
2363 (@pxref{Characters,,Character Constants}), any whitespace means the same
2364 as exactly one space.
2370 There are two ways of rendering comments to @command{@value{AS}}. In both
2371 cases the comment is equivalent to one space.
2373 Anything from @samp{/*} through the next @samp{*/} is a comment.
2374 This means you may not nest these comments.
2378 The only way to include a newline ('\n') in a comment
2379 is to use this sort of comment.
2382 /* This sort of comment does not nest. */
2385 @cindex line comment character
2386 Anything from a @dfn{line comment} character up to the next newline is
2387 considered a comment and is ignored. The line comment character is target
2388 specific, and some targets multiple comment characters. Some targets also have
2389 line comment characters that only work if they are the first character on a
2390 line. Some targets use a sequence of two characters to introduce a line
2391 comment. Some targets can also change their line comment characters depending
2392 upon command line options that have been used. For more details see the
2393 @emph{Syntax} section in the documentation for individual targets.
2395 If the line comment character is the hash sign (@samp{#}) then it still has the
2396 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2397 to specify logical line numbers:
2400 @cindex lines starting with @code{#}
2401 @cindex logical line numbers
2402 To be compatible with past assemblers, lines that begin with @samp{#} have a
2403 special interpretation. Following the @samp{#} should be an absolute
2404 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2405 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2406 new logical file name. The rest of the line, if any, should be whitespace.
2408 If the first non-whitespace characters on the line are not numeric,
2409 the line is ignored. (Just like a comment.)
2412 # This is an ordinary comment.
2413 # 42-6 "new_file_name" # New logical file name
2414 # This is logical line # 36.
2416 This feature is deprecated, and may disappear from future versions
2417 of @command{@value{AS}}.
2422 @cindex characters used in symbols
2423 @ifclear SPECIAL-SYMS
2424 A @dfn{symbol} is one or more characters chosen from the set of all
2425 letters (both upper and lower case), digits and the three characters
2431 A @dfn{symbol} is one or more characters chosen from the set of all
2432 letters (both upper and lower case), digits and the three characters
2433 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2439 On most machines, you can also use @code{$} in symbol names; exceptions
2440 are noted in @ref{Machine Dependencies}.
2442 No symbol may begin with a digit. Case is significant.
2443 There is no length limit: all characters are significant. Symbols are
2444 delimited by characters not in that set, or by the beginning of a file
2445 (since the source program must end with a newline, the end of a file is
2446 not a possible symbol delimiter). @xref{Symbols}.
2447 @cindex length of symbols
2452 @cindex statements, structure of
2453 @cindex line separator character
2454 @cindex statement separator character
2456 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2457 @dfn{line separator character}. The line separator character is target
2458 specific and described in the @emph{Syntax} section of each
2459 target's documentation. Not all targets support a line separator character.
2460 The newline or line separator character is considered to be part of the
2461 preceding statement. Newlines and separators within character constants are an
2462 exception: they do not end statements.
2464 @cindex newline, required at file end
2465 @cindex EOF, newline must precede
2466 It is an error to end any statement with end-of-file: the last
2467 character of any input file should be a newline.@refill
2469 An empty statement is allowed, and may include whitespace. It is ignored.
2471 @cindex instructions and directives
2472 @cindex directives and instructions
2473 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2474 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2476 A statement begins with zero or more labels, optionally followed by a
2477 key symbol which determines what kind of statement it is. The key
2478 symbol determines the syntax of the rest of the statement. If the
2479 symbol begins with a dot @samp{.} then the statement is an assembler
2480 directive: typically valid for any computer. If the symbol begins with
2481 a letter the statement is an assembly language @dfn{instruction}: it
2482 assembles into a machine language instruction.
2484 Different versions of @command{@value{AS}} for different computers
2485 recognize different instructions. In fact, the same symbol may
2486 represent a different instruction in a different computer's assembly
2490 @cindex @code{:} (label)
2491 @cindex label (@code{:})
2492 A label is a symbol immediately followed by a colon (@code{:}).
2493 Whitespace before a label or after a colon is permitted, but you may not
2494 have whitespace between a label's symbol and its colon. @xref{Labels}.
2497 For HPPA targets, labels need not be immediately followed by a colon, but
2498 the definition of a label must begin in column zero. This also implies that
2499 only one label may be defined on each line.
2503 label: .directive followed by something
2504 another_label: # This is an empty statement.
2505 instruction operand_1, operand_2, @dots{}
2512 A constant is a number, written so that its value is known by
2513 inspection, without knowing any context. Like this:
2516 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2517 .ascii "Ring the bell\7" # A string constant.
2518 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2519 .float 0f-314159265358979323846264338327\
2520 95028841971.693993751E-40 # - pi, a flonum.
2525 * Characters:: Character Constants
2526 * Numbers:: Number Constants
2530 @subsection Character Constants
2532 @cindex character constants
2533 @cindex constants, character
2534 There are two kinds of character constants. A @dfn{character} stands
2535 for one character in one byte and its value may be used in
2536 numeric expressions. String constants (properly called string
2537 @emph{literals}) are potentially many bytes and their values may not be
2538 used in arithmetic expressions.
2542 * Chars:: Characters
2546 @subsubsection Strings
2548 @cindex string constants
2549 @cindex constants, string
2550 A @dfn{string} is written between double-quotes. It may contain
2551 double-quotes or null characters. The way to get special characters
2552 into a string is to @dfn{escape} these characters: precede them with
2553 a backslash @samp{\} character. For example @samp{\\} represents
2554 one backslash: the first @code{\} is an escape which tells
2555 @command{@value{AS}} to interpret the second character literally as a backslash
2556 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2557 escape character). The complete list of escapes follows.
2559 @cindex escape codes, character
2560 @cindex character escape codes
2563 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2565 @cindex @code{\b} (backspace character)
2566 @cindex backspace (@code{\b})
2568 Mnemonic for backspace; for ASCII this is octal code 010.
2571 @c Mnemonic for EOText; for ASCII this is octal code 004.
2573 @cindex @code{\f} (formfeed character)
2574 @cindex formfeed (@code{\f})
2576 Mnemonic for FormFeed; for ASCII this is octal code 014.
2578 @cindex @code{\n} (newline character)
2579 @cindex newline (@code{\n})
2581 Mnemonic for newline; for ASCII this is octal code 012.
2584 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2586 @cindex @code{\r} (carriage return character)
2587 @cindex carriage return (@code{\r})
2589 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2592 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2593 @c other assemblers.
2595 @cindex @code{\t} (tab)
2596 @cindex tab (@code{\t})
2598 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2601 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2602 @c @item \x @var{digit} @var{digit} @var{digit}
2603 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2605 @cindex @code{\@var{ddd}} (octal character code)
2606 @cindex octal character code (@code{\@var{ddd}})
2607 @item \ @var{digit} @var{digit} @var{digit}
2608 An octal character code. The numeric code is 3 octal digits.
2609 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2610 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2612 @cindex @code{\@var{xd...}} (hex character code)
2613 @cindex hex character code (@code{\@var{xd...}})
2614 @item \@code{x} @var{hex-digits...}
2615 A hex character code. All trailing hex digits are combined. Either upper or
2616 lower case @code{x} works.
2618 @cindex @code{\\} (@samp{\} character)
2619 @cindex backslash (@code{\\})
2621 Represents one @samp{\} character.
2624 @c Represents one @samp{'} (accent acute) character.
2625 @c This is needed in single character literals
2626 @c (@xref{Characters,,Character Constants}.) to represent
2629 @cindex @code{\"} (doublequote character)
2630 @cindex doublequote (@code{\"})
2632 Represents one @samp{"} character. Needed in strings to represent
2633 this character, because an unescaped @samp{"} would end the string.
2635 @item \ @var{anything-else}
2636 Any other character when escaped by @kbd{\} gives a warning, but
2637 assembles as if the @samp{\} was not present. The idea is that if
2638 you used an escape sequence you clearly didn't want the literal
2639 interpretation of the following character. However @command{@value{AS}} has no
2640 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2641 code and warns you of the fact.
2644 Which characters are escapable, and what those escapes represent,
2645 varies widely among assemblers. The current set is what we think
2646 the BSD 4.2 assembler recognizes, and is a subset of what most C
2647 compilers recognize. If you are in doubt, do not use an escape
2651 @subsubsection Characters
2653 @cindex single character constant
2654 @cindex character, single
2655 @cindex constant, single character
2656 A single character may be written as a single quote immediately
2657 followed by that character. The same escapes apply to characters as
2658 to strings. So if you want to write the character backslash, you
2659 must write @kbd{'\\} where the first @code{\} escapes the second
2660 @code{\}. As you can see, the quote is an acute accent, not a
2661 grave accent. A newline
2663 @ifclear abnormal-separator
2664 (or semicolon @samp{;})
2666 @ifset abnormal-separator
2668 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2673 immediately following an acute accent is taken as a literal character
2674 and does not count as the end of a statement. The value of a character
2675 constant in a numeric expression is the machine's byte-wide code for
2676 that character. @command{@value{AS}} assumes your character code is ASCII:
2677 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2680 @subsection Number Constants
2682 @cindex constants, number
2683 @cindex number constants
2684 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2685 are stored in the target machine. @emph{Integers} are numbers that
2686 would fit into an @code{int} in the C language. @emph{Bignums} are
2687 integers, but they are stored in more than 32 bits. @emph{Flonums}
2688 are floating point numbers, described below.
2691 * Integers:: Integers
2696 * Bit Fields:: Bit Fields
2702 @subsubsection Integers
2704 @cindex constants, integer
2706 @cindex binary integers
2707 @cindex integers, binary
2708 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2709 the binary digits @samp{01}.
2711 @cindex octal integers
2712 @cindex integers, octal
2713 An octal integer is @samp{0} followed by zero or more of the octal
2714 digits (@samp{01234567}).
2716 @cindex decimal integers
2717 @cindex integers, decimal
2718 A decimal integer starts with a non-zero digit followed by zero or
2719 more digits (@samp{0123456789}).
2721 @cindex hexadecimal integers
2722 @cindex integers, hexadecimal
2723 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2724 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2726 Integers have the usual values. To denote a negative integer, use
2727 the prefix operator @samp{-} discussed under expressions
2728 (@pxref{Prefix Ops,,Prefix Operators}).
2731 @subsubsection Bignums
2734 @cindex constants, bignum
2735 A @dfn{bignum} has the same syntax and semantics as an integer
2736 except that the number (or its negative) takes more than 32 bits to
2737 represent in binary. The distinction is made because in some places
2738 integers are permitted while bignums are not.
2741 @subsubsection Flonums
2743 @cindex floating point numbers
2744 @cindex constants, floating point
2746 @cindex precision, floating point
2747 A @dfn{flonum} represents a floating point number. The translation is
2748 indirect: a decimal floating point number from the text is converted by
2749 @command{@value{AS}} to a generic binary floating point number of more than
2750 sufficient precision. This generic floating point number is converted
2751 to a particular computer's floating point format (or formats) by a
2752 portion of @command{@value{AS}} specialized to that computer.
2754 A flonum is written by writing (in order)
2759 (@samp{0} is optional on the HPPA.)
2763 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2765 @kbd{e} is recommended. Case is not important.
2767 @c FIXME: verify if flonum syntax really this vague for most cases
2768 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2769 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2772 On the H8/300, Renesas / SuperH SH,
2773 and AMD 29K architectures, the letter must be
2774 one of the letters @samp{DFPRSX} (in upper or lower case).
2776 On the ARC, the letter must be one of the letters @samp{DFRS}
2777 (in upper or lower case).
2779 On the Intel 960 architecture, the letter must be
2780 one of the letters @samp{DFT} (in upper or lower case).
2782 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2786 One of the letters @samp{DFRS} (in upper or lower case).
2789 One of the letters @samp{DFPRSX} (in upper or lower case).
2792 The letter @samp{E} (upper case only).
2795 One of the letters @samp{DFT} (in upper or lower case).
2800 An optional sign: either @samp{+} or @samp{-}.
2803 An optional @dfn{integer part}: zero or more decimal digits.
2806 An optional @dfn{fractional part}: @samp{.} followed by zero
2807 or more decimal digits.
2810 An optional exponent, consisting of:
2814 An @samp{E} or @samp{e}.
2815 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2816 @c principle this can perfectly well be different on different targets.
2818 Optional sign: either @samp{+} or @samp{-}.
2820 One or more decimal digits.
2825 At least one of the integer part or the fractional part must be
2826 present. The floating point number has the usual base-10 value.
2828 @command{@value{AS}} does all processing using integers. Flonums are computed
2829 independently of any floating point hardware in the computer running
2830 @command{@value{AS}}.
2834 @c Bit fields are written as a general facility but are also controlled
2835 @c by a conditional-compilation flag---which is as of now (21mar91)
2836 @c turned on only by the i960 config of GAS.
2838 @subsubsection Bit Fields
2841 @cindex constants, bit field
2842 You can also define numeric constants as @dfn{bit fields}.
2843 Specify two numbers separated by a colon---
2845 @var{mask}:@var{value}
2848 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2851 The resulting number is then packed
2853 @c this conditional paren in case bit fields turned on elsewhere than 960
2854 (in host-dependent byte order)
2856 into a field whose width depends on which assembler directive has the
2857 bit-field as its argument. Overflow (a result from the bitwise and
2858 requiring more binary digits to represent) is not an error; instead,
2859 more constants are generated, of the specified width, beginning with the
2860 least significant digits.@refill
2862 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2863 @code{.short}, and @code{.word} accept bit-field arguments.
2868 @chapter Sections and Relocation
2873 * Secs Background:: Background
2874 * Ld Sections:: Linker Sections
2875 * As Sections:: Assembler Internal Sections
2876 * Sub-Sections:: Sub-Sections
2880 @node Secs Background
2883 Roughly, a section is a range of addresses, with no gaps; all data
2884 ``in'' those addresses is treated the same for some particular purpose.
2885 For example there may be a ``read only'' section.
2887 @cindex linker, and assembler
2888 @cindex assembler, and linker
2889 The linker @code{@value{LD}} reads many object files (partial programs) and
2890 combines their contents to form a runnable program. When @command{@value{AS}}
2891 emits an object file, the partial program is assumed to start at address 0.
2892 @code{@value{LD}} assigns the final addresses for the partial program, so that
2893 different partial programs do not overlap. This is actually an
2894 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2897 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2898 addresses. These blocks slide to their run-time addresses as rigid
2899 units; their length does not change and neither does the order of bytes
2900 within them. Such a rigid unit is called a @emph{section}. Assigning
2901 run-time addresses to sections is called @dfn{relocation}. It includes
2902 the task of adjusting mentions of object-file addresses so they refer to
2903 the proper run-time addresses.
2905 For the H8/300, and for the Renesas / SuperH SH,
2906 @command{@value{AS}} pads sections if needed to
2907 ensure they end on a word (sixteen bit) boundary.
2910 @cindex standard assembler sections
2911 An object file written by @command{@value{AS}} has at least three sections, any
2912 of which may be empty. These are named @dfn{text}, @dfn{data} and
2917 When it generates COFF or ELF output,
2919 @command{@value{AS}} can also generate whatever other named sections you specify
2920 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2921 If you do not use any directives that place output in the @samp{.text}
2922 or @samp{.data} sections, these sections still exist, but are empty.
2927 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2929 @command{@value{AS}} can also generate whatever other named sections you
2930 specify using the @samp{.space} and @samp{.subspace} directives. See
2931 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2932 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2933 assembler directives.
2936 Additionally, @command{@value{AS}} uses different names for the standard
2937 text, data, and bss sections when generating SOM output. Program text
2938 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2939 BSS into @samp{$BSS$}.
2943 Within the object file, the text section starts at address @code{0}, the
2944 data section follows, and the bss section follows the data section.
2947 When generating either SOM or ELF output files on the HPPA, the text
2948 section starts at address @code{0}, the data section at address
2949 @code{0x4000000}, and the bss section follows the data section.
2952 To let @code{@value{LD}} know which data changes when the sections are
2953 relocated, and how to change that data, @command{@value{AS}} also writes to the
2954 object file details of the relocation needed. To perform relocation
2955 @code{@value{LD}} must know, each time an address in the object
2959 Where in the object file is the beginning of this reference to
2962 How long (in bytes) is this reference?
2964 Which section does the address refer to? What is the numeric value of
2966 (@var{address}) @minus{} (@var{start-address of section})?
2969 Is the reference to an address ``Program-Counter relative''?
2972 @cindex addresses, format of
2973 @cindex section-relative addressing
2974 In fact, every address @command{@value{AS}} ever uses is expressed as
2976 (@var{section}) + (@var{offset into section})
2979 Further, most expressions @command{@value{AS}} computes have this section-relative
2982 (For some object formats, such as SOM for the HPPA, some expressions are
2983 symbol-relative instead.)
2986 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2987 @var{N} into section @var{secname}.''
2989 Apart from text, data and bss sections you need to know about the
2990 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2991 addresses in the absolute section remain unchanged. For example, address
2992 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2993 @code{@value{LD}}. Although the linker never arranges two partial programs'
2994 data sections with overlapping addresses after linking, @emph{by definition}
2995 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2996 part of a program is always the same address when the program is running as
2997 address @code{@{absolute@ 239@}} in any other part of the program.
2999 The idea of sections is extended to the @dfn{undefined} section. Any
3000 address whose section is unknown at assembly time is by definition
3001 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3002 Since numbers are always defined, the only way to generate an undefined
3003 address is to mention an undefined symbol. A reference to a named
3004 common block would be such a symbol: its value is unknown at assembly
3005 time so it has section @emph{undefined}.
3007 By analogy the word @emph{section} is used to describe groups of sections in
3008 the linked program. @code{@value{LD}} puts all partial programs' text
3009 sections in contiguous addresses in the linked program. It is
3010 customary to refer to the @emph{text section} of a program, meaning all
3011 the addresses of all partial programs' text sections. Likewise for
3012 data and bss sections.
3014 Some sections are manipulated by @code{@value{LD}}; others are invented for
3015 use of @command{@value{AS}} and have no meaning except during assembly.
3018 @section Linker Sections
3019 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3024 @cindex named sections
3025 @cindex sections, named
3026 @item named sections
3029 @cindex text section
3030 @cindex data section
3034 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3035 separate but equal sections. Anything you can say of one section is
3038 When the program is running, however, it is
3039 customary for the text section to be unalterable. The
3040 text section is often shared among processes: it contains
3041 instructions, constants and the like. The data section of a running
3042 program is usually alterable: for example, C variables would be stored
3043 in the data section.
3048 This section contains zeroed bytes when your program begins running. It
3049 is used to hold uninitialized variables or common storage. The length of
3050 each partial program's bss section is important, but because it starts
3051 out containing zeroed bytes there is no need to store explicit zero
3052 bytes in the object file. The bss section was invented to eliminate
3053 those explicit zeros from object files.
3055 @cindex absolute section
3056 @item absolute section
3057 Address 0 of this section is always ``relocated'' to runtime address 0.
3058 This is useful if you want to refer to an address that @code{@value{LD}} must
3059 not change when relocating. In this sense we speak of absolute
3060 addresses being ``unrelocatable'': they do not change during relocation.
3062 @cindex undefined section
3063 @item undefined section
3064 This ``section'' is a catch-all for address references to objects not in
3065 the preceding sections.
3066 @c FIXME: ref to some other doc on obj-file formats could go here.
3069 @cindex relocation example
3070 An idealized example of three relocatable sections follows.
3072 The example uses the traditional section names @samp{.text} and @samp{.data}.
3074 Memory addresses are on the horizontal axis.
3078 @c END TEXI2ROFF-KILL
3081 partial program # 1: |ttttt|dddd|00|
3088 partial program # 2: |TTT|DDD|000|
3091 +--+---+-----+--+----+---+-----+~~
3092 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3093 +--+---+-----+--+----+---+-----+~~
3095 addresses: 0 @dots{}
3102 \line{\it Partial program \#1: \hfil}
3103 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3104 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3106 \line{\it Partial program \#2: \hfil}
3107 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3108 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3110 \line{\it linked program: \hfil}
3111 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3112 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3113 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3114 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3116 \line{\it addresses: \hfil}
3120 @c END TEXI2ROFF-KILL
3123 @section Assembler Internal Sections
3125 @cindex internal assembler sections
3126 @cindex sections in messages, internal
3127 These sections are meant only for the internal use of @command{@value{AS}}. They
3128 have no meaning at run-time. You do not really need to know about these
3129 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3130 warning messages, so it might be helpful to have an idea of their
3131 meanings to @command{@value{AS}}. These sections are used to permit the
3132 value of every expression in your assembly language program to be a
3133 section-relative address.
3136 @cindex assembler internal logic error
3137 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3138 An internal assembler logic error has been found. This means there is a
3139 bug in the assembler.
3141 @cindex expr (internal section)
3143 The assembler stores complex expression internally as combinations of
3144 symbols. When it needs to represent an expression as a symbol, it puts
3145 it in the expr section.
3147 @c FIXME item transfer[t] vector preload
3148 @c FIXME item transfer[t] vector postload
3149 @c FIXME item register
3153 @section Sub-Sections
3155 @cindex numbered subsections
3156 @cindex grouping data
3162 fall into two sections: text and data.
3164 You may have separate groups of
3166 data in named sections
3170 data in named sections
3176 that you want to end up near to each other in the object file, even though they
3177 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3178 use @dfn{subsections} for this purpose. Within each section, there can be
3179 numbered subsections with values from 0 to 8192. Objects assembled into the
3180 same subsection go into the object file together with other objects in the same
3181 subsection. For example, a compiler might want to store constants in the text
3182 section, but might not want to have them interspersed with the program being
3183 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3184 section of code being output, and a @samp{.text 1} before each group of
3185 constants being output.
3187 Subsections are optional. If you do not use subsections, everything
3188 goes in subsection number zero.
3191 Each subsection is zero-padded up to a multiple of four bytes.
3192 (Subsections may be padded a different amount on different flavors
3193 of @command{@value{AS}}.)
3197 On the H8/300 platform, each subsection is zero-padded to a word
3198 boundary (two bytes).
3199 The same is true on the Renesas SH.
3202 @c FIXME section padding (alignment)?
3203 @c Rich Pixley says padding here depends on target obj code format; that
3204 @c doesn't seem particularly useful to say without further elaboration,
3205 @c so for now I say nothing about it. If this is a generic BFD issue,
3206 @c these paragraphs might need to vanish from this manual, and be
3207 @c discussed in BFD chapter of binutils (or some such).
3211 Subsections appear in your object file in numeric order, lowest numbered
3212 to highest. (All this to be compatible with other people's assemblers.)
3213 The object file contains no representation of subsections; @code{@value{LD}} and
3214 other programs that manipulate object files see no trace of them.
3215 They just see all your text subsections as a text section, and all your
3216 data subsections as a data section.
3218 To specify which subsection you want subsequent statements assembled
3219 into, use a numeric argument to specify it, in a @samp{.text
3220 @var{expression}} or a @samp{.data @var{expression}} statement.
3223 When generating COFF output, you
3228 can also use an extra subsection
3229 argument with arbitrary named sections: @samp{.section @var{name},
3234 When generating ELF output, you
3239 can also use the @code{.subsection} directive (@pxref{SubSection})
3240 to specify a subsection: @samp{.subsection @var{expression}}.
3242 @var{Expression} should be an absolute expression
3243 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3244 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3245 begins in @code{text 0}. For instance:
3247 .text 0 # The default subsection is text 0 anyway.
3248 .ascii "This lives in the first text subsection. *"
3250 .ascii "But this lives in the second text subsection."
3252 .ascii "This lives in the data section,"
3253 .ascii "in the first data subsection."
3255 .ascii "This lives in the first text section,"
3256 .ascii "immediately following the asterisk (*)."
3259 Each section has a @dfn{location counter} incremented by one for every byte
3260 assembled into that section. Because subsections are merely a convenience
3261 restricted to @command{@value{AS}} there is no concept of a subsection location
3262 counter. There is no way to directly manipulate a location counter---but the
3263 @code{.align} directive changes it, and any label definition captures its
3264 current value. The location counter of the section where statements are being
3265 assembled is said to be the @dfn{active} location counter.
3268 @section bss Section
3271 @cindex common variable storage
3272 The bss section is used for local common variable storage.
3273 You may allocate address space in the bss section, but you may
3274 not dictate data to load into it before your program executes. When
3275 your program starts running, all the contents of the bss
3276 section are zeroed bytes.
3278 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3279 @ref{Lcomm,,@code{.lcomm}}.
3281 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3282 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3285 When assembling for a target which supports multiple sections, such as ELF or
3286 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3287 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3288 section. Typically the section will only contain symbol definitions and
3289 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3296 Symbols are a central concept: the programmer uses symbols to name
3297 things, the linker uses symbols to link, and the debugger uses symbols
3301 @cindex debuggers, and symbol order
3302 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3303 the same order they were declared. This may break some debuggers.
3308 * Setting Symbols:: Giving Symbols Other Values
3309 * Symbol Names:: Symbol Names
3310 * Dot:: The Special Dot Symbol
3311 * Symbol Attributes:: Symbol Attributes
3318 A @dfn{label} is written as a symbol immediately followed by a colon
3319 @samp{:}. The symbol then represents the current value of the
3320 active location counter, and is, for example, a suitable instruction
3321 operand. You are warned if you use the same symbol to represent two
3322 different locations: the first definition overrides any other
3326 On the HPPA, the usual form for a label need not be immediately followed by a
3327 colon, but instead must start in column zero. Only one label may be defined on
3328 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3329 provides a special directive @code{.label} for defining labels more flexibly.
3332 @node Setting Symbols
3333 @section Giving Symbols Other Values
3335 @cindex assigning values to symbols
3336 @cindex symbol values, assigning
3337 A symbol can be given an arbitrary value by writing a symbol, followed
3338 by an equals sign @samp{=}, followed by an expression
3339 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3340 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3341 equals sign @samp{=}@samp{=} here represents an equivalent of the
3342 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3345 Blackfin does not support symbol assignment with @samp{=}.
3349 @section Symbol Names
3351 @cindex symbol names
3352 @cindex names, symbol
3353 @ifclear SPECIAL-SYMS
3354 Symbol names begin with a letter or with one of @samp{._}. On most
3355 machines, you can also use @code{$} in symbol names; exceptions are
3356 noted in @ref{Machine Dependencies}. That character may be followed by any
3357 string of digits, letters, dollar signs (unless otherwise noted for a
3358 particular target machine), and underscores.
3362 Symbol names begin with a letter or with one of @samp{._}. On the
3363 Renesas SH you can also use @code{$} in symbol names. That
3364 character may be followed by any string of digits, letters, dollar signs (save
3365 on the H8/300), and underscores.
3369 Case of letters is significant: @code{foo} is a different symbol name
3372 Each symbol has exactly one name. Each name in an assembly language program
3373 refers to exactly one symbol. You may use that symbol name any number of times
3376 @subheading Local Symbol Names
3378 @cindex local symbol names
3379 @cindex symbol names, local
3380 A local symbol is any symbol beginning with certain local label prefixes.
3381 By default, the local label prefix is @samp{.L} for ELF systems or
3382 @samp{L} for traditional a.out systems, but each target may have its own
3383 set of local label prefixes.
3385 On the HPPA local symbols begin with @samp{L$}.
3388 Local symbols are defined and used within the assembler, but they are
3389 normally not saved in object files. Thus, they are not visible when debugging.
3390 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols:
3391 @option{-L}}) to retain the local symbols in the object files.
3393 @subheading Local Labels
3395 @cindex local labels
3396 @cindex temporary symbol names
3397 @cindex symbol names, temporary
3398 Local labels help compilers and programmers use names temporarily.
3399 They create symbols which are guaranteed to be unique over the entire scope of
3400 the input source code and which can be referred to by a simple notation.
3401 To define a local label, write a label of the form @samp{@b{N}:} (where @b{N}
3402 represents any positive integer). To refer to the most recent previous
3403 definition of that label write @samp{@b{N}b}, using the same number as when
3404 you defined the label. To refer to the next definition of a local label, write
3405 @samp{@b{N}f}---the @samp{b} stands for ``backwards'' and the @samp{f} stands
3408 There is no restriction on how you can use these labels, and you can reuse them
3409 too. So that it is possible to repeatedly define the same local label (using
3410 the same number @samp{@b{N}}), although you can only refer to the most recently
3411 defined local label of that number (for a backwards reference) or the next
3412 definition of a specific local label for a forward reference. It is also worth
3413 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3414 implemented in a slightly more efficient manner than the others.
3425 Which is the equivalent of:
3428 label_1: branch label_3
3429 label_2: branch label_1
3430 label_3: branch label_4
3431 label_4: branch label_3
3434 Local label names are only a notational device. They are immediately
3435 transformed into more conventional symbol names before the assembler uses them.
3436 The symbol names are stored in the symbol table, appear in error messages, and
3437 are optionally emitted to the object file. The names are constructed using
3441 @item @emph{local label prefix}
3442 All local symbols begin with the system-specific local label prefix.
3443 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3444 that start with the local label prefix. These labels are
3445 used for symbols you are never intended to see. If you use the
3446 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3447 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3448 you may use them in debugging.
3451 This is the number that was used in the local label definition. So if the
3452 label is written @samp{55:} then the number is @samp{55}.
3455 This unusual character is included so you do not accidentally invent a symbol
3456 of the same name. The character has ASCII value of @samp{\002} (control-B).
3458 @item @emph{ordinal number}
3459 This is a serial number to keep the labels distinct. The first definition of
3460 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3461 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3462 the number @samp{1} and its 15th definition gets @samp{15} as well.
3465 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3466 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3468 @subheading Dollar Local Labels
3469 @cindex dollar local symbols
3471 @code{@value{AS}} also supports an even more local form of local labels called
3472 dollar labels. These labels go out of scope (i.e., they become undefined) as
3473 soon as a non-local label is defined. Thus they remain valid for only a small
3474 region of the input source code. Normal local labels, by contrast, remain in
3475 scope for the entire file, or until they are redefined by another occurrence of
3476 the same local label.
3478 Dollar labels are defined in exactly the same way as ordinary local labels,
3479 except that they have a dollar sign suffix to their numeric value, e.g.,
3482 They can also be distinguished from ordinary local labels by their transformed
3483 names which use ASCII character @samp{\001} (control-A) as the magic character
3484 to distinguish them from ordinary labels. For example, the fifth definition of
3485 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3488 @section The Special Dot Symbol
3490 @cindex dot (symbol)
3491 @cindex @code{.} (symbol)
3492 @cindex current address
3493 @cindex location counter
3494 The special symbol @samp{.} refers to the current address that
3495 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3496 .long .} defines @code{melvin} to contain its own address.
3497 Assigning a value to @code{.} is treated the same as a @code{.org}
3499 @ifclear no-space-dir
3500 Thus, the expression @samp{.=.+4} is the same as saying
3504 @node Symbol Attributes
3505 @section Symbol Attributes
3507 @cindex symbol attributes
3508 @cindex attributes, symbol
3509 Every symbol has, as well as its name, the attributes ``Value'' and
3510 ``Type''. Depending on output format, symbols can also have auxiliary
3513 The detailed definitions are in @file{a.out.h}.
3516 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3517 all these attributes, and probably won't warn you. This makes the
3518 symbol an externally defined symbol, which is generally what you
3522 * Symbol Value:: Value
3523 * Symbol Type:: Type
3526 * a.out Symbols:: Symbol Attributes: @code{a.out}
3530 * a.out Symbols:: Symbol Attributes: @code{a.out}
3533 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3538 * COFF Symbols:: Symbol Attributes for COFF
3541 * SOM Symbols:: Symbol Attributes for SOM
3548 @cindex value of a symbol
3549 @cindex symbol value
3550 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3551 location in the text, data, bss or absolute sections the value is the
3552 number of addresses from the start of that section to the label.
3553 Naturally for text, data and bss sections the value of a symbol changes
3554 as @code{@value{LD}} changes section base addresses during linking. Absolute
3555 symbols' values do not change during linking: that is why they are
3558 The value of an undefined symbol is treated in a special way. If it is
3559 0 then the symbol is not defined in this assembler source file, and
3560 @code{@value{LD}} tries to determine its value from other files linked into the
3561 same program. You make this kind of symbol simply by mentioning a symbol
3562 name without defining it. A non-zero value represents a @code{.comm}
3563 common declaration. The value is how much common storage to reserve, in
3564 bytes (addresses). The symbol refers to the first address of the
3570 @cindex type of a symbol
3572 The type attribute of a symbol contains relocation (section)
3573 information, any flag settings indicating that a symbol is external, and
3574 (optionally), other information for linkers and debuggers. The exact
3575 format depends on the object-code output format in use.
3580 @c The following avoids a "widow" subsection title. @group would be
3581 @c better if it were available outside examples.
3584 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3586 @cindex @code{b.out} symbol attributes
3587 @cindex symbol attributes, @code{b.out}
3588 These symbol attributes appear only when @command{@value{AS}} is configured for
3589 one of the Berkeley-descended object output formats---@code{a.out} or
3595 @subsection Symbol Attributes: @code{a.out}
3597 @cindex @code{a.out} symbol attributes
3598 @cindex symbol attributes, @code{a.out}
3604 @subsection Symbol Attributes: @code{a.out}
3606 @cindex @code{a.out} symbol attributes
3607 @cindex symbol attributes, @code{a.out}
3611 * Symbol Desc:: Descriptor
3612 * Symbol Other:: Other
3616 @subsubsection Descriptor
3618 @cindex descriptor, of @code{a.out} symbol
3619 This is an arbitrary 16-bit value. You may establish a symbol's
3620 descriptor value by using a @code{.desc} statement
3621 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3622 @command{@value{AS}}.
3625 @subsubsection Other
3627 @cindex other attribute, of @code{a.out} symbol
3628 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3633 @subsection Symbol Attributes for COFF
3635 @cindex COFF symbol attributes
3636 @cindex symbol attributes, COFF
3638 The COFF format supports a multitude of auxiliary symbol attributes;
3639 like the primary symbol attributes, they are set between @code{.def} and
3640 @code{.endef} directives.
3642 @subsubsection Primary Attributes
3644 @cindex primary attributes, COFF symbols
3645 The symbol name is set with @code{.def}; the value and type,
3646 respectively, with @code{.val} and @code{.type}.
3648 @subsubsection Auxiliary Attributes
3650 @cindex auxiliary attributes, COFF symbols
3651 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3652 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3653 table information for COFF.
3658 @subsection Symbol Attributes for SOM
3660 @cindex SOM symbol attributes
3661 @cindex symbol attributes, SOM
3663 The SOM format for the HPPA supports a multitude of symbol attributes set with
3664 the @code{.EXPORT} and @code{.IMPORT} directives.
3666 The attributes are described in @cite{HP9000 Series 800 Assembly
3667 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3668 @code{EXPORT} assembler directive documentation.
3672 @chapter Expressions
3676 @cindex numeric values
3677 An @dfn{expression} specifies an address or numeric value.
3678 Whitespace may precede and/or follow an expression.
3680 The result of an expression must be an absolute number, or else an offset into
3681 a particular section. If an expression is not absolute, and there is not
3682 enough information when @command{@value{AS}} sees the expression to know its
3683 section, a second pass over the source program might be necessary to interpret
3684 the expression---but the second pass is currently not implemented.
3685 @command{@value{AS}} aborts with an error message in this situation.
3688 * Empty Exprs:: Empty Expressions
3689 * Integer Exprs:: Integer Expressions
3693 @section Empty Expressions
3695 @cindex empty expressions
3696 @cindex expressions, empty
3697 An empty expression has no value: it is just whitespace or null.
3698 Wherever an absolute expression is required, you may omit the
3699 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3700 is compatible with other assemblers.
3703 @section Integer Expressions
3705 @cindex integer expressions
3706 @cindex expressions, integer
3707 An @dfn{integer expression} is one or more @emph{arguments} delimited
3708 by @emph{operators}.
3711 * Arguments:: Arguments
3712 * Operators:: Operators
3713 * Prefix Ops:: Prefix Operators
3714 * Infix Ops:: Infix Operators
3718 @subsection Arguments
3720 @cindex expression arguments
3721 @cindex arguments in expressions
3722 @cindex operands in expressions
3723 @cindex arithmetic operands
3724 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3725 contexts arguments are sometimes called ``arithmetic operands''. In
3726 this manual, to avoid confusing them with the ``instruction operands'' of
3727 the machine language, we use the term ``argument'' to refer to parts of
3728 expressions only, reserving the word ``operand'' to refer only to machine
3729 instruction operands.
3731 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3732 @var{section} is one of text, data, bss, absolute,
3733 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3736 Numbers are usually integers.
3738 A number can be a flonum or bignum. In this case, you are warned
3739 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3740 these 32 bits are an integer. You may write integer-manipulating
3741 instructions that act on exotic constants, compatible with other
3744 @cindex subexpressions
3745 Subexpressions are a left parenthesis @samp{(} followed by an integer
3746 expression, followed by a right parenthesis @samp{)}; or a prefix
3747 operator followed by an argument.
3750 @subsection Operators
3752 @cindex operators, in expressions
3753 @cindex arithmetic functions
3754 @cindex functions, in expressions
3755 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3756 operators are followed by an argument. Infix operators appear
3757 between their arguments. Operators may be preceded and/or followed by
3761 @subsection Prefix Operator
3763 @cindex prefix operators
3764 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3765 one argument, which must be absolute.
3767 @c the tex/end tex stuff surrounding this small table is meant to make
3768 @c it align, on the printed page, with the similar table in the next
3769 @c section (which is inside an enumerate).
3771 \global\advance\leftskip by \itemindent
3776 @dfn{Negation}. Two's complement negation.
3778 @dfn{Complementation}. Bitwise not.
3782 \global\advance\leftskip by -\itemindent
3786 @subsection Infix Operators
3788 @cindex infix operators
3789 @cindex operators, permitted arguments
3790 @dfn{Infix operators} take two arguments, one on either side. Operators
3791 have precedence, but operations with equal precedence are performed left
3792 to right. Apart from @code{+} or @option{-}, both arguments must be
3793 absolute, and the result is absolute.
3796 @cindex operator precedence
3797 @cindex precedence of operators
3804 @dfn{Multiplication}.
3807 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3813 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3816 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3820 Intermediate precedence
3825 @dfn{Bitwise Inclusive Or}.
3831 @dfn{Bitwise Exclusive Or}.
3834 @dfn{Bitwise Or Not}.
3841 @cindex addition, permitted arguments
3842 @cindex plus, permitted arguments
3843 @cindex arguments for addition
3845 @dfn{Addition}. If either argument is absolute, the result has the section of
3846 the other argument. You may not add together arguments from different
3849 @cindex subtraction, permitted arguments
3850 @cindex minus, permitted arguments
3851 @cindex arguments for subtraction
3853 @dfn{Subtraction}. If the right argument is absolute, the
3854 result has the section of the left argument.
3855 If both arguments are in the same section, the result is absolute.
3856 You may not subtract arguments from different sections.
3857 @c FIXME is there still something useful to say about undefined - undefined ?
3859 @cindex comparison expressions
3860 @cindex expressions, comparison
3865 @dfn{Is Not Equal To}
3869 @dfn{Is Greater Than}
3871 @dfn{Is Greater Than Or Equal To}
3873 @dfn{Is Less Than Or Equal To}
3875 The comparison operators can be used as infix operators. A true results has a
3876 value of -1 whereas a false result has a value of 0. Note, these operators
3877 perform signed comparisons.
3880 @item Lowest Precedence
3889 These two logical operations can be used to combine the results of sub
3890 expressions. Note, unlike the comparison operators a true result returns a
3891 value of 1 but a false results does still return 0. Also note that the logical
3892 or operator has a slightly lower precedence than logical and.
3897 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3898 address; you can only have a defined section in one of the two arguments.
3901 @chapter Assembler Directives
3903 @cindex directives, machine independent
3904 @cindex pseudo-ops, machine independent
3905 @cindex machine independent directives
3906 All assembler directives have names that begin with a period (@samp{.}).
3907 The rest of the name is letters, usually in lower case.
3909 This chapter discusses directives that are available regardless of the
3910 target machine configuration for the @sc{gnu} assembler.
3912 Some machine configurations provide additional directives.
3913 @xref{Machine Dependencies}.
3916 @ifset machine-directives
3917 @xref{Machine Dependencies}, for additional directives.
3922 * Abort:: @code{.abort}
3924 * ABORT (COFF):: @code{.ABORT}
3927 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3928 * Altmacro:: @code{.altmacro}
3929 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3930 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3931 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3932 * Byte:: @code{.byte @var{expressions}}
3933 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
3934 * Comm:: @code{.comm @var{symbol} , @var{length} }
3935 * Data:: @code{.data @var{subsection}}
3937 * Def:: @code{.def @var{name}}
3940 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3946 * Double:: @code{.double @var{flonums}}
3947 * Eject:: @code{.eject}
3948 * Else:: @code{.else}
3949 * Elseif:: @code{.elseif}
3952 * Endef:: @code{.endef}
3955 * Endfunc:: @code{.endfunc}
3956 * Endif:: @code{.endif}
3957 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3958 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3959 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
3961 * Error:: @code{.error @var{string}}
3962 * Exitm:: @code{.exitm}
3963 * Extern:: @code{.extern}
3964 * Fail:: @code{.fail}
3965 * File:: @code{.file}
3966 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3967 * Float:: @code{.float @var{flonums}}
3968 * Func:: @code{.func}
3969 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3971 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
3972 * Hidden:: @code{.hidden @var{names}}
3975 * hword:: @code{.hword @var{expressions}}
3976 * Ident:: @code{.ident}
3977 * If:: @code{.if @var{absolute expression}}
3978 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
3979 * Include:: @code{.include "@var{file}"}
3980 * Int:: @code{.int @var{expressions}}
3982 * Internal:: @code{.internal @var{names}}
3985 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3986 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3987 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3988 * Lflags:: @code{.lflags}
3989 @ifclear no-line-dir
3990 * Line:: @code{.line @var{line-number}}
3993 * Linkonce:: @code{.linkonce [@var{type}]}
3994 * List:: @code{.list}
3995 * Ln:: @code{.ln @var{line-number}}
3996 * Loc:: @code{.loc @var{fileno} @var{lineno}}
3997 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
3999 * Local:: @code{.local @var{names}}
4002 * Long:: @code{.long @var{expressions}}
4004 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4007 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4008 * MRI:: @code{.mri @var{val}}
4009 * Noaltmacro:: @code{.noaltmacro}
4010 * Nolist:: @code{.nolist}
4011 * Octa:: @code{.octa @var{bignums}}
4012 * Org:: @code{.org @var{new-lc}, @var{fill}}
4013 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4015 * PopSection:: @code{.popsection}
4016 * Previous:: @code{.previous}
4019 * Print:: @code{.print @var{string}}
4021 * Protected:: @code{.protected @var{names}}
4024 * Psize:: @code{.psize @var{lines}, @var{columns}}
4025 * Purgem:: @code{.purgem @var{name}}
4027 * PushSection:: @code{.pushsection @var{name}}
4030 * Quad:: @code{.quad @var{bignums}}
4031 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4032 * Rept:: @code{.rept @var{count}}
4033 * Sbttl:: @code{.sbttl "@var{subheading}"}
4035 * Scl:: @code{.scl @var{class}}
4038 * Section:: @code{.section @var{name}[, @var{flags}]}
4041 * Set:: @code{.set @var{symbol}, @var{expression}}
4042 * Short:: @code{.short @var{expressions}}
4043 * Single:: @code{.single @var{flonums}}
4045 * Size:: @code{.size [@var{name} , @var{expression}]}
4047 @ifclear no-space-dir
4048 * Skip:: @code{.skip @var{size} , @var{fill}}
4051 * Sleb128:: @code{.sleb128 @var{expressions}}
4052 @ifclear no-space-dir
4053 * Space:: @code{.space @var{size} , @var{fill}}
4056 * Stab:: @code{.stabd, .stabn, .stabs}
4059 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4060 * Struct:: @code{.struct @var{expression}}
4062 * SubSection:: @code{.subsection}
4063 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4067 * Tag:: @code{.tag @var{structname}}
4070 * Text:: @code{.text @var{subsection}}
4071 * Title:: @code{.title "@var{heading}"}
4073 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4076 * Uleb128:: @code{.uleb128 @var{expressions}}
4078 * Val:: @code{.val @var{addr}}
4082 * Version:: @code{.version "@var{string}"}
4083 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4084 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4087 * Warning:: @code{.warning @var{string}}
4088 * Weak:: @code{.weak @var{names}}
4089 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4090 * Word:: @code{.word @var{expressions}}
4091 * Deprecated:: Deprecated Directives
4095 @section @code{.abort}
4097 @cindex @code{abort} directive
4098 @cindex stopping the assembly
4099 This directive stops the assembly immediately. It is for
4100 compatibility with other assemblers. The original idea was that the
4101 assembly language source would be piped into the assembler. If the sender
4102 of the source quit, it could use this directive tells @command{@value{AS}} to
4103 quit also. One day @code{.abort} will not be supported.
4107 @section @code{.ABORT} (COFF)
4109 @cindex @code{ABORT} directive
4110 When producing COFF output, @command{@value{AS}} accepts this directive as a
4111 synonym for @samp{.abort}.
4114 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4120 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4122 @cindex padding the location counter
4123 @cindex @code{align} directive
4124 Pad the location counter (in the current subsection) to a particular storage
4125 boundary. The first expression (which must be absolute) is the alignment
4126 required, as described below.
4128 The second expression (also absolute) gives the fill value to be stored in the
4129 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4130 padding bytes are normally zero. However, on some systems, if the section is
4131 marked as containing code and the fill value is omitted, the space is filled
4132 with no-op instructions.
4134 The third expression is also absolute, and is also optional. If it is present,
4135 it is the maximum number of bytes that should be skipped by this alignment
4136 directive. If doing the alignment would require skipping more bytes than the
4137 specified maximum, then the alignment is not done at all. You can omit the
4138 fill value (the second argument) entirely by simply using two commas after the
4139 required alignment; this can be useful if you want the alignment to be filled
4140 with no-op instructions when appropriate.
4142 The way the required alignment is specified varies from system to system.
4143 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or32,
4144 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4145 alignment request in bytes. For example @samp{.align 8} advances
4146 the location counter until it is a multiple of 8. If the location counter
4147 is already a multiple of 8, no change is needed. For the tic54x, the
4148 first expression is the alignment request in words.
4150 For other systems, including ppc, i386 using a.out format, arm and
4151 strongarm, it is the
4152 number of low-order zero bits the location counter must have after
4153 advancement. For example @samp{.align 3} advances the location
4154 counter until it a multiple of 8. If the location counter is already a
4155 multiple of 8, no change is needed.
4157 This inconsistency is due to the different behaviors of the various
4158 native assemblers for these systems which GAS must emulate.
4159 GAS also provides @code{.balign} and @code{.p2align} directives,
4160 described later, which have a consistent behavior across all
4161 architectures (but are specific to GAS).
4164 @section @code{.altmacro}
4165 Enable alternate macro mode, enabling:
4168 @item LOCAL @var{name} [ , @dots{} ]
4169 One additional directive, @code{LOCAL}, is available. It is used to
4170 generate a string replacement for each of the @var{name} arguments, and
4171 replace any instances of @var{name} in each macro expansion. The
4172 replacement string is unique in the assembly, and different for each
4173 separate macro expansion. @code{LOCAL} allows you to write macros that
4174 define symbols, without fear of conflict between separate macro expansions.
4176 @item String delimiters
4177 You can write strings delimited in these other ways besides
4178 @code{"@var{string}"}:
4181 @item '@var{string}'
4182 You can delimit strings with single-quote characters.
4184 @item <@var{string}>
4185 You can delimit strings with matching angle brackets.
4188 @item single-character string escape
4189 To include any single character literally in a string (even if the
4190 character would otherwise have some special meaning), you can prefix the
4191 character with @samp{!} (an exclamation mark). For example, you can
4192 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4194 @item Expression results as strings
4195 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4196 and use the result as a string.
4200 @section @code{.ascii "@var{string}"}@dots{}
4202 @cindex @code{ascii} directive
4203 @cindex string literals
4204 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4205 separated by commas. It assembles each string (with no automatic
4206 trailing zero byte) into consecutive addresses.
4209 @section @code{.asciz "@var{string}"}@dots{}
4211 @cindex @code{asciz} directive
4212 @cindex zero-terminated strings
4213 @cindex null-terminated strings
4214 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4215 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4218 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4220 @cindex padding the location counter given number of bytes
4221 @cindex @code{balign} directive
4222 Pad the location counter (in the current subsection) to a particular
4223 storage boundary. The first expression (which must be absolute) is the
4224 alignment request in bytes. For example @samp{.balign 8} advances
4225 the location counter until it is a multiple of 8. If the location counter
4226 is already a multiple of 8, no change is needed.
4228 The second expression (also absolute) gives the fill value to be stored in the
4229 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4230 padding bytes are normally zero. However, on some systems, if the section is
4231 marked as containing code and the fill value is omitted, the space is filled
4232 with no-op instructions.
4234 The third expression is also absolute, and is also optional. If it is present,
4235 it is the maximum number of bytes that should be skipped by this alignment
4236 directive. If doing the alignment would require skipping more bytes than the
4237 specified maximum, then the alignment is not done at all. You can omit the
4238 fill value (the second argument) entirely by simply using two commas after the
4239 required alignment; this can be useful if you want the alignment to be filled
4240 with no-op instructions when appropriate.
4242 @cindex @code{balignw} directive
4243 @cindex @code{balignl} directive
4244 The @code{.balignw} and @code{.balignl} directives are variants of the
4245 @code{.balign} directive. The @code{.balignw} directive treats the fill
4246 pattern as a two byte word value. The @code{.balignl} directives treats the
4247 fill pattern as a four byte longword value. For example, @code{.balignw
4248 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4249 filled in with the value 0x368d (the exact placement of the bytes depends upon
4250 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4254 @section @code{.byte @var{expressions}}
4256 @cindex @code{byte} directive
4257 @cindex integers, one byte
4258 @code{.byte} expects zero or more expressions, separated by commas.
4259 Each expression is assembled into the next byte.
4261 @node CFI directives
4262 @section @code{.cfi_sections @var{section_list}}
4263 @cindex @code{cfi_sections} directive
4264 @code{.cfi_sections} may be used to specify whether CFI directives
4265 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4266 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4267 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4268 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4269 directive is not used is @code{.cfi_sections .eh_frame}.
4271 @section @code{.cfi_startproc [simple]}
4272 @cindex @code{cfi_startproc} directive
4273 @code{.cfi_startproc} is used at the beginning of each function that
4274 should have an entry in @code{.eh_frame}. It initializes some internal
4275 data structures. Don't forget to close the function by
4276 @code{.cfi_endproc}.
4278 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4279 it also emits some architecture dependent initial CFI instructions.
4281 @section @code{.cfi_endproc}
4282 @cindex @code{cfi_endproc} directive
4283 @code{.cfi_endproc} is used at the end of a function where it closes its
4284 unwind entry previously opened by
4285 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4287 @section @code{.cfi_personality @var{encoding} [, @var{exp}]}
4288 @code{.cfi_personality} defines personality routine and its encoding.
4289 @var{encoding} must be a constant determining how the personality
4290 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4291 argument is not present, otherwise second argument should be
4292 a constant or a symbol name. When using indirect encodings,
4293 the symbol provided should be the location where personality
4294 can be loaded from, not the personality routine itself.
4295 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4296 no personality routine.
4298 @section @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4299 @code{.cfi_lsda} defines LSDA and its encoding.
4300 @var{encoding} must be a constant determining how the LSDA
4301 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4302 argument is not present, otherwise second argument should be a constant
4303 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4306 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4307 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4308 address from @var{register} and add @var{offset} to it}.
4310 @section @code{.cfi_def_cfa_register @var{register}}
4311 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4312 now on @var{register} will be used instead of the old one. Offset
4315 @section @code{.cfi_def_cfa_offset @var{offset}}
4316 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4317 remains the same, but @var{offset} is new. Note that it is the
4318 absolute offset that will be added to a defined register to compute
4321 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4322 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4323 value that is added/substracted from the previous offset.
4325 @section @code{.cfi_offset @var{register}, @var{offset}}
4326 Previous value of @var{register} is saved at offset @var{offset} from
4329 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4330 Previous value of @var{register} is saved at offset @var{offset} from
4331 the current CFA register. This is transformed to @code{.cfi_offset}
4332 using the known displacement of the CFA register from the CFA.
4333 This is often easier to use, because the number will match the
4334 code it's annotating.
4336 @section @code{.cfi_register @var{register1}, @var{register2}}
4337 Previous value of @var{register1} is saved in register @var{register2}.
4339 @section @code{.cfi_restore @var{register}}
4340 @code{.cfi_restore} says that the rule for @var{register} is now the
4341 same as it was at the beginning of the function, after all initial
4342 instruction added by @code{.cfi_startproc} were executed.
4344 @section @code{.cfi_undefined @var{register}}
4345 From now on the previous value of @var{register} can't be restored anymore.
4347 @section @code{.cfi_same_value @var{register}}
4348 Current value of @var{register} is the same like in the previous frame,
4349 i.e. no restoration needed.
4351 @section @code{.cfi_remember_state},
4352 First save all current rules for all registers by @code{.cfi_remember_state},
4353 then totally screw them up by subsequent @code{.cfi_*} directives and when
4354 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4355 the previous saved state.
4357 @section @code{.cfi_return_column @var{register}}
4358 Change return column @var{register}, i.e. the return address is either
4359 directly in @var{register} or can be accessed by rules for @var{register}.
4361 @section @code{.cfi_signal_frame}
4362 Mark current function as signal trampoline.
4364 @section @code{.cfi_window_save}
4365 SPARC register window has been saved.
4367 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4368 Allows the user to add arbitrary bytes to the unwind info. One
4369 might use this to add OS-specific CFI opcodes, or generic CFI
4370 opcodes that GAS does not yet support.
4372 @section @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4373 The current value of @var{register} is @var{label}. The value of @var{label}
4374 will be encoded in the output file according to @var{encoding}; see the
4375 description of @code{.cfi_personality} for details on this encoding.
4377 The usefulness of equating a register to a fixed label is probably
4378 limited to the return address register. Here, it can be useful to
4379 mark a code segment that has only one return address which is reached
4380 by a direct branch and no copy of the return address exists in memory
4381 or another register.
4384 @section @code{.comm @var{symbol} , @var{length} }
4386 @cindex @code{comm} directive
4387 @cindex symbol, common
4388 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4389 common symbol in one object file may be merged with a defined or common symbol
4390 of the same name in another object file. If @code{@value{LD}} does not see a
4391 definition for the symbol--just one or more common symbols--then it will
4392 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4393 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4394 the same name, and they do not all have the same size, it will allocate space
4395 using the largest size.
4398 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4399 an optional third argument. This is the desired alignment of the symbol,
4400 specified for ELF as a byte boundary (for example, an alignment of 16 means
4401 that the least significant 4 bits of the address should be zero), and for PE
4402 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4403 boundary). The alignment must be an absolute expression, and it must be a
4404 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4405 common symbol, it will use the alignment when placing the symbol. If no
4406 alignment is specified, @command{@value{AS}} will set the alignment to the
4407 largest power of two less than or equal to the size of the symbol, up to a
4408 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4409 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4410 @samp{--section-alignment} option; image file sections in PE are aligned to
4411 multiples of 4096, which is far too large an alignment for ordinary variables.
4412 It is rather the default alignment for (non-debug) sections within object
4413 (@samp{*.o}) files, which are less strictly aligned.}.
4417 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4418 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4422 @section @code{.data @var{subsection}}
4424 @cindex @code{data} directive
4425 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4426 end of the data subsection numbered @var{subsection} (which is an
4427 absolute expression). If @var{subsection} is omitted, it defaults
4432 @section @code{.def @var{name}}
4434 @cindex @code{def} directive
4435 @cindex COFF symbols, debugging
4436 @cindex debugging COFF symbols
4437 Begin defining debugging information for a symbol @var{name}; the
4438 definition extends until the @code{.endef} directive is encountered.
4441 This directive is only observed when @command{@value{AS}} is configured for COFF
4442 format output; when producing @code{b.out}, @samp{.def} is recognized,
4449 @section @code{.desc @var{symbol}, @var{abs-expression}}
4451 @cindex @code{desc} directive
4452 @cindex COFF symbol descriptor
4453 @cindex symbol descriptor, COFF
4454 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4455 to the low 16 bits of an absolute expression.
4458 The @samp{.desc} directive is not available when @command{@value{AS}} is
4459 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4460 object format. For the sake of compatibility, @command{@value{AS}} accepts
4461 it, but produces no output, when configured for COFF.
4467 @section @code{.dim}
4469 @cindex @code{dim} directive
4470 @cindex COFF auxiliary symbol information
4471 @cindex auxiliary symbol information, COFF
4472 This directive is generated by compilers to include auxiliary debugging
4473 information in the symbol table. It is only permitted inside
4474 @code{.def}/@code{.endef} pairs.
4477 @samp{.dim} is only meaningful when generating COFF format output; when
4478 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4484 @section @code{.double @var{flonums}}
4486 @cindex @code{double} directive
4487 @cindex floating point numbers (double)
4488 @code{.double} expects zero or more flonums, separated by commas. It
4489 assembles floating point numbers.
4491 The exact kind of floating point numbers emitted depends on how
4492 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4496 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4497 in @sc{ieee} format.
4502 @section @code{.eject}
4504 @cindex @code{eject} directive
4505 @cindex new page, in listings
4506 @cindex page, in listings
4507 @cindex listing control: new page
4508 Force a page break at this point, when generating assembly listings.
4511 @section @code{.else}
4513 @cindex @code{else} directive
4514 @code{.else} is part of the @command{@value{AS}} support for conditional
4515 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4516 of code to be assembled if the condition for the preceding @code{.if}
4520 @section @code{.elseif}
4522 @cindex @code{elseif} directive
4523 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4524 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4525 @code{.if} block that would otherwise fill the entire @code{.else} section.
4528 @section @code{.end}
4530 @cindex @code{end} directive
4531 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4532 process anything in the file past the @code{.end} directive.
4536 @section @code{.endef}
4538 @cindex @code{endef} directive
4539 This directive flags the end of a symbol definition begun with
4543 @samp{.endef} is only meaningful when generating COFF format output; if
4544 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4545 directive but ignores it.
4550 @section @code{.endfunc}
4551 @cindex @code{endfunc} directive
4552 @code{.endfunc} marks the end of a function specified with @code{.func}.
4555 @section @code{.endif}
4557 @cindex @code{endif} directive
4558 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4559 it marks the end of a block of code that is only assembled
4560 conditionally. @xref{If,,@code{.if}}.
4563 @section @code{.equ @var{symbol}, @var{expression}}
4565 @cindex @code{equ} directive
4566 @cindex assigning values to symbols
4567 @cindex symbols, assigning values to
4568 This directive sets the value of @var{symbol} to @var{expression}.
4569 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4572 The syntax for @code{equ} on the HPPA is
4573 @samp{@var{symbol} .equ @var{expression}}.
4577 The syntax for @code{equ} on the Z80 is
4578 @samp{@var{symbol} equ @var{expression}}.
4579 On the Z80 it is an eror if @var{symbol} is already defined,
4580 but the symbol is not protected from later redefinition.
4581 Compare @ref{Equiv}.
4585 @section @code{.equiv @var{symbol}, @var{expression}}
4586 @cindex @code{equiv} directive
4587 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4588 the assembler will signal an error if @var{symbol} is already defined. Note a
4589 symbol which has been referenced but not actually defined is considered to be
4592 Except for the contents of the error message, this is roughly equivalent to
4599 plus it protects the symbol from later redefinition.
4602 @section @code{.eqv @var{symbol}, @var{expression}}
4603 @cindex @code{eqv} directive
4604 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
4605 evaluate the expression or any part of it immediately. Instead each time
4606 the resulting symbol is used in an expression, a snapshot of its current
4610 @section @code{.err}
4611 @cindex @code{err} directive
4612 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4613 message and, unless the @option{-Z} option was used, it will not generate an
4614 object file. This can be used to signal an error in conditionally compiled code.
4617 @section @code{.error "@var{string}"}
4618 @cindex error directive
4620 Similarly to @code{.err}, this directive emits an error, but you can specify a
4621 string that will be emitted as the error message. If you don't specify the
4622 message, it defaults to @code{".error directive invoked in source file"}.
4623 @xref{Errors, ,Error and Warning Messages}.
4626 .error "This code has not been assembled and tested."
4630 @section @code{.exitm}
4631 Exit early from the current macro definition. @xref{Macro}.
4634 @section @code{.extern}
4636 @cindex @code{extern} directive
4637 @code{.extern} is accepted in the source program---for compatibility
4638 with other assemblers---but it is ignored. @command{@value{AS}} treats
4639 all undefined symbols as external.
4642 @section @code{.fail @var{expression}}
4644 @cindex @code{fail} directive
4645 Generates an error or a warning. If the value of the @var{expression} is 500
4646 or more, @command{@value{AS}} will print a warning message. If the value is less
4647 than 500, @command{@value{AS}} will print an error message. The message will
4648 include the value of @var{expression}. This can occasionally be useful inside
4649 complex nested macros or conditional assembly.
4652 @section @code{.file}
4653 @cindex @code{file} directive
4655 @ifclear no-file-dir
4656 There are two different versions of the @code{.file} directive. Targets
4657 that support DWARF2 line number information use the DWARF2 version of
4658 @code{.file}. Other targets use the default version.
4660 @subheading Default Version
4662 @cindex logical file name
4663 @cindex file name, logical
4664 This version of the @code{.file} directive tells @command{@value{AS}} that we
4665 are about to start a new logical file. The syntax is:
4671 @var{string} is the new file name. In general, the filename is
4672 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4673 to specify an empty file name, you must give the quotes--@code{""}. This
4674 statement may go away in future: it is only recognized to be compatible with
4675 old @command{@value{AS}} programs.
4677 @subheading DWARF2 Version
4680 When emitting DWARF2 line number information, @code{.file} assigns filenames
4681 to the @code{.debug_line} file name table. The syntax is:
4684 .file @var{fileno} @var{filename}
4687 The @var{fileno} operand should be a unique positive integer to use as the
4688 index of the entry in the table. The @var{filename} operand is a C string
4691 The detail of filename indices is exposed to the user because the filename
4692 table is shared with the @code{.debug_info} section of the DWARF2 debugging
4693 information, and thus the user must know the exact indices that table
4697 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4699 @cindex @code{fill} directive
4700 @cindex writing patterns in memory
4701 @cindex patterns, writing in memory
4702 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4703 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4704 may be zero or more. @var{Size} may be zero or more, but if it is
4705 more than 8, then it is deemed to have the value 8, compatible with
4706 other people's assemblers. The contents of each @var{repeat} bytes
4707 is taken from an 8-byte number. The highest order 4 bytes are
4708 zero. The lowest order 4 bytes are @var{value} rendered in the
4709 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4710 Each @var{size} bytes in a repetition is taken from the lowest order
4711 @var{size} bytes of this number. Again, this bizarre behavior is
4712 compatible with other people's assemblers.
4714 @var{size} and @var{value} are optional.
4715 If the second comma and @var{value} are absent, @var{value} is
4716 assumed zero. If the first comma and following tokens are absent,
4717 @var{size} is assumed to be 1.
4720 @section @code{.float @var{flonums}}
4722 @cindex floating point numbers (single)
4723 @cindex @code{float} directive
4724 This directive assembles zero or more flonums, separated by commas. It
4725 has the same effect as @code{.single}.
4727 The exact kind of floating point numbers emitted depends on how
4728 @command{@value{AS}} is configured.
4729 @xref{Machine Dependencies}.
4733 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4734 in @sc{ieee} format.
4739 @section @code{.func @var{name}[,@var{label}]}
4740 @cindex @code{func} directive
4741 @code{.func} emits debugging information to denote function @var{name}, and
4742 is ignored unless the file is assembled with debugging enabled.
4743 Only @samp{--gstabs[+]} is currently supported.
4744 @var{label} is the entry point of the function and if omitted @var{name}
4745 prepended with the @samp{leading char} is used.
4746 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4747 All functions are currently defined to have @code{void} return type.
4748 The function must be terminated with @code{.endfunc}.
4751 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4753 @cindex @code{global} directive
4754 @cindex symbol, making visible to linker
4755 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4756 @var{symbol} in your partial program, its value is made available to
4757 other partial programs that are linked with it. Otherwise,
4758 @var{symbol} takes its attributes from a symbol of the same name
4759 from another file linked into the same program.
4761 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4762 compatibility with other assemblers.
4765 On the HPPA, @code{.global} is not always enough to make it accessible to other
4766 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4767 @xref{HPPA Directives, ,HPPA Assembler Directives}.
4772 @section @code{.gnu_attribute @var{tag},@var{value}}
4773 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
4776 @section @code{.hidden @var{names}}
4778 @cindex @code{hidden} directive
4780 This is one of the ELF visibility directives. The other two are
4781 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4782 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4784 This directive overrides the named symbols default visibility (which is set by
4785 their binding: local, global or weak). The directive sets the visibility to
4786 @code{hidden} which means that the symbols are not visible to other components.
4787 Such symbols are always considered to be @code{protected} as well.
4791 @section @code{.hword @var{expressions}}
4793 @cindex @code{hword} directive
4794 @cindex integers, 16-bit
4795 @cindex numbers, 16-bit
4796 @cindex sixteen bit integers
4797 This expects zero or more @var{expressions}, and emits
4798 a 16 bit number for each.
4801 This directive is a synonym for @samp{.short}; depending on the target
4802 architecture, it may also be a synonym for @samp{.word}.
4806 This directive is a synonym for @samp{.short}.
4809 This directive is a synonym for both @samp{.short} and @samp{.word}.
4814 @section @code{.ident}
4816 @cindex @code{ident} directive
4818 This directive is used by some assemblers to place tags in object files. The
4819 behavior of this directive varies depending on the target. When using the
4820 a.out object file format, @command{@value{AS}} simply accepts the directive for
4821 source-file compatibility with existing assemblers, but does not emit anything
4822 for it. When using COFF, comments are emitted to the @code{.comment} or
4823 @code{.rdata} section, depending on the target. When using ELF, comments are
4824 emitted to the @code{.comment} section.
4827 @section @code{.if @var{absolute expression}}
4829 @cindex conditional assembly
4830 @cindex @code{if} directive
4831 @code{.if} marks the beginning of a section of code which is only
4832 considered part of the source program being assembled if the argument
4833 (which must be an @var{absolute expression}) is non-zero. The end of
4834 the conditional section of code must be marked by @code{.endif}
4835 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4836 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4837 If you have several conditions to check, @code{.elseif} may be used to avoid
4838 nesting blocks if/else within each subsequent @code{.else} block.
4840 The following variants of @code{.if} are also supported:
4842 @cindex @code{ifdef} directive
4843 @item .ifdef @var{symbol}
4844 Assembles the following section of code if the specified @var{symbol}
4845 has been defined. Note a symbol which has been referenced but not yet defined
4846 is considered to be undefined.
4848 @cindex @code{ifb} directive
4849 @item .ifb @var{text}
4850 Assembles the following section of code if the operand is blank (empty).
4852 @cindex @code{ifc} directive
4853 @item .ifc @var{string1},@var{string2}
4854 Assembles the following section of code if the two strings are the same. The
4855 strings may be optionally quoted with single quotes. If they are not quoted,
4856 the first string stops at the first comma, and the second string stops at the
4857 end of the line. Strings which contain whitespace should be quoted. The
4858 string comparison is case sensitive.
4860 @cindex @code{ifeq} directive
4861 @item .ifeq @var{absolute expression}
4862 Assembles the following section of code if the argument is zero.
4864 @cindex @code{ifeqs} directive
4865 @item .ifeqs @var{string1},@var{string2}
4866 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4868 @cindex @code{ifge} directive
4869 @item .ifge @var{absolute expression}
4870 Assembles the following section of code if the argument is greater than or
4873 @cindex @code{ifgt} directive
4874 @item .ifgt @var{absolute expression}
4875 Assembles the following section of code if the argument is greater than zero.
4877 @cindex @code{ifle} directive
4878 @item .ifle @var{absolute expression}
4879 Assembles the following section of code if the argument is less than or equal
4882 @cindex @code{iflt} directive
4883 @item .iflt @var{absolute expression}
4884 Assembles the following section of code if the argument is less than zero.
4886 @cindex @code{ifnb} directive
4887 @item .ifnb @var{text}
4888 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
4889 following section of code if the operand is non-blank (non-empty).
4891 @cindex @code{ifnc} directive
4892 @item .ifnc @var{string1},@var{string2}.
4893 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4894 following section of code if the two strings are not the same.
4896 @cindex @code{ifndef} directive
4897 @cindex @code{ifnotdef} directive
4898 @item .ifndef @var{symbol}
4899 @itemx .ifnotdef @var{symbol}
4900 Assembles the following section of code if the specified @var{symbol}
4901 has not been defined. Both spelling variants are equivalent. Note a symbol
4902 which has been referenced but not yet defined is considered to be undefined.
4904 @cindex @code{ifne} directive
4905 @item .ifne @var{absolute expression}
4906 Assembles the following section of code if the argument is not equal to zero
4907 (in other words, this is equivalent to @code{.if}).
4909 @cindex @code{ifnes} directive
4910 @item .ifnes @var{string1},@var{string2}
4911 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4912 following section of code if the two strings are not the same.
4916 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4918 @cindex @code{incbin} directive
4919 @cindex binary files, including
4920 The @code{incbin} directive includes @var{file} verbatim at the current
4921 location. You can control the search paths used with the @samp{-I} command-line
4922 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4925 The @var{skip} argument skips a number of bytes from the start of the
4926 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4927 read. Note that the data is not aligned in any way, so it is the user's
4928 responsibility to make sure that proper alignment is provided both before and
4929 after the @code{incbin} directive.
4932 @section @code{.include "@var{file}"}
4934 @cindex @code{include} directive
4935 @cindex supporting files, including
4936 @cindex files, including
4937 This directive provides a way to include supporting files at specified
4938 points in your source program. The code from @var{file} is assembled as
4939 if it followed the point of the @code{.include}; when the end of the
4940 included file is reached, assembly of the original file continues. You
4941 can control the search paths used with the @samp{-I} command-line option
4942 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4946 @section @code{.int @var{expressions}}
4948 @cindex @code{int} directive
4949 @cindex integers, 32-bit
4950 Expect zero or more @var{expressions}, of any section, separated by commas.
4951 For each expression, emit a number that, at run time, is the value of that
4952 expression. The byte order and bit size of the number depends on what kind
4953 of target the assembly is for.
4957 On most forms of the H8/300, @code{.int} emits 16-bit
4958 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
4965 @section @code{.internal @var{names}}
4967 @cindex @code{internal} directive
4969 This is one of the ELF visibility directives. The other two are
4970 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
4971 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4973 This directive overrides the named symbols default visibility (which is set by
4974 their binding: local, global or weak). The directive sets the visibility to
4975 @code{internal} which means that the symbols are considered to be @code{hidden}
4976 (i.e., not visible to other components), and that some extra, processor specific
4977 processing must also be performed upon the symbols as well.
4981 @section @code{.irp @var{symbol},@var{values}}@dots{}
4983 @cindex @code{irp} directive
4984 Evaluate a sequence of statements assigning different values to @var{symbol}.
4985 The sequence of statements starts at the @code{.irp} directive, and is
4986 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
4987 set to @var{value}, and the sequence of statements is assembled. If no
4988 @var{value} is listed, the sequence of statements is assembled once, with
4989 @var{symbol} set to the null string. To refer to @var{symbol} within the
4990 sequence of statements, use @var{\symbol}.
4992 For example, assembling
5000 is equivalent to assembling
5008 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5011 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5013 @cindex @code{irpc} directive
5014 Evaluate a sequence of statements assigning different values to @var{symbol}.
5015 The sequence of statements starts at the @code{.irpc} directive, and is
5016 terminated by an @code{.endr} directive. For each character in @var{value},
5017 @var{symbol} is set to the character, and the sequence of statements is
5018 assembled. If no @var{value} is listed, the sequence of statements is
5019 assembled once, with @var{symbol} set to the null string. To refer to
5020 @var{symbol} within the sequence of statements, use @var{\symbol}.
5022 For example, assembling
5030 is equivalent to assembling
5038 For some caveats with the spelling of @var{symbol}, see also the discussion
5042 @section @code{.lcomm @var{symbol} , @var{length}}
5044 @cindex @code{lcomm} directive
5045 @cindex local common symbols
5046 @cindex symbols, local common
5047 Reserve @var{length} (an absolute expression) bytes for a local common
5048 denoted by @var{symbol}. The section and value of @var{symbol} are
5049 those of the new local common. The addresses are allocated in the bss
5050 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5051 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5052 not visible to @code{@value{LD}}.
5055 Some targets permit a third argument to be used with @code{.lcomm}. This
5056 argument specifies the desired alignment of the symbol in the bss section.
5060 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5061 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5065 @section @code{.lflags}
5067 @cindex @code{lflags} directive (ignored)
5068 @command{@value{AS}} accepts this directive, for compatibility with other
5069 assemblers, but ignores it.
5071 @ifclear no-line-dir
5073 @section @code{.line @var{line-number}}
5075 @cindex @code{line} directive
5076 @cindex logical line number
5078 Change the logical line number. @var{line-number} must be an absolute
5079 expression. The next line has that logical line number. Therefore any other
5080 statements on the current line (after a statement separator character) are
5081 reported as on logical line number @var{line-number} @minus{} 1. One day
5082 @command{@value{AS}} will no longer support this directive: it is recognized only
5083 for compatibility with existing assembler programs.
5086 Even though this is a directive associated with the @code{a.out} or
5087 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5088 when producing COFF output, and treats @samp{.line} as though it
5089 were the COFF @samp{.ln} @emph{if} it is found outside a
5090 @code{.def}/@code{.endef} pair.
5092 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5093 used by compilers to generate auxiliary symbol information for
5098 @section @code{.linkonce [@var{type}]}
5100 @cindex @code{linkonce} directive
5101 @cindex common sections
5102 Mark the current section so that the linker only includes a single copy of it.
5103 This may be used to include the same section in several different object files,
5104 but ensure that the linker will only include it once in the final output file.
5105 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5106 Duplicate sections are detected based on the section name, so it should be
5109 This directive is only supported by a few object file formats; as of this
5110 writing, the only object file format which supports it is the Portable
5111 Executable format used on Windows NT.
5113 The @var{type} argument is optional. If specified, it must be one of the
5114 following strings. For example:
5118 Not all types may be supported on all object file formats.
5122 Silently discard duplicate sections. This is the default.
5125 Warn if there are duplicate sections, but still keep only one copy.
5128 Warn if any of the duplicates have different sizes.
5131 Warn if any of the duplicates do not have exactly the same contents.
5135 @section @code{.list}
5137 @cindex @code{list} directive
5138 @cindex listing control, turning on
5139 Control (in conjunction with the @code{.nolist} directive) whether or
5140 not assembly listings are generated. These two directives maintain an
5141 internal counter (which is zero initially). @code{.list} increments the
5142 counter, and @code{.nolist} decrements it. Assembly listings are
5143 generated whenever the counter is greater than zero.
5145 By default, listings are disabled. When you enable them (with the
5146 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5147 the initial value of the listing counter is one.
5150 @section @code{.ln @var{line-number}}
5152 @cindex @code{ln} directive
5153 @ifclear no-line-dir
5154 @samp{.ln} is a synonym for @samp{.line}.
5157 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5158 must be an absolute expression. The next line has that logical
5159 line number, so any other statements on the current line (after a
5160 statement separator character @code{;}) are reported as on logical
5161 line number @var{line-number} @minus{} 1.
5164 This directive is accepted, but ignored, when @command{@value{AS}} is
5165 configured for @code{b.out}; its effect is only associated with COFF
5171 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5172 @cindex @code{loc} directive
5173 When emitting DWARF2 line number information,
5174 the @code{.loc} directive will add a row to the @code{.debug_line} line
5175 number matrix corresponding to the immediately following assembly
5176 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5177 arguments will be applied to the @code{.debug_line} state machine before
5180 The @var{options} are a sequence of the following tokens in any order:
5184 This option will set the @code{basic_block} register in the
5185 @code{.debug_line} state machine to @code{true}.
5188 This option will set the @code{prologue_end} register in the
5189 @code{.debug_line} state machine to @code{true}.
5191 @item epilogue_begin
5192 This option will set the @code{epilogue_begin} register in the
5193 @code{.debug_line} state machine to @code{true}.
5195 @item is_stmt @var{value}
5196 This option will set the @code{is_stmt} register in the
5197 @code{.debug_line} state machine to @code{value}, which must be
5200 @item isa @var{value}
5201 This directive will set the @code{isa} register in the @code{.debug_line}
5202 state machine to @var{value}, which must be an unsigned integer.
5204 @item discriminator @var{value}
5205 This directive will set the @code{discriminator} register in the @code{.debug_line}
5206 state machine to @var{value}, which must be an unsigned integer.
5210 @node Loc_mark_labels
5211 @section @code{.loc_mark_labels @var{enable}}
5212 @cindex @code{loc_mark_labels} directive
5213 When emitting DWARF2 line number information,
5214 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5215 to the @code{.debug_line} line number matrix with the @code{basic_block}
5216 register in the state machine set whenever a code label is seen.
5217 The @var{enable} argument should be either 1 or 0, to enable or disable
5218 this function respectively.
5222 @section @code{.local @var{names}}
5224 @cindex @code{local} directive
5225 This directive, which is available for ELF targets, marks each symbol in
5226 the comma-separated list of @code{names} as a local symbol so that it
5227 will not be externally visible. If the symbols do not already exist,
5228 they will be created.
5230 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5231 accept an alignment argument, which is the case for most ELF targets,
5232 the @code{.local} directive can be used in combination with @code{.comm}
5233 (@pxref{Comm}) to define aligned local common data.
5237 @section @code{.long @var{expressions}}
5239 @cindex @code{long} directive
5240 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5243 @c no one seems to know what this is for or whether this description is
5244 @c what it really ought to do
5246 @section @code{.lsym @var{symbol}, @var{expression}}
5248 @cindex @code{lsym} directive
5249 @cindex symbol, not referenced in assembly
5250 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5251 the hash table, ensuring it cannot be referenced by name during the
5252 rest of the assembly. This sets the attributes of the symbol to be
5253 the same as the expression value:
5255 @var{other} = @var{descriptor} = 0
5256 @var{type} = @r{(section of @var{expression})}
5257 @var{value} = @var{expression}
5260 The new symbol is not flagged as external.
5264 @section @code{.macro}
5267 The commands @code{.macro} and @code{.endm} allow you to define macros that
5268 generate assembly output. For example, this definition specifies a macro
5269 @code{sum} that puts a sequence of numbers into memory:
5272 .macro sum from=0, to=5
5281 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5293 @item .macro @var{macname}
5294 @itemx .macro @var{macname} @var{macargs} @dots{}
5295 @cindex @code{macro} directive
5296 Begin the definition of a macro called @var{macname}. If your macro
5297 definition requires arguments, specify their names after the macro name,
5298 separated by commas or spaces. You can qualify the macro argument to
5299 indicate whether all invocations must specify a non-blank value (through
5300 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5301 (through @samp{:@code{vararg}}). You can supply a default value for any
5302 macro argument by following the name with @samp{=@var{deflt}}. You
5303 cannot define two macros with the same @var{macname} unless it has been
5304 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5305 definitions. For example, these are all valid @code{.macro} statements:
5309 Begin the definition of a macro called @code{comm}, which takes no
5312 @item .macro plus1 p, p1
5313 @itemx .macro plus1 p p1
5314 Either statement begins the definition of a macro called @code{plus1},
5315 which takes two arguments; within the macro definition, write
5316 @samp{\p} or @samp{\p1} to evaluate the arguments.
5318 @item .macro reserve_str p1=0 p2
5319 Begin the definition of a macro called @code{reserve_str}, with two
5320 arguments. The first argument has a default value, but not the second.
5321 After the definition is complete, you can call the macro either as
5322 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5323 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5324 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5325 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5327 @item .macro m p1:req, p2=0, p3:vararg
5328 Begin the definition of a macro called @code{m}, with at least three
5329 arguments. The first argument must always have a value specified, but
5330 not the second, which instead has a default value. The third formal
5331 will get assigned all remaining arguments specified at invocation time.
5333 When you call a macro, you can specify the argument values either by
5334 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5335 @samp{sum to=17, from=9}.
5339 Note that since each of the @var{macargs} can be an identifier exactly
5340 as any other one permitted by the target architecture, there may be
5341 occasional problems if the target hand-crafts special meanings to certain
5342 characters when they occur in a special position. For example, if the colon
5343 (@code{:}) is generally permitted to be part of a symbol name, but the
5344 architecture specific code special-cases it when occurring as the final
5345 character of a symbol (to denote a label), then the macro parameter
5346 replacement code will have no way of knowing that and consider the whole
5347 construct (including the colon) an identifier, and check only this
5348 identifier for being the subject to parameter substitution. So for example
5349 this macro definition:
5357 might not work as expected. Invoking @samp{label foo} might not create a label
5358 called @samp{foo} but instead just insert the text @samp{\l:} into the
5359 assembler source, probably generating an error about an unrecognised
5362 Similarly problems might occur with the period character (@samp{.})
5363 which is often allowed inside opcode names (and hence identifier names). So
5364 for example constructing a macro to build an opcode from a base name and a
5365 length specifier like this:
5368 .macro opcode base length
5373 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5374 instruction but instead generate some kind of error as the assembler tries to
5375 interpret the text @samp{\base.\length}.
5377 There are several possible ways around this problem:
5380 @item Insert white space
5381 If it is possible to use white space characters then this is the simplest
5390 @item Use @samp{\()}
5391 The string @samp{\()} can be used to separate the end of a macro argument from
5392 the following text. eg:
5395 .macro opcode base length
5400 @item Use the alternate macro syntax mode
5401 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5402 used as a separator. eg:
5412 Note: this problem of correctly identifying string parameters to pseudo ops
5413 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5414 and @code{.irpc} (@pxref{Irpc}) as well.
5417 @cindex @code{endm} directive
5418 Mark the end of a macro definition.
5421 @cindex @code{exitm} directive
5422 Exit early from the current macro definition.
5424 @cindex number of macros executed
5425 @cindex macros, count executed
5427 @command{@value{AS}} maintains a counter of how many macros it has
5428 executed in this pseudo-variable; you can copy that number to your
5429 output with @samp{\@@}, but @emph{only within a macro definition}.
5431 @item LOCAL @var{name} [ , @dots{} ]
5432 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5433 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5434 @xref{Altmacro,,@code{.altmacro}}.
5438 @section @code{.mri @var{val}}
5440 @cindex @code{mri} directive
5441 @cindex MRI mode, temporarily
5442 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5443 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5444 affects code assembled until the next @code{.mri} directive, or until the end
5445 of the file. @xref{M, MRI mode, MRI mode}.
5448 @section @code{.noaltmacro}
5449 Disable alternate macro mode. @xref{Altmacro}.
5452 @section @code{.nolist}
5454 @cindex @code{nolist} directive
5455 @cindex listing control, turning off
5456 Control (in conjunction with the @code{.list} directive) whether or
5457 not assembly listings are generated. These two directives maintain an
5458 internal counter (which is zero initially). @code{.list} increments the
5459 counter, and @code{.nolist} decrements it. Assembly listings are
5460 generated whenever the counter is greater than zero.
5463 @section @code{.octa @var{bignums}}
5465 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5466 @cindex @code{octa} directive
5467 @cindex integer, 16-byte
5468 @cindex sixteen byte integer
5469 This directive expects zero or more bignums, separated by commas. For each
5470 bignum, it emits a 16-byte integer.
5472 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5473 hence @emph{octa}-word for 16 bytes.
5476 @section @code{.org @var{new-lc} , @var{fill}}
5478 @cindex @code{org} directive
5479 @cindex location counter, advancing
5480 @cindex advancing location counter
5481 @cindex current address, advancing
5482 Advance the location counter of the current section to
5483 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5484 expression with the same section as the current subsection. That is,
5485 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5486 wrong section, the @code{.org} directive is ignored. To be compatible
5487 with former assemblers, if the section of @var{new-lc} is absolute,
5488 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5489 is the same as the current subsection.
5491 @code{.org} may only increase the location counter, or leave it
5492 unchanged; you cannot use @code{.org} to move the location counter
5495 @c double negative used below "not undefined" because this is a specific
5496 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5497 @c section. doc@cygnus.com 18feb91
5498 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5499 may not be undefined. If you really detest this restriction we eagerly await
5500 a chance to share your improved assembler.
5502 Beware that the origin is relative to the start of the section, not
5503 to the start of the subsection. This is compatible with other
5504 people's assemblers.
5506 When the location counter (of the current subsection) is advanced, the
5507 intervening bytes are filled with @var{fill} which should be an
5508 absolute expression. If the comma and @var{fill} are omitted,
5509 @var{fill} defaults to zero.
5512 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5514 @cindex padding the location counter given a power of two
5515 @cindex @code{p2align} directive
5516 Pad the location counter (in the current subsection) to a particular
5517 storage boundary. The first expression (which must be absolute) is the
5518 number of low-order zero bits the location counter must have after
5519 advancement. For example @samp{.p2align 3} advances the location
5520 counter until it a multiple of 8. If the location counter is already a
5521 multiple of 8, no change is needed.
5523 The second expression (also absolute) gives the fill value to be stored in the
5524 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5525 padding bytes are normally zero. However, on some systems, if the section is
5526 marked as containing code and the fill value is omitted, the space is filled
5527 with no-op instructions.
5529 The third expression is also absolute, and is also optional. If it is present,
5530 it is the maximum number of bytes that should be skipped by this alignment
5531 directive. If doing the alignment would require skipping more bytes than the
5532 specified maximum, then the alignment is not done at all. You can omit the
5533 fill value (the second argument) entirely by simply using two commas after the
5534 required alignment; this can be useful if you want the alignment to be filled
5535 with no-op instructions when appropriate.
5537 @cindex @code{p2alignw} directive
5538 @cindex @code{p2alignl} directive
5539 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5540 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5541 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5542 fill pattern as a four byte longword value. For example, @code{.p2alignw
5543 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5544 filled in with the value 0x368d (the exact placement of the bytes depends upon
5545 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5550 @section @code{.popsection}
5552 @cindex @code{popsection} directive
5553 @cindex Section Stack
5554 This is one of the ELF section stack manipulation directives. The others are
5555 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5556 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5559 This directive replaces the current section (and subsection) with the top
5560 section (and subsection) on the section stack. This section is popped off the
5566 @section @code{.previous}
5568 @cindex @code{previous} directive
5569 @cindex Section Stack
5570 This is one of the ELF section stack manipulation directives. The others are
5571 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5572 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5573 (@pxref{PopSection}).
5575 This directive swaps the current section (and subsection) with most recently
5576 referenced section/subsection pair prior to this one. Multiple
5577 @code{.previous} directives in a row will flip between two sections (and their
5578 subsections). For example:
5590 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
5596 # Now in section A subsection 1
5600 # Now in section B subsection 0
5603 # Now in section B subsection 1
5606 # Now in section B subsection 0
5610 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
5611 section B and 0x9abc into subsection 1 of section B.
5613 In terms of the section stack, this directive swaps the current section with
5614 the top section on the section stack.
5618 @section @code{.print @var{string}}
5620 @cindex @code{print} directive
5621 @command{@value{AS}} will print @var{string} on the standard output during
5622 assembly. You must put @var{string} in double quotes.
5626 @section @code{.protected @var{names}}
5628 @cindex @code{protected} directive
5630 This is one of the ELF visibility directives. The other two are
5631 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5633 This directive overrides the named symbols default visibility (which is set by
5634 their binding: local, global or weak). The directive sets the visibility to
5635 @code{protected} which means that any references to the symbols from within the
5636 components that defines them must be resolved to the definition in that
5637 component, even if a definition in another component would normally preempt
5642 @section @code{.psize @var{lines} , @var{columns}}
5644 @cindex @code{psize} directive
5645 @cindex listing control: paper size
5646 @cindex paper size, for listings
5647 Use this directive to declare the number of lines---and, optionally, the
5648 number of columns---to use for each page, when generating listings.
5650 If you do not use @code{.psize}, listings use a default line-count
5651 of 60. You may omit the comma and @var{columns} specification; the
5652 default width is 200 columns.
5654 @command{@value{AS}} generates formfeeds whenever the specified number of
5655 lines is exceeded (or whenever you explicitly request one, using
5658 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5659 those explicitly specified with @code{.eject}.
5662 @section @code{.purgem @var{name}}
5664 @cindex @code{purgem} directive
5665 Undefine the macro @var{name}, so that later uses of the string will not be
5666 expanded. @xref{Macro}.
5670 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
5672 @cindex @code{pushsection} directive
5673 @cindex Section Stack
5674 This is one of the ELF section stack manipulation directives. The others are
5675 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5676 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5679 This directive pushes the current section (and subsection) onto the
5680 top of the section stack, and then replaces the current section and
5681 subsection with @code{name} and @code{subsection}. The optional
5682 @code{flags}, @code{type} and @code{arguments} are treated the same
5683 as in the @code{.section} (@pxref{Section}) directive.
5687 @section @code{.quad @var{bignums}}
5689 @cindex @code{quad} directive
5690 @code{.quad} expects zero or more bignums, separated by commas. For
5691 each bignum, it emits
5693 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5694 warning message; and just takes the lowest order 8 bytes of the bignum.
5695 @cindex eight-byte integer
5696 @cindex integer, 8-byte
5698 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5699 hence @emph{quad}-word for 8 bytes.
5702 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5703 warning message; and just takes the lowest order 16 bytes of the bignum.
5704 @cindex sixteen-byte integer
5705 @cindex integer, 16-byte
5709 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
5711 @cindex @code{reloc} directive
5712 Generate a relocation at @var{offset} of type @var{reloc_name} with value
5713 @var{expression}. If @var{offset} is a number, the relocation is generated in
5714 the current section. If @var{offset} is an expression that resolves to a
5715 symbol plus offset, the relocation is generated in the given symbol's section.
5716 @var{expression}, if present, must resolve to a symbol plus addend or to an
5717 absolute value, but note that not all targets support an addend. e.g. ELF REL
5718 targets such as i386 store an addend in the section contents rather than in the
5719 relocation. This low level interface does not support addends stored in the
5723 @section @code{.rept @var{count}}
5725 @cindex @code{rept} directive
5726 Repeat the sequence of lines between the @code{.rept} directive and the next
5727 @code{.endr} directive @var{count} times.
5729 For example, assembling
5737 is equivalent to assembling
5746 @section @code{.sbttl "@var{subheading}"}
5748 @cindex @code{sbttl} directive
5749 @cindex subtitles for listings
5750 @cindex listing control: subtitle
5751 Use @var{subheading} as the title (third line, immediately after the
5752 title line) when generating assembly listings.
5754 This directive affects subsequent pages, as well as the current page if
5755 it appears within ten lines of the top of a page.
5759 @section @code{.scl @var{class}}
5761 @cindex @code{scl} directive
5762 @cindex symbol storage class (COFF)
5763 @cindex COFF symbol storage class
5764 Set the storage-class value for a symbol. This directive may only be
5765 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5766 whether a symbol is static or external, or it may record further
5767 symbolic debugging information.
5770 The @samp{.scl} directive is primarily associated with COFF output; when
5771 configured to generate @code{b.out} output format, @command{@value{AS}}
5772 accepts this directive but ignores it.
5778 @section @code{.section @var{name}}
5780 @cindex named section
5781 Use the @code{.section} directive to assemble the following code into a section
5784 This directive is only supported for targets that actually support arbitrarily
5785 named sections; on @code{a.out} targets, for example, it is not accepted, even
5786 with a standard @code{a.out} section name.
5790 @c only print the extra heading if both COFF and ELF are set
5791 @subheading COFF Version
5794 @cindex @code{section} directive (COFF version)
5795 For COFF targets, the @code{.section} directive is used in one of the following
5799 .section @var{name}[, "@var{flags}"]
5800 .section @var{name}[, @var{subsection}]
5803 If the optional argument is quoted, it is taken as flags to use for the
5804 section. Each flag is a single character. The following flags are recognized:
5807 bss section (uninitialized data)
5809 section is not loaded
5819 shared section (meaningful for PE targets)
5821 ignored. (For compatibility with the ELF version)
5823 section is not readable (meaningful for PE targets)
5825 single-digit power-of-two section alignment (GNU extension)
5828 If no flags are specified, the default flags depend upon the section name. If
5829 the section name is not recognized, the default will be for the section to be
5830 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
5831 from the section, rather than adding them, so if they are used on their own it
5832 will be as if no flags had been specified at all.
5834 If the optional argument to the @code{.section} directive is not quoted, it is
5835 taken as a subsection number (@pxref{Sub-Sections}).
5840 @c only print the extra heading if both COFF and ELF are set
5841 @subheading ELF Version
5844 @cindex Section Stack
5845 This is one of the ELF section stack manipulation directives. The others are
5846 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
5847 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
5848 @code{.previous} (@pxref{Previous}).
5850 @cindex @code{section} directive (ELF version)
5851 For ELF targets, the @code{.section} directive is used like this:
5854 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
5857 The optional @var{flags} argument is a quoted string which may contain any
5858 combination of the following characters:
5861 section is allocatable
5863 section is excluded from executable and shared library.
5867 section is executable
5869 section is mergeable
5871 section contains zero terminated strings
5873 section is a member of a section group
5875 section is used for thread-local-storage
5877 section is a member of the previously-current section's group, if any
5880 The optional @var{type} argument may contain one of the following constants:
5883 section contains data
5885 section does not contain data (i.e., section only occupies space)
5887 section contains data which is used by things other than the program
5889 section contains an array of pointers to init functions
5891 section contains an array of pointers to finish functions
5892 @item @@preinit_array
5893 section contains an array of pointers to pre-init functions
5896 Many targets only support the first three section types.
5898 Note on targets where the @code{@@} character is the start of a comment (eg
5899 ARM) then another character is used instead. For example the ARM port uses the
5902 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
5903 be specified as well as an extra argument---@var{entsize}---like this:
5906 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
5909 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
5910 constants, each @var{entsize} octets long. Sections with both @code{M} and
5911 @code{S} must contain zero terminated strings where each character is
5912 @var{entsize} bytes long. The linker may remove duplicates within sections with
5913 the same name, same entity size and same flags. @var{entsize} must be an
5914 absolute expression. For sections with both @code{M} and @code{S}, a string
5915 which is a suffix of a larger string is considered a duplicate. Thus
5916 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
5917 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
5919 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
5920 be present along with an additional field like this:
5923 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
5926 The @var{GroupName} field specifies the name of the section group to which this
5927 particular section belongs. The optional linkage field can contain:
5930 indicates that only one copy of this section should be retained
5935 Note: if both the @var{M} and @var{G} flags are present then the fields for
5936 the Merge flag should come first, like this:
5939 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
5942 If @var{flags} contains the @code{?} symbol then it may not also contain the
5943 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
5944 present. Instead, @code{?} says to consider the section that's current before
5945 this directive. If that section used @code{G}, then the new section will use
5946 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
5947 If not, then the @code{?} symbol has no effect.
5949 If no flags are specified, the default flags depend upon the section name. If
5950 the section name is not recognized, the default will be for the section to have
5951 none of the above flags: it will not be allocated in memory, nor writable, nor
5952 executable. The section will contain data.
5954 For ELF targets, the assembler supports another type of @code{.section}
5955 directive for compatibility with the Solaris assembler:
5958 .section "@var{name}"[, @var{flags}...]
5961 Note that the section name is quoted. There may be a sequence of comma
5965 section is allocatable
5969 section is executable
5971 section is excluded from executable and shared library.
5973 section is used for thread local storage
5976 This directive replaces the current section and subsection. See the
5977 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
5978 some examples of how this directive and the other section stack directives
5984 @section @code{.set @var{symbol}, @var{expression}}
5986 @cindex @code{set} directive
5987 @cindex symbol value, setting
5988 Set the value of @var{symbol} to @var{expression}. This
5989 changes @var{symbol}'s value and type to conform to
5990 @var{expression}. If @var{symbol} was flagged as external, it remains
5991 flagged (@pxref{Symbol Attributes}).
5993 You may @code{.set} a symbol many times in the same assembly.
5995 If you @code{.set} a global symbol, the value stored in the object
5996 file is the last value stored into it.
5999 On Z80 @code{set} is a real instruction, use
6000 @samp{@var{symbol} defl @var{expression}} instead.
6004 @section @code{.short @var{expressions}}
6006 @cindex @code{short} directive
6008 @code{.short} is normally the same as @samp{.word}.
6009 @xref{Word,,@code{.word}}.
6011 In some configurations, however, @code{.short} and @code{.word} generate
6012 numbers of different lengths. @xref{Machine Dependencies}.
6016 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6019 This expects zero or more @var{expressions}, and emits
6020 a 16 bit number for each.
6025 @section @code{.single @var{flonums}}
6027 @cindex @code{single} directive
6028 @cindex floating point numbers (single)
6029 This directive assembles zero or more flonums, separated by commas. It
6030 has the same effect as @code{.float}.
6032 The exact kind of floating point numbers emitted depends on how
6033 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6037 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6038 numbers in @sc{ieee} format.
6044 @section @code{.size}
6046 This directive is used to set the size associated with a symbol.
6050 @c only print the extra heading if both COFF and ELF are set
6051 @subheading COFF Version
6054 @cindex @code{size} directive (COFF version)
6055 For COFF targets, the @code{.size} directive is only permitted inside
6056 @code{.def}/@code{.endef} pairs. It is used like this:
6059 .size @var{expression}
6063 @samp{.size} is only meaningful when generating COFF format output; when
6064 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6071 @c only print the extra heading if both COFF and ELF are set
6072 @subheading ELF Version
6075 @cindex @code{size} directive (ELF version)
6076 For ELF targets, the @code{.size} directive is used like this:
6079 .size @var{name} , @var{expression}
6082 This directive sets the size associated with a symbol @var{name}.
6083 The size in bytes is computed from @var{expression} which can make use of label
6084 arithmetic. This directive is typically used to set the size of function
6089 @ifclear no-space-dir
6091 @section @code{.skip @var{size} , @var{fill}}
6093 @cindex @code{skip} directive
6094 @cindex filling memory
6095 This directive emits @var{size} bytes, each of value @var{fill}. Both
6096 @var{size} and @var{fill} are absolute expressions. If the comma and
6097 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6102 @section @code{.sleb128 @var{expressions}}
6104 @cindex @code{sleb128} directive
6105 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6106 compact, variable length representation of numbers used by the DWARF
6107 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6109 @ifclear no-space-dir
6111 @section @code{.space @var{size} , @var{fill}}
6113 @cindex @code{space} directive
6114 @cindex filling memory
6115 This directive emits @var{size} bytes, each of value @var{fill}. Both
6116 @var{size} and @var{fill} are absolute expressions. If the comma
6117 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6122 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6123 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6124 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6125 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6133 @section @code{.stabd, .stabn, .stabs}
6135 @cindex symbolic debuggers, information for
6136 @cindex @code{stab@var{x}} directives
6137 There are three directives that begin @samp{.stab}.
6138 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6139 The symbols are not entered in the @command{@value{AS}} hash table: they
6140 cannot be referenced elsewhere in the source file.
6141 Up to five fields are required:
6145 This is the symbol's name. It may contain any character except
6146 @samp{\000}, so is more general than ordinary symbol names. Some
6147 debuggers used to code arbitrarily complex structures into symbol names
6151 An absolute expression. The symbol's type is set to the low 8 bits of
6152 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6153 and debuggers choke on silly bit patterns.
6156 An absolute expression. The symbol's ``other'' attribute is set to the
6157 low 8 bits of this expression.
6160 An absolute expression. The symbol's descriptor is set to the low 16
6161 bits of this expression.
6164 An absolute expression which becomes the symbol's value.
6167 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6168 or @code{.stabs} statement, the symbol has probably already been created;
6169 you get a half-formed symbol in your object file. This is
6170 compatible with earlier assemblers!
6173 @cindex @code{stabd} directive
6174 @item .stabd @var{type} , @var{other} , @var{desc}
6176 The ``name'' of the symbol generated is not even an empty string.
6177 It is a null pointer, for compatibility. Older assemblers used a
6178 null pointer so they didn't waste space in object files with empty
6181 The symbol's value is set to the location counter,
6182 relocatably. When your program is linked, the value of this symbol
6183 is the address of the location counter when the @code{.stabd} was
6186 @cindex @code{stabn} directive
6187 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6188 The name of the symbol is set to the empty string @code{""}.
6190 @cindex @code{stabs} directive
6191 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6192 All five fields are specified.
6198 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6199 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6201 @cindex string, copying to object file
6202 @cindex string8, copying to object file
6203 @cindex string16, copying to object file
6204 @cindex string32, copying to object file
6205 @cindex string64, copying to object file
6206 @cindex @code{string} directive
6207 @cindex @code{string8} directive
6208 @cindex @code{string16} directive
6209 @cindex @code{string32} directive
6210 @cindex @code{string64} directive
6212 Copy the characters in @var{str} to the object file. You may specify more than
6213 one string to copy, separated by commas. Unless otherwise specified for a
6214 particular machine, the assembler marks the end of each string with a 0 byte.
6215 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6217 The variants @code{string16}, @code{string32} and @code{string64} differ from
6218 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6219 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6220 are stored in target endianness byte order.
6226 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6227 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6232 @section @code{.struct @var{expression}}
6234 @cindex @code{struct} directive
6235 Switch to the absolute section, and set the section offset to @var{expression},
6236 which must be an absolute expression. You might use this as follows:
6245 This would define the symbol @code{field1} to have the value 0, the symbol
6246 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6247 value 8. Assembly would be left in the absolute section, and you would need to
6248 use a @code{.section} directive of some sort to change to some other section
6249 before further assembly.
6253 @section @code{.subsection @var{name}}
6255 @cindex @code{subsection} directive
6256 @cindex Section Stack
6257 This is one of the ELF section stack manipulation directives. The others are
6258 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6259 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6262 This directive replaces the current subsection with @code{name}. The current
6263 section is not changed. The replaced subsection is put onto the section stack
6264 in place of the then current top of stack subsection.
6269 @section @code{.symver}
6270 @cindex @code{symver} directive
6271 @cindex symbol versioning
6272 @cindex versions of symbols
6273 Use the @code{.symver} directive to bind symbols to specific version nodes
6274 within a source file. This is only supported on ELF platforms, and is
6275 typically used when assembling files to be linked into a shared library.
6276 There are cases where it may make sense to use this in objects to be bound
6277 into an application itself so as to override a versioned symbol from a
6280 For ELF targets, the @code{.symver} directive can be used like this:
6282 .symver @var{name}, @var{name2@@nodename}
6284 If the symbol @var{name} is defined within the file
6285 being assembled, the @code{.symver} directive effectively creates a symbol
6286 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6287 just don't try and create a regular alias is that the @var{@@} character isn't
6288 permitted in symbol names. The @var{name2} part of the name is the actual name
6289 of the symbol by which it will be externally referenced. The name @var{name}
6290 itself is merely a name of convenience that is used so that it is possible to
6291 have definitions for multiple versions of a function within a single source
6292 file, and so that the compiler can unambiguously know which version of a
6293 function is being mentioned. The @var{nodename} portion of the alias should be
6294 the name of a node specified in the version script supplied to the linker when
6295 building a shared library. If you are attempting to override a versioned
6296 symbol from a shared library, then @var{nodename} should correspond to the
6297 nodename of the symbol you are trying to override.
6299 If the symbol @var{name} is not defined within the file being assembled, all
6300 references to @var{name} will be changed to @var{name2@@nodename}. If no
6301 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6304 Another usage of the @code{.symver} directive is:
6306 .symver @var{name}, @var{name2@@@@nodename}
6308 In this case, the symbol @var{name} must exist and be defined within
6309 the file being assembled. It is similar to @var{name2@@nodename}. The
6310 difference is @var{name2@@@@nodename} will also be used to resolve
6311 references to @var{name2} by the linker.
6313 The third usage of the @code{.symver} directive is:
6315 .symver @var{name}, @var{name2@@@@@@nodename}
6317 When @var{name} is not defined within the
6318 file being assembled, it is treated as @var{name2@@nodename}. When
6319 @var{name} is defined within the file being assembled, the symbol
6320 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6325 @section @code{.tag @var{structname}}
6327 @cindex COFF structure debugging
6328 @cindex structure debugging, COFF
6329 @cindex @code{tag} directive
6330 This directive is generated by compilers to include auxiliary debugging
6331 information in the symbol table. It is only permitted inside
6332 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6333 definitions in the symbol table with instances of those structures.
6336 @samp{.tag} is only used when generating COFF format output; when
6337 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6343 @section @code{.text @var{subsection}}
6345 @cindex @code{text} directive
6346 Tells @command{@value{AS}} to assemble the following statements onto the end of
6347 the text subsection numbered @var{subsection}, which is an absolute
6348 expression. If @var{subsection} is omitted, subsection number zero
6352 @section @code{.title "@var{heading}"}
6354 @cindex @code{title} directive
6355 @cindex listing control: title line
6356 Use @var{heading} as the title (second line, immediately after the
6357 source file name and pagenumber) when generating assembly listings.
6359 This directive affects subsequent pages, as well as the current page if
6360 it appears within ten lines of the top of a page.
6364 @section @code{.type}
6366 This directive is used to set the type of a symbol.
6370 @c only print the extra heading if both COFF and ELF are set
6371 @subheading COFF Version
6374 @cindex COFF symbol type
6375 @cindex symbol type, COFF
6376 @cindex @code{type} directive (COFF version)
6377 For COFF targets, this directive is permitted only within
6378 @code{.def}/@code{.endef} pairs. It is used like this:
6384 This records the integer @var{int} as the type attribute of a symbol table
6388 @samp{.type} is associated only with COFF format output; when
6389 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6390 directive but ignores it.
6396 @c only print the extra heading if both COFF and ELF are set
6397 @subheading ELF Version
6400 @cindex ELF symbol type
6401 @cindex symbol type, ELF
6402 @cindex @code{type} directive (ELF version)
6403 For ELF targets, the @code{.type} directive is used like this:
6406 .type @var{name} , @var{type description}
6409 This sets the type of symbol @var{name} to be either a
6410 function symbol or an object symbol. There are five different syntaxes
6411 supported for the @var{type description} field, in order to provide
6412 compatibility with various other assemblers.
6414 Because some of the characters used in these syntaxes (such as @samp{@@} and
6415 @samp{#}) are comment characters for some architectures, some of the syntaxes
6416 below do not work on all architectures. The first variant will be accepted by
6417 the GNU assembler on all architectures so that variant should be used for
6418 maximum portability, if you do not need to assemble your code with other
6421 The syntaxes supported are:
6424 .type <name> STT_<TYPE_IN_UPPER_CASE>
6425 .type <name>,#<type>
6426 .type <name>,@@<type>
6427 .type <name>,%<type>
6428 .type <name>,"<type>"
6431 The types supported are:
6436 Mark the symbol as being a function name.
6439 @itemx gnu_indirect_function
6440 Mark the symbol as an indirect function when evaluated during reloc
6441 processing. (This is only supported on assemblers targeting GNU systems).
6445 Mark the symbol as being a data object.
6449 Mark the symbol as being a thead-local data object.
6453 Mark the symbol as being a common data object.
6457 Does not mark the symbol in any way. It is supported just for completeness.
6459 @item gnu_unique_object
6460 Marks the symbol as being a globally unique data object. The dynamic linker
6461 will make sure that in the entire process there is just one symbol with this
6462 name and type in use. (This is only supported on assemblers targeting GNU
6467 Note: Some targets support extra types in addition to those listed above.
6473 @section @code{.uleb128 @var{expressions}}
6475 @cindex @code{uleb128} directive
6476 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6477 compact, variable length representation of numbers used by the DWARF
6478 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6482 @section @code{.val @var{addr}}
6484 @cindex @code{val} directive
6485 @cindex COFF value attribute
6486 @cindex value attribute, COFF
6487 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6488 records the address @var{addr} as the value attribute of a symbol table
6492 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6493 configured for @code{b.out}, it accepts this directive but ignores it.
6499 @section @code{.version "@var{string}"}
6501 @cindex @code{version} directive
6502 This directive creates a @code{.note} section and places into it an ELF
6503 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6508 @section @code{.vtable_entry @var{table}, @var{offset}}
6510 @cindex @code{vtable_entry} directive
6511 This directive finds or creates a symbol @code{table} and creates a
6512 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6515 @section @code{.vtable_inherit @var{child}, @var{parent}}
6517 @cindex @code{vtable_inherit} directive
6518 This directive finds the symbol @code{child} and finds or creates the symbol
6519 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6520 parent whose addend is the value of the child symbol. As a special case the
6521 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6525 @section @code{.warning "@var{string}"}
6526 @cindex warning directive
6527 Similar to the directive @code{.error}
6528 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6531 @section @code{.weak @var{names}}
6533 @cindex @code{weak} directive
6534 This directive sets the weak attribute on the comma separated list of symbol
6535 @code{names}. If the symbols do not already exist, they will be created.
6537 On COFF targets other than PE, weak symbols are a GNU extension. This
6538 directive sets the weak attribute on the comma separated list of symbol
6539 @code{names}. If the symbols do not already exist, they will be created.
6541 On the PE target, weak symbols are supported natively as weak aliases.
6542 When a weak symbol is created that is not an alias, GAS creates an
6543 alternate symbol to hold the default value.
6546 @section @code{.weakref @var{alias}, @var{target}}
6548 @cindex @code{weakref} directive
6549 This directive creates an alias to the target symbol that enables the symbol to
6550 be referenced with weak-symbol semantics, but without actually making it weak.
6551 If direct references or definitions of the symbol are present, then the symbol
6552 will not be weak, but if all references to it are through weak references, the
6553 symbol will be marked as weak in the symbol table.
6555 The effect is equivalent to moving all references to the alias to a separate
6556 assembly source file, renaming the alias to the symbol in it, declaring the
6557 symbol as weak there, and running a reloadable link to merge the object files
6558 resulting from the assembly of the new source file and the old source file that
6559 had the references to the alias removed.
6561 The alias itself never makes to the symbol table, and is entirely handled
6562 within the assembler.
6565 @section @code{.word @var{expressions}}
6567 @cindex @code{word} directive
6568 This directive expects zero or more @var{expressions}, of any section,
6569 separated by commas.
6572 For each expression, @command{@value{AS}} emits a 32-bit number.
6575 For each expression, @command{@value{AS}} emits a 16-bit number.
6580 The size of the number emitted, and its byte order,
6581 depend on what target computer the assembly is for.
6584 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
6585 @c happen---32-bit addressability, period; no long/short jumps.
6586 @ifset DIFF-TBL-KLUGE
6587 @cindex difference tables altered
6588 @cindex altered difference tables
6590 @emph{Warning: Special Treatment to support Compilers}
6594 Machines with a 32-bit address space, but that do less than 32-bit
6595 addressing, require the following special treatment. If the machine of
6596 interest to you does 32-bit addressing (or doesn't require it;
6597 @pxref{Machine Dependencies}), you can ignore this issue.
6600 In order to assemble compiler output into something that works,
6601 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6602 Directives of the form @samp{.word sym1-sym2} are often emitted by
6603 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6604 directive of the form @samp{.word sym1-sym2}, and the difference between
6605 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6606 creates a @dfn{secondary jump table}, immediately before the next label.
6607 This secondary jump table is preceded by a short-jump to the
6608 first byte after the secondary table. This short-jump prevents the flow
6609 of control from accidentally falling into the new table. Inside the
6610 table is a long-jump to @code{sym2}. The original @samp{.word}
6611 contains @code{sym1} minus the address of the long-jump to
6614 If there were several occurrences of @samp{.word sym1-sym2} before the
6615 secondary jump table, all of them are adjusted. If there was a
6616 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6617 long-jump to @code{sym4} is included in the secondary jump table,
6618 and the @code{.word} directives are adjusted to contain @code{sym3}
6619 minus the address of the long-jump to @code{sym4}; and so on, for as many
6620 entries in the original jump table as necessary.
6623 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6624 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6625 assembly language programmers.
6628 @c end DIFF-TBL-KLUGE
6631 @section Deprecated Directives
6633 @cindex deprecated directives
6634 @cindex obsolescent directives
6635 One day these directives won't work.
6636 They are included for compatibility with older assemblers.
6643 @node Object Attributes
6644 @chapter Object Attributes
6645 @cindex object attributes
6647 @command{@value{AS}} assembles source files written for a specific architecture
6648 into object files for that architecture. But not all object files are alike.
6649 Many architectures support incompatible variations. For instance, floating
6650 point arguments might be passed in floating point registers if the object file
6651 requires hardware floating point support---or floating point arguments might be
6652 passed in integer registers if the object file supports processors with no
6653 hardware floating point unit. Or, if two objects are built for different
6654 generations of the same architecture, the combination may require the
6655 newer generation at run-time.
6657 This information is useful during and after linking. At link time,
6658 @command{@value{LD}} can warn about incompatible object files. After link
6659 time, tools like @command{gdb} can use it to process the linked file
6662 Compatibility information is recorded as a series of object attributes. Each
6663 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
6664 string, and indicates who sets the meaning of the tag. The tag is an integer,
6665 and indicates what property the attribute describes. The value may be a string
6666 or an integer, and indicates how the property affects this object. Missing
6667 attributes are the same as attributes with a zero value or empty string value.
6669 Object attributes were developed as part of the ABI for the ARM Architecture.
6670 The file format is documented in @cite{ELF for the ARM Architecture}.
6673 * GNU Object Attributes:: @sc{gnu} Object Attributes
6674 * Defining New Object Attributes:: Defining New Object Attributes
6677 @node GNU Object Attributes
6678 @section @sc{gnu} Object Attributes
6680 The @code{.gnu_attribute} directive records an object attribute
6681 with vendor @samp{gnu}.
6683 Except for @samp{Tag_compatibility}, which has both an integer and a string for
6684 its value, @sc{gnu} attributes have a string value if the tag number is odd and
6685 an integer value if the tag number is even. The second bit (@code{@var{tag} &
6686 2} is set for architecture-independent attributes and clear for
6687 architecture-dependent ones.
6689 @subsection Common @sc{gnu} attributes
6691 These attributes are valid on all architectures.
6694 @item Tag_compatibility (32)
6695 The compatibility attribute takes an integer flag value and a vendor name. If
6696 the flag value is 0, the file is compatible with other toolchains. If it is 1,
6697 then the file is only compatible with the named toolchain. If it is greater
6698 than 1, the file can only be processed by other toolchains under some private
6699 arrangement indicated by the flag value and the vendor name.
6702 @subsection MIPS Attributes
6705 @item Tag_GNU_MIPS_ABI_FP (4)
6706 The floating-point ABI used by this object file. The value will be:
6710 0 for files not affected by the floating-point ABI.
6712 1 for files using the hardware floating-point with a standard double-precision
6715 2 for files using the hardware floating-point ABI with a single-precision FPU.
6717 3 for files using the software floating-point ABI.
6719 4 for files using the hardware floating-point ABI with 64-bit wide
6720 double-precision floating-point registers and 32-bit wide general
6725 @subsection PowerPC Attributes
6728 @item Tag_GNU_Power_ABI_FP (4)
6729 The floating-point ABI used by this object file. The value will be:
6733 0 for files not affected by the floating-point ABI.
6735 1 for files using double-precision hardware floating-point ABI.
6737 2 for files using the software floating-point ABI.
6739 3 for files using single-precision hardware floating-point ABI.
6742 @item Tag_GNU_Power_ABI_Vector (8)
6743 The vector ABI used by this object file. The value will be:
6747 0 for files not affected by the vector ABI.
6749 1 for files using general purpose registers to pass vectors.
6751 2 for files using AltiVec registers to pass vectors.
6753 3 for files using SPE registers to pass vectors.
6757 @node Defining New Object Attributes
6758 @section Defining New Object Attributes
6760 If you want to define a new @sc{gnu} object attribute, here are the places you
6761 will need to modify. New attributes should be discussed on the @samp{binutils}
6766 This manual, which is the official register of attributes.
6768 The header for your architecture @file{include/elf}, to define the tag.
6770 The @file{bfd} support file for your architecture, to merge the attribute
6771 and issue any appropriate link warnings.
6773 Test cases in @file{ld/testsuite} for merging and link warnings.
6775 @file{binutils/readelf.c} to display your attribute.
6777 GCC, if you want the compiler to mark the attribute automatically.
6783 @node Machine Dependencies
6784 @chapter Machine Dependent Features
6786 @cindex machine dependencies
6787 The machine instruction sets are (almost by definition) different on
6788 each machine where @command{@value{AS}} runs. Floating point representations
6789 vary as well, and @command{@value{AS}} often supports a few additional
6790 directives or command-line options for compatibility with other
6791 assemblers on a particular platform. Finally, some versions of
6792 @command{@value{AS}} support special pseudo-instructions for branch
6795 This chapter discusses most of these differences, though it does not
6796 include details on any machine's instruction set. For details on that
6797 subject, see the hardware manufacturer's manual.
6801 * Alpha-Dependent:: Alpha Dependent Features
6804 * ARC-Dependent:: ARC Dependent Features
6807 * ARM-Dependent:: ARM Dependent Features
6810 * AVR-Dependent:: AVR Dependent Features
6813 * Blackfin-Dependent:: Blackfin Dependent Features
6816 * CR16-Dependent:: CR16 Dependent Features
6819 * CRIS-Dependent:: CRIS Dependent Features
6822 * D10V-Dependent:: D10V Dependent Features
6825 * D30V-Dependent:: D30V Dependent Features
6828 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6831 * HPPA-Dependent:: HPPA Dependent Features
6834 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
6837 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
6840 * i860-Dependent:: Intel 80860 Dependent Features
6843 * i960-Dependent:: Intel 80960 Dependent Features
6846 * IA-64-Dependent:: Intel IA-64 Dependent Features
6849 * IP2K-Dependent:: IP2K Dependent Features
6852 * LM32-Dependent:: LM32 Dependent Features
6855 * M32C-Dependent:: M32C Dependent Features
6858 * M32R-Dependent:: M32R Dependent Features
6861 * M68K-Dependent:: M680x0 Dependent Features
6864 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
6867 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
6870 * MIPS-Dependent:: MIPS Dependent Features
6873 * MMIX-Dependent:: MMIX Dependent Features
6876 * MSP430-Dependent:: MSP430 Dependent Features
6879 * NS32K-Dependent:: NS32K Dependent Features
6882 * SH-Dependent:: Renesas / SuperH SH Dependent Features
6883 * SH64-Dependent:: SuperH SH64 Dependent Features
6886 * PDP-11-Dependent:: PDP-11 Dependent Features
6889 * PJ-Dependent:: picoJava Dependent Features
6892 * PPC-Dependent:: PowerPC Dependent Features
6895 * RX-Dependent:: RX Dependent Features
6898 * S/390-Dependent:: IBM S/390 Dependent Features
6901 * SCORE-Dependent:: SCORE Dependent Features
6904 * Sparc-Dependent:: SPARC Dependent Features
6907 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
6910 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
6913 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
6916 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
6919 * V850-Dependent:: V850 Dependent Features
6922 * Xtensa-Dependent:: Xtensa Dependent Features
6925 * Z80-Dependent:: Z80 Dependent Features
6928 * Z8000-Dependent:: Z8000 Dependent Features
6931 * Vax-Dependent:: VAX Dependent Features
6938 @c The following major nodes are *sections* in the GENERIC version, *chapters*
6939 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
6940 @c peculiarity: to preserve cross-references, there must be a node called
6941 @c "Machine Dependencies". Hence the conditional nodenames in each
6942 @c major node below. Node defaulting in makeinfo requires adjacency of
6943 @c node and sectioning commands; hence the repetition of @chapter BLAH
6944 @c in both conditional blocks.
6947 @include c-alpha.texi
6963 @include c-bfin.texi
6967 @include c-cr16.texi
6971 @include c-cris.texi
6976 @node Machine Dependencies
6977 @chapter Machine Dependent Features
6979 The machine instruction sets are different on each Renesas chip family,
6980 and there are also some syntax differences among the families. This
6981 chapter describes the specific @command{@value{AS}} features for each
6985 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6986 * SH-Dependent:: Renesas SH Dependent Features
6993 @include c-d10v.texi
6997 @include c-d30v.texi
7001 @include c-h8300.texi
7005 @include c-hppa.texi
7009 @include c-i370.texi
7013 @include c-i386.texi
7017 @include c-i860.texi
7021 @include c-i960.texi
7025 @include c-ia64.texi
7029 @include c-ip2k.texi
7033 @include c-lm32.texi
7037 @include c-m32c.texi
7041 @include c-m32r.texi
7045 @include c-m68k.texi
7049 @include c-m68hc11.texi
7053 @include c-microblaze.texi
7057 @include c-mips.texi
7061 @include c-mmix.texi
7065 @include c-msp430.texi
7069 @include c-ns32k.texi
7073 @include c-pdp11.texi
7089 @include c-s390.texi
7093 @include c-score.texi
7098 @include c-sh64.texi
7102 @include c-sparc.texi
7106 @include c-tic54x.texi
7110 @include c-tic6x.texi
7114 @include c-tilegx.texi
7118 @include c-tilepro.texi
7134 @include c-v850.texi
7138 @include c-xtensa.texi
7142 @c reverse effect of @down at top of generic Machine-Dep chapter
7146 @node Reporting Bugs
7147 @chapter Reporting Bugs
7148 @cindex bugs in assembler
7149 @cindex reporting bugs in assembler
7151 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7153 Reporting a bug may help you by bringing a solution to your problem, or it may
7154 not. But in any case the principal function of a bug report is to help the
7155 entire community by making the next version of @command{@value{AS}} work better.
7156 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7158 In order for a bug report to serve its purpose, you must include the
7159 information that enables us to fix the bug.
7162 * Bug Criteria:: Have you found a bug?
7163 * Bug Reporting:: How to report bugs
7167 @section Have You Found a Bug?
7168 @cindex bug criteria
7170 If you are not sure whether you have found a bug, here are some guidelines:
7173 @cindex fatal signal
7174 @cindex assembler crash
7175 @cindex crash of assembler
7177 If the assembler gets a fatal signal, for any input whatever, that is a
7178 @command{@value{AS}} bug. Reliable assemblers never crash.
7180 @cindex error on valid input
7182 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7184 @cindex invalid input
7186 If @command{@value{AS}} does not produce an error message for invalid input, that
7187 is a bug. However, you should note that your idea of ``invalid input'' might
7188 be our idea of ``an extension'' or ``support for traditional practice''.
7191 If you are an experienced user of assemblers, your suggestions for improvement
7192 of @command{@value{AS}} are welcome in any case.
7196 @section How to Report Bugs
7198 @cindex assembler bugs, reporting
7200 A number of companies and individuals offer support for @sc{gnu} products. If
7201 you obtained @command{@value{AS}} from a support organization, we recommend you
7202 contact that organization first.
7204 You can find contact information for many support companies and
7205 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7209 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7213 The fundamental principle of reporting bugs usefully is this:
7214 @strong{report all the facts}. If you are not sure whether to state a
7215 fact or leave it out, state it!
7217 Often people omit facts because they think they know what causes the problem
7218 and assume that some details do not matter. Thus, you might assume that the
7219 name of a symbol you use in an example does not matter. Well, probably it does
7220 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7221 happens to fetch from the location where that name is stored in memory;
7222 perhaps, if the name were different, the contents of that location would fool
7223 the assembler into doing the right thing despite the bug. Play it safe and
7224 give a specific, complete example. That is the easiest thing for you to do,
7225 and the most helpful.
7227 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7228 it is new to us. Therefore, always write your bug reports on the assumption
7229 that the bug has not been reported previously.
7231 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7232 bell?'' This cannot help us fix a bug, so it is basically useless. We
7233 respond by asking for enough details to enable us to investigate.
7234 You might as well expedite matters by sending them to begin with.
7236 To enable us to fix the bug, you should include all these things:
7240 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
7241 it with the @samp{--version} argument.
7243 Without this, we will not know whether there is any point in looking for
7244 the bug in the current version of @command{@value{AS}}.
7247 Any patches you may have applied to the @command{@value{AS}} source.
7250 The type of machine you are using, and the operating system name and
7254 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
7258 The command arguments you gave the assembler to assemble your example and
7259 observe the bug. To guarantee you will not omit something important, list them
7260 all. A copy of the Makefile (or the output from make) is sufficient.
7262 If we were to try to guess the arguments, we would probably guess wrong
7263 and then we might not encounter the bug.
7266 A complete input file that will reproduce the bug. If the bug is observed when
7267 the assembler is invoked via a compiler, send the assembler source, not the
7268 high level language source. Most compilers will produce the assembler source
7269 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
7270 the options @samp{-v --save-temps}; this will save the assembler source in a
7271 file with an extension of @file{.s}, and also show you exactly how
7272 @command{@value{AS}} is being run.
7275 A description of what behavior you observe that you believe is
7276 incorrect. For example, ``It gets a fatal signal.''
7278 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
7279 will certainly notice it. But if the bug is incorrect output, we might not
7280 notice unless it is glaringly wrong. You might as well not give us a chance to
7283 Even if the problem you experience is a fatal signal, you should still say so
7284 explicitly. Suppose something strange is going on, such as, your copy of
7285 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
7286 library on your system. (This has happened!) Your copy might crash and ours
7287 would not. If you told us to expect a crash, then when ours fails to crash, we
7288 would know that the bug was not happening for us. If you had not told us to
7289 expect a crash, then we would not be able to draw any conclusion from our
7293 If you wish to suggest changes to the @command{@value{AS}} source, send us context
7294 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
7295 option. Always send diffs from the old file to the new file. If you even
7296 discuss something in the @command{@value{AS}} source, refer to it by context, not
7299 The line numbers in our development sources will not match those in your
7300 sources. Your line numbers would convey no useful information to us.
7303 Here are some things that are not necessary:
7307 A description of the envelope of the bug.
7309 Often people who encounter a bug spend a lot of time investigating
7310 which changes to the input file will make the bug go away and which
7311 changes will not affect it.
7313 This is often time consuming and not very useful, because the way we
7314 will find the bug is by running a single example under the debugger
7315 with breakpoints, not by pure deduction from a series of examples.
7316 We recommend that you save your time for something else.
7318 Of course, if you can find a simpler example to report @emph{instead}
7319 of the original one, that is a convenience for us. Errors in the
7320 output will be easier to spot, running under the debugger will take
7321 less time, and so on.
7323 However, simplification is not vital; if you do not want to do this,
7324 report the bug anyway and send us the entire test case you used.
7327 A patch for the bug.
7329 A patch for the bug does help us if it is a good one. But do not omit
7330 the necessary information, such as the test case, on the assumption that
7331 a patch is all we need. We might see problems with your patch and decide
7332 to fix the problem another way, or we might not understand it at all.
7334 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
7335 construct an example that will make the program follow a certain path through
7336 the code. If you do not send us the example, we will not be able to construct
7337 one, so we will not be able to verify that the bug is fixed.
7339 And if we cannot understand what bug you are trying to fix, or why your
7340 patch should be an improvement, we will not install it. A test case will
7341 help us to understand.
7344 A guess about what the bug is or what it depends on.
7346 Such guesses are usually wrong. Even we cannot guess right about such
7347 things without first using the debugger to find the facts.
7350 @node Acknowledgements
7351 @chapter Acknowledgements
7353 If you have contributed to GAS and your name isn't listed here,
7354 it is not meant as a slight. We just don't know about it. Send mail to the
7355 maintainer, and we'll correct the situation. Currently
7357 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
7359 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
7362 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
7363 information and the 68k series machines, most of the preprocessing pass, and
7364 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
7366 K. Richard Pixley maintained GAS for a while, adding various enhancements and
7367 many bug fixes, including merging support for several processors, breaking GAS
7368 up to handle multiple object file format back ends (including heavy rewrite,
7369 testing, an integration of the coff and b.out back ends), adding configuration
7370 including heavy testing and verification of cross assemblers and file splits
7371 and renaming, converted GAS to strictly ANSI C including full prototypes, added
7372 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
7373 port (including considerable amounts of reverse engineering), a SPARC opcode
7374 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
7375 assertions and made them work, much other reorganization, cleanup, and lint.
7377 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
7378 in format-specific I/O modules.
7380 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
7381 has done much work with it since.
7383 The Intel 80386 machine description was written by Eliot Dresselhaus.
7385 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
7387 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
7388 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
7390 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
7391 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
7392 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
7393 support a.out format.
7395 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
7396 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
7397 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
7398 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
7401 John Gilmore built the AMD 29000 support, added @code{.include} support, and
7402 simplified the configuration of which versions accept which directives. He
7403 updated the 68k machine description so that Motorola's opcodes always produced
7404 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
7405 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
7406 cross-compilation support, and one bug in relaxation that took a week and
7407 required the proverbial one-bit fix.
7409 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
7410 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
7411 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
7412 PowerPC assembler, and made a few other minor patches.
7414 Steve Chamberlain made GAS able to generate listings.
7416 Hewlett-Packard contributed support for the HP9000/300.
7418 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
7419 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
7420 formats). This work was supported by both the Center for Software Science at
7421 the University of Utah and Cygnus Support.
7423 Support for ELF format files has been worked on by Mark Eichin of Cygnus
7424 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
7425 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
7426 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
7427 and some initial 64-bit support).
7429 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
7431 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
7432 support for openVMS/Alpha.
7434 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
7437 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
7438 Inc.@: added support for Xtensa processors.
7440 Several engineers at Cygnus Support have also provided many small bug fixes and
7441 configuration enhancements.
7443 Jon Beniston added support for the Lattice Mico32 architecture.
7445 Many others have contributed large or small bugfixes and enhancements. If
7446 you have contributed significant work and are not mentioned on this list, and
7447 want to be, let us know. Some of the history has been lost; we are not
7448 intentionally leaving anyone out.
7450 @node GNU Free Documentation License
7451 @appendix GNU Free Documentation License
7455 @unnumbered AS Index