1 \input texinfo.tex @c -*-texinfo-*-
4 @setfilename gccinstall.info
5 @settitle Installing GCC
10 @include gcc-common.texi
12 @c Specify title for specific html page
14 @settitle Installing GCC
17 @settitle Host/Target specific installation notes for GCC
19 @ifset prerequisiteshtml
20 @settitle Prerequisites for GCC
23 @settitle Downloading GCC
26 @settitle Installing GCC: Configuration
29 @settitle Installing GCC: Building
32 @settitle Installing GCC: Testing
34 @ifset finalinstallhtml
35 @settitle Installing GCC: Final installation
38 @settitle Installing GCC: Binaries
41 @settitle Installing GCC: Old documentation
44 @settitle Installing GCC: GNU Free Documentation License
47 @c Copyright (C) 1988, 1989, 1992, 1993, 1994, 1995, 1996, 1997, 1998,
48 @c 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009 Free Software Foundation, Inc.
49 @c *** Converted to texinfo by Dean Wakerley, dean@wakerley.com
51 @c IMPORTANT: whenever you modify this file, run `install.texi2html' to
52 @c test the generation of HTML documents for the gcc.gnu.org web pages.
54 @c Do not use @footnote{} in this file as it breaks install.texi2html!
56 @c Include everything if we're not making html
60 @set prerequisiteshtml
71 @c Part 2 Summary Description and Copyright
73 Copyright @copyright{} 1988, 1989, 1992, 1993, 1994, 1995, 1996, 1997,
74 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007,
75 2008 Free Software Foundation, Inc.
77 Permission is granted to copy, distribute and/or modify this document
78 under the terms of the GNU Free Documentation License, Version 1.2 or
79 any later version published by the Free Software Foundation; with no
80 Invariant Sections, the Front-Cover texts being (a) (see below), and
81 with the Back-Cover Texts being (b) (see below). A copy of the
82 license is included in the section entitled ``@uref{./gfdl.html,,GNU
83 Free Documentation License}''.
85 (a) The FSF's Front-Cover Text is:
89 (b) The FSF's Back-Cover Text is:
91 You have freedom to copy and modify this GNU Manual, like GNU
92 software. Copies published by the Free Software Foundation raise
93 funds for GNU development.
98 @dircategory Software development
100 * gccinstall: (gccinstall). Installing the GNU Compiler Collection.
103 @c Part 3 Titlepage and Copyright
105 @title Installing GCC
108 @c The following two commands start the copyright page.
110 @vskip 0pt plus 1filll
114 @c Part 4 Top node, Master Menu, and/or Table of Contents
117 @comment node-name, next, Previous, up
120 * Installing GCC:: This document describes the generic installation
121 procedure for GCC as well as detailing some target
122 specific installation instructions.
124 * Specific:: Host/target specific installation notes for GCC.
125 * Binaries:: Where to get pre-compiled binaries.
127 * Old:: Old installation documentation.
129 * GNU Free Documentation License:: How you can copy and share this manual.
130 * Concept Index:: This index has two entries.
138 @c Part 5 The Body of the Document
139 @c ***Installing GCC**********************************************************
141 @comment node-name, next, previous, up
142 @node Installing GCC, Binaries, , Top
146 @chapter Installing GCC
149 The latest version of this document is always available at
150 @uref{http://gcc.gnu.org/install/,,http://gcc.gnu.org/install/}.
152 This document describes the generic installation procedure for GCC as well
153 as detailing some target specific installation instructions.
155 GCC includes several components that previously were separate distributions
156 with their own installation instructions. This document supersedes all
157 package specific installation instructions.
159 @emph{Before} starting the build/install procedure please check the
161 @ref{Specific, host/target specific installation notes}.
164 @uref{specific.html,,host/target specific installation notes}.
166 We recommend you browse the entire generic installation instructions before
169 Lists of successful builds for released versions of GCC are
170 available at @uref{http://gcc.gnu.org/buildstat.html}.
171 These lists are updated as new information becomes available.
173 The installation procedure itself is broken into five steps.
178 * Downloading the source::
181 * Testing:: (optional)
188 @uref{prerequisites.html,,Prerequisites}
190 @uref{download.html,,Downloading the source}
192 @uref{configure.html,,Configuration}
194 @uref{build.html,,Building}
196 @uref{test.html,,Testing} (optional)
198 @uref{finalinstall.html,,Final install}
202 Please note that GCC does not support @samp{make uninstall} and probably
203 won't do so in the near future as this would open a can of worms. Instead,
204 we suggest that you install GCC into a directory of its own and simply
205 remove that directory when you do not need that specific version of GCC
206 any longer, and, if shared libraries are installed there as well, no
207 more binaries exist that use them.
210 There are also some @uref{old.html,,old installation instructions},
211 which are mostly obsolete but still contain some information which has
212 not yet been merged into the main part of this manual.
220 @uref{./index.html,,Return to the GCC Installation page}
226 @c ***Prerequisites**************************************************
228 @comment node-name, next, previous, up
229 @node Prerequisites, Downloading the source, , Installing GCC
231 @ifset prerequisiteshtml
233 @chapter Prerequisites
235 @cindex Prerequisites
237 GCC requires that various tools and packages be available for use in the
238 build procedure. Modifying GCC sources requires additional tools
241 @heading Tools/packages necessary for building GCC
243 @item ISO C90 compiler
244 Necessary to bootstrap GCC, although versions of GCC prior
245 to 3.4 also allow bootstrapping with a traditional (K&R) C compiler.
247 To build all languages in a cross-compiler or other configuration where
248 3-stage bootstrap is not performed, you need to start with an existing
249 GCC binary (version 2.95 or later) because source code for language
250 frontends other than C might use GCC extensions.
254 In order to build the Ada compiler (GNAT) you must already have GNAT
255 installed because portions of the Ada frontend are written in Ada (with
256 GNAT extensions.) Refer to the Ada installation instructions for more
257 specific information.
259 @item A ``working'' POSIX compatible shell, or GNU bash
261 Necessary when running @command{configure} because some
262 @command{/bin/sh} shells have bugs and may crash when configuring the
263 target libraries. In other cases, @command{/bin/sh} or @command{ksh}
264 have disastrous corner-case performance problems. This
265 can cause target @command{configure} runs to literally take days to
266 complete in some cases.
268 So on some platforms @command{/bin/ksh} is sufficient, on others it
269 isn't. See the host/target specific instructions for your platform, or
270 use @command{bash} to be sure. Then set @env{CONFIG_SHELL} in your
271 environment to your ``good'' shell prior to running
272 @command{configure}/@command{make}.
274 @command{zsh} is not a fully compliant POSIX shell and will not
275 work when configuring GCC@.
277 @item A POSIX or SVR4 awk
279 Necessary for creating some of the generated source files for GCC@.
280 If in doubt, use a recent GNU awk version, as some of the older ones
281 are broken. GNU awk version 3.1.5 is known to work.
285 Necessary in some circumstances, optional in others. See the
286 host/target specific instructions for your platform for the exact
289 @item gzip version 1.2.4 (or later) or
290 @itemx bzip2 version 1.0.2 (or later)
292 Necessary to uncompress GCC @command{tar} files when source code is
293 obtained via FTP mirror sites.
295 @item GNU make version 3.80 (or later)
297 You must have GNU make installed to build GCC@.
299 @item GNU tar version 1.14 (or later)
301 Necessary (only on some platforms) to untar the source code. Many
302 systems' @command{tar} programs will also work, only try GNU
303 @command{tar} if you have problems.
305 @item GNU Multiple Precision Library (GMP) version 4.2 (or later)
307 Necessary to build GCC@. If you do not have it installed in your
308 library search path, you will have to configure with the
309 @option{--with-gmp} configure option. See also @option{--with-gmp-lib}
310 and @option{--with-gmp-include}. Alternatively, if a GMP source
311 distribution is found in a subdirectory of your GCC sources named
312 @file{gmp}, it will be built together with GCC@.
314 @item MPFR Library version 2.3.2 (or later)
316 Necessary to build GCC@. It can be downloaded from
317 @uref{http://www.mpfr.org/}. The @option{--with-mpfr} configure
318 option should be used if your MPFR Library is not installed in your
319 default library search path. See also @option{--with-mpfr-lib} and
320 @option{--with-mpfr-include}. Alternatively, if a MPFR source
321 distribution is found in a subdirectory of your GCC sources named
322 @file{mpfr}, it will be built together with GCC@.
324 @item Parma Polyhedra Library (PPL) version 0.10
326 Necessary to build GCC with the Graphite loop optimizations.
327 It can be downloaded from @uref{http://www.cs.unipr.it/ppl/Download/}.
329 The @option{--with-ppl} configure option should be used if PPL is not
330 installed in your default library search path.
332 @item CLooG-PPL version 0.15
334 Necessary to build GCC with the Graphite loop optimizations. It can
335 be downloaded from @uref{ftp://gcc.gnu.org/pub/gcc/infrastructure/}.
336 The code in @file{cloog-ppl-0.15.tar.gz} comes from a branch of CLooG
337 available from @uref{http://repo.or.cz/w/cloog-ppl.git}. CLooG-PPL
338 should be configured with @option{--with-ppl}.
340 The @option{--with-cloog} configure option should be used if CLooG is
341 not installed in your default library search path.
343 @item @command{jar}, or InfoZIP (@command{zip} and @command{unzip})
345 Necessary to build libgcj, the GCJ runtime.
347 @item MPC Library version 0.6.0 (or later)
349 Optional when building GCC@. Having this library will enable
350 additional optimizations on complex numbers. It can be downloaded
351 from @uref{http://www.multiprecision.org/mpc/}. The
352 @option{--with-mpc} configure option should be used if your MPC
353 Library is not installed in your default library search path. See
354 also @option{--with-mpc-lib} and @option{--with-mpc-include}.
355 Alternatively, if an MPC source distribution is found in a
356 subdirectory of your GCC sources named @file{mpc}, it will be built
361 @heading Tools/packages necessary for modifying GCC
363 @item autoconf version 2.64
364 @itemx GNU m4 version 1.4.6 (or later)
366 Necessary when modifying @file{configure.ac}, @file{aclocal.m4}, etc.@:
367 to regenerate @file{configure} and @file{config.in} files.
369 @item automake version 1.11
371 Necessary when modifying a @file{Makefile.am} file to regenerate its
372 associated @file{Makefile.in}.
374 Much of GCC does not use automake, so directly edit the @file{Makefile.in}
375 file. Specifically this applies to the @file{gcc}, @file{intl},
376 @file{libcpp}, @file{libiberty}, @file{libobjc} directories as well
377 as any of their subdirectories.
379 For directories that use automake, GCC requires the latest release in
380 the 1.11 series, which is currently 1.11. When regenerating a directory
381 to a newer version, please update all the directories using an older 1.11
382 to the latest released version.
384 @item gettext version 0.14.5 (or later)
386 Needed to regenerate @file{gcc.pot}.
388 @item gperf version 2.7.2 (or later)
390 Necessary when modifying @command{gperf} input files, e.g.@:
391 @file{gcc/cp/cfns.gperf} to regenerate its associated header file, e.g.@:
392 @file{gcc/cp/cfns.h}.
398 Necessary to run the GCC testsuite; see the section on testing for details.
400 @item autogen version 5.5.4 (or later) and
401 @itemx guile version 1.4.1 (or later)
403 Necessary to regenerate @file{fixinc/fixincl.x} from
404 @file{fixinc/inclhack.def} and @file{fixinc/*.tpl}.
406 Necessary to run @samp{make check} for @file{fixinc}.
408 Necessary to regenerate the top level @file{Makefile.in} file from
409 @file{Makefile.tpl} and @file{Makefile.def}.
411 @item Flex version 2.5.4 (or later)
413 Necessary when modifying @file{*.l} files.
415 Necessary to build GCC during development because the generated output
416 files are not included in the SVN repository. They are included in
419 @item Texinfo version 4.7 (or later)
421 Necessary for running @command{makeinfo} when modifying @file{*.texi}
422 files to test your changes.
424 Necessary for running @command{make dvi} or @command{make pdf} to
425 create printable documentation in DVI or PDF format. Texinfo version
426 4.8 or later is required for @command{make pdf}.
428 Necessary to build GCC documentation during development because the
429 generated output files are not included in the SVN repository. They are
430 included in releases.
432 @item @TeX{} (any working version)
434 Necessary for running @command{texi2dvi} and @command{texi2pdf}, which
435 are used when running @command{make dvi} or @command{make pdf} to create
436 DVI or PDF files, respectively.
438 @item SVN (any version)
439 @itemx SSH (any version)
441 Necessary to access the SVN repository. Public releases and weekly
442 snapshots of the development sources are also available via FTP@.
444 @item Perl version 5.6.1 (or later)
446 Necessary when regenerating @file{Makefile} dependencies in libiberty.
447 Necessary when regenerating @file{libiberty/functions.texi}.
448 Necessary when generating manpages from Texinfo manuals.
449 Necessary when targetting Darwin, building libstdc++,
450 and not using @option{--disable-symvers}.
451 Used by various scripts to generate some files included in SVN (mainly
452 Unicode-related and rarely changing) from source tables.
454 @item GNU diffutils version 2.7 (or later)
456 Useful when submitting patches for the GCC source code.
458 @item patch version 2.5.4 (or later)
460 Necessary when applying patches, created with @command{diff}, to one's
466 If you wish to modify @file{.java} files in libjava, you will need to
467 configure with @option{--enable-java-maintainer-mode}, and you will need
468 to have executables named @command{ecj1} and @command{gjavah} in your path.
469 The @command{ecj1} executable should run the Eclipse Java compiler via
470 the GCC-specific entry point. You can download a suitable jar from
471 @uref{ftp://sourceware.org/pub/java/}, or by running the script
472 @command{contrib/download_ecj}.
474 @item antlr.jar version 2.7.1 (or later)
477 If you wish to build the @command{gjdoc} binary in libjava, you will
478 need to have an @file{antlr.jar} library available. The library is
479 searched in system locations but can be configured with
480 @option{--with-antlr-jar=} instead. When configuring with
481 @option{--enable-java-maintainer-mode}, you will need to have one of
482 the executables named @command{cantlr}, @command{runantlr} or
483 @command{antlr} in your path.
492 @uref{./index.html,,Return to the GCC Installation page}
496 @c ***Downloading the source**************************************************
498 @comment node-name, next, previous, up
499 @node Downloading the source, Configuration, Prerequisites, Installing GCC
503 @chapter Downloading GCC
505 @cindex Downloading GCC
506 @cindex Downloading the Source
508 GCC is distributed via @uref{http://gcc.gnu.org/svn.html,,SVN} and FTP
509 tarballs compressed with @command{gzip} or
510 @command{bzip2}. It is possible to download a full distribution or specific
513 Please refer to the @uref{http://gcc.gnu.org/releases.html,,releases web page}
514 for information on how to obtain GCC@.
516 The full distribution includes the C, C++, Objective-C, Fortran, Java,
517 and Ada (in the case of GCC 3.1 and later) compilers. The full
518 distribution also includes runtime libraries for C++, Objective-C,
519 Fortran, and Java. In GCC 3.0 and later versions, the GNU compiler
520 testsuites are also included in the full distribution.
522 If you choose to download specific components, you must download the core
523 GCC distribution plus any language specific distributions you wish to
524 use. The core distribution includes the C language front end as well as the
525 shared components. Each language has a tarball which includes the language
526 front end as well as the language runtime (when appropriate).
528 Unpack the core distribution as well as any language specific
529 distributions in the same directory.
531 If you also intend to build binutils (either to upgrade an existing
532 installation or for use in place of the corresponding tools of your
533 OS), unpack the binutils distribution either in the same directory or
534 a separate one. In the latter case, add symbolic links to any
535 components of the binutils you intend to build alongside the compiler
536 (@file{bfd}, @file{binutils}, @file{gas}, @file{gprof}, @file{ld},
537 @file{opcodes}, @dots{}) to the directory containing the GCC sources.
539 Likewise the GMP, MPFR and MPC libraries can be automatically built
540 together with GCC. Unpack the GMP, MPFR and/or MPC source
541 distributions in the directory containing the GCC sources and rename
542 their directories to @file{gmp}, @file{mpfr} and @file{mpc},
543 respectively (or use symbolic links with the same name).
550 @uref{./index.html,,Return to the GCC Installation page}
554 @c ***Configuration***********************************************************
556 @comment node-name, next, previous, up
557 @node Configuration, Building, Downloading the source, Installing GCC
561 @chapter Installing GCC: Configuration
563 @cindex Configuration
564 @cindex Installing GCC: Configuration
566 Like most GNU software, GCC must be configured before it can be built.
567 This document describes the recommended configuration procedure
568 for both native and cross targets.
570 We use @var{srcdir} to refer to the toplevel source directory for
571 GCC; we use @var{objdir} to refer to the toplevel build/object directory.
573 If you obtained the sources via SVN, @var{srcdir} must refer to the top
574 @file{gcc} directory, the one where the @file{MAINTAINERS} can be found,
575 and not its @file{gcc} subdirectory, otherwise the build will fail.
577 If either @var{srcdir} or @var{objdir} is located on an automounted NFS
578 file system, the shell's built-in @command{pwd} command will return
579 temporary pathnames. Using these can lead to various sorts of build
580 problems. To avoid this issue, set the @env{PWDCMD} environment
581 variable to an automounter-aware @command{pwd} command, e.g.,
582 @command{pawd} or @samp{amq -w}, during the configuration and build
585 First, we @strong{highly} recommend that GCC be built into a
586 separate directory than the sources which does @strong{not} reside
587 within the source tree. This is how we generally build GCC; building
588 where @var{srcdir} == @var{objdir} should still work, but doesn't
589 get extensive testing; building where @var{objdir} is a subdirectory
590 of @var{srcdir} is unsupported.
592 If you have previously built GCC in the same directory for a
593 different target machine, do @samp{make distclean} to delete all files
594 that might be invalid. One of the files this deletes is @file{Makefile};
595 if @samp{make distclean} complains that @file{Makefile} does not exist
596 or issues a message like ``don't know how to make distclean'' it probably
597 means that the directory is already suitably clean. However, with the
598 recommended method of building in a separate @var{objdir}, you should
599 simply use a different @var{objdir} for each target.
601 Second, when configuring a native system, either @command{cc} or
602 @command{gcc} must be in your path or you must set @env{CC} in
603 your environment before running configure. Otherwise the configuration
607 Note that the bootstrap compiler and the resulting GCC must be link
608 compatible, else the bootstrap will fail with linker errors about
609 incompatible object file formats. Several multilibed targets are
610 affected by this requirement, see
612 @ref{Specific, host/target specific installation notes}.
615 @uref{specific.html,,host/target specific installation notes}.
624 % @var{srcdir}/configure [@var{options}] [@var{target}]
627 @heading Distributor options
629 If you will be distributing binary versions of GCC, with modifications
630 to the source code, you should use the options described in this
631 section to make clear that your version contains modifications.
634 @item --with-pkgversion=@var{version}
635 Specify a string that identifies your package. You may wish
636 to include a build number or build date. This version string will be
637 included in the output of @command{gcc --version}. This suffix does
638 not replace the default version string, only the @samp{GCC} part.
640 The default value is @samp{GCC}.
642 @item --with-bugurl=@var{url}
643 Specify the URL that users should visit if they wish to report a bug.
644 You are of course welcome to forward bugs reported to you to the FSF,
645 if you determine that they are not bugs in your modifications.
647 The default value refers to the FSF's GCC bug tracker.
651 @heading Target specification
654 GCC has code to correctly determine the correct value for @var{target}
655 for nearly all native systems. Therefore, we highly recommend you not
656 provide a configure target when configuring a native compiler.
659 @var{target} must be specified as @option{--target=@var{target}}
660 when configuring a cross compiler; examples of valid targets would be
661 m68k-elf, sh-elf, etc.
664 Specifying just @var{target} instead of @option{--target=@var{target}}
665 implies that the host defaults to @var{target}.
669 @heading Options specification
671 Use @var{options} to override several configure time options for
672 GCC@. A list of supported @var{options} follows; @samp{configure
673 --help} may list other options, but those not listed below may not
674 work and should not normally be used.
676 Note that each @option{--enable} option has a corresponding
677 @option{--disable} option and that each @option{--with} option has a
678 corresponding @option{--without} option.
681 @item --prefix=@var{dirname}
682 Specify the toplevel installation
683 directory. This is the recommended way to install the tools into a directory
684 other than the default. The toplevel installation directory defaults to
687 We @strong{highly} recommend against @var{dirname} being the same or a
688 subdirectory of @var{objdir} or vice versa. If specifying a directory
689 beneath a user's home directory tree, some shells will not expand
690 @var{dirname} correctly if it contains the @samp{~} metacharacter; use
693 The following standard @command{autoconf} options are supported. Normally you
694 should not need to use these options.
696 @item --exec-prefix=@var{dirname}
697 Specify the toplevel installation directory for architecture-dependent
698 files. The default is @file{@var{prefix}}.
700 @item --bindir=@var{dirname}
701 Specify the installation directory for the executables called by users
702 (such as @command{gcc} and @command{g++}). The default is
703 @file{@var{exec-prefix}/bin}.
705 @item --libdir=@var{dirname}
706 Specify the installation directory for object code libraries and
707 internal data files of GCC@. The default is @file{@var{exec-prefix}/lib}.
709 @item --libexecdir=@var{dirname}
710 Specify the installation directory for internal executables of GCC@.
711 The default is @file{@var{exec-prefix}/libexec}.
713 @item --with-slibdir=@var{dirname}
714 Specify the installation directory for the shared libgcc library. The
715 default is @file{@var{libdir}}.
717 @item --datarootdir=@var{dirname}
718 Specify the root of the directory tree for read-only architecture-independent
719 data files referenced by GCC@. The default is @file{@var{prefix}/share}.
721 @item --infodir=@var{dirname}
722 Specify the installation directory for documentation in info format.
723 The default is @file{@var{datarootdir}/info}.
725 @item --datadir=@var{dirname}
726 Specify the installation directory for some architecture-independent
727 data files referenced by GCC@. The default is @file{@var{datarootdir}}.
729 @item --docdir=@var{dirname}
730 Specify the installation directory for documentation files (other
731 than Info) for GCC@. The default is @file{@var{datarootdir}/doc}.
733 @item --htmldir=@var{dirname}
734 Specify the installation directory for HTML documentation files.
735 The default is @file{@var{docdir}}.
737 @item --pdfdir=@var{dirname}
738 Specify the installation directory for PDF documentation files.
739 The default is @file{@var{docdir}}.
741 @item --mandir=@var{dirname}
742 Specify the installation directory for manual pages. The default is
743 @file{@var{datarootdir}/man}. (Note that the manual pages are only extracts
744 from the full GCC manuals, which are provided in Texinfo format. The manpages
745 are derived by an automatic conversion process from parts of the full
748 @item --with-gxx-include-dir=@var{dirname}
750 the installation directory for G++ header files. The default depends
751 on other configuration options, and differs between cross and native
756 @item --program-prefix=@var{prefix}
757 GCC supports some transformations of the names of its programs when
758 installing them. This option prepends @var{prefix} to the names of
759 programs to install in @var{bindir} (see above). For example, specifying
760 @option{--program-prefix=foo-} would result in @samp{gcc}
761 being installed as @file{/usr/local/bin/foo-gcc}.
763 @item --program-suffix=@var{suffix}
764 Appends @var{suffix} to the names of programs to install in @var{bindir}
765 (see above). For example, specifying @option{--program-suffix=-3.1}
766 would result in @samp{gcc} being installed as
767 @file{/usr/local/bin/gcc-3.1}.
769 @item --program-transform-name=@var{pattern}
770 Applies the @samp{sed} script @var{pattern} to be applied to the names
771 of programs to install in @var{bindir} (see above). @var{pattern} has to
772 consist of one or more basic @samp{sed} editing commands, separated by
773 semicolons. For example, if you want the @samp{gcc} program name to be
774 transformed to the installed program @file{/usr/local/bin/myowngcc} and
775 the @samp{g++} program name to be transformed to
776 @file{/usr/local/bin/gspecial++} without changing other program names,
777 you could use the pattern
778 @option{--program-transform-name='s/^gcc$/myowngcc/; s/^g++$/gspecial++/'}
779 to achieve this effect.
781 All three options can be combined and used together, resulting in more
782 complex conversion patterns. As a basic rule, @var{prefix} (and
783 @var{suffix}) are prepended (appended) before further transformations
784 can happen with a special transformation script @var{pattern}.
786 As currently implemented, this option only takes effect for native
787 builds; cross compiler binaries' names are not transformed even when a
788 transformation is explicitly asked for by one of these options.
790 For native builds, some of the installed programs are also installed
791 with the target alias in front of their name, as in
792 @samp{i686-pc-linux-gnu-gcc}. All of the above transformations happen
793 before the target alias is prepended to the name---so, specifying
794 @option{--program-prefix=foo-} and @option{program-suffix=-3.1}, the
795 resulting binary would be installed as
796 @file{/usr/local/bin/i686-pc-linux-gnu-foo-gcc-3.1}.
798 As a last shortcoming, none of the installed Ada programs are
799 transformed yet, which will be fixed in some time.
801 @item --with-local-prefix=@var{dirname}
803 installation directory for local include files. The default is
804 @file{/usr/local}. Specify this option if you want the compiler to
805 search directory @file{@var{dirname}/include} for locally installed
806 header files @emph{instead} of @file{/usr/local/include}.
808 You should specify @option{--with-local-prefix} @strong{only} if your
809 site has a different convention (not @file{/usr/local}) for where to put
812 The default value for @option{--with-local-prefix} is @file{/usr/local}
813 regardless of the value of @option{--prefix}. Specifying
814 @option{--prefix} has no effect on which directory GCC searches for
815 local header files. This may seem counterintuitive, but actually it is
818 The purpose of @option{--prefix} is to specify where to @emph{install
819 GCC}. The local header files in @file{/usr/local/include}---if you put
820 any in that directory---are not part of GCC@. They are part of other
821 programs---perhaps many others. (GCC installs its own header files in
822 another directory which is based on the @option{--prefix} value.)
824 Both the local-prefix include directory and the GCC-prefix include
825 directory are part of GCC's ``system include'' directories. Although these
826 two directories are not fixed, they need to be searched in the proper
827 order for the correct processing of the include_next directive. The
828 local-prefix include directory is searched before the GCC-prefix
829 include directory. Another characteristic of system include directories
830 is that pedantic warnings are turned off for headers in these directories.
832 Some autoconf macros add @option{-I @var{directory}} options to the
833 compiler command line, to ensure that directories containing installed
834 packages' headers are searched. When @var{directory} is one of GCC's
835 system include directories, GCC will ignore the option so that system
836 directories continue to be processed in the correct order. This
837 may result in a search order different from what was specified but the
838 directory will still be searched.
840 GCC automatically searches for ordinary libraries using
841 @env{GCC_EXEC_PREFIX}. Thus, when the same installation prefix is
842 used for both GCC and packages, GCC will automatically search for
843 both headers and libraries. This provides a configuration that is
844 easy to use. GCC behaves in a manner similar to that when it is
845 installed as a system compiler in @file{/usr}.
847 Sites that need to install multiple versions of GCC may not want to
848 use the above simple configuration. It is possible to use the
849 @option{--program-prefix}, @option{--program-suffix} and
850 @option{--program-transform-name} options to install multiple versions
851 into a single directory, but it may be simpler to use different prefixes
852 and the @option{--with-local-prefix} option to specify the location of the
853 site-specific files for each version. It will then be necessary for
854 users to specify explicitly the location of local site libraries
855 (e.g., with @env{LIBRARY_PATH}).
857 The same value can be used for both @option{--with-local-prefix} and
858 @option{--prefix} provided it is not @file{/usr}. This can be used
859 to avoid the default search of @file{/usr/local/include}.
861 @strong{Do not} specify @file{/usr} as the @option{--with-local-prefix}!
862 The directory you use for @option{--with-local-prefix} @strong{must not}
863 contain any of the system's standard header files. If it did contain
864 them, certain programs would be miscompiled (including GNU Emacs, on
865 certain targets), because this would override and nullify the header
866 file corrections made by the @command{fixincludes} script.
868 Indications are that people who use this option use it based on mistaken
869 ideas of what it is for. People use it as if it specified where to
870 install part of GCC@. Perhaps they make this assumption because
871 installing GCC creates the directory.
873 @item --enable-shared[=@var{package}[,@dots{}]]
874 Build shared versions of libraries, if shared libraries are supported on
875 the target platform. Unlike GCC 2.95.x and earlier, shared libraries
876 are enabled by default on all platforms that support shared libraries.
878 If a list of packages is given as an argument, build shared libraries
879 only for the listed packages. For other packages, only static libraries
880 will be built. Package names currently recognized in the GCC tree are
881 @samp{libgcc} (also known as @samp{gcc}), @samp{libstdc++} (not
882 @samp{libstdc++-v3}), @samp{libffi}, @samp{zlib}, @samp{boehm-gc},
883 @samp{ada}, @samp{libada}, @samp{libjava} and @samp{libobjc}.
884 Note @samp{libiberty} does not support shared libraries at all.
886 Use @option{--disable-shared} to build only static libraries. Note that
887 @option{--disable-shared} does not accept a list of package names as
888 argument, only @option{--enable-shared} does.
890 @item @anchor{with-gnu-as}--with-gnu-as
891 Specify that the compiler should assume that the
892 assembler it finds is the GNU assembler. However, this does not modify
893 the rules to find an assembler and will result in confusion if the
894 assembler found is not actually the GNU assembler. (Confusion may also
895 result if the compiler finds the GNU assembler but has not been
896 configured with @option{--with-gnu-as}.) If you have more than one
897 assembler installed on your system, you may want to use this option in
898 connection with @option{--with-as=@var{pathname}} or
899 @option{--with-build-time-tools=@var{pathname}}.
901 The following systems are the only ones where it makes a difference
902 whether you use the GNU assembler. On any other system,
903 @option{--with-gnu-as} has no effect.
906 @item @samp{hppa1.0-@var{any}-@var{any}}
907 @item @samp{hppa1.1-@var{any}-@var{any}}
908 @item @samp{sparc-sun-solaris2.@var{any}}
909 @item @samp{sparc64-@var{any}-solaris2.@var{any}}
912 @item @anchor{with-as}--with-as=@var{pathname}
913 Specify that the compiler should use the assembler pointed to by
914 @var{pathname}, rather than the one found by the standard rules to find
915 an assembler, which are:
918 Unless GCC is being built with a cross compiler, check the
919 @file{@var{libexec}/gcc/@var{target}/@var{version}} directory.
920 @var{libexec} defaults to @file{@var{exec-prefix}/libexec};
921 @var{exec-prefix} defaults to @var{prefix}, which
922 defaults to @file{/usr/local} unless overridden by the
923 @option{--prefix=@var{pathname}} switch described above. @var{target}
924 is the target system triple, such as @samp{sparc-sun-solaris2.7}, and
925 @var{version} denotes the GCC version, such as 3.0.
928 If the target system is the same that you are building on, check
929 operating system specific directories (e.g.@: @file{/usr/ccs/bin} on
933 Check in the @env{PATH} for a tool whose name is prefixed by the
934 target system triple.
937 Check in the @env{PATH} for a tool whose name is not prefixed by the
938 target system triple, if the host and target system triple are
939 the same (in other words, we use a host tool if it can be used for
943 You may want to use @option{--with-as} if no assembler
944 is installed in the directories listed above, or if you have multiple
945 assemblers installed and want to choose one that is not found by the
948 @item @anchor{with-gnu-ld}--with-gnu-ld
949 Same as @uref{#with-gnu-as,,@option{--with-gnu-as}}
952 @item --with-ld=@var{pathname}
953 Same as @uref{#with-as,,@option{--with-as}}
957 Specify that stabs debugging
958 information should be used instead of whatever format the host normally
959 uses. Normally GCC uses the same debug format as the host system.
961 On MIPS based systems and on Alphas, you must specify whether you want
962 GCC to create the normal ECOFF debugging format, or to use BSD-style
963 stabs passed through the ECOFF symbol table. The normal ECOFF debug
964 format cannot fully handle languages other than C@. BSD stabs format can
965 handle other languages, but it only works with the GNU debugger GDB@.
967 Normally, GCC uses the ECOFF debugging format by default; if you
968 prefer BSD stabs, specify @option{--with-stabs} when you configure GCC@.
970 No matter which default you choose when you configure GCC, the user
971 can use the @option{-gcoff} and @option{-gstabs+} options to specify explicitly
972 the debug format for a particular compilation.
974 @option{--with-stabs} is meaningful on the ISC system on the 386, also, if
975 @option{--with-gas} is used. It selects use of stabs debugging
976 information embedded in COFF output. This kind of debugging information
977 supports C++ well; ordinary COFF debugging information does not.
979 @option{--with-stabs} is also meaningful on 386 systems running SVR4. It
980 selects use of stabs debugging information embedded in ELF output. The
981 C++ compiler currently (2.6.0) does not support the DWARF debugging
982 information normally used on 386 SVR4 platforms; stabs provide a
983 workable alternative. This requires gas and gdb, as the normal SVR4
984 tools can not generate or interpret stabs.
986 @item --disable-multilib
987 Specify that multiple target
988 libraries to support different target variants, calling
989 conventions, etc.@: should not be built. The default is to build a
990 predefined set of them.
992 Some targets provide finer-grained control over which multilibs are built
993 (e.g., @option{--disable-softfloat}):
999 fpu, 26bit, underscore, interwork, biendian, nofmult.
1002 softfloat, m68881, m68000, m68020.
1005 single-float, biendian, softfloat.
1007 @item powerpc*-*-*, rs6000*-*-*
1008 aix64, pthread, softfloat, powercpu, powerpccpu, powerpcos, biendian,
1013 @item --with-multilib-list=@var{list}
1014 @itemx --without-multilib-list
1015 Specify what multilibs to build.
1016 Currently only implemented for sh*-*-*.
1018 @var{list} is a comma separated list of CPU names. These must be of the
1019 form @code{sh*} or @code{m*} (in which case they match the compiler option
1020 for that processor). The list should not contain any endian options -
1021 these are handled by @option{--with-endian}.
1023 If @var{list} is empty, then there will be no multilibs for extra
1024 processors. The multilib for the secondary endian remains enabled.
1026 As a special case, if an entry in the list starts with a @code{!}
1027 (exclamation point), then it is added to the list of excluded multilibs.
1028 Entries of this sort should be compatible with @samp{MULTILIB_EXCLUDES}
1029 (once the leading @code{!} has been stripped).
1031 If @option{--with-multilib-list} is not given, then a default set of
1032 multilibs is selected based on the value of @option{--target}. This is
1033 usually the complete set of libraries, but some targets imply a more
1036 Example 1: to configure a compiler for SH4A only, but supporting both
1037 endians, with little endian being the default:
1039 --with-cpu=sh4a --with-endian=little,big --with-multilib-list=
1042 Example 2: to configure a compiler for both SH4A and SH4AL-DSP, but with
1043 only little endian SH4AL:
1045 --with-cpu=sh4a --with-endian=little,big --with-multilib-list=sh4al,!mb/m4al
1048 @item --with-endian=@var{endians}
1049 Specify what endians to use.
1050 Currently only implemented for sh*-*-*.
1052 @var{endians} may be one of the following:
1055 Use big endian exclusively.
1057 Use little endian exclusively.
1059 Use big endian by default. Provide a multilib for little endian.
1061 Use little endian by default. Provide a multilib for big endian.
1064 @item --enable-threads
1065 Specify that the target
1066 supports threads. This affects the Objective-C compiler and runtime
1067 library, and exception handling for other languages like C++ and Java.
1068 On some systems, this is the default.
1070 In general, the best (and, in many cases, the only known) threading
1071 model available will be configured for use. Beware that on some
1072 systems, GCC has not been taught what threading models are generally
1073 available for the system. In this case, @option{--enable-threads} is an
1074 alias for @option{--enable-threads=single}.
1076 @item --disable-threads
1077 Specify that threading support should be disabled for the system.
1078 This is an alias for @option{--enable-threads=single}.
1080 @item --enable-threads=@var{lib}
1082 @var{lib} is the thread support library. This affects the Objective-C
1083 compiler and runtime library, and exception handling for other languages
1084 like C++ and Java. The possibilities for @var{lib} are:
1092 Ada tasking support. For non-Ada programs, this setting is equivalent
1093 to @samp{single}. When used in conjunction with the Ada run time, it
1094 causes GCC to use the same thread primitives as Ada uses. This option
1095 is necessary when using both Ada and the back end exception handling,
1096 which is the default for most Ada targets.
1098 Generic MACH thread support, known to work on NeXTSTEP@. (Please note
1099 that the file needed to support this configuration, @file{gthr-mach.h}, is
1100 missing and thus this setting will cause a known bootstrap failure.)
1102 This is an alias for @samp{single}.
1104 Generic POSIX/Unix98 thread support.
1106 Generic POSIX/Unix95 thread support.
1108 RTEMS thread support.
1110 Disable thread support, should work for all platforms.
1112 Sun Solaris 2 thread support.
1114 VxWorks thread support.
1116 Microsoft Win32 API thread support.
1118 Novell Kernel Services thread support.
1122 Specify that the target supports TLS (Thread Local Storage). Usually
1123 configure can correctly determine if TLS is supported. In cases where
1124 it guesses incorrectly, TLS can be explicitly enabled or disabled with
1125 @option{--enable-tls} or @option{--disable-tls}. This can happen if
1126 the assembler supports TLS but the C library does not, or if the
1127 assumptions made by the configure test are incorrect.
1130 Specify that the target does not support TLS.
1131 This is an alias for @option{--enable-tls=no}.
1133 @item --with-cpu=@var{cpu}
1134 @itemx --with-cpu-32=@var{cpu}
1135 @itemx --with-cpu-64=@var{cpu}
1136 Specify which cpu variant the compiler should generate code for by default.
1137 @var{cpu} will be used as the default value of the @option{-mcpu=} switch.
1138 This option is only supported on some targets, including ARM, i386, M68k,
1139 PowerPC, and SPARC@. The @option{--with-cpu-32} and
1140 @option{--with-cpu-64} options specify separate default CPUs for
1141 32-bit and 64-bit modes; these options are only supported for i386,
1144 @item --with-schedule=@var{cpu}
1145 @itemx --with-arch=@var{cpu}
1146 @itemx --with-arch-32=@var{cpu}
1147 @itemx --with-arch-64=@var{cpu}
1148 @itemx --with-tune=@var{cpu}
1149 @itemx --with-tune-32=@var{cpu}
1150 @itemx --with-tune-64=@var{cpu}
1151 @itemx --with-abi=@var{abi}
1152 @itemx --with-fpu=@var{type}
1153 @itemx --with-float=@var{type}
1154 These configure options provide default values for the @option{-mschedule=},
1155 @option{-march=}, @option{-mtune=}, @option{-mabi=}, and @option{-mfpu=}
1156 options and for @option{-mhard-float} or @option{-msoft-float}. As with
1157 @option{--with-cpu}, which switches will be accepted and acceptable values
1158 of the arguments depend on the target.
1160 @item --with-mode=@var{mode}
1161 Specify if the compiler should default to @option{-marm} or @option{-mthumb}.
1162 This option is only supported on ARM targets.
1164 @item --with-divide=@var{type}
1165 Specify how the compiler should generate code for checking for
1166 division by zero. This option is only supported on the MIPS target.
1167 The possibilities for @var{type} are:
1170 Division by zero checks use conditional traps (this is the default on
1171 systems that support conditional traps).
1173 Division by zero checks use the break instruction.
1176 @c If you make --with-llsc the default for additional targets,
1177 @c update the --with-llsc description in the MIPS section below.
1180 On MIPS targets, make @option{-mllsc} the default when no
1181 @option{-mno-lsc} option is passed. This is the default for
1182 Linux-based targets, as the kernel will emulate them if the ISA does
1185 @item --without-llsc
1186 On MIPS targets, make @option{-mno-llsc} the default when no
1187 @option{-mllsc} option is passed.
1190 On MIPS targets, make @option{-msynci} the default when no
1191 @option{-mno-synci} option is passed.
1193 @item --without-synci
1194 On MIPS targets, make @option{-mno-synci} the default when no
1195 @option{-msynci} option is passed. This is the default.
1197 @item --with-mips-plt
1198 On MIPS targets, make use of copy relocations and PLTs.
1199 These features are extensions to the traditional
1200 SVR4-based MIPS ABIs and require support from GNU binutils
1201 and the runtime C library.
1203 @item --enable-__cxa_atexit
1204 Define if you want to use __cxa_atexit, rather than atexit, to
1205 register C++ destructors for local statics and global objects.
1206 This is essential for fully standards-compliant handling of
1207 destructors, but requires __cxa_atexit in libc. This option is currently
1208 only available on systems with GNU libc. When enabled, this will cause
1209 @option{-fuse-cxa-atexit} to be passed by default.
1211 @item --enable-target-optspace
1213 libraries should be optimized for code space instead of code speed.
1214 This is the default for the m32r platform.
1217 Specify that a user visible @command{cpp} program should not be installed.
1219 @item --with-cpp-install-dir=@var{dirname}
1220 Specify that the user visible @command{cpp} program should be installed
1221 in @file{@var{prefix}/@var{dirname}/cpp}, in addition to @var{bindir}.
1223 @item --enable-initfini-array
1224 Force the use of sections @code{.init_array} and @code{.fini_array}
1225 (instead of @code{.init} and @code{.fini}) for constructors and
1226 destructors. Option @option{--disable-initfini-array} has the
1227 opposite effect. If neither option is specified, the configure script
1228 will try to guess whether the @code{.init_array} and
1229 @code{.fini_array} sections are supported and, if they are, use them.
1231 @item --enable-build-with-cxx
1232 Build GCC using a C++ compiler rather than a C compiler. This is an
1233 experimental option which may become the default in a later release.
1235 @item --enable-maintainer-mode
1236 The build rules that
1237 regenerate the GCC master message catalog @file{gcc.pot} are normally
1238 disabled. This is because it can only be rebuilt if the complete source
1239 tree is present. If you have changed the sources and want to rebuild the
1240 catalog, configuring with @option{--enable-maintainer-mode} will enable
1241 this. Note that you need a recent version of the @code{gettext} tools
1244 @item --disable-bootstrap
1245 For a native build, the default configuration is to perform
1246 a 3-stage bootstrap of the compiler when @samp{make} is invoked,
1247 testing that GCC can compile itself correctly. If you want to disable
1248 this process, you can configure with @option{--disable-bootstrap}.
1250 @item --enable-bootstrap
1251 In special cases, you may want to perform a 3-stage build
1252 even if the target and host triplets are different.
1253 This could happen when the host can run code compiled for
1254 the target (e.g.@: host is i686-linux, target is i486-linux).
1255 Starting from GCC 4.2, to do this you have to configure explicitly
1256 with @option{--enable-bootstrap}.
1258 @item --enable-generated-files-in-srcdir
1259 Neither the .c and .h files that are generated from Bison and flex nor the
1260 info manuals and man pages that are built from the .texi files are present
1261 in the SVN development tree. When building GCC from that development tree,
1262 or from one of our snapshots, those generated files are placed in your
1263 build directory, which allows for the source to be in a readonly
1266 If you configure with @option{--enable-generated-files-in-srcdir} then those
1267 generated files will go into the source directory. This is mainly intended
1268 for generating release or prerelease tarballs of the GCC sources, since it
1269 is not a requirement that the users of source releases to have flex, Bison,
1272 @item --enable-version-specific-runtime-libs
1274 that runtime libraries should be installed in the compiler specific
1275 subdirectory (@file{@var{libdir}/gcc}) rather than the usual places. In
1276 addition, @samp{libstdc++}'s include files will be installed into
1277 @file{@var{libdir}} unless you overruled it by using
1278 @option{--with-gxx-include-dir=@var{dirname}}. Using this option is
1279 particularly useful if you intend to use several versions of GCC in
1280 parallel. This is currently supported by @samp{libgfortran},
1281 @samp{libjava}, @samp{libmudflap}, @samp{libstdc++}, and @samp{libobjc}.
1283 @item --enable-languages=@var{lang1},@var{lang2},@dots{}
1284 Specify that only a particular subset of compilers and
1285 their runtime libraries should be built. For a list of valid values for
1286 @var{langN} you can issue the following command in the
1287 @file{gcc} directory of your GCC source tree:@*
1289 grep language= */config-lang.in
1291 Currently, you can use any of the following:
1292 @code{all}, @code{ada}, @code{c}, @code{c++}, @code{fortran}, @code{java},
1293 @code{objc}, @code{obj-c++}.
1294 Building the Ada compiler has special requirements, see below.
1295 If you do not pass this flag, or specify the option @code{all}, then all
1296 default languages available in the @file{gcc} sub-tree will be configured.
1297 Ada and Objective-C++ are not default languages; the rest are.
1298 Re-defining @code{LANGUAGES} when calling @samp{make} @strong{does not}
1299 work anymore, as those language sub-directories might not have been
1302 @item --enable-stage1-languages=@var{lang1},@var{lang2},@dots{}
1303 Specify that a particular subset of compilers and their runtime
1304 libraries should be built with the system C compiler during stage 1 of
1305 the bootstrap process, rather than only in later stages with the
1306 bootstrapped C compiler. The list of valid values is the same as for
1307 @option{--enable-languages}, and the option @code{all} will select all
1308 of the languages enabled by @option{--enable-languages}. This option is
1309 primarily useful for GCC development; for instance, when a development
1310 version of the compiler cannot bootstrap due to compiler bugs, or when
1311 one is debugging front ends other than the C front end. When this
1312 option is used, one can then build the target libraries for the
1313 specified languages with the stage-1 compiler by using @command{make
1314 stage1-bubble all-target}, or run the testsuite on the stage-1 compiler
1315 for the specified languages using @command{make stage1-start check-gcc}.
1317 @item --disable-libada
1318 Specify that the run-time libraries and tools used by GNAT should not
1319 be built. This can be useful for debugging, or for compatibility with
1320 previous Ada build procedures, when it was required to explicitly
1321 do a @samp{make -C gcc gnatlib_and_tools}.
1323 @item --disable-libssp
1324 Specify that the run-time libraries for stack smashing protection
1325 should not be built.
1327 @item --disable-libgomp
1328 Specify that the run-time libraries used by GOMP should not be built.
1331 Specify that the compiler should
1332 use DWARF 2 debugging information as the default.
1334 @item --enable-targets=all
1335 @itemx --enable-targets=@var{target_list}
1336 Some GCC targets, e.g.@: powerpc64-linux, build bi-arch compilers.
1337 These are compilers that are able to generate either 64-bit or 32-bit
1338 code. Typically, the corresponding 32-bit target, e.g.@:
1339 powerpc-linux for powerpc64-linux, only generates 32-bit code. This
1340 option enables the 32-bit target to be a bi-arch compiler, which is
1341 useful when you want a bi-arch compiler that defaults to 32-bit, and
1342 you are building a bi-arch or multi-arch binutils in a combined tree.
1343 Currently, this option only affects sparc-linux, powerpc-linux and
1346 @item --enable-secureplt
1347 This option enables @option{-msecure-plt} by default for powerpc-linux.
1349 @xref{RS/6000 and PowerPC Options,, RS/6000 and PowerPC Options, gcc,
1350 Using the GNU Compiler Collection (GCC)},
1353 See ``RS/6000 and PowerPC Options'' in the main manual
1357 This option enables @option{-mcld} by default for 32-bit x86 targets.
1359 @xref{i386 and x86-64 Options,, i386 and x86-64 Options, gcc,
1360 Using the GNU Compiler Collection (GCC)},
1363 See ``i386 and x86-64 Options'' in the main manual
1366 @item --enable-win32-registry
1367 @itemx --enable-win32-registry=@var{key}
1368 @itemx --disable-win32-registry
1369 The @option{--enable-win32-registry} option enables Microsoft Windows-hosted GCC
1370 to look up installations paths in the registry using the following key:
1373 @code{HKEY_LOCAL_MACHINE\SOFTWARE\Free Software Foundation\@var{key}}
1376 @var{key} defaults to GCC version number, and can be overridden by the
1377 @option{--enable-win32-registry=@var{key}} option. Vendors and distributors
1378 who use custom installers are encouraged to provide a different key,
1379 perhaps one comprised of vendor name and GCC version number, to
1380 avoid conflict with existing installations. This feature is enabled
1381 by default, and can be disabled by @option{--disable-win32-registry}
1382 option. This option has no effect on the other hosts.
1385 Specify that the machine does not have a floating point unit. This
1386 option only applies to @samp{m68k-sun-sunos@var{n}}. On any other
1387 system, @option{--nfp} has no effect.
1389 @item --enable-werror
1390 @itemx --disable-werror
1391 @itemx --enable-werror=yes
1392 @itemx --enable-werror=no
1393 When you specify this option, it controls whether certain files in the
1394 compiler are built with @option{-Werror} in bootstrap stage2 and later.
1395 If you don't specify it, @option{-Werror} is turned on for the main
1396 development trunk. However it defaults to off for release branches and
1397 final releases. The specific files which get @option{-Werror} are
1398 controlled by the Makefiles.
1400 @item --enable-checking
1401 @itemx --enable-checking=@var{list}
1402 When you specify this option, the compiler is built to perform internal
1403 consistency checks of the requested complexity. This does not change the
1404 generated code, but adds error checking within the compiler. This will
1405 slow down the compiler and may only work properly if you are building
1406 the compiler with GCC@. This is @samp{yes} by default when building
1407 from SVN or snapshots, but @samp{release} for releases. The default
1408 for building the stage1 compiler is @samp{yes}. More control
1409 over the checks may be had by specifying @var{list}. The categories of
1410 checks available are @samp{yes} (most common checks
1411 @samp{assert,misc,tree,gc,rtlflag,runtime}), @samp{no} (no checks at
1412 all), @samp{all} (all but @samp{valgrind}), @samp{release} (cheapest
1413 checks @samp{assert,runtime}) or @samp{none} (same as @samp{no}).
1414 Individual checks can be enabled with these flags @samp{assert},
1415 @samp{df}, @samp{fold}, @samp{gc}, @samp{gcac} @samp{misc}, @samp{rtl},
1416 @samp{rtlflag}, @samp{runtime}, @samp{tree}, and @samp{valgrind}.
1418 The @samp{valgrind} check requires the external @command{valgrind}
1419 simulator, available from @uref{http://valgrind.org/}. The
1420 @samp{df}, @samp{rtl}, @samp{gcac} and @samp{valgrind} checks are very expensive.
1421 To disable all checking, @samp{--disable-checking} or
1422 @samp{--enable-checking=none} must be explicitly requested. Disabling
1423 assertions will make the compiler and runtime slightly faster but
1424 increase the risk of undetected internal errors causing wrong code to be
1427 @item --disable-stage1-checking
1428 @item --enable-stage1-checking
1429 @itemx --enable-stage1-checking=@var{list}
1430 If no @option{--enable-checking} option is specified the stage1
1431 compiler will be built with @samp{yes} checking enabled, otherwise
1432 the stage1 checking flags are the same as specified by
1433 @option{--enable-checking}. To build the stage1 compiler with
1434 different checking options use @option{--enable-stage1-checking}.
1435 The list of checking options is the same as for @option{--enable-checking}.
1436 If your system is too slow or too small to bootstrap a released compiler
1437 with checking for stage1 enabled, you can use @samp{--disable-stage1-checking}
1438 to disable checking for the stage1 compiler.
1440 @item --enable-coverage
1441 @itemx --enable-coverage=@var{level}
1442 With this option, the compiler is built to collect self coverage
1443 information, every time it is run. This is for internal development
1444 purposes, and only works when the compiler is being built with gcc. The
1445 @var{level} argument controls whether the compiler is built optimized or
1446 not, values are @samp{opt} and @samp{noopt}. For coverage analysis you
1447 want to disable optimization, for performance analysis you want to
1448 enable optimization. When coverage is enabled, the default level is
1449 without optimization.
1451 @item --enable-gather-detailed-mem-stats
1452 When this option is specified more detailed information on memory
1453 allocation is gathered. This information is printed when using
1454 @option{-fmem-report}.
1457 @itemx --with-gc=@var{choice}
1458 With this option you can specify the garbage collector implementation
1459 used during the compilation process. @var{choice} can be one of
1460 @samp{page} and @samp{zone}, where @samp{page} is the default.
1463 @itemx --disable-nls
1464 The @option{--enable-nls} option enables Native Language Support (NLS),
1465 which lets GCC output diagnostics in languages other than American
1466 English. Native Language Support is enabled by default if not doing a
1467 canadian cross build. The @option{--disable-nls} option disables NLS@.
1469 @item --with-included-gettext
1470 If NLS is enabled, the @option{--with-included-gettext} option causes the build
1471 procedure to prefer its copy of GNU @command{gettext}.
1473 @item --with-catgets
1474 If NLS is enabled, and if the host lacks @code{gettext} but has the
1475 inferior @code{catgets} interface, the GCC build procedure normally
1476 ignores @code{catgets} and instead uses GCC's copy of the GNU
1477 @code{gettext} library. The @option{--with-catgets} option causes the
1478 build procedure to use the host's @code{catgets} in this situation.
1480 @item --with-libiconv-prefix=@var{dir}
1481 Search for libiconv header files in @file{@var{dir}/include} and
1482 libiconv library files in @file{@var{dir}/lib}.
1484 @item --enable-obsolete
1485 Enable configuration for an obsoleted system. If you attempt to
1486 configure GCC for a system (build, host, or target) which has been
1487 obsoleted, and you do not specify this flag, configure will halt with an
1490 All support for systems which have been obsoleted in one release of GCC
1491 is removed entirely in the next major release, unless someone steps
1492 forward to maintain the port.
1494 @item --enable-decimal-float
1495 @itemx --enable-decimal-float=yes
1496 @itemx --enable-decimal-float=no
1497 @itemx --enable-decimal-float=bid
1498 @itemx --enable-decimal-float=dpd
1499 @itemx --disable-decimal-float
1500 Enable (or disable) support for the C decimal floating point extension
1501 that is in the IEEE 754-2008 standard. This is enabled by default only
1502 on PowerPC, i386, and x86_64 GNU/Linux systems. Other systems may also
1503 support it, but require the user to specifically enable it. You can
1504 optionally control which decimal floating point format is used (either
1505 @samp{bid} or @samp{dpd}). The @samp{bid} (binary integer decimal)
1506 format is default on i386 and x86_64 systems, and the @samp{dpd}
1507 (densely packed decimal) format is default on PowerPC systems.
1509 @item --enable-fixed-point
1510 @itemx --disable-fixed-point
1511 Enable (or disable) support for C fixed-point arithmetic.
1512 This option is enabled by default for some targets (such as MIPS) which
1513 have hardware-support for fixed-point operations. On other targets, you
1514 may enable this option manually.
1516 @item --with-long-double-128
1517 Specify if @code{long double} type should be 128-bit by default on selected
1518 GNU/Linux architectures. If using @code{--without-long-double-128},
1519 @code{long double} will be by default 64-bit, the same as @code{double} type.
1520 When neither of these configure options are used, the default will be
1521 128-bit @code{long double} when built against GNU C Library 2.4 and later,
1522 64-bit @code{long double} otherwise.
1524 @item --with-gmp=@var{pathname}
1525 @itemx --with-gmp-include=@var{pathname}
1526 @itemx --with-gmp-lib=@var{pathname}
1527 @itemx --with-mpfr=@var{pathname}
1528 @itemx --with-mpfr-include=@var{pathname}
1529 @itemx --with-mpfr-lib=@var{pathname}
1530 @itemx --with-mpc=@var{pathname}
1531 @itemx --with-mpc-include=@var{pathname}
1532 @itemx --with-mpc-lib=@var{pathname}
1533 If you do not have GMP (the GNU Multiple Precision library), the MPFR
1534 library and/or the MPC library installed in a standard location and
1535 you want to build GCC, you can explicitly specify the directory where
1536 they are installed (@samp{--with-gmp=@var{gmpinstalldir}},
1537 @samp{--with-mpfr=@var{mpfrinstalldir}},
1538 @samp{--with-mpc=@var{mpcinstalldir}}). The
1539 @option{--with-gmp=@var{gmpinstalldir}} option is shorthand for
1540 @option{--with-gmp-lib=@var{gmpinstalldir}/lib} and
1541 @option{--with-gmp-include=@var{gmpinstalldir}/include}. Likewise the
1542 @option{--with-mpfr=@var{mpfrinstalldir}} option is shorthand for
1543 @option{--with-mpfr-lib=@var{mpfrinstalldir}/lib} and
1544 @option{--with-mpfr-include=@var{mpfrinstalldir}/include}, also the
1545 @option{--with-mpc=@var{mpcinstalldir}} option is shorthand for
1546 @option{--with-mpc-lib=@var{mpcinstalldir}/lib} and
1547 @option{--with-mpc-include=@var{mpcinstalldir}/include}. If these
1548 shorthand assumptions are not correct, you can use the explicit
1549 include and lib options directly.
1551 @item --with-ppl=@var{pathname}
1552 @itemx --with-ppl-include=@var{pathname}
1553 @itemx --with-ppl-lib=@var{pathname}
1554 @itemx --with-cloog=@var{pathname}
1555 @itemx --with-cloog-include=@var{pathname}
1556 @itemx --with-cloog-lib=@var{pathname}
1557 If you do not have PPL (the Parma Polyhedra Library) and the CLooG
1558 libraries installed in a standard location and you want to build GCC,
1559 you can explicitly specify the directory where they are installed
1560 (@samp{--with-ppl=@var{pplinstalldir}},
1561 @samp{--with-cloog=@var{clooginstalldir}}). The
1562 @option{--with-ppl=@var{pplinstalldir}} option is shorthand for
1563 @option{--with-ppl-lib=@var{pplinstalldir}/lib} and
1564 @option{--with-ppl-include=@var{pplinstalldir}/include}. Likewise the
1565 @option{--with-cloog=@var{clooginstalldir}} option is shorthand for
1566 @option{--with-cloog-lib=@var{clooginstalldir}/lib} and
1567 @option{--with-cloog-include=@var{clooginstalldir}/include}. If these
1568 shorthand assumptions are not correct, you can use the explicit
1569 include and lib options directly.
1571 @item --with-host-libstdcxx=@var{linker-args}
1572 If you are linking with a static copy of PPL, you can use this option
1573 to specify how the linker should find the standard C++ library used
1574 internally by PPL. Typical values of @var{linker-args} might be
1575 @samp{-lstdc++} or @samp{-Wl,-Bstatic,-lstdc++,-Bdynamic -lm}. If you are
1576 linking with a shared copy of PPL, you probably do not need this
1577 option; shared library dependencies will cause the linker to search
1578 for the standard C++ library automatically.
1580 @item --with-stage1-ldflags=@var{flags}
1581 This option may be used to set linker flags to be used when linking
1582 stage 1 of GCC. These are also used when linking GCC if configured with
1583 @option{--disable-bootstrap}. By default no special flags are used.
1585 @item --with-stage1-libs=@var{libs}
1586 This option may be used to set libraries to be used when linking stage 1
1587 of GCC. These are also used when linking GCC if configured with
1588 @option{--disable-bootstrap}. The default is the argument to
1589 @option{--with-host-libstdcxx}, if specified.
1591 @item --with-boot-ldflags=@var{flags}
1592 This option may be used to set linker flags to be used when linking
1593 stage 2 and later when bootstrapping GCC. By default no special flags
1596 @item --with-boot-libs=@var{libs}
1597 This option may be used to set libraries to be used when linking stage 2
1598 and later when bootstrapping GCC. The default is the argument to
1599 @option{--with-host-libstdcxx}, if specified.
1601 @item --with-debug-prefix-map=@var{map}
1602 Convert source directory names using @option{-fdebug-prefix-map} when
1603 building runtime libraries. @samp{@var{map}} is a space-separated
1604 list of maps of the form @samp{@var{old}=@var{new}}.
1606 @item --enable-linker-build-id
1607 Tells GCC to pass @option{--build-id} option to the linker for all final
1608 links (links performed without the @option{-r} or @option{--relocatable}
1609 option), if the linker supports it. If you specify
1610 @option{--enable-linker-build-id}, but your linker does not
1611 support @option{--build-id} option, a warning is issued and the
1612 @option{--enable-linker-build-id} option is ignored. The default is off.
1614 @item --enable-gnu-unique-object
1615 @itemx --disable-gnu-unique-object
1616 Tells GCC to use the gnu_unique_object relocation for C++ template
1617 static data members and inline function local statics. Enabled by
1618 default for a native toolchain with an assembler that accepts it and
1619 GLIBC 2.11 or above, otherwise disabled.
1623 @subheading Cross-Compiler-Specific Options
1624 The following options only apply to building cross compilers.
1626 @item --with-sysroot
1627 @itemx --with-sysroot=@var{dir}
1628 Tells GCC to consider @var{dir} as the root of a tree that contains a
1629 (subset of) the root filesystem of the target operating system.
1630 Target system headers, libraries and run-time object files will be
1631 searched in there. The specified directory is not copied into the
1632 install tree, unlike the options @option{--with-headers} and
1633 @option{--with-libs} that this option obsoletes. The default value,
1634 in case @option{--with-sysroot} is not given an argument, is
1635 @option{$@{gcc_tooldir@}/sys-root}. If the specified directory is a
1636 subdirectory of @option{$@{exec_prefix@}}, then it will be found relative to
1637 the GCC binaries if the installation tree is moved.
1639 @item --with-build-sysroot
1640 @itemx --with-build-sysroot=@var{dir}
1641 Tells GCC to consider @var{dir} as the system root (see
1642 @option{--with-sysroot}) while building target libraries, instead of
1643 the directory specified with @option{--with-sysroot}. This option is
1644 only useful when you are already using @option{--with-sysroot}. You
1645 can use @option{--with-build-sysroot} when you are configuring with
1646 @option{--prefix} set to a directory that is different from the one in
1647 which you are installing GCC and your target libraries.
1649 This option affects the system root for the compiler used to build
1650 target libraries (which runs on the build system); it does not affect
1651 the compiler which is used to build GCC itself.
1653 @item --with-headers
1654 @itemx --with-headers=@var{dir}
1655 Deprecated in favor of @option{--with-sysroot}.
1656 Specifies that target headers are available when building a cross compiler.
1657 The @var{dir} argument specifies a directory which has the target include
1658 files. These include files will be copied into the @file{gcc} install
1659 directory. @emph{This option with the @var{dir} argument is required} when
1660 building a cross compiler, if @file{@var{prefix}/@var{target}/sys-include}
1661 doesn't pre-exist. If @file{@var{prefix}/@var{target}/sys-include} does
1662 pre-exist, the @var{dir} argument may be omitted. @command{fixincludes}
1663 will be run on these files to make them compatible with GCC@.
1665 @item --without-headers
1666 Tells GCC not use any target headers from a libc when building a cross
1667 compiler. When crossing to GNU/Linux, you need the headers so GCC
1668 can build the exception handling for libgcc.
1671 @itemx --with-libs=``@var{dir1} @var{dir2} @dots{} @var{dirN}''
1672 Deprecated in favor of @option{--with-sysroot}.
1673 Specifies a list of directories which contain the target runtime
1674 libraries. These libraries will be copied into the @file{gcc} install
1675 directory. If the directory list is omitted, this option has no
1679 Specifies that @samp{newlib} is
1680 being used as the target C library. This causes @code{__eprintf} to be
1681 omitted from @file{libgcc.a} on the assumption that it will be provided by
1684 @item --with-build-time-tools=@var{dir}
1685 Specifies where to find the set of target tools (assembler, linker, etc.)
1686 that will be used while building GCC itself. This option can be useful
1687 if the directory layouts are different between the system you are building
1688 GCC on, and the system where you will deploy it.
1690 For example, on an @samp{ia64-hp-hpux} system, you may have the GNU
1691 assembler and linker in @file{/usr/bin}, and the native tools in a
1692 different path, and build a toolchain that expects to find the
1693 native tools in @file{/usr/bin}.
1695 When you use this option, you should ensure that @var{dir} includes
1696 @command{ar}, @command{as}, @command{ld}, @command{nm},
1697 @command{ranlib} and @command{strip} if necessary, and possibly
1698 @command{objdump}. Otherwise, GCC may use an inconsistent set of
1702 @subheading Java-Specific Options
1704 The following option applies to the build of the Java front end.
1707 @item --disable-libgcj
1708 Specify that the run-time libraries
1709 used by GCJ should not be built. This is useful in case you intend
1710 to use GCJ with some other run-time, or you're going to install it
1711 separately, or it just happens not to build on your particular
1712 machine. In general, if the Java front end is enabled, the GCJ
1713 libraries will be enabled too, unless they're known to not work on
1714 the target platform. If GCJ is enabled but @samp{libgcj} isn't built, you
1715 may need to port it; in this case, before modifying the top-level
1716 @file{configure.in} so that @samp{libgcj} is enabled by default on this platform,
1717 you may use @option{--enable-libgcj} to override the default.
1721 The following options apply to building @samp{libgcj}.
1723 @subsubheading General Options
1726 @item --enable-java-maintainer-mode
1727 By default the @samp{libjava} build will not attempt to compile the
1728 @file{.java} source files to @file{.class}. Instead, it will use the
1729 @file{.class} files from the source tree. If you use this option you
1730 must have executables named @command{ecj1} and @command{gjavah} in your path
1731 for use by the build. You must use this option if you intend to
1732 modify any @file{.java} files in @file{libjava}.
1734 @item --with-java-home=@var{dirname}
1735 This @samp{libjava} option overrides the default value of the
1736 @samp{java.home} system property. It is also used to set
1737 @samp{sun.boot.class.path} to @file{@var{dirname}/lib/rt.jar}. By
1738 default @samp{java.home} is set to @file{@var{prefix}} and
1739 @samp{sun.boot.class.path} to
1740 @file{@var{datadir}/java/libgcj-@var{version}.jar}.
1742 @item --with-ecj-jar=@var{filename}
1743 This option can be used to specify the location of an external jar
1744 file containing the Eclipse Java compiler. A specially modified
1745 version of this compiler is used by @command{gcj} to parse
1746 @file{.java} source files. If this option is given, the
1747 @samp{libjava} build will create and install an @file{ecj1} executable
1748 which uses this jar file at runtime.
1750 If this option is not given, but an @file{ecj.jar} file is found in
1751 the topmost source tree at configure time, then the @samp{libgcj}
1752 build will create and install @file{ecj1}, and will also install the
1753 discovered @file{ecj.jar} into a suitable place in the install tree.
1755 If @file{ecj1} is not installed, then the user will have to supply one
1756 on his path in order for @command{gcj} to properly parse @file{.java}
1757 source files. A suitable jar is available from
1758 @uref{ftp://sourceware.org/pub/java/}.
1760 @item --disable-getenv-properties
1761 Don't set system properties from @env{GCJ_PROPERTIES}.
1763 @item --enable-hash-synchronization
1764 Use a global hash table for monitor locks. Ordinarily,
1765 @samp{libgcj}'s @samp{configure} script automatically makes
1766 the correct choice for this option for your platform. Only use
1767 this if you know you need the library to be configured differently.
1769 @item --enable-interpreter
1770 Enable the Java interpreter. The interpreter is automatically
1771 enabled by default on all platforms that support it. This option
1772 is really only useful if you want to disable the interpreter
1773 (using @option{--disable-interpreter}).
1775 @item --disable-java-net
1776 Disable java.net. This disables the native part of java.net only,
1777 using non-functional stubs for native method implementations.
1779 @item --disable-jvmpi
1780 Disable JVMPI support.
1782 @item --disable-libgcj-bc
1783 Disable BC ABI compilation of certain parts of libgcj. By default,
1784 some portions of libgcj are compiled with @option{-findirect-dispatch}
1785 and @option{-fno-indirect-classes}, allowing them to be overridden at
1788 If @option{--disable-libgcj-bc} is specified, libgcj is built without
1789 these options. This allows the compile-time linker to resolve
1790 dependencies when statically linking to libgcj. However it makes it
1791 impossible to override the affected portions of libgcj at run-time.
1793 @item --enable-reduced-reflection
1794 Build most of libgcj with @option{-freduced-reflection}. This reduces
1795 the size of libgcj at the expense of not being able to do accurate
1796 reflection on the classes it contains. This option is safe if you
1797 know that code using libgcj will never use reflection on the standard
1798 runtime classes in libgcj (including using serialization, RMI or CORBA).
1801 Enable runtime eCos target support.
1803 @item --without-libffi
1804 Don't use @samp{libffi}. This will disable the interpreter and JNI
1805 support as well, as these require @samp{libffi} to work.
1807 @item --enable-libgcj-debug
1808 Enable runtime debugging code.
1810 @item --enable-libgcj-multifile
1811 If specified, causes all @file{.java} source files to be
1812 compiled into @file{.class} files in one invocation of
1813 @samp{gcj}. This can speed up build time, but is more
1814 resource-intensive. If this option is unspecified or
1815 disabled, @samp{gcj} is invoked once for each @file{.java}
1816 file to compile into a @file{.class} file.
1818 @item --with-libiconv-prefix=DIR
1819 Search for libiconv in @file{DIR/include} and @file{DIR/lib}.
1821 @item --enable-sjlj-exceptions
1822 Force use of the @code{setjmp}/@code{longjmp}-based scheme for exceptions.
1823 @samp{configure} ordinarily picks the correct value based on the platform.
1824 Only use this option if you are sure you need a different setting.
1826 @item --with-system-zlib
1827 Use installed @samp{zlib} rather than that included with GCC@.
1829 @item --with-win32-nlsapi=ansi, unicows or unicode
1830 Indicates how MinGW @samp{libgcj} translates between UNICODE
1831 characters and the Win32 API@.
1833 @item --enable-java-home
1834 If enabled, this creates a JPackage compatible SDK environment during install.
1835 Note that if --enable-java-home is used, --with-arch-directory=ARCH must also
1838 @item --with-arch-directory=ARCH
1839 Specifies the name to use for the @file{jre/lib/ARCH} directory in the SDK
1840 environment created when --enable-java-home is passed. Typical names for this
1841 directory include i386, amd64, ia64, etc.
1843 @item --with-os-directory=DIR
1844 Specifies the OS directory for the SDK include directory. This is set to auto
1845 detect, and is typically 'linux'.
1847 @item --with-origin-name=NAME
1848 Specifies the JPackage origin name. This defaults to the 'gcj' in
1851 @item --with-arch-suffix=SUFFIX
1852 Specifies the suffix for the sdk directory. Defaults to the empty string.
1853 Examples include '.x86_64' in 'java-1.5.0-gcj-1.5.0.0.x86_64'.
1855 @item --with-jvm-root-dir=DIR
1856 Specifies where to install the SDK. Default is $(prefix)/lib/jvm.
1858 @item --with-jvm-jar-dir=DIR
1859 Specifies where to install jars. Default is $(prefix)/lib/jvm-exports.
1861 @item --with-python-dir=DIR
1862 Specifies where to install the Python modules used for aot-compile. DIR should
1863 not include the prefix used in installation. For example, if the Python modules
1864 are to be installed in /usr/lib/python2.5/site-packages, then
1865 --with-python-dir=/lib/python2.5/site-packages should be passed. If this is
1866 not specified, then the Python modules are installed in $(prefix)/share/python.
1868 @item --enable-aot-compile-rpm
1869 Adds aot-compile-rpm to the list of installed scripts.
1873 Use the single-byte @code{char} and the Win32 A functions natively,
1874 translating to and from UNICODE when using these functions. If
1875 unspecified, this is the default.
1878 Use the @code{WCHAR} and Win32 W functions natively. Adds
1879 @code{-lunicows} to @file{libgcj.spec} to link with @samp{libunicows}.
1880 @file{unicows.dll} needs to be deployed on Microsoft Windows 9X machines
1881 running built executables. @file{libunicows.a}, an open-source
1882 import library around Microsoft's @code{unicows.dll}, is obtained from
1883 @uref{http://libunicows.sourceforge.net/}, which also gives details
1884 on getting @file{unicows.dll} from Microsoft.
1887 Use the @code{WCHAR} and Win32 W functions natively. Does @emph{not}
1888 add @code{-lunicows} to @file{libgcj.spec}. The built executables will
1889 only run on Microsoft Windows NT and above.
1893 @subsubheading AWT-Specific Options
1897 Use the X Window System.
1899 @item --enable-java-awt=PEER(S)
1900 Specifies the AWT peer library or libraries to build alongside
1901 @samp{libgcj}. If this option is unspecified or disabled, AWT
1902 will be non-functional. Current valid values are @option{gtk} and
1903 @option{xlib}. Multiple libraries should be separated by a
1904 comma (i.e.@: @option{--enable-java-awt=gtk,xlib}).
1906 @item --enable-gtk-cairo
1907 Build the cairo Graphics2D implementation on GTK@.
1909 @item --enable-java-gc=TYPE
1910 Choose garbage collector. Defaults to @option{boehm} if unspecified.
1912 @item --disable-gtktest
1913 Do not try to compile and run a test GTK+ program.
1915 @item --disable-glibtest
1916 Do not try to compile and run a test GLIB program.
1918 @item --with-libart-prefix=PFX
1919 Prefix where libart is installed (optional).
1921 @item --with-libart-exec-prefix=PFX
1922 Exec prefix where libart is installed (optional).
1924 @item --disable-libarttest
1925 Do not try to compile and run a test libart program.
1934 @uref{./index.html,,Return to the GCC Installation page}
1938 @c ***Building****************************************************************
1940 @comment node-name, next, previous, up
1941 @node Building, Testing, Configuration, Installing GCC
1947 @cindex Installing GCC: Building
1949 Now that GCC is configured, you are ready to build the compiler and
1952 Some commands executed when making the compiler may fail (return a
1953 nonzero status) and be ignored by @command{make}. These failures, which
1954 are often due to files that were not found, are expected, and can safely
1957 It is normal to have compiler warnings when compiling certain files.
1958 Unless you are a GCC developer, you can generally ignore these warnings
1959 unless they cause compilation to fail. Developers should attempt to fix
1960 any warnings encountered, however they can temporarily continue past
1961 warnings-as-errors by specifying the configure flag
1962 @option{--disable-werror}.
1964 On certain old systems, defining certain environment variables such as
1965 @env{CC} can interfere with the functioning of @command{make}.
1967 If you encounter seemingly strange errors when trying to build the
1968 compiler in a directory other than the source directory, it could be
1969 because you have previously configured the compiler in the source
1970 directory. Make sure you have done all the necessary preparations.
1972 If you build GCC on a BSD system using a directory stored in an old System
1973 V file system, problems may occur in running @command{fixincludes} if the
1974 System V file system doesn't support symbolic links. These problems
1975 result in a failure to fix the declaration of @code{size_t} in
1976 @file{sys/types.h}. If you find that @code{size_t} is a signed type and
1977 that type mismatches occur, this could be the cause.
1979 The solution is not to use such a directory for building GCC@.
1981 Similarly, when building from SVN or snapshots, or if you modify
1982 @file{*.l} files, you need the Flex lexical analyzer generator
1983 installed. If you do not modify @file{*.l} files, releases contain
1984 the Flex-generated files and you do not need Flex installed to build
1985 them. There is still one Flex-based lexical analyzer (part of the
1986 build machinery, not of GCC itself) that is used even if you only
1987 build the C front end.
1989 When building from SVN or snapshots, or if you modify Texinfo
1990 documentation, you need version 4.7 or later of Texinfo installed if you
1991 want Info documentation to be regenerated. Releases contain Info
1992 documentation pre-built for the unmodified documentation in the release.
1994 @section Building a native compiler
1996 For a native build, the default configuration is to perform
1997 a 3-stage bootstrap of the compiler when @samp{make} is invoked.
1998 This will build the entire GCC system and ensure that it compiles
1999 itself correctly. It can be disabled with the @option{--disable-bootstrap}
2000 parameter to @samp{configure}, but bootstrapping is suggested because
2001 the compiler will be tested more completely and could also have
2004 The bootstrapping process will complete the following steps:
2008 Build tools necessary to build the compiler.
2011 Perform a 3-stage bootstrap of the compiler. This includes building
2012 three times the target tools for use by the compiler such as binutils
2013 (bfd, binutils, gas, gprof, ld, and opcodes) if they have been
2014 individually linked or moved into the top level GCC source tree before
2018 Perform a comparison test of the stage2 and stage3 compilers.
2021 Build runtime libraries using the stage3 compiler from the previous step.
2025 If you are short on disk space you might consider @samp{make
2026 bootstrap-lean} instead. The sequence of compilation is the
2027 same described above, but object files from the stage1 and
2028 stage2 of the 3-stage bootstrap of the compiler are deleted as
2029 soon as they are no longer needed.
2031 If you wish to use non-default GCC flags when compiling the stage2
2032 and stage3 compilers, set @code{BOOT_CFLAGS} on the command line when
2033 doing @samp{make}. For example, if you want to save additional space
2034 during the bootstrap and in the final installation as well, you can
2035 build the compiler binaries without debugging information as in the
2036 following example. This will save roughly 40% of disk space both for
2037 the bootstrap and the final installation. (Libraries will still contain
2038 debugging information.)
2041 make BOOT_CFLAGS='-O' bootstrap
2044 You can place non-default optimization flags into @code{BOOT_CFLAGS}; they
2045 are less well tested here than the default of @samp{-g -O2}, but should
2046 still work. In a few cases, you may find that you need to specify special
2047 flags such as @option{-msoft-float} here to complete the bootstrap; or,
2048 if the native compiler miscompiles the stage1 compiler, you may need
2049 to work around this, by choosing @code{BOOT_CFLAGS} to avoid the parts
2050 of the stage1 compiler that were miscompiled, or by using @samp{make
2051 bootstrap4} to increase the number of stages of bootstrap.
2053 @code{BOOT_CFLAGS} does not apply to bootstrapped target libraries.
2054 Since these are always compiled with the compiler currently being
2055 bootstrapped, you can use @code{CFLAGS_FOR_TARGET} to modify their
2056 compilation flags, as for non-bootstrapped target libraries.
2057 Again, if the native compiler miscompiles the stage1 compiler, you may
2058 need to work around this by avoiding non-working parts of the stage1
2059 compiler. Use @code{STAGE1_TFLAGS} to this end.
2061 If you used the flag @option{--enable-languages=@dots{}} to restrict
2062 the compilers to be built, only those you've actually enabled will be
2063 built. This will of course only build those runtime libraries, for
2064 which the particular compiler has been built. Please note,
2065 that re-defining @env{LANGUAGES} when calling @samp{make}
2066 @strong{does not} work anymore!
2068 If the comparison of stage2 and stage3 fails, this normally indicates
2069 that the stage2 compiler has compiled GCC incorrectly, and is therefore
2070 a potentially serious bug which you should investigate and report. (On
2071 a few systems, meaningful comparison of object files is impossible; they
2072 always appear ``different''. If you encounter this problem, you will
2073 need to disable comparison in the @file{Makefile}.)
2075 If you do not want to bootstrap your compiler, you can configure with
2076 @option{--disable-bootstrap}. In particular cases, you may want to
2077 bootstrap your compiler even if the target system is not the same as
2078 the one you are building on: for example, you could build a
2079 @code{powerpc-unknown-linux-gnu} toolchain on a
2080 @code{powerpc64-unknown-linux-gnu} host. In this case, pass
2081 @option{--enable-bootstrap} to the configure script.
2083 @code{BUILD_CONFIG} can be used to bring in additional customization to
2084 the build. It can be set to a whitespace-separated list of names. For
2085 each such @code{NAME}, top-level @file{config/@code{NAME}.mk} will be
2086 included by the top-level @file{Makefile}, bringing in any settings it
2087 contains. Some examples are:
2090 @item @samp{bootstrap-O1}
2091 Removes any @option{-O}-started option from @code{BOOT_CFLAGS}, and adds
2092 @option{-O1} to it. @samp{BUILD_CONFIG=bootstrap-O1} is equivalent to
2093 @samp{BOOT_CFLAGS='-g -O1'}.
2095 @item @samp{bootstrap-O3}
2096 Analogous to @code{bootstrap-O1}.
2098 @item @samp{bootstrap-debug}
2099 Verifies that the compiler generates the same executable code, whether
2100 or not it is asked to emit debug information. To this end, this option
2101 builds stage2 host programs without debug information, and uses
2102 @file{contrib/compare-debug} to compare them with the stripped stage3
2103 object files. If @code{BOOT_CFLAGS} is overridden so as to not enable
2104 debug information, stage2 will have it, and stage3 won't. This option
2105 is enabled by default when GCC bootstrapping is enabled: in addition to
2106 better test coverage, it makes default bootstraps faster and leaner.
2108 @item @samp{bootstrap-debug-big}
2109 In addition to the checking performed by @code{bootstrap-debug}, this
2110 option saves internal compiler dumps during stage2 and stage3 and
2111 compares them as well, which helps catch additional potential problems,
2112 but at a great cost in terms of disk space.
2114 @item @samp{bootstrap-debug-lean}
2115 This option saves disk space compared with @code{bootstrap-debug-big},
2116 but at the expense of some recompilation. Instead of saving the dumps
2117 of stage2 and stage3 until the final compare, it uses
2118 @option{-fcompare-debug} to generate, compare and remove the dumps
2119 during stage3, repeating the compilation that already took place in
2120 stage2, whose dumps were not saved.
2122 @item @samp{bootstrap-debug-lib}
2123 This option tests executable code invariance over debug information
2124 generation on target libraries, just like @code{bootstrap-debug-lean}
2125 tests it on host programs. It builds stage3 libraries with
2126 @option{-fcompare-debug}, and it can be used along with any of the
2127 @code{bootstrap-debug} options above.
2129 There aren't @code{-lean} or @code{-big} counterparts to this option
2130 because most libraries are only build in stage3, so bootstrap compares
2131 would not get significant coverage. Moreover, the few libraries built
2132 in stage2 are used in stage3 host programs, so we wouldn't want to
2133 compile stage2 libraries with different options for comparison purposes.
2135 @item @samp{bootstrap-debug-ckovw}
2136 Arranges for error messages to be issued if the compiler built on any
2137 stage is run without the option @option{-fcompare-debug}. This is
2138 useful to verify the full @option{-fcompare-debug} testing coverage. It
2139 must be used along with @code{bootstrap-debug-lean} and
2140 @code{bootstrap-debug-lib}.
2142 @item @samp{bootstrap-time}
2143 Arranges for the run time of each program started by the GCC driver,
2144 built in any stage, to be logged to @file{time.log}, in the top level of
2149 @section Building a cross compiler
2151 When building a cross compiler, it is not generally possible to do a
2152 3-stage bootstrap of the compiler. This makes for an interesting problem
2153 as parts of GCC can only be built with GCC@.
2155 To build a cross compiler, we first recommend building and installing a
2156 native compiler. You can then use the native GCC compiler to build the
2157 cross compiler. The installed native compiler needs to be GCC version
2160 If the cross compiler is to be built with support for the Java
2161 programming language and the ability to compile .java source files is
2162 desired, the installed native compiler used to build the cross
2163 compiler needs to be the same GCC version as the cross compiler. In
2164 addition the cross compiler needs to be configured with
2165 @option{--with-ecj-jar=@dots{}}.
2167 Assuming you have already installed a native copy of GCC and configured
2168 your cross compiler, issue the command @command{make}, which performs the
2173 Build host tools necessary to build the compiler.
2176 Build target tools for use by the compiler such as binutils (bfd,
2177 binutils, gas, gprof, ld, and opcodes)
2178 if they have been individually linked or moved into the top level GCC source
2179 tree before configuring.
2182 Build the compiler (single stage only).
2185 Build runtime libraries using the compiler from the previous step.
2188 Note that if an error occurs in any step the make process will exit.
2190 If you are not building GNU binutils in the same source tree as GCC,
2191 you will need a cross-assembler and cross-linker installed before
2192 configuring GCC@. Put them in the directory
2193 @file{@var{prefix}/@var{target}/bin}. Here is a table of the tools
2194 you should put in this directory:
2198 This should be the cross-assembler.
2201 This should be the cross-linker.
2204 This should be the cross-archiver: a program which can manipulate
2205 archive files (linker libraries) in the target machine's format.
2208 This should be a program to construct a symbol table in an archive file.
2211 The installation of GCC will find these programs in that directory,
2212 and copy or link them to the proper place to for the cross-compiler to
2213 find them when run later.
2215 The easiest way to provide these files is to build the Binutils package.
2216 Configure it with the same @option{--host} and @option{--target}
2217 options that you use for configuring GCC, then build and install
2218 them. They install their executables automatically into the proper
2219 directory. Alas, they do not support all the targets that GCC
2222 If you are not building a C library in the same source tree as GCC,
2223 you should also provide the target libraries and headers before
2224 configuring GCC, specifying the directories with
2225 @option{--with-sysroot} or @option{--with-headers} and
2226 @option{--with-libs}. Many targets also require ``start files'' such
2227 as @file{crt0.o} and
2228 @file{crtn.o} which are linked into each executable. There may be several
2229 alternatives for @file{crt0.o}, for use with profiling or other
2230 compilation options. Check your target's definition of
2231 @code{STARTFILE_SPEC} to find out what start files it uses.
2233 @section Building in parallel
2235 GNU Make 3.79 and above, which is necessary to build GCC, support
2236 building in parallel. To activate this, you can use @samp{make -j 2}
2237 instead of @samp{make}. You can also specify a bigger number, and
2238 in most cases using a value greater than the number of processors in
2239 your machine will result in fewer and shorter I/O latency hits, thus
2240 improving overall throughput; this is especially true for slow drives
2241 and network filesystems.
2243 @section Building the Ada compiler
2245 In order to build GNAT, the Ada compiler, you need a working GNAT
2246 compiler (GCC version 3.4 or later).
2247 This includes GNAT tools such as @command{gnatmake} and
2248 @command{gnatlink}, since the Ada front end is written in Ada and
2249 uses some GNAT-specific extensions.
2251 In order to build a cross compiler, it is suggested to install
2252 the new compiler as native first, and then use it to build the cross
2255 @command{configure} does not test whether the GNAT installation works
2256 and has a sufficiently recent version; if too old a GNAT version is
2257 installed, the build will fail unless @option{--enable-languages} is
2258 used to disable building the Ada front end.
2260 @env{ADA_INCLUDE_PATH} and @env{ADA_OBJECT_PATH} environment variables
2261 must not be set when building the Ada compiler, the Ada tools, or the
2262 Ada runtime libraries. You can check that your build environment is clean
2263 by verifying that @samp{gnatls -v} lists only one explicit path in each
2266 @section Building with profile feedback
2268 It is possible to use profile feedback to optimize the compiler itself. This
2269 should result in a faster compiler binary. Experiments done on x86 using gcc
2270 3.3 showed approximately 7 percent speedup on compiling C programs. To
2271 bootstrap the compiler with profile feedback, use @code{make profiledbootstrap}.
2273 When @samp{make profiledbootstrap} is run, it will first build a @code{stage1}
2274 compiler. This compiler is used to build a @code{stageprofile} compiler
2275 instrumented to collect execution counts of instruction and branch
2276 probabilities. Then runtime libraries are compiled with profile collected.
2277 Finally a @code{stagefeedback} compiler is built using the information collected.
2279 Unlike standard bootstrap, several additional restrictions apply. The
2280 compiler used to build @code{stage1} needs to support a 64-bit integral type.
2281 It is recommended to only use GCC for this. Also parallel make is currently
2282 not supported since collisions in profile collecting may occur.
2289 @uref{./index.html,,Return to the GCC Installation page}
2293 @c ***Testing*****************************************************************
2295 @comment node-name, next, previous, up
2296 @node Testing, Final install, Building, Installing GCC
2300 @chapter Installing GCC: Testing
2303 @cindex Installing GCC: Testing
2306 Before you install GCC, we encourage you to run the testsuites and to
2307 compare your results with results from a similar configuration that have
2308 been submitted to the
2309 @uref{http://gcc.gnu.org/ml/gcc-testresults/,,gcc-testresults mailing list}.
2310 Some of these archived results are linked from the build status lists
2311 at @uref{http://gcc.gnu.org/buildstat.html}, although not everyone who
2312 reports a successful build runs the testsuites and submits the results.
2313 This step is optional and may require you to download additional software,
2314 but it can give you confidence in your new GCC installation or point out
2315 problems before you install and start using your new GCC@.
2317 First, you must have @uref{download.html,,downloaded the testsuites}.
2318 These are part of the full distribution, but if you downloaded the
2319 ``core'' compiler plus any front ends, you must download the testsuites
2322 Second, you must have the testing tools installed. This includes
2323 @uref{http://www.gnu.org/software/dejagnu/,,DejaGnu}, Tcl, and Expect;
2324 the DejaGnu site has links to these.
2326 If the directories where @command{runtest} and @command{expect} were
2327 installed are not in the @env{PATH}, you may need to set the following
2328 environment variables appropriately, as in the following example (which
2329 assumes that DejaGnu has been installed under @file{/usr/local}):
2332 TCL_LIBRARY = /usr/local/share/tcl8.0
2333 DEJAGNULIBS = /usr/local/share/dejagnu
2336 (On systems such as Cygwin, these paths are required to be actual
2337 paths, not mounts or links; presumably this is due to some lack of
2338 portability in the DejaGnu code.)
2341 Finally, you can run the testsuite (which may take a long time):
2343 cd @var{objdir}; make -k check
2346 This will test various components of GCC, such as compiler
2347 front ends and runtime libraries. While running the testsuite, DejaGnu
2348 might emit some harmless messages resembling
2349 @samp{WARNING: Couldn't find the global config file.} or
2350 @samp{WARNING: Couldn't find tool init file} that can be ignored.
2352 If you are testing a cross-compiler, you may want to run the testsuite
2353 on a simulator as described at @uref{http://gcc.gnu.org/simtest-howto.html}.
2355 @section How can you run the testsuite on selected tests?
2357 In order to run sets of tests selectively, there are targets
2358 @samp{make check-gcc} and @samp{make check-g++}
2359 in the @file{gcc} subdirectory of the object directory. You can also
2360 just run @samp{make check} in a subdirectory of the object directory.
2363 A more selective way to just run all @command{gcc} execute tests in the
2367 make check-gcc RUNTESTFLAGS="execute.exp @var{other-options}"
2370 Likewise, in order to run only the @command{g++} ``old-deja'' tests in
2371 the testsuite with filenames matching @samp{9805*}, you would use
2374 make check-g++ RUNTESTFLAGS="old-deja.exp=9805* @var{other-options}"
2377 The @file{*.exp} files are located in the testsuite directories of the GCC
2378 source, the most important ones being @file{compile.exp},
2379 @file{execute.exp}, @file{dg.exp} and @file{old-deja.exp}.
2380 To get a list of the possible @file{*.exp} files, pipe the
2381 output of @samp{make check} into a file and look at the
2382 @samp{Running @dots{} .exp} lines.
2384 @section Passing options and running multiple testsuites
2386 You can pass multiple options to the testsuite using the
2387 @samp{--target_board} option of DejaGNU, either passed as part of
2388 @samp{RUNTESTFLAGS}, or directly to @command{runtest} if you prefer to
2389 work outside the makefiles. For example,
2392 make check-g++ RUNTESTFLAGS="--target_board=unix/-O3/-fmerge-constants"
2395 will run the standard @command{g++} testsuites (``unix'' is the target name
2396 for a standard native testsuite situation), passing
2397 @samp{-O3 -fmerge-constants} to the compiler on every test, i.e.,
2398 slashes separate options.
2400 You can run the testsuites multiple times using combinations of options
2401 with a syntax similar to the brace expansion of popular shells:
2404 @dots{}"--target_board=arm-sim\@{-mhard-float,-msoft-float\@}\@{-O1,-O2,-O3,\@}"
2407 (Note the empty option caused by the trailing comma in the final group.)
2408 The following will run each testsuite eight times using the @samp{arm-sim}
2409 target, as if you had specified all possible combinations yourself:
2412 --target_board=arm-sim/-mhard-float/-O1
2413 --target_board=arm-sim/-mhard-float/-O2
2414 --target_board=arm-sim/-mhard-float/-O3
2415 --target_board=arm-sim/-mhard-float
2416 --target_board=arm-sim/-msoft-float/-O1
2417 --target_board=arm-sim/-msoft-float/-O2
2418 --target_board=arm-sim/-msoft-float/-O3
2419 --target_board=arm-sim/-msoft-float
2422 They can be combined as many times as you wish, in arbitrary ways. This
2426 @dots{}"--target_board=unix/-Wextra\@{-O3,-fno-strength\@}\@{-fomit-frame,\@}"
2429 will generate four combinations, all involving @samp{-Wextra}.
2431 The disadvantage to this method is that the testsuites are run in serial,
2432 which is a waste on multiprocessor systems. For users with GNU Make and
2433 a shell which performs brace expansion, you can run the testsuites in
2434 parallel by having the shell perform the combinations and @command{make}
2435 do the parallel runs. Instead of using @samp{--target_board}, use a
2436 special makefile target:
2439 make -j@var{N} check-@var{testsuite}//@var{test-target}/@var{option1}/@var{option2}/@dots{}
2445 make -j3 check-gcc//sh-hms-sim/@{-m1,-m2,-m3,-m3e,-m4@}/@{,-nofpu@}
2448 will run three concurrent ``make-gcc'' testsuites, eventually testing all
2449 ten combinations as described above. Note that this is currently only
2450 supported in the @file{gcc} subdirectory. (To see how this works, try
2451 typing @command{echo} before the example given here.)
2454 @section Additional testing for Java Class Libraries
2456 The Java runtime tests can be executed via @samp{make check}
2457 in the @file{@var{target}/libjava/testsuite} directory in
2460 The @uref{http://sourceware.org/mauve/,,Mauve Project} provides
2461 a suite of tests for the Java Class Libraries. This suite can be run
2462 as part of libgcj testing by placing the Mauve tree within the libjava
2463 testsuite at @file{libjava/testsuite/libjava.mauve/mauve}, or by
2464 specifying the location of that tree when invoking @samp{make}, as in
2465 @samp{make MAUVEDIR=~/mauve check}.
2467 @section How to interpret test results
2469 The result of running the testsuite are various @file{*.sum} and @file{*.log}
2470 files in the testsuite subdirectories. The @file{*.log} files contain a
2471 detailed log of the compiler invocations and the corresponding
2472 results, the @file{*.sum} files summarize the results. These summaries
2473 contain status codes for all tests:
2477 PASS: the test passed as expected
2479 XPASS: the test unexpectedly passed
2481 FAIL: the test unexpectedly failed
2483 XFAIL: the test failed as expected
2485 UNSUPPORTED: the test is not supported on this platform
2487 ERROR: the testsuite detected an error
2489 WARNING: the testsuite detected a possible problem
2492 It is normal for some tests to report unexpected failures. At the
2493 current time the testing harness does not allow fine grained control
2494 over whether or not a test is expected to fail. This problem should
2495 be fixed in future releases.
2498 @section Submitting test results
2500 If you want to report the results to the GCC project, use the
2501 @file{contrib/test_summary} shell script. Start it in the @var{objdir} with
2504 @var{srcdir}/contrib/test_summary -p your_commentary.txt \
2505 -m gcc-testresults@@gcc.gnu.org |sh
2508 This script uses the @command{Mail} program to send the results, so
2509 make sure it is in your @env{PATH}. The file @file{your_commentary.txt} is
2510 prepended to the testsuite summary and should contain any special
2511 remarks you have on your results or your build environment. Please
2512 do not edit the testsuite result block or the subject line, as these
2513 messages may be automatically processed.
2520 @uref{./index.html,,Return to the GCC Installation page}
2524 @c ***Final install***********************************************************
2526 @comment node-name, next, previous, up
2527 @node Final install, , Testing, Installing GCC
2529 @ifset finalinstallhtml
2531 @chapter Installing GCC: Final installation
2534 Now that GCC has been built (and optionally tested), you can install it with
2536 cd @var{objdir}; make install
2539 We strongly recommend to install into a target directory where there is
2540 no previous version of GCC present. Also, the GNAT runtime should not
2541 be stripped, as this would break certain features of the debugger that
2542 depend on this debugging information (catching Ada exceptions for
2545 That step completes the installation of GCC; user level binaries can
2546 be found in @file{@var{prefix}/bin} where @var{prefix} is the value
2547 you specified with the @option{--prefix} to configure (or
2548 @file{/usr/local} by default). (If you specified @option{--bindir},
2549 that directory will be used instead; otherwise, if you specified
2550 @option{--exec-prefix}, @file{@var{exec-prefix}/bin} will be used.)
2551 Headers for the C++ and Java libraries are installed in
2552 @file{@var{prefix}/include}; libraries in @file{@var{libdir}}
2553 (normally @file{@var{prefix}/lib}); internal parts of the compiler in
2554 @file{@var{libdir}/gcc} and @file{@var{libexecdir}/gcc}; documentation
2555 in info format in @file{@var{infodir}} (normally
2556 @file{@var{prefix}/info}).
2558 When installing cross-compilers, GCC's executables
2559 are not only installed into @file{@var{bindir}}, that
2560 is, @file{@var{exec-prefix}/bin}, but additionally into
2561 @file{@var{exec-prefix}/@var{target-alias}/bin}, if that directory
2562 exists. Typically, such @dfn{tooldirs} hold target-specific
2563 binutils, including assembler and linker.
2565 Installation into a temporary staging area or into a @command{chroot}
2566 jail can be achieved with the command
2569 make DESTDIR=@var{path-to-rootdir} install
2572 @noindent where @var{path-to-rootdir} is the absolute path of
2573 a directory relative to which all installation paths will be
2574 interpreted. Note that the directory specified by @code{DESTDIR}
2575 need not exist yet; it will be created if necessary.
2577 There is a subtle point with tooldirs and @code{DESTDIR}:
2578 If you relocate a cross-compiler installation with
2579 e.g.@: @samp{DESTDIR=@var{rootdir}}, then the directory
2580 @file{@var{rootdir}/@var{exec-prefix}/@var{target-alias}/bin} will
2581 be filled with duplicated GCC executables only if it already exists,
2582 it will not be created otherwise. This is regarded as a feature,
2583 not as a bug, because it gives slightly more control to the packagers
2584 using the @code{DESTDIR} feature.
2586 If you are bootstrapping a released version of GCC then please
2587 quickly review the build status page for your release, available from
2588 @uref{http://gcc.gnu.org/buildstat.html}.
2589 If your system is not listed for the version of GCC that you built,
2591 @email{gcc@@gcc.gnu.org} indicating
2592 that you successfully built and installed GCC@.
2593 Include the following information:
2597 Output from running @file{@var{srcdir}/config.guess}. Do not send
2598 that file itself, just the one-line output from running it.
2601 The output of @samp{gcc -v} for your newly installed @command{gcc}.
2602 This tells us which version of GCC you built and the options you passed to
2606 Whether you enabled all languages or a subset of them. If you used a
2607 full distribution then this information is part of the configure
2608 options in the output of @samp{gcc -v}, but if you downloaded the
2609 ``core'' compiler plus additional front ends then it isn't apparent
2610 which ones you built unless you tell us about it.
2613 If the build was for GNU/Linux, also include:
2616 The distribution name and version (e.g., Red Hat 7.1 or Debian 2.2.3);
2617 this information should be available from @file{/etc/issue}.
2620 The version of the Linux kernel, available from @samp{uname --version}
2624 The version of glibc you used; for RPM-based systems like Red Hat,
2625 Mandrake, and SuSE type @samp{rpm -q glibc} to get the glibc version,
2626 and on systems like Debian and Progeny use @samp{dpkg -l libc6}.
2628 For other systems, you can include similar information if you think it is
2632 Any other information that you think would be useful to people building
2633 GCC on the same configuration. The new entry in the build status list
2634 will include a link to the archived copy of your message.
2637 We'd also like to know if the
2639 @ref{Specific, host/target specific installation notes}
2642 @uref{specific.html,,host/target specific installation notes}
2644 didn't include your host/target information or if that information is
2645 incomplete or out of date. Send a note to
2646 @email{gcc@@gcc.gnu.org} detailing how the information should be changed.
2648 If you find a bug, please report it following the
2649 @uref{../bugs.html,,bug reporting guidelines}.
2651 If you want to print the GCC manuals, do @samp{cd @var{objdir}; make
2652 dvi}. You will need to have @command{texi2dvi} (version at least 4.7)
2653 and @TeX{} installed. This creates a number of @file{.dvi} files in
2654 subdirectories of @file{@var{objdir}}; these may be converted for
2655 printing with programs such as @command{dvips}. Alternately, by using
2656 @samp{make pdf} in place of @samp{make dvi}, you can create documentation
2657 in the form of @file{.pdf} files; this requires @command{texi2pdf}, which
2658 is included with Texinfo version 4.8 and later. You can also
2659 @uref{http://shop.fsf.org/,,buy printed manuals from the
2660 Free Software Foundation}, though such manuals may not be for the most
2661 recent version of GCC@.
2663 If you would like to generate online HTML documentation, do @samp{cd
2664 @var{objdir}; make html} and HTML will be generated for the gcc manuals in
2665 @file{@var{objdir}/gcc/HTML}.
2672 @uref{./index.html,,Return to the GCC Installation page}
2676 @c ***Binaries****************************************************************
2678 @comment node-name, next, previous, up
2679 @node Binaries, Specific, Installing GCC, Top
2683 @chapter Installing GCC: Binaries
2686 @cindex Installing GCC: Binaries
2688 We are often asked about pre-compiled versions of GCC@. While we cannot
2689 provide these for all platforms, below you'll find links to binaries for
2690 various platforms where creating them by yourself is not easy due to various
2693 Please note that we did not create these binaries, nor do we
2694 support them. If you have any problems installing them, please
2695 contact their makers.
2702 @uref{http://www.bullfreeware.com,,Bull's Freeware and Shareware Archive for AIX};
2705 @uref{http://pware.hvcc.edu,,Hudson Valley Community College Open Source Software for IBM System p};
2708 @uref{http://www.perzl.org/aix,,AIX 5L and 6 Open Source Packages}.
2712 DOS---@uref{http://www.delorie.com/djgpp/,,DJGPP}.
2715 Renesas H8/300[HS]---@uref{http://h8300-hms.sourceforge.net/,,GNU
2716 Development Tools for the Renesas H8/300[HS] Series}.
2722 @uref{http://hpux.cs.utah.edu/,,HP-UX Porting Center};
2725 @uref{ftp://sunsite.informatik.rwth-aachen.de/pub/packages/gcc_hpux/,,Binaries for HP-UX 11.00 at Aachen University of Technology}.
2729 Motorola 68HC11/68HC12---@uref{http://www.gnu-m68hc11.org,,GNU
2730 Development Tools for the Motorola 68HC11/68HC12}.
2733 @uref{http://www.sco.com/skunkware/devtools/index.html#gcc,,SCO
2734 OpenServer/Unixware}.
2737 Solaris 2 (SPARC, Intel)---@uref{http://www.sunfreeware.com/,,Sunfreeware}.
2740 SGI---@uref{http://freeware.sgi.com/,,SGI Freeware}.
2746 The @uref{http://sourceware.org/cygwin/,,Cygwin} project;
2748 The @uref{http://www.mingw.org/,,MinGW} project.
2752 @uref{ftp://ftp.thewrittenword.com/packages/by-name/,,The
2753 Written Word} offers binaries for
2754 AIX 4.3.3, 5.1 and 5.2,
2756 Tru64 UNIX 4.0D and 5.1,
2758 HP-UX 10.20, 11.00, and 11.11, and
2759 Solaris/SPARC 2.5.1, 2.6, 7, 8, 9 and 10.
2762 @uref{http://www.openpkg.org/,,OpenPKG} offers binaries for quite a
2763 number of platforms.
2766 The @uref{http://gcc.gnu.org/wiki/GFortranBinaries,,GFortran Wiki} has
2767 links to GNU Fortran binaries for several platforms.
2770 In addition to those specific offerings, you can get a binary
2771 distribution CD-ROM from the
2772 @uref{http://www.gnu.org/order/order.html,,Free Software Foundation}.
2773 It contains binaries for a number of platforms, and
2774 includes not only GCC, but other stuff as well. The current CD does
2775 not contain the latest version of GCC, but it should allow
2776 bootstrapping the compiler. An updated version of that disk is in the
2784 @uref{./index.html,,Return to the GCC Installation page}
2788 @c ***Specific****************************************************************
2790 @comment node-name, next, previous, up
2791 @node Specific, Old, Binaries, Top
2795 @chapter Host/target specific installation notes for GCC
2798 @cindex Specific installation notes
2799 @cindex Target specific installation
2800 @cindex Host specific installation
2801 @cindex Target specific installation notes
2803 Please read this document carefully @emph{before} installing the
2804 GNU Compiler Collection on your machine.
2806 Note that this list of install notes is @emph{not} a list of supported
2807 hosts or targets. Not all supported hosts and targets are listed
2808 here, only the ones that require host-specific or target-specific
2814 @uref{#alpha-x-x,,alpha*-*-*}
2816 @uref{#alpha-dec-osf,,alpha*-dec-osf*}
2818 @uref{#arc-x-elf,,arc-*-elf}
2820 @uref{#arm-x-elf,,arm-*-elf}
2824 @uref{#bfin,,Blackfin}
2828 @uref{#x-x-freebsd,,*-*-freebsd*}
2830 @uref{#h8300-hms,,h8300-hms}
2832 @uref{#hppa-hp-hpux,,hppa*-hp-hpux*}
2834 @uref{#hppa-hp-hpux10,,hppa*-hp-hpux10}
2836 @uref{#hppa-hp-hpux11,,hppa*-hp-hpux11}
2838 @uref{#x-x-linux-gnu,,*-*-linux-gnu}
2840 @uref{#ix86-x-linux,,i?86-*-linux*}
2842 @uref{#ix86-x-solaris210,,i?86-*-solaris2.10}
2844 @uref{#ia64-x-linux,,ia64-*-linux}
2846 @uref{#ia64-x-hpux,,ia64-*-hpux*}
2848 @uref{#x-ibm-aix,,*-ibm-aix*}
2850 @uref{#iq2000-x-elf,,iq2000-*-elf}
2852 @uref{#m32c-x-elf,,m32c-*-elf}
2854 @uref{#m32r-x-elf,,m32r-*-elf}
2856 @uref{#m6811-elf,,m6811-elf}
2858 @uref{#m6812-elf,,m6812-elf}
2860 @uref{#m68k-x-x,,m68k-*-*}
2862 @uref{#m68k-uclinux,,m68k-uclinux}
2864 @uref{#mep-x-elf,,mep-*-elf}
2866 @uref{#mips-x-x,,mips-*-*}
2868 @uref{#mips-sgi-irix5,,mips-sgi-irix5}
2870 @uref{#mips-sgi-irix6,,mips-sgi-irix6}
2872 @uref{#powerpc-x-x,,powerpc*-*-*}
2874 @uref{#powerpc-x-darwin,,powerpc-*-darwin*}
2876 @uref{#powerpc-x-elf,,powerpc-*-elf}
2878 @uref{#powerpc-x-linux-gnu,,powerpc*-*-linux-gnu*}
2880 @uref{#powerpc-x-netbsd,,powerpc-*-netbsd*}
2882 @uref{#powerpc-x-eabisim,,powerpc-*-eabisim}
2884 @uref{#powerpc-x-eabi,,powerpc-*-eabi}
2886 @uref{#powerpcle-x-elf,,powerpcle-*-elf}
2888 @uref{#powerpcle-x-eabisim,,powerpcle-*-eabisim}
2890 @uref{#powerpcle-x-eabi,,powerpcle-*-eabi}
2892 @uref{#s390-x-linux,,s390-*-linux*}
2894 @uref{#s390x-x-linux,,s390x-*-linux*}
2896 @uref{#s390x-ibm-tpf,,s390x-ibm-tpf*}
2898 @uref{#x-x-solaris2,,*-*-solaris2*}
2900 @uref{#sparc-sun-solaris2,,sparc-sun-solaris2*}
2902 @uref{#sparc-sun-solaris27,,sparc-sun-solaris2.7}
2904 @uref{#sparc-x-linux,,sparc-*-linux*}
2906 @uref{#sparc64-x-solaris2,,sparc64-*-solaris2*}
2908 @uref{#sparcv9-x-solaris2,,sparcv9-*-solaris2*}
2910 @uref{#x-x-vxworks,,*-*-vxworks*}
2912 @uref{#x86-64-x-x,,x86_64-*-*, amd64-*-*}
2914 @uref{#xtensa-x-elf,,xtensa*-*-elf}
2916 @uref{#xtensa-x-linux,,xtensa*-*-linux*}
2918 @uref{#windows,,Microsoft Windows}
2920 @uref{#x-x-cygwin,,*-*-cygwin}
2922 @uref{#x-x-interix,,*-*-interix}
2924 @uref{#x-x-mingw32,,*-*-mingw32}
2928 @uref{#older,,Older systems}
2933 @uref{#elf,,all ELF targets} (SVR4, Solaris 2, etc.)
2939 <!-- -------- host/target specific issues start here ---------------- -->
2942 @heading @anchor{alpha-x-x}alpha*-*-*
2944 This section contains general configuration information for all
2945 alpha-based platforms using ELF (in particular, ignore this section for
2946 DEC OSF/1, Digital UNIX and Tru64 UNIX)@. In addition to reading this
2947 section, please read all other sections that match your target.
2949 We require binutils 2.11.2 or newer.
2950 Previous binutils releases had a number of problems with DWARF 2
2951 debugging information, not the least of which is incorrect linking of
2957 @heading @anchor{alpha-dec-osf}alpha*-dec-osf*
2958 Systems using processors that implement the DEC Alpha architecture and
2959 are running the DEC/Compaq Unix (DEC OSF/1, Digital UNIX, or Compaq
2960 Tru64 UNIX) operating system, for example the DEC Alpha AXP systems.
2962 As of GCC 3.2, versions before @code{alpha*-dec-osf4} are no longer
2963 supported. (These are the versions which identify themselves as DEC
2966 In Digital Unix V4.0, virtual memory exhausted bootstrap failures
2967 may be fixed by configuring with @option{--with-gc=simple},
2968 reconfiguring Kernel Virtual Memory and Swap parameters
2969 per the @command{/usr/sbin/sys_check} Tuning Suggestions,
2970 or applying the patch in
2971 @uref{http://gcc.gnu.org/ml/gcc/2002-08/msg00822.html}.
2973 In Tru64 UNIX V5.1, Compaq introduced a new assembler that does not
2974 currently (2001-06-13) work with @command{mips-tfile}. As a workaround,
2975 we need to use the old assembler, invoked via the barely documented
2976 @option{-oldas} option. To bootstrap GCC, you either need to use the
2980 % CC=cc @var{srcdir}/configure [@var{options}] [@var{target}]
2983 or you can use a copy of GCC 2.95.3 or higher built on Tru64 UNIX V4.0:
2986 % CC=gcc -Wa,-oldas @var{srcdir}/configure [@var{options}] [@var{target}]
2989 As of GNU binutils 2.11.2, neither GNU @command{as} nor GNU @command{ld}
2990 are supported on Tru64 UNIX, so you must not configure GCC with
2991 @option{--with-gnu-as} or @option{--with-gnu-ld}.
2993 GCC writes a @samp{.verstamp} directive to the assembler output file
2994 unless it is built as a cross-compiler. It gets the version to use from
2995 the system header file @file{/usr/include/stamp.h}. If you install a
2996 new version of DEC Unix, you should rebuild GCC to pick up the new version
2999 @samp{make compare} may fail on old versions of DEC Unix unless you add
3000 @option{-save-temps} to @code{BOOT_CFLAGS}. On these systems, the name
3001 of the assembler input file is stored in the object file, and that makes
3002 comparison fail if it differs between the @code{stage1} and
3003 @code{stage2} compilations. The option @option{-save-temps} forces a
3004 fixed name to be used for the assembler input file, instead of a
3005 randomly chosen name in @file{/tmp}. Do not add @option{-save-temps}
3006 unless the comparisons fail without that option. If you add
3007 @option{-save-temps}, you will have to manually delete the @samp{.i} and
3008 @samp{.s} files after each series of compilations.
3010 GCC now supports both the native (ECOFF) debugging format used by DBX
3011 and GDB and an encapsulated STABS format for use only with GDB@. See the
3012 discussion of the @option{--with-stabs} option of @file{configure} above
3013 for more information on these formats and how to select them.
3015 There is a bug in DEC's assembler that produces incorrect line numbers
3016 for ECOFF format when the @samp{.align} directive is used. To work
3017 around this problem, GCC will not emit such alignment directives
3018 while writing ECOFF format debugging information even if optimization is
3019 being performed. Unfortunately, this has the very undesirable
3020 side-effect that code addresses when @option{-O} is specified are
3021 different depending on whether or not @option{-g} is also specified.
3023 To avoid this behavior, specify @option{-gstabs+} and use GDB instead of
3024 DBX@. DEC is now aware of this problem with the assembler and hopes to
3025 provide a fix shortly.
3030 @heading @anchor{arc-x-elf}arc-*-elf
3031 Argonaut ARC processor.
3032 This configuration is intended for embedded systems.
3037 @heading @anchor{arm-x-elf}arm-*-elf
3038 ARM-family processors. Subtargets that use the ELF object format
3039 require GNU binutils 2.13 or newer. Such subtargets include:
3040 @code{arm-*-freebsd}, @code{arm-*-netbsdelf}, @code{arm-*-*linux}
3041 and @code{arm-*-rtems}.
3046 @heading @anchor{avr}avr
3048 ATMEL AVR-family micro controllers. These are used in embedded
3049 applications. There are no standard Unix configurations.
3051 @xref{AVR Options,, AVR Options, gcc, Using the GNU Compiler
3055 See ``AVR Options'' in the main manual
3057 for the list of supported MCU types.
3059 Use @samp{configure --target=avr --enable-languages="c"} to configure GCC@.
3061 Further installation notes and other useful information about AVR tools
3062 can also be obtained from:
3066 @uref{http://www.nongnu.org/avr/,,http://www.nongnu.org/avr/}
3068 @uref{http://home.overta.ru/users/denisc/,,http://home.overta.ru/users/denisc/}
3070 @uref{http://www.amelek.gda.pl/avr/,,http://www.amelek.gda.pl/avr/}
3073 We @emph{strongly} recommend using binutils 2.13 or newer.
3075 The following error:
3077 Error: register required
3080 indicates that you should upgrade to a newer version of the binutils.
3085 @heading @anchor{bfin}Blackfin
3087 The Blackfin processor, an Analog Devices DSP.
3089 @xref{Blackfin Options,, Blackfin Options, gcc, Using the GNU Compiler
3093 See ``Blackfin Options'' in the main manual
3096 More information, and a version of binutils with support for this processor,
3097 is available at @uref{http://blackfin.uclinux.org}
3102 @heading @anchor{cris}CRIS
3104 CRIS is the CPU architecture in Axis Communications ETRAX system-on-a-chip
3105 series. These are used in embedded applications.
3108 @xref{CRIS Options,, CRIS Options, gcc, Using the GNU Compiler
3112 See ``CRIS Options'' in the main manual
3114 for a list of CRIS-specific options.
3116 There are a few different CRIS targets:
3119 Mainly for monolithic embedded systems. Includes a multilib for the
3120 @samp{v10} core used in @samp{ETRAX 100 LX}.
3121 @item cris-axis-linux-gnu
3122 A GNU/Linux port for the CRIS architecture, currently targeting
3123 @samp{ETRAX 100 LX} by default.
3126 For @code{cris-axis-elf} you need binutils 2.11
3127 or newer. For @code{cris-axis-linux-gnu} you need binutils 2.12 or newer.
3129 Pre-packaged tools can be obtained from
3130 @uref{ftp://ftp.axis.com/pub/axis/tools/cris/compiler-kit/}. More
3131 information about this platform is available at
3132 @uref{http://developer.axis.com/}.
3137 @heading @anchor{crx}CRX
3139 The CRX CompactRISC architecture is a low-power 32-bit architecture with
3140 fast context switching and architectural extensibility features.
3143 @xref{CRX Options,, CRX Options, gcc, Using and Porting the GNU Compiler
3148 See ``CRX Options'' in the main manual for a list of CRX-specific options.
3151 Use @samp{configure --target=crx-elf --enable-languages=c,c++} to configure
3152 GCC@ for building a CRX cross-compiler. The option @samp{--target=crx-elf}
3153 is also used to build the @samp{newlib} C library for CRX.
3155 It is also possible to build libstdc++-v3 for the CRX architecture. This
3156 needs to be done in a separate step with the following configure settings:
3157 @samp{gcc/libstdc++-v3/configure --host=crx-elf --with-newlib
3158 --enable-sjlj-exceptions --enable-cxx-flags='-fexceptions -frtti'}
3163 @heading @anchor{dos}DOS
3165 Please have a look at the @uref{binaries.html,,binaries page}.
3167 You cannot install GCC by itself on MSDOS; it will not compile under
3168 any MSDOS compiler except itself. You need to get the complete
3169 compilation package DJGPP, which includes binaries as well as sources,
3170 and includes all the necessary compilation tools and libraries.
3175 @heading @anchor{x-x-freebsd}*-*-freebsd*
3177 The version of binutils installed in @file{/usr/bin} probably works with
3178 this release of GCC@. However, on FreeBSD 4, bootstrapping against the
3179 latest FSF binutils is known to improve overall testsuite results; and,
3180 on FreeBSD/alpha, using binutils 2.14 or later is required to build libjava.
3182 Support for FreeBSD 1 was discontinued in GCC 3.2.
3184 Support for FreeBSD 2 will be discontinued after GCC 3.4. The
3185 following was true for GCC 3.1 but the current status is unknown.
3186 For FreeBSD 2 or any mutant a.out versions of FreeBSD 3: All
3187 configuration support and files as shipped with GCC 2.95 are still in
3188 place. FreeBSD 2.2.7 has been known to bootstrap completely; however,
3189 it is unknown which version of binutils was used (it is assumed that it
3190 was the system copy in @file{/usr/bin}) and C++ EH failures were noted.
3192 For FreeBSD using the ELF file format: DWARF 2 debugging is now the
3193 default for all CPU architectures. It had been the default on
3194 FreeBSD/alpha since its inception. You may use @option{-gstabs} instead
3195 of @option{-g}, if you really want the old debugging format. There are
3196 no known issues with mixing object files and libraries with different
3197 debugging formats. Otherwise, this release of GCC should now match more
3198 of the configuration used in the stock FreeBSD configuration of GCC@. In
3199 particular, @option{--enable-threads} is now configured by default.
3200 However, as a general user, do not attempt to replace the system
3201 compiler with this release. Known to bootstrap and check with good
3202 results on FreeBSD 4.9-STABLE and 5-CURRENT@. In the past, known to
3203 bootstrap and check with good results on FreeBSD 3.0, 3.4, 4.0, 4.2,
3204 4.3, 4.4, 4.5, 4.8-STABLE@.
3206 In principle, @option{--enable-threads} is now compatible with
3207 @option{--enable-libgcj} on FreeBSD@. However, it has only been built
3208 and tested on @samp{i386-*-freebsd[45]} and @samp{alpha-*-freebsd[45]}.
3210 library may be incorrectly built (symbols are missing at link time).
3211 There is a rare timing-based startup hang (probably involves an
3212 assumption about the thread library). Multi-threaded boehm-gc (required for
3213 libjava) exposes severe threaded signal-handling bugs on FreeBSD before
3214 4.5-RELEASE@. Other CPU architectures
3215 supported by FreeBSD will require additional configuration tuning in, at
3216 the very least, both boehm-gc and libffi.
3218 Shared @file{libgcc_s.so} is now built and installed by default.
3223 @heading @anchor{h8300-hms}h8300-hms
3224 Renesas H8/300 series of processors.
3226 Please have a look at the @uref{binaries.html,,binaries page}.
3228 The calling convention and structure layout has changed in release 2.6.
3229 All code must be recompiled. The calling convention now passes the
3230 first three arguments in function calls in registers. Structures are no
3231 longer a multiple of 2 bytes.
3236 @heading @anchor{hppa-hp-hpux}hppa*-hp-hpux*
3237 Support for HP-UX version 9 and older was discontinued in GCC 3.4.
3239 We require using gas/binutils on all hppa platforms. Version 2.19 or
3240 later is recommended.
3242 It may be helpful to configure GCC with the
3243 @uref{./configure.html#with-gnu-as,,@option{--with-gnu-as}} and
3244 @option{--with-as=@dots{}} options to ensure that GCC can find GAS@.
3246 The HP assembler should not be used with GCC. It is rarely tested and may
3247 not work. It shouldn't be used with any languages other than C due to its
3250 Specifically, @option{-g} does not work (HP-UX uses a peculiar debugging
3251 format which GCC does not know about). It also inserts timestamps
3252 into each object file it creates, causing the 3-stage comparison test to
3253 fail during a bootstrap. You should be able to continue by saying
3254 @samp{make all-host all-target} after getting the failure from @samp{make}.
3256 Various GCC features are not supported. For example, it does not support weak
3257 symbols or alias definitions. As a result, explicit template instantiations
3258 are required when using C++. This makes it difficult if not impossible to
3259 build many C++ applications.
3261 There are two default scheduling models for instructions. These are
3262 PROCESSOR_7100LC and PROCESSOR_8000. They are selected from the pa-risc
3263 architecture specified for the target machine when configuring.
3264 PROCESSOR_8000 is the default. PROCESSOR_7100LC is selected when
3265 the target is a @samp{hppa1*} machine.
3267 The PROCESSOR_8000 model is not well suited to older processors. Thus,
3268 it is important to completely specify the machine architecture when
3269 configuring if you want a model other than PROCESSOR_8000. The macro
3270 TARGET_SCHED_DEFAULT can be defined in BOOT_CFLAGS if a different
3271 default scheduling model is desired.
3273 As of GCC 4.0, GCC uses the UNIX 95 namespace for HP-UX 10.10
3274 through 11.00, and the UNIX 98 namespace for HP-UX 11.11 and later.
3275 This namespace change might cause problems when bootstrapping with
3276 an earlier version of GCC or the HP compiler as essentially the same
3277 namespace is required for an entire build. This problem can be avoided
3278 in a number of ways. With HP cc, @env{UNIX_STD} can be set to @samp{95}
3279 or @samp{98}. Another way is to add an appropriate set of predefines
3280 to @env{CC}. The description for the @option{munix=} option contains
3281 a list of the predefines used with each standard.
3283 More specific information to @samp{hppa*-hp-hpux*} targets follows.
3288 @heading @anchor{hppa-hp-hpux10}hppa*-hp-hpux10
3290 For hpux10.20, we @emph{highly} recommend you pick up the latest sed patch
3291 @code{PHCO_19798} from HP@. HP has two sites which provide patches free of
3297 <a href="http://us.itrc.hp.com/service/home/home.do">US, Canada, Asia-Pacific, and
3301 @uref{http://us.itrc.hp.com/service/home/home.do,,} US, Canada, Asia-Pacific,
3305 @uref{http://europe.itrc.hp.com/service/home/home.do,,} Europe.
3308 The C++ ABI has changed incompatibly in GCC 4.0. COMDAT subspaces are
3309 used for one-only code and data. This resolves many of the previous
3310 problems in using C++ on this target. However, the ABI is not compatible
3311 with the one implemented under HP-UX 11 using secondary definitions.
3316 @heading @anchor{hppa-hp-hpux11}hppa*-hp-hpux11
3318 GCC 3.0 and up support HP-UX 11. GCC 2.95.x is not supported and cannot
3319 be used to compile GCC 3.0 and up.
3321 The libffi and libjava libraries haven't been ported to 64-bit HP-UX@
3324 Refer to @uref{binaries.html,,binaries} for information about obtaining
3325 precompiled GCC binaries for HP-UX@. Precompiled binaries must be obtained
3326 to build the Ada language as it can't be bootstrapped using C@. Ada is
3327 only available for the 32-bit PA-RISC runtime.
3329 Starting with GCC 3.4 an ISO C compiler is required to bootstrap. The
3330 bundled compiler supports only traditional C; you will need either HP's
3331 unbundled compiler, or a binary distribution of GCC@.
3333 It is possible to build GCC 3.3 starting with the bundled HP compiler,
3334 but the process requires several steps. GCC 3.3 can then be used to
3335 build later versions. The fastjar program contains ISO C code and
3336 can't be built with the HP bundled compiler. This problem can be
3337 avoided by not building the Java language. For example, use the
3338 @option{--enable-languages="c,c++,f77,objc"} option in your configure
3341 There are several possible approaches to building the distribution.
3342 Binutils can be built first using the HP tools. Then, the GCC
3343 distribution can be built. The second approach is to build GCC
3344 first using the HP tools, then build binutils, then rebuild GCC@.
3345 There have been problems with various binary distributions, so it
3346 is best not to start from a binary distribution.
3348 On 64-bit capable systems, there are two distinct targets. Different
3349 installation prefixes must be used if both are to be installed on
3350 the same system. The @samp{hppa[1-2]*-hp-hpux11*} target generates code
3351 for the 32-bit PA-RISC runtime architecture and uses the HP linker.
3352 The @samp{hppa64-hp-hpux11*} target generates 64-bit code for the
3353 PA-RISC 2.0 architecture.
3355 The script config.guess now selects the target type based on the compiler
3356 detected during configuration. You must define @env{PATH} or @env{CC} so
3357 that configure finds an appropriate compiler for the initial bootstrap.
3358 When @env{CC} is used, the definition should contain the options that are
3359 needed whenever @env{CC} is used.
3361 Specifically, options that determine the runtime architecture must be
3362 in @env{CC} to correctly select the target for the build. It is also
3363 convenient to place many other compiler options in @env{CC}. For example,
3364 @env{CC="cc -Ac +DA2.0W -Wp,-H16376 -D_CLASSIC_TYPES -D_HPUX_SOURCE"}
3365 can be used to bootstrap the GCC 3.3 branch with the HP compiler in
3366 64-bit K&R/bundled mode. The @option{+DA2.0W} option will result in
3367 the automatic selection of the @samp{hppa64-hp-hpux11*} target. The
3368 macro definition table of cpp needs to be increased for a successful
3369 build with the HP compiler. _CLASSIC_TYPES and _HPUX_SOURCE need to
3370 be defined when building with the bundled compiler, or when using the
3371 @option{-Ac} option. These defines aren't necessary with @option{-Ae}.
3373 It is best to explicitly configure the @samp{hppa64-hp-hpux11*} target
3374 with the @option{--with-ld=@dots{}} option. This overrides the standard
3375 search for ld. The two linkers supported on this target require different
3376 commands. The default linker is determined during configuration. As a
3377 result, it's not possible to switch linkers in the middle of a GCC build.
3378 This has been reported to sometimes occur in unified builds of binutils
3381 A recent linker patch must be installed for the correct operation of
3382 GCC 3.3 and later. @code{PHSS_26559} and @code{PHSS_24304} are the
3383 oldest linker patches that are known to work. They are for HP-UX
3384 11.00 and 11.11, respectively. @code{PHSS_24303}, the companion to
3385 @code{PHSS_24304}, might be usable but it hasn't been tested. These
3386 patches have been superseded. Consult the HP patch database to obtain
3387 the currently recommended linker patch for your system.
3389 The patches are necessary for the support of weak symbols on the
3390 32-bit port, and for the running of initializers and finalizers. Weak
3391 symbols are implemented using SOM secondary definition symbols. Prior
3392 to HP-UX 11, there are bugs in the linker support for secondary symbols.
3393 The patches correct a problem of linker core dumps creating shared
3394 libraries containing secondary symbols, as well as various other
3395 linking issues involving secondary symbols.
3397 GCC 3.3 uses the ELF DT_INIT_ARRAY and DT_FINI_ARRAY capabilities to
3398 run initializers and finalizers on the 64-bit port. The 32-bit port
3399 uses the linker @option{+init} and @option{+fini} options for the same
3400 purpose. The patches correct various problems with the +init/+fini
3401 options, including program core dumps. Binutils 2.14 corrects a
3402 problem on the 64-bit port resulting from HP's non-standard use of
3403 the .init and .fini sections for array initializers and finalizers.
3405 Although the HP and GNU linkers are both supported for the
3406 @samp{hppa64-hp-hpux11*} target, it is strongly recommended that the
3407 HP linker be used for link editing on this target.
3409 At this time, the GNU linker does not support the creation of long
3410 branch stubs. As a result, it can't successfully link binaries
3411 containing branch offsets larger than 8 megabytes. In addition,
3412 there are problems linking shared libraries, linking executables
3413 with @option{-static}, and with dwarf2 unwind and exception support.
3414 It also doesn't provide stubs for internal calls to global functions
3415 in shared libraries, so these calls can't be overloaded.
3417 The HP dynamic loader does not support GNU symbol versioning, so symbol
3418 versioning is not supported. It may be necessary to disable symbol
3419 versioning with @option{--disable-symvers} when using GNU ld.
3421 POSIX threads are the default. The optional DCE thread library is not
3422 supported, so @option{--enable-threads=dce} does not work.
3427 @heading @anchor{x-x-linux-gnu}*-*-linux-gnu
3429 Versions of libstdc++-v3 starting with 3.2.1 require bug fixes present
3430 in glibc 2.2.5 and later. More information is available in the
3431 libstdc++-v3 documentation.
3436 @heading @anchor{ix86-x-linux}i?86-*-linux*
3438 As of GCC 3.3, binutils 2.13.1 or later is required for this platform.
3439 See @uref{http://gcc.gnu.org/PR10877,,bug 10877} for more information.
3441 If you receive Signal 11 errors when building on GNU/Linux, then it is
3442 possible you have a hardware problem. Further information on this can be
3443 found on @uref{http://www.bitwizard.nl/sig11/,,www.bitwizard.nl}.
3448 @heading @anchor{ix86-x-solaris210}i?86-*-solaris2.10
3449 Use this for Solaris 10 or later on x86 and x86-64 systems. This
3450 configuration is supported by GCC 4.0 and later versions only.
3452 It is recommended that you configure GCC to use the GNU assembler in
3453 @file{/usr/sfw/bin/gas} but the Sun linker, using the options
3454 @option{--with-gnu-as --with-as=/usr/sfw/bin/gas --without-gnu-ld
3455 --with-ld=/usr/ccs/bin/ld}.
3460 @heading @anchor{ia64-x-linux}ia64-*-linux
3461 IA-64 processor (also known as IPF, or Itanium Processor Family)
3464 If you are using the installed system libunwind library with
3465 @option{--with-system-libunwind}, then you must use libunwind 0.98 or
3468 None of the following versions of GCC has an ABI that is compatible
3469 with any of the other versions in this list, with the exception that
3470 Red Hat 2.96 and Trillian 000171 are compatible with each other:
3471 3.1, 3.0.2, 3.0.1, 3.0, Red Hat 2.96, and Trillian 000717.
3472 This primarily affects C++ programs and programs that create shared libraries.
3473 GCC 3.1 or later is recommended for compiling linux, the kernel.
3474 As of version 3.1 GCC is believed to be fully ABI compliant, and hence no
3475 more major ABI changes are expected.
3480 @heading @anchor{ia64-x-hpux}ia64-*-hpux*
3481 Building GCC on this target requires the GNU Assembler. The bundled HP
3482 assembler will not work. To prevent GCC from using the wrong assembler,
3483 the option @option{--with-gnu-as} may be necessary.
3485 The GCC libunwind library has not been ported to HPUX@. This means that for
3486 GCC versions 3.2.3 and earlier, @option{--enable-libunwind-exceptions}
3487 is required to build GCC@. For GCC 3.3 and later, this is the default.
3488 For gcc 3.4.3 and later, @option{--enable-libunwind-exceptions} is
3489 removed and the system libunwind library will always be used.
3493 <!-- rs6000-ibm-aix*, powerpc-ibm-aix* -->
3495 @heading @anchor{x-ibm-aix}*-ibm-aix*
3496 Support for AIX version 3 and older was discontinued in GCC 3.4.
3497 Support for AIX version 4.2 and older was discontinued in GCC 4.5.
3499 ``out of memory'' bootstrap failures may indicate a problem with
3500 process resource limits (ulimit). Hard limits are configured in the
3501 @file{/etc/security/limits} system configuration file.
3503 GCC can bootstrap with recent versions of IBM XLC, but bootstrapping
3504 with an earlier release of GCC is recommended. Bootstrapping with XLC
3505 requires a larger data segment, which can be enabled through the
3506 @var{LDR_CNTRL} environment variable, e.g.,
3509 % LDR_CNTRL=MAXDATA=0x50000000
3513 One can start with a pre-compiled version of GCC to build from
3514 sources. One may delete GCC's ``fixed'' header files when starting
3515 with a version of GCC built for an earlier release of AIX.
3517 To speed up the configuration phases of bootstrapping and installing GCC,
3518 one may use GNU Bash instead of AIX @command{/bin/sh}, e.g.,
3521 % CONFIG_SHELL=/opt/freeware/bin/bash
3522 % export CONFIG_SHELL
3525 and then proceed as described in @uref{build.html,,the build
3526 instructions}, where we strongly recommend specifying an absolute path
3527 to invoke @var{srcdir}/configure.
3529 Because GCC on AIX is built as a 32-bit executable by default,
3530 (although it can generate 64-bit programs) the GMP and MPFR libraries
3531 required by gfortran must be 32-bit libraries. Building GMP and MPFR
3532 as static archive libraries works better than shared libraries.
3534 Errors involving @code{alloca} when building GCC generally are due
3535 to an incorrect definition of @code{CC} in the Makefile or mixing files
3536 compiled with the native C compiler and GCC@. During the stage1 phase of
3537 the build, the native AIX compiler @strong{must} be invoked as @command{cc}
3538 (not @command{xlc}). Once @command{configure} has been informed of
3539 @command{xlc}, one needs to use @samp{make distclean} to remove the
3540 configure cache files and ensure that @env{CC} environment variable
3541 does not provide a definition that will confuse @command{configure}.
3542 If this error occurs during stage2 or later, then the problem most likely
3543 is the version of Make (see above).
3545 The native @command{as} and @command{ld} are recommended for bootstrapping
3546 on AIX@. The GNU Assembler, GNU Linker, and GNU Binutils version 2.20
3547 is required to bootstrap on AIX 5@. The native AIX tools do
3548 interoperate with GCC@.
3550 Building @file{libstdc++.a} requires a fix for an AIX Assembler bug
3551 APAR IY26685 (AIX 4.3) or APAR IY25528 (AIX 5.1). It also requires a
3552 fix for another AIX Assembler bug and a co-dependent AIX Archiver fix
3553 referenced as APAR IY53606 (AIX 5.2) or as APAR IY54774 (AIX 5.1)
3555 @samp{libstdc++} in GCC 3.4 increments the major version number of the
3556 shared object and GCC installation places the @file{libstdc++.a}
3557 shared library in a common location which will overwrite the and GCC
3558 3.3 version of the shared library. Applications either need to be
3559 re-linked against the new shared library or the GCC 3.1 and GCC 3.3
3560 versions of the @samp{libstdc++} shared object needs to be available
3561 to the AIX runtime loader. The GCC 3.1 @samp{libstdc++.so.4}, if
3562 present, and GCC 3.3 @samp{libstdc++.so.5} shared objects can be
3563 installed for runtime dynamic loading using the following steps to set
3564 the @samp{F_LOADONLY} flag in the shared object for @emph{each}
3565 multilib @file{libstdc++.a} installed:
3567 Extract the shared objects from the currently installed
3568 @file{libstdc++.a} archive:
3570 % ar -x libstdc++.a libstdc++.so.4 libstdc++.so.5
3573 Enable the @samp{F_LOADONLY} flag so that the shared object will be
3574 available for runtime dynamic loading, but not linking:
3576 % strip -e libstdc++.so.4 libstdc++.so.5
3579 Archive the runtime-only shared object in the GCC 3.4
3580 @file{libstdc++.a} archive:
3582 % ar -q libstdc++.a libstdc++.so.4 libstdc++.so.5
3585 Linking executables and shared libraries may produce warnings of
3586 duplicate symbols. The assembly files generated by GCC for AIX always
3587 have included multiple symbol definitions for certain global variable
3588 and function declarations in the original program. The warnings should
3589 not prevent the linker from producing a correct library or runnable
3592 AIX 4.3 utilizes a ``large format'' archive to support both 32-bit and
3593 64-bit object modules. The routines provided in AIX 4.3.0 and AIX 4.3.1
3594 to parse archive libraries did not handle the new format correctly.
3595 These routines are used by GCC and result in error messages during
3596 linking such as ``not a COFF file''. The version of the routines shipped
3597 with AIX 4.3.1 should work for a 32-bit environment. The @option{-g}
3598 option of the archive command may be used to create archives of 32-bit
3599 objects using the original ``small format''. A correct version of the
3600 routines is shipped with AIX 4.3.2 and above.
3602 Some versions of the AIX binder (linker) can fail with a relocation
3603 overflow severe error when the @option{-bbigtoc} option is used to link
3604 GCC-produced object files into an executable that overflows the TOC@. A fix
3605 for APAR IX75823 (OVERFLOW DURING LINK WHEN USING GCC AND -BBIGTOC) is
3606 available from IBM Customer Support and from its
3607 @uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com}
3608 website as PTF U455193.
3610 The AIX 4.3.2.1 linker (bos.rte.bind_cmds Level 4.3.2.1) will dump core
3611 with a segmentation fault when invoked by any version of GCC@. A fix for
3612 APAR IX87327 is available from IBM Customer Support and from its
3613 @uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com}
3614 website as PTF U461879. This fix is incorporated in AIX 4.3.3 and above.
3616 The initial assembler shipped with AIX 4.3.0 generates incorrect object
3617 files. A fix for APAR IX74254 (64BIT DISASSEMBLED OUTPUT FROM COMPILER FAILS
3618 TO ASSEMBLE/BIND) is available from IBM Customer Support and from its
3619 @uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com}
3620 website as PTF U453956. This fix is incorporated in AIX 4.3.1 and above.
3622 AIX provides National Language Support (NLS)@. Compilers and assemblers
3623 use NLS to support locale-specific representations of various data
3624 formats including floating-point numbers (e.g., @samp{.} vs @samp{,} for
3625 separating decimal fractions). There have been problems reported where
3626 GCC does not produce the same floating-point formats that the assembler
3627 expects. If one encounters this problem, set the @env{LANG}
3628 environment variable to @samp{C} or @samp{En_US}.
3630 A default can be specified with the @option{-mcpu=@var{cpu_type}}
3631 switch and using the configure option @option{--with-cpu-@var{cpu_type}}.
3636 @heading @anchor{iq2000-x-elf}iq2000-*-elf
3637 Vitesse IQ2000 processors. These are used in embedded
3638 applications. There are no standard Unix configurations.
3643 @heading @anchor{m32c-x-elf}m32c-*-elf
3644 Renesas M32C processor.
3645 This configuration is intended for embedded systems.
3650 @heading @anchor{m32r-x-elf}m32r-*-elf
3651 Renesas M32R processor.
3652 This configuration is intended for embedded systems.
3657 @heading @anchor{m6811-elf}m6811-elf
3658 Motorola 68HC11 family micro controllers. These are used in embedded
3659 applications. There are no standard Unix configurations.
3664 @heading @anchor{m6812-elf}m6812-elf
3665 Motorola 68HC12 family micro controllers. These are used in embedded
3666 applications. There are no standard Unix configurations.
3671 @heading @anchor{m68k-x-x}m68k-*-*
3673 @samp{m68k-*-elf*}, @samp{m68k-*-rtems}, @samp{m68k-*-uclinux} and
3675 build libraries for both M680x0 and ColdFire processors. If you only
3676 need the M680x0 libraries, you can omit the ColdFire ones by passing
3677 @option{--with-arch=m68k} to @command{configure}. Alternatively, you
3678 can omit the M680x0 libraries by passing @option{--with-arch=cf} to
3679 @command{configure}. These targets default to 5206 or 5475 code as
3680 appropriate for the target system when
3681 configured with @option{--with-arch=cf} and 68020 code otherwise.
3683 The @samp{m68k-*-netbsd} and
3684 @samp{m68k-*-openbsd} targets also support the @option{--with-arch}
3685 option. They will generate ColdFire CFV4e code when configured with
3686 @option{--with-arch=cf} and 68020 code otherwise.
3688 You can override the default processors listed above by configuring
3689 with @option{--with-cpu=@var{target}}. This @var{target} can either
3690 be a @option{-mcpu} argument or one of the following values:
3691 @samp{m68000}, @samp{m68010}, @samp{m68020}, @samp{m68030},
3692 @samp{m68040}, @samp{m68060}, @samp{m68020-40} and @samp{m68020-60}.
3697 @heading @anchor{m68k-x-uclinux}m68k-*-uclinux
3698 GCC 4.3 changed the uClinux configuration so that it uses the
3699 @samp{m68k-linux-gnu} ABI rather than the @samp{m68k-elf} ABI.
3700 It also added improved support for C++ and flat shared libraries,
3701 both of which were ABI changes. However, you can still use the
3702 original ABI by configuring for @samp{m68k-uclinuxoldabi} or
3703 @samp{m68k-@var{vendor}-uclinuxoldabi}.
3709 @heading @anchor{mep-x-elf}mep-*-elf
3710 Toshiba Media embedded Processor.
3711 This configuration is intended for embedded systems.
3716 @heading @anchor{mips-x-x}mips-*-*
3717 If on a MIPS system you get an error message saying ``does not have gp
3718 sections for all it's [sic] sectons [sic]'', don't worry about it. This
3719 happens whenever you use GAS with the MIPS linker, but there is not
3720 really anything wrong, and it is okay to use the output file. You can
3721 stop such warnings by installing the GNU linker.
3723 It would be nice to extend GAS to produce the gp tables, but they are
3724 optional, and there should not be a warning about their absence.
3726 The libstdc++ atomic locking routines for MIPS targets requires MIPS II
3727 and later. A patch went in just after the GCC 3.3 release to
3728 make @samp{mips*-*-*} use the generic implementation instead. You can also
3729 configure for @samp{mipsel-elf} as a workaround. The
3730 @samp{mips*-*-linux*} target continues to use the MIPS II routines. More
3731 work on this is expected in future releases.
3733 @c If you make --with-llsc the default for another target, please also
3734 @c update the description of the --with-llsc option.
3736 The built-in @code{__sync_*} functions are available on MIPS II and
3737 later systems and others that support the @samp{ll}, @samp{sc} and
3738 @samp{sync} instructions. This can be overridden by passing
3739 @option{--with-llsc} or @option{--without-llsc} when configuring GCC.
3740 Since the Linux kernel emulates these instructions if they are
3741 missing, the default for @samp{mips*-*-linux*} targets is
3742 @option{--with-llsc}. The @option{--with-llsc} and
3743 @option{--without-llsc} configure options may be overridden at compile
3744 time by passing the @option{-mllsc} or @option{-mno-llsc} options to
3747 MIPS systems check for division by zero (unless
3748 @option{-mno-check-zero-division} is passed to the compiler) by
3749 generating either a conditional trap or a break instruction. Using
3750 trap results in smaller code, but is only supported on MIPS II and
3751 later. Also, some versions of the Linux kernel have a bug that
3752 prevents trap from generating the proper signal (@code{SIGFPE}). To enable
3753 the use of break, use the @option{--with-divide=breaks}
3754 @command{configure} option when configuring GCC@. The default is to
3755 use traps on systems that support them.
3757 Cross-compilers for the MIPS as target using the MIPS assembler
3758 currently do not work, because the auxiliary programs
3759 @file{mips-tdump.c} and @file{mips-tfile.c} can't be compiled on
3760 anything but a MIPS@. It does work to cross compile for a MIPS
3761 if you use the GNU assembler and linker.
3763 The assembler from GNU binutils 2.17 and earlier has a bug in the way
3764 it sorts relocations for REL targets (o32, o64, EABI). This can cause
3765 bad code to be generated for simple C++ programs. Also the linker
3766 from GNU binutils versions prior to 2.17 has a bug which causes the
3767 runtime linker stubs in very large programs, like @file{libgcj.so}, to
3768 be incorrectly generated. GNU Binutils 2.18 and later (and snapshots
3769 made after Nov. 9, 2006) should be free from both of these problems.
3774 @heading @anchor{mips-sgi-irix5}mips-sgi-irix5
3776 In order to compile GCC on an SGI running IRIX 5, the @samp{compiler_dev.hdr}
3777 subsystem must be installed from the IDO CD-ROM supplied by SGI@.
3778 It is also available for download from
3779 @uref{ftp://ftp.sgi.com/sgi/IRIX5.3/iris-development-option-5.3.tardist}.
3781 If you use the MIPS C compiler to bootstrap, it may be necessary
3782 to increase its table size for switch statements with the
3783 @option{-Wf,-XNg1500} option. If you use the @option{-O2}
3784 optimization option, you also need to use @option{-Olimit 3000}.
3786 To enable debugging under IRIX 5, you must use GNU binutils 2.15 or
3787 later, and use the @option{--with-gnu-ld} @command{configure} option
3788 when configuring GCC@. You need to use GNU @command{ar} and @command{nm},
3789 also distributed with GNU binutils.
3791 Some users have reported that @command{/bin/sh} will hang during bootstrap.
3792 This problem can be avoided by running the commands:
3795 % CONFIG_SHELL=/bin/ksh
3796 % export CONFIG_SHELL
3799 before starting the build.
3804 @heading @anchor{mips-sgi-irix6}mips-sgi-irix6
3806 If you are using SGI's MIPSpro @command{cc} as your bootstrap compiler, you must
3807 ensure that the N32 ABI is in use. To test this, compile a simple C
3808 file with @command{cc} and then run @command{file} on the
3809 resulting object file. The output should look like:
3812 test.o: ELF N32 MSB @dots{}
3818 test.o: ELF 32-bit MSB @dots{}
3824 test.o: ELF 64-bit MSB @dots{}
3827 then your version of @command{cc} uses the O32 or N64 ABI by default. You
3828 should set the environment variable @env{CC} to @samp{cc -n32}
3829 before configuring GCC@.
3831 If you want the resulting @command{gcc} to run on old 32-bit systems
3832 with the MIPS R4400 CPU, you need to ensure that only code for the @samp{mips3}
3833 instruction set architecture (ISA) is generated. While GCC 3.x does
3834 this correctly, both GCC 2.95 and SGI's MIPSpro @command{cc} may change
3835 the ISA depending on the machine where GCC is built. Using one of them
3836 as the bootstrap compiler may result in @samp{mips4} code, which won't run at
3837 all on @samp{mips3}-only systems. For the test program above, you should see:
3840 test.o: ELF N32 MSB mips-3 @dots{}
3846 test.o: ELF N32 MSB mips-4 @dots{}
3849 instead, you should set the environment variable @env{CC} to @samp{cc
3850 -n32 -mips3} or @samp{gcc -mips3} respectively before configuring GCC@.
3852 MIPSpro C 7.4 may cause bootstrap failures, due to a bug when inlining
3853 @code{memcmp}. Either add @code{-U__INLINE_INTRINSICS} to the @env{CC}
3854 environment variable as a workaround or upgrade to MIPSpro C 7.4.1m.
3856 GCC on IRIX 6 is usually built to support the N32, O32 and N64 ABIs. If
3857 you build GCC on a system that doesn't have the N64 libraries installed
3858 or cannot run 64-bit binaries,
3859 you need to configure with @option{--disable-multilib} so GCC doesn't
3860 try to use them. This will disable building the O32 libraries, too.
3861 Look for @file{/usr/lib64/libc.so.1} to see if you
3862 have the 64-bit libraries installed.
3864 To enable debugging for the O32 ABI, you must use GNU @command{as} from
3865 GNU binutils 2.15 or later. You may also use GNU @command{ld}, but
3866 this is not required and currently causes some problems with Ada.
3868 The @option{--enable-libgcj}
3869 option is disabled by default: IRIX 6 uses a very low default limit
3870 (20480) for the command line length. Although @command{libtool} contains a
3871 workaround for this problem, at least the N64 @samp{libgcj} is known not
3872 to build despite this, running into an internal error of the native
3873 @command{ld}. A sure fix is to increase this limit (@samp{ncargs}) to
3874 its maximum of 262144 bytes. If you have root access, you can use the
3875 @command{systune} command to do this.
3877 @code{wchar_t} support in @samp{libstdc++} is not available for old
3878 IRIX 6.5.x releases, @math{x < 19}. The problem cannot be autodetected
3879 and in order to build GCC for such targets you need to configure with
3880 @option{--disable-wchar_t}.
3882 See @uref{http://freeware.sgi.com/} for more
3883 information about using GCC on IRIX platforms.
3888 @heading @anchor{moxie-x-elf}moxie-*-elf
3889 The moxie processor. See @uref{http://moxielogic.org/} for more
3890 information about this processor.
3895 @heading @anchor{powerpc-x-x}powerpc-*-*
3897 You can specify a default version for the @option{-mcpu=@var{cpu_type}}
3898 switch by using the configure option @option{--with-cpu-@var{cpu_type}}.
3901 @uref{ftp://ftp.kernel.org/pub/linux/devel/binutils,,binutils 2.15}
3902 or newer for a working GCC@.
3907 @heading @anchor{powerpc-x-darwin}powerpc-*-darwin*
3908 PowerPC running Darwin (Mac OS X kernel).
3910 Pre-installed versions of Mac OS X may not include any developer tools,
3911 meaning that you will not be able to build GCC from source. Tool
3912 binaries are available at
3913 @uref{http://developer.apple.com/darwin/projects/compiler/} (free
3914 registration required).
3916 This version of GCC requires at least cctools-590.36. The
3917 cctools-590.36 package referenced from
3918 @uref{http://gcc.gnu.org/ml/gcc/2006-03/msg00507.html} will not work
3919 on systems older than 10.3.9 (aka darwin7.9.0).
3924 @heading @anchor{powerpc-x-elf}powerpc-*-elf
3925 PowerPC system in big endian mode, running System V.4.
3930 @heading @anchor{powerpc-x-linux-gnu}powerpc*-*-linux-gnu*
3932 PowerPC system in big endian mode running Linux.
3937 @heading @anchor{powerpc-x-netbsd}powerpc-*-netbsd*
3938 PowerPC system in big endian mode running NetBSD@.
3943 @heading @anchor{powerpc-x-eabisim}powerpc-*-eabisim
3944 Embedded PowerPC system in big endian mode for use in running under the
3950 @heading @anchor{powerpc-x-eabi}powerpc-*-eabi
3951 Embedded PowerPC system in big endian mode.
3956 @heading @anchor{powerpcle-x-elf}powerpcle-*-elf
3957 PowerPC system in little endian mode, running System V.4.
3962 @heading @anchor{powerpcle-x-eabisim}powerpcle-*-eabisim
3963 Embedded PowerPC system in little endian mode for use in running under
3969 @heading @anchor{powerpcle-x-eabi}powerpcle-*-eabi
3970 Embedded PowerPC system in little endian mode.
3975 @heading @anchor{s390-x-linux}s390-*-linux*
3976 S/390 system running GNU/Linux for S/390@.
3981 @heading @anchor{s390x-x-linux}s390x-*-linux*
3982 zSeries system (64-bit) running GNU/Linux for zSeries@.
3987 @heading @anchor{s390x-ibm-tpf}s390x-ibm-tpf*
3988 zSeries system (64-bit) running TPF@. This platform is
3989 supported as cross-compilation target only.
3994 @c Please use Solaris 2 to refer to all release of Solaris, starting
3995 @c with 2.0 until 2.6, 7, 8, etc. Solaris 1 was a marketing name for
3996 @c SunOS 4 releases which we don't use to avoid confusion. Solaris
3997 @c alone is too unspecific and must be avoided.
3998 @heading @anchor{x-x-solaris2}*-*-solaris2*
4000 Sun does not ship a C compiler with Solaris 2. To bootstrap and install
4001 GCC you first have to install a pre-built compiler, see the
4002 @uref{binaries.html,,binaries page} for details.
4004 The Solaris 2 @command{/bin/sh} will often fail to configure
4005 @file{libstdc++-v3}, @file{boehm-gc} or @file{libjava}. We therefore
4006 recommend using the following initial sequence of commands
4009 % CONFIG_SHELL=/bin/ksh
4010 % export CONFIG_SHELL
4013 and proceed as described in @uref{configure.html,,the configure instructions}.
4014 In addition we strongly recommend specifying an absolute path to invoke
4015 @var{srcdir}/configure.
4017 Solaris 2 comes with a number of optional OS packages. Some of these
4018 are needed to use GCC fully, namely @code{SUNWarc},
4019 @code{SUNWbtool}, @code{SUNWesu}, @code{SUNWhea}, @code{SUNWlibm},
4020 @code{SUNWsprot}, and @code{SUNWtoo}. If you did not install all
4021 optional packages when installing Solaris 2, you will need to verify that
4022 the packages that GCC needs are installed.
4024 To check whether an optional package is installed, use
4025 the @command{pkginfo} command. To add an optional package, use the
4026 @command{pkgadd} command. For further details, see the Solaris 2
4029 Trying to use the linker and other tools in
4030 @file{/usr/ucb} to install GCC has been observed to cause trouble.
4031 For example, the linker may hang indefinitely. The fix is to remove
4032 @file{/usr/ucb} from your @env{PATH}.
4034 The build process works more smoothly with the legacy Sun tools so, if you
4035 have @file{/usr/xpg4/bin} in your @env{PATH}, we recommend that you place
4036 @file{/usr/bin} before @file{/usr/xpg4/bin} for the duration of the build.
4038 We recommend the use of GNU binutils 2.14 or later, or the vendor tools
4039 (Sun @command{as}, Sun @command{ld}). Note that your mileage may vary
4040 if you use a combination of the GNU tools and the Sun tools: while the
4041 combination GNU @command{as} + Sun @command{ld} should reasonably work,
4042 the reverse combination Sun @command{as} + GNU @command{ld} is known to
4043 cause memory corruption at runtime in some cases for C++ programs.
4045 The stock GNU binutils 2.15 release is broken on this platform because of a
4046 single bug. It has been fixed on the 2.15 branch in the CVS repository.
4047 You can obtain a working version by checking out the binutils-2_15-branch
4048 from the CVS repository or applying the patch
4049 @uref{http://sourceware.org/ml/binutils-cvs/2004-09/msg00036.html} to the
4052 We recommend the use of GNU binutils 2.16 or later in conjunction with GCC
4053 4.x, or the vendor tools (Sun @command{as}, Sun @command{ld}). However,
4054 for Solaris 10 and above, an additional patch is required in order for the
4055 GNU linker to be able to cope with a new flavor of shared libraries. You
4056 can obtain a working version by checking out the binutils-2_16-branch from
4057 the CVS repository or applying the patch
4058 @uref{http://sourceware.org/ml/binutils-cvs/2005-07/msg00122.html} to the
4061 Sun bug 4296832 turns up when compiling X11 headers with GCC 2.95 or
4062 newer: @command{g++} will complain that types are missing. These headers
4063 assume that omitting the type means @code{int}; this assumption worked for
4064 C89 but is wrong for C++, and is now wrong for C99 also.
4066 @command{g++} accepts such (invalid) constructs with the option
4067 @option{-fpermissive}; it will assume that any missing type is @code{int}
4068 (as defined by C89).
4070 There are patches for Solaris 7 (108376-21 or newer for SPARC,
4071 108377-20 for Intel), and Solaris 8 (108652-24 or newer for SPARC,
4072 108653-22 for Intel) that fix this bug.
4074 Sun bug 4927647 sometimes causes random spurious testsuite failures
4075 related to missing diagnostic output. This bug doesn't affect GCC
4076 itself, rather it is a kernel bug triggered by the @command{expect}
4077 program which is used only by the GCC testsuite driver. When the bug
4078 causes the @command{expect} program to miss anticipated output, extra
4079 testsuite failures appear.
4081 There are patches for Solaris 8 (117350-12 or newer for SPARC,
4082 117351-12 or newer for Intel) and Solaris 9 (117171-11 or newer for
4083 SPARC, 117172-11 or newer for Intel) that address this problem.
4088 @heading @anchor{sparc-sun-solaris2}sparc-sun-solaris2*
4090 When GCC is configured to use binutils 2.14 or later the binaries
4091 produced are smaller than the ones produced using Sun's native tools;
4092 this difference is quite significant for binaries containing debugging
4095 Starting with Solaris 7, the operating system is capable of executing
4096 64-bit SPARC V9 binaries. GCC 3.1 and later properly supports
4097 this; the @option{-m64} option enables 64-bit code generation.
4098 However, if all you want is code tuned for the UltraSPARC CPU, you
4099 should try the @option{-mtune=ultrasparc} option instead, which produces
4100 code that, unlike full 64-bit code, can still run on non-UltraSPARC
4103 When configuring on a Solaris 7 or later system that is running a kernel
4104 that supports only 32-bit binaries, one must configure with
4105 @option{--disable-multilib}, since we will not be able to build the
4106 64-bit target libraries.
4108 GCC 3.3 and GCC 3.4 trigger code generation bugs in earlier versions of
4109 the GNU compiler (especially GCC 3.0.x versions), which lead to the
4110 miscompilation of the stage1 compiler and the subsequent failure of the
4111 bootstrap process. A workaround is to use GCC 3.2.3 as an intermediary
4112 stage, i.e.@: to bootstrap that compiler with the base compiler and then
4113 use it to bootstrap the final compiler.
4115 GCC 3.4 triggers a code generation bug in versions 5.4 (Sun ONE Studio 7)
4116 and 5.5 (Sun ONE Studio 8) of the Sun compiler, which causes a bootstrap
4117 failure in form of a miscompilation of the stage1 compiler by the Sun
4118 compiler. This is Sun bug 4974440. This is fixed with patch 112760-07.
4120 GCC 3.4 changed the default debugging format from STABS to DWARF-2 for
4121 32-bit code on Solaris 7 and later. If you use the Sun assembler, this
4122 change apparently runs afoul of Sun bug 4910101 (which is referenced as
4123 an x86-only problem by Sun, probably because they do not use DWARF-2).
4124 A symptom of the problem is that you cannot compile C++ programs like
4125 @command{groff} 1.19.1 without getting messages similar to the following:
4128 ld: warning: relocation error: R_SPARC_UA32: @dots{}
4129 external symbolic relocation against non-allocatable section
4130 .debug_info cannot be processed at runtime: relocation ignored.
4133 To work around this problem, compile with @option{-gstabs+} instead of
4136 When configuring the GNU Multiple Precision Library (GMP) or the MPFR
4137 library on a Solaris 7 or later system, the canonical target triplet
4138 must be specified as the @command{build} parameter on the configure
4139 line. This triplet can be obtained by invoking ./config.guess in
4140 the toplevel source directory of GCC (and not that of GMP or MPFR).
4141 For example on a Solaris 7 system:
4144 % ./configure --build=sparc-sun-solaris2.7 --prefix=xxx
4150 @heading @anchor{sparc-sun-solaris27}sparc-sun-solaris2.7
4152 Sun patch 107058-01 (1999-01-13) for Solaris 7/SPARC triggers a bug in
4153 the dynamic linker. This problem (Sun bug 4210064) affects GCC 2.8
4154 and later, including all EGCS releases. Sun formerly recommended
4155 107058-01 for all Solaris 7 users, but around 1999-09-01 it started to
4156 recommend it only for people who use Sun's compilers.
4158 Here are some workarounds to this problem:
4161 Do not install Sun patch 107058-01 until after Sun releases a
4162 complete patch for bug 4210064. This is the simplest course to take,
4163 unless you must also use Sun's C compiler. Unfortunately 107058-01
4164 is preinstalled on some new Solaris 7-based hosts, so you may have to
4168 Copy the original, unpatched Solaris 7
4169 @command{/usr/ccs/bin/as} into
4170 @command{/usr/local/libexec/gcc/sparc-sun-solaris2.7/3.4/as},
4171 adjusting the latter name to fit your local conventions and software
4175 Install Sun patch 106950-03 (1999-05-25) or later. Nobody with
4176 both 107058-01 and 106950-03 installed has reported the bug with GCC
4177 and Sun's dynamic linker. This last course of action is riskiest,
4178 for two reasons. First, you must install 106950 on all hosts that
4179 run code generated by GCC; it doesn't suffice to install it only on
4180 the hosts that run GCC itself. Second, Sun says that 106950-03 is
4181 only a partial fix for bug 4210064, but Sun doesn't know whether the
4182 partial fix is adequate for GCC@. Revision -08 or later should fix
4183 the bug. The current (as of 2004-05-23) revision is -24, and is included in
4184 the Solaris 7 Recommended Patch Cluster.
4187 GCC 3.3 triggers a bug in version 5.0 Alpha 03/27/98 of the Sun assembler,
4188 which causes a bootstrap failure when linking the 64-bit shared version of
4189 libgcc. A typical error message is:
4192 ld: fatal: relocation error: R_SPARC_32: file libgcc/sparcv9/_muldi3.o:
4193 symbol <unknown>: offset 0xffffffff7ec133e7 is non-aligned.
4196 This bug has been fixed in the final 5.0 version of the assembler.
4198 A similar problem was reported for version Sun WorkShop 6 99/08/18 of the
4199 Sun assembler, which causes a bootstrap failure with GCC 4.0.0:
4202 ld: fatal: relocation error: R_SPARC_DISP32:
4203 file .libs/libstdc++.lax/libsupc++convenience.a/vterminate.o:
4204 symbol <unknown>: offset 0xfccd33ad is non-aligned
4207 This bug has been fixed in more recent revisions of the assembler.
4212 @heading @anchor{sparc-x-linux}sparc-*-linux*
4214 GCC versions 3.0 and higher require binutils 2.11.2 and glibc 2.2.4
4215 or newer on this platform. All earlier binutils and glibc
4216 releases mishandled unaligned relocations on @code{sparc-*-*} targets.
4222 @heading @anchor{sparc64-x-solaris2}sparc64-*-solaris2*
4224 When configuring the GNU Multiple Precision Library (GMP) or the
4225 MPFR library, the canonical target triplet must be specified as
4226 the @command{build} parameter on the configure line. For example
4227 on a Solaris 7 system:
4230 % ./configure --build=sparc64-sun-solaris2.7 --prefix=xxx
4233 The following compiler flags must be specified in the configure
4234 step in order to bootstrap this target with the Sun compiler:
4237 % CC="cc -xarch=v9 -xildoff" @var{srcdir}/configure [@var{options}] [@var{target}]
4240 @option{-xarch=v9} specifies the SPARC-V9 architecture to the Sun toolchain
4241 and @option{-xildoff} turns off the incremental linker.
4246 @heading @anchor{sparcv9-x-solaris2}sparcv9-*-solaris2*
4248 This is a synonym for sparc64-*-solaris2*.
4253 @heading @anchor{x-x-vxworks}*-*-vxworks*
4254 Support for VxWorks is in flux. At present GCC supports @emph{only} the
4255 very recent VxWorks 5.5 (aka Tornado 2.2) release, and only on PowerPC@.
4256 We welcome patches for other architectures supported by VxWorks 5.5.
4257 Support for VxWorks AE would also be welcome; we believe this is merely
4258 a matter of writing an appropriate ``configlette'' (see below). We are
4259 not interested in supporting older, a.out or COFF-based, versions of
4262 VxWorks comes with an older version of GCC installed in
4263 @file{@var{$WIND_BASE}/host}; we recommend you do not overwrite it.
4264 Choose an installation @var{prefix} entirely outside @var{$WIND_BASE}.
4265 Before running @command{configure}, create the directories @file{@var{prefix}}
4266 and @file{@var{prefix}/bin}. Link or copy the appropriate assembler,
4267 linker, etc.@: into @file{@var{prefix}/bin}, and set your @var{PATH} to
4268 include that directory while running both @command{configure} and
4271 You must give @command{configure} the
4272 @option{--with-headers=@var{$WIND_BASE}/target/h} switch so that it can
4273 find the VxWorks system headers. Since VxWorks is a cross compilation
4274 target only, you must also specify @option{--target=@var{target}}.
4275 @command{configure} will attempt to create the directory
4276 @file{@var{prefix}/@var{target}/sys-include} and copy files into it;
4277 make sure the user running @command{configure} has sufficient privilege
4280 GCC's exception handling runtime requires a special ``configlette''
4281 module, @file{contrib/gthr_supp_vxw_5x.c}. Follow the instructions in
4282 that file to add the module to your kernel build. (Future versions of
4283 VxWorks will incorporate this module.)
4288 @heading @anchor{x86-64-x-x}x86_64-*-*, amd64-*-*
4290 GCC supports the x86-64 architecture implemented by the AMD64 processor
4291 (amd64-*-* is an alias for x86_64-*-*) on GNU/Linux, FreeBSD and NetBSD@.
4292 On GNU/Linux the default is a bi-arch compiler which is able to generate
4293 both 64-bit x86-64 and 32-bit x86 code (via the @option{-m32} switch).
4298 @heading @anchor{xtensa-x-elf}xtensa*-*-elf
4300 This target is intended for embedded Xtensa systems using the
4301 @samp{newlib} C library. It uses ELF but does not support shared
4302 objects. Designed-defined instructions specified via the
4303 Tensilica Instruction Extension (TIE) language are only supported
4304 through inline assembly.
4306 The Xtensa configuration information must be specified prior to
4307 building GCC@. The @file{include/xtensa-config.h} header
4308 file contains the configuration information. If you created your
4309 own Xtensa configuration with the Xtensa Processor Generator, the
4310 downloaded files include a customized copy of this header file,
4311 which you can use to replace the default header file.
4316 @heading @anchor{xtensa-x-linux}xtensa*-*-linux*
4318 This target is for Xtensa systems running GNU/Linux. It supports ELF
4319 shared objects and the GNU C library (glibc). It also generates
4320 position-independent code (PIC) regardless of whether the
4321 @option{-fpic} or @option{-fPIC} options are used. In other
4322 respects, this target is the same as the
4323 @uref{#xtensa*-*-elf,,@samp{xtensa*-*-elf}} target.
4328 @heading @anchor{windows}Microsoft Windows
4330 @subheading Intel 16-bit versions
4331 The 16-bit versions of Microsoft Windows, such as Windows 3.1, are not
4334 However, the 32-bit port has limited support for Microsoft
4335 Windows 3.11 in the Win32s environment, as a target only. See below.
4337 @subheading Intel 32-bit versions
4339 The 32-bit versions of Windows, including Windows 95, Windows NT, Windows
4340 XP, and Windows Vista, are supported by several different target
4341 platforms. These targets differ in which Windows subsystem they target
4342 and which C libraries are used.
4345 @item Cygwin @uref{#x-x-cygwin,,*-*-cygwin}: Cygwin provides a user-space
4346 Linux API emulation layer in the Win32 subsystem.
4347 @item Interix @uref{#x-x-interix,,*-*-interix}: The Interix subsystem
4348 provides native support for POSIX.
4349 @item MinGW @uref{#x-x-mingw32,,*-*-mingw32}: MinGW is a native GCC port for
4350 the Win32 subsystem that provides a subset of POSIX.
4351 @item MKS i386-pc-mks: NuTCracker from MKS. See
4352 @uref{http://www.mkssoftware.com/} for more information.
4355 @subheading Intel 64-bit versions
4357 GCC contains support for x86-64 using the mingw-w64
4358 runtime library, available from @uref{http://mingw-w64.sourceforge.net/}.
4359 This library should be used with the target triple x86_64-pc-mingw32.
4361 Presently Windows for Itanium is not supported.
4363 @subheading Windows CE
4365 Windows CE is supported as a target only on ARM (arm-wince-pe), Hitachi
4366 SuperH (sh-wince-pe), and MIPS (mips-wince-pe).
4368 @subheading Other Windows Platforms
4370 GCC no longer supports Windows NT on the Alpha or PowerPC.
4372 GCC no longer supports the Windows POSIX subsystem. However, it does
4373 support the Interix subsystem. See above.
4375 Old target names including *-*-winnt and *-*-windowsnt are no longer used.
4377 PW32 (i386-pc-pw32) support was never completed, and the project seems to
4378 be inactive. See @uref{http://pw32.sourceforge.net/} for more information.
4380 UWIN support has been removed due to a lack of maintenance.
4385 @heading @anchor{x-x-cygwin}*-*-cygwin
4387 Ports of GCC are included with the
4388 @uref{http://www.cygwin.com/,,Cygwin environment}.
4390 GCC will build under Cygwin without modification; it does not build
4391 with Microsoft's C++ compiler and there are no plans to make it do so.
4393 Cygwin can be compiled with i?86-pc-cygwin.
4398 @heading @anchor{x-x-interix}*-*-interix
4400 The Interix target is used by OpenNT, Interix, Services For UNIX (SFU),
4401 and Subsystem for UNIX-based Applications (SUA). Applications compiled
4402 with this target run in the Interix subsystem, which is separate from
4403 the Win32 subsystem. This target was last known to work in GCC 3.3.
4405 For more information, see @uref{http://www.interix.com/}.
4410 @heading @anchor{x-x-mingw32}*-*-mingw32
4412 GCC will build with and support only MinGW runtime 3.12 and later.
4413 Earlier versions of headers are incompatible with the new default semantics
4414 of @code{extern inline} in @code{-std=c99} and @code{-std=gnu99} modes.
4419 @heading @anchor{os2}OS/2
4421 GCC does not currently support OS/2. However, Andrew Zabolotny has been
4422 working on a generic OS/2 port with pgcc. The current code can be found
4423 at @uref{http://www.goof.com/pcg/os2/,,http://www.goof.com/pcg/os2/}.
4428 @heading @anchor{older}Older systems
4430 GCC contains support files for many older (1980s and early
4431 1990s) Unix variants. For the most part, support for these systems
4432 has not been deliberately removed, but it has not been maintained for
4433 several years and may suffer from bitrot.
4435 Starting with GCC 3.1, each release has a list of ``obsoleted'' systems.
4436 Support for these systems is still present in that release, but
4437 @command{configure} will fail unless the @option{--enable-obsolete}
4438 option is given. Unless a maintainer steps forward, support for these
4439 systems will be removed from the next release of GCC@.
4441 Support for old systems as hosts for GCC can cause problems if the
4442 workarounds for compiler, library and operating system bugs affect the
4443 cleanliness or maintainability of the rest of GCC@. In some cases, to
4444 bring GCC up on such a system, if still possible with current GCC, may
4445 require first installing an old version of GCC which did work on that
4446 system, and using it to compile a more recent GCC, to avoid bugs in the
4447 vendor compiler. Old releases of GCC 1 and GCC 2 are available in the
4448 @file{old-releases} directory on the @uref{../mirrors.html,,GCC mirror
4449 sites}. Header bugs may generally be avoided using
4450 @command{fixincludes}, but bugs or deficiencies in libraries and the
4451 operating system may still cause problems.
4453 Support for older systems as targets for cross-compilation is less
4454 problematic than support for them as hosts for GCC; if an enthusiast
4455 wishes to make such a target work again (including resurrecting any of
4456 the targets that never worked with GCC 2, starting from the last
4457 version before they were removed), patches
4458 @uref{../contribute.html,,following the usual requirements} would be
4459 likely to be accepted, since they should not affect the support for more
4462 For some systems, old versions of GNU binutils may also be useful,
4463 and are available from @file{pub/binutils/old-releases} on
4464 @uref{http://sourceware.org/mirrors.html,,sourceware.org mirror sites}.
4466 Some of the information on specific systems above relates to
4467 such older systems, but much of the information
4468 about GCC on such systems (which may no longer be applicable to
4469 current GCC) is to be found in the GCC texinfo manual.
4474 @heading @anchor{elf}all ELF targets (SVR4, Solaris 2, etc.)
4476 C++ support is significantly better on ELF targets if you use the
4477 @uref{./configure.html#with-gnu-ld,,GNU linker}; duplicate copies of
4478 inlines, vtables and template instantiations will be discarded
4487 @uref{./index.html,,Return to the GCC Installation page}
4491 @c ***Old documentation******************************************************
4493 @include install-old.texi
4499 @uref{./index.html,,Return to the GCC Installation page}
4503 @c ***GFDL********************************************************************
4511 @uref{./index.html,,Return to the GCC Installation page}
4515 @c ***************************************************************************
4516 @c Part 6 The End of the Document
4518 @comment node-name, next, previous, up
4519 @node Concept Index, , GNU Free Documentation License, Top
4523 @unnumbered Concept Index