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.
350 @heading Tools/packages necessary for modifying GCC
352 @item autoconf version 2.59
353 @itemx GNU m4 version 1.4 (or later)
355 Necessary when modifying @file{configure.ac}, @file{aclocal.m4}, etc.@:
356 to regenerate @file{configure} and @file{config.in} files.
358 @item automake version 1.9.6
360 Necessary when modifying a @file{Makefile.am} file to regenerate its
361 associated @file{Makefile.in}.
363 Much of GCC does not use automake, so directly edit the @file{Makefile.in}
364 file. Specifically this applies to the @file{gcc}, @file{intl},
365 @file{libcpp}, @file{libiberty}, @file{libobjc} directories as well
366 as any of their subdirectories.
368 For directories that use automake, GCC requires the latest release in
369 the 1.9.x series, which is currently 1.9.6. When regenerating a directory
370 to a newer version, please update all the directories using an older 1.9.x
371 to the latest released version.
373 @item gettext version 0.14.5 (or later)
375 Needed to regenerate @file{gcc.pot}.
377 @item gperf version 2.7.2 (or later)
379 Necessary when modifying @command{gperf} input files, e.g.@:
380 @file{gcc/cp/cfns.gperf} to regenerate its associated header file, e.g.@:
381 @file{gcc/cp/cfns.h}.
387 Necessary to run the GCC testsuite; see the section on testing for details.
389 @item autogen version 5.5.4 (or later) and
390 @itemx guile version 1.4.1 (or later)
392 Necessary to regenerate @file{fixinc/fixincl.x} from
393 @file{fixinc/inclhack.def} and @file{fixinc/*.tpl}.
395 Necessary to run @samp{make check} for @file{fixinc}.
397 Necessary to regenerate the top level @file{Makefile.in} file from
398 @file{Makefile.tpl} and @file{Makefile.def}.
400 @item Flex version 2.5.4 (or later)
402 Necessary when modifying @file{*.l} files.
404 Necessary to build GCC during development because the generated output
405 files are not included in the SVN repository. They are included in
408 @item Texinfo version 4.7 (or later)
410 Necessary for running @command{makeinfo} when modifying @file{*.texi}
411 files to test your changes.
413 Necessary for running @command{make dvi} or @command{make pdf} to
414 create printable documentation in DVI or PDF format. Texinfo version
415 4.8 or later is required for @command{make pdf}.
417 Necessary to build GCC documentation during development because the
418 generated output files are not included in the SVN repository. They are
419 included in releases.
421 @item @TeX{} (any working version)
423 Necessary for running @command{texi2dvi} and @command{texi2pdf}, which
424 are used when running @command{make dvi} or @command{make pdf} to create
425 DVI or PDF files, respectively.
427 @item SVN (any version)
428 @itemx SSH (any version)
430 Necessary to access the SVN repository. Public releases and weekly
431 snapshots of the development sources are also available via FTP@.
433 @item Perl version 5.6.1 (or later)
435 Necessary when regenerating @file{Makefile} dependencies in libiberty.
436 Necessary when regenerating @file{libiberty/functions.texi}.
437 Necessary when generating manpages from Texinfo manuals.
438 Necessary when targetting Darwin, building libstdc++,
439 and not using @option{--disable-symvers}.
440 Used by various scripts to generate some files included in SVN (mainly
441 Unicode-related and rarely changing) from source tables.
443 @item GNU diffutils version 2.7 (or later)
445 Useful when submitting patches for the GCC source code.
447 @item patch version 2.5.4 (or later)
449 Necessary when applying patches, created with @command{diff}, to one's
455 If you wish to modify @file{.java} files in libjava, you will need to
456 configure with @option{--enable-java-maintainer-mode}, and you will need
457 to have executables named @command{ecj1} and @command{gjavah} in your path.
458 The @command{ecj1} executable should run the Eclipse Java compiler via
459 the GCC-specific entry point. You can download a suitable jar from
460 @uref{ftp://sourceware.org/pub/java/}, or by running the script
461 @command{contrib/download_ecj}.
463 @item antlr.jar version 2.7.1 (or later)
466 If you wish to build the @command{gjdoc} binary in libjava, you will
467 need to have an @file{antlr.jar} library available. The library is
468 searched in system locations but can be configured with
469 @option{--with-antlr-jar=} instead. When configuring with
470 @option{--enable-java-maintainer-mode}, you will need to have one of
471 the executables named @command{cantlr}, @command{runantlr} or
472 @command{antlr} in your path.
481 @uref{./index.html,,Return to the GCC Installation page}
485 @c ***Downloading the source**************************************************
487 @comment node-name, next, previous, up
488 @node Downloading the source, Configuration, Prerequisites, Installing GCC
492 @chapter Downloading GCC
494 @cindex Downloading GCC
495 @cindex Downloading the Source
497 GCC is distributed via @uref{http://gcc.gnu.org/svn.html,,SVN} and FTP
498 tarballs compressed with @command{gzip} or
499 @command{bzip2}. It is possible to download a full distribution or specific
502 Please refer to the @uref{http://gcc.gnu.org/releases.html,,releases web page}
503 for information on how to obtain GCC@.
505 The full distribution includes the C, C++, Objective-C, Fortran, Java,
506 and Ada (in the case of GCC 3.1 and later) compilers. The full
507 distribution also includes runtime libraries for C++, Objective-C,
508 Fortran, and Java. In GCC 3.0 and later versions, the GNU compiler
509 testsuites are also included in the full distribution.
511 If you choose to download specific components, you must download the core
512 GCC distribution plus any language specific distributions you wish to
513 use. The core distribution includes the C language front end as well as the
514 shared components. Each language has a tarball which includes the language
515 front end as well as the language runtime (when appropriate).
517 Unpack the core distribution as well as any language specific
518 distributions in the same directory.
520 If you also intend to build binutils (either to upgrade an existing
521 installation or for use in place of the corresponding tools of your
522 OS), unpack the binutils distribution either in the same directory or
523 a separate one. In the latter case, add symbolic links to any
524 components of the binutils you intend to build alongside the compiler
525 (@file{bfd}, @file{binutils}, @file{gas}, @file{gprof}, @file{ld},
526 @file{opcodes}, @dots{}) to the directory containing the GCC sources.
528 Likewise, the GMP and MPFR libraries can be automatically built together
529 with GCC. Unpack the GMP and/or MPFR source distributions in the
530 directory containing the GCC sources and rename their directories to
531 @file{gmp} and @file{mpfr}, respectively (or use symbolic links with the
539 @uref{./index.html,,Return to the GCC Installation page}
543 @c ***Configuration***********************************************************
545 @comment node-name, next, previous, up
546 @node Configuration, Building, Downloading the source, Installing GCC
550 @chapter Installing GCC: Configuration
552 @cindex Configuration
553 @cindex Installing GCC: Configuration
555 Like most GNU software, GCC must be configured before it can be built.
556 This document describes the recommended configuration procedure
557 for both native and cross targets.
559 We use @var{srcdir} to refer to the toplevel source directory for
560 GCC; we use @var{objdir} to refer to the toplevel build/object directory.
562 If you obtained the sources via SVN, @var{srcdir} must refer to the top
563 @file{gcc} directory, the one where the @file{MAINTAINERS} can be found,
564 and not its @file{gcc} subdirectory, otherwise the build will fail.
566 If either @var{srcdir} or @var{objdir} is located on an automounted NFS
567 file system, the shell's built-in @command{pwd} command will return
568 temporary pathnames. Using these can lead to various sorts of build
569 problems. To avoid this issue, set the @env{PWDCMD} environment
570 variable to an automounter-aware @command{pwd} command, e.g.,
571 @command{pawd} or @samp{amq -w}, during the configuration and build
574 First, we @strong{highly} recommend that GCC be built into a
575 separate directory than the sources which does @strong{not} reside
576 within the source tree. This is how we generally build GCC; building
577 where @var{srcdir} == @var{objdir} should still work, but doesn't
578 get extensive testing; building where @var{objdir} is a subdirectory
579 of @var{srcdir} is unsupported.
581 If you have previously built GCC in the same directory for a
582 different target machine, do @samp{make distclean} to delete all files
583 that might be invalid. One of the files this deletes is @file{Makefile};
584 if @samp{make distclean} complains that @file{Makefile} does not exist
585 or issues a message like ``don't know how to make distclean'' it probably
586 means that the directory is already suitably clean. However, with the
587 recommended method of building in a separate @var{objdir}, you should
588 simply use a different @var{objdir} for each target.
590 Second, when configuring a native system, either @command{cc} or
591 @command{gcc} must be in your path or you must set @env{CC} in
592 your environment before running configure. Otherwise the configuration
596 Note that the bootstrap compiler and the resulting GCC must be link
597 compatible, else the bootstrap will fail with linker errors about
598 incompatible object file formats. Several multilibed targets are
599 affected by this requirement, see
601 @ref{Specific, host/target specific installation notes}.
604 @uref{specific.html,,host/target specific installation notes}.
613 % @var{srcdir}/configure [@var{options}] [@var{target}]
616 @heading Distributor options
618 If you will be distributing binary versions of GCC, with modifications
619 to the source code, you should use the options described in this
620 section to make clear that your version contains modifications.
623 @item --with-pkgversion=@var{version}
624 Specify a string that identifies your package. You may wish
625 to include a build number or build date. This version string will be
626 included in the output of @command{gcc --version}. This suffix does
627 not replace the default version string, only the @samp{GCC} part.
629 The default value is @samp{GCC}.
631 @item --with-bugurl=@var{url}
632 Specify the URL that users should visit if they wish to report a bug.
633 You are of course welcome to forward bugs reported to you to the FSF,
634 if you determine that they are not bugs in your modifications.
636 The default value refers to the FSF's GCC bug tracker.
640 @heading Target specification
643 GCC has code to correctly determine the correct value for @var{target}
644 for nearly all native systems. Therefore, we highly recommend you not
645 provide a configure target when configuring a native compiler.
648 @var{target} must be specified as @option{--target=@var{target}}
649 when configuring a cross compiler; examples of valid targets would be
650 m68k-elf, sh-elf, etc.
653 Specifying just @var{target} instead of @option{--target=@var{target}}
654 implies that the host defaults to @var{target}.
658 @heading Options specification
660 Use @var{options} to override several configure time options for
661 GCC@. A list of supported @var{options} follows; @samp{configure
662 --help} may list other options, but those not listed below may not
663 work and should not normally be used.
665 Note that each @option{--enable} option has a corresponding
666 @option{--disable} option and that each @option{--with} option has a
667 corresponding @option{--without} option.
670 @item --prefix=@var{dirname}
671 Specify the toplevel installation
672 directory. This is the recommended way to install the tools into a directory
673 other than the default. The toplevel installation directory defaults to
676 We @strong{highly} recommend against @var{dirname} being the same or a
677 subdirectory of @var{objdir} or vice versa. If specifying a directory
678 beneath a user's home directory tree, some shells will not expand
679 @var{dirname} correctly if it contains the @samp{~} metacharacter; use
682 The following standard @command{autoconf} options are supported. Normally you
683 should not need to use these options.
685 @item --exec-prefix=@var{dirname}
686 Specify the toplevel installation directory for architecture-dependent
687 files. The default is @file{@var{prefix}}.
689 @item --bindir=@var{dirname}
690 Specify the installation directory for the executables called by users
691 (such as @command{gcc} and @command{g++}). The default is
692 @file{@var{exec-prefix}/bin}.
694 @item --libdir=@var{dirname}
695 Specify the installation directory for object code libraries and
696 internal data files of GCC@. The default is @file{@var{exec-prefix}/lib}.
698 @item --libexecdir=@var{dirname}
699 Specify the installation directory for internal executables of GCC@.
700 The default is @file{@var{exec-prefix}/libexec}.
702 @item --with-slibdir=@var{dirname}
703 Specify the installation directory for the shared libgcc library. The
704 default is @file{@var{libdir}}.
706 @item --infodir=@var{dirname}
707 Specify the installation directory for documentation in info format.
708 The default is @file{@var{prefix}/info}.
710 @item --datadir=@var{dirname}
711 Specify the installation directory for some architecture-independent
712 data files referenced by GCC@. The default is @file{@var{prefix}/share}.
714 @item --mandir=@var{dirname}
715 Specify the installation directory for manual pages. The default is
716 @file{@var{prefix}/man}. (Note that the manual pages are only extracts from
717 the full GCC manuals, which are provided in Texinfo format. The manpages
718 are derived by an automatic conversion process from parts of the full
721 @item --with-gxx-include-dir=@var{dirname}
723 the installation directory for G++ header files. The default depends
724 on other configuration options, and differs between cross and native
729 @item --program-prefix=@var{prefix}
730 GCC supports some transformations of the names of its programs when
731 installing them. This option prepends @var{prefix} to the names of
732 programs to install in @var{bindir} (see above). For example, specifying
733 @option{--program-prefix=foo-} would result in @samp{gcc}
734 being installed as @file{/usr/local/bin/foo-gcc}.
736 @item --program-suffix=@var{suffix}
737 Appends @var{suffix} to the names of programs to install in @var{bindir}
738 (see above). For example, specifying @option{--program-suffix=-3.1}
739 would result in @samp{gcc} being installed as
740 @file{/usr/local/bin/gcc-3.1}.
742 @item --program-transform-name=@var{pattern}
743 Applies the @samp{sed} script @var{pattern} to be applied to the names
744 of programs to install in @var{bindir} (see above). @var{pattern} has to
745 consist of one or more basic @samp{sed} editing commands, separated by
746 semicolons. For example, if you want the @samp{gcc} program name to be
747 transformed to the installed program @file{/usr/local/bin/myowngcc} and
748 the @samp{g++} program name to be transformed to
749 @file{/usr/local/bin/gspecial++} without changing other program names,
750 you could use the pattern
751 @option{--program-transform-name='s/^gcc$/myowngcc/; s/^g++$/gspecial++/'}
752 to achieve this effect.
754 All three options can be combined and used together, resulting in more
755 complex conversion patterns. As a basic rule, @var{prefix} (and
756 @var{suffix}) are prepended (appended) before further transformations
757 can happen with a special transformation script @var{pattern}.
759 As currently implemented, this option only takes effect for native
760 builds; cross compiler binaries' names are not transformed even when a
761 transformation is explicitly asked for by one of these options.
763 For native builds, some of the installed programs are also installed
764 with the target alias in front of their name, as in
765 @samp{i686-pc-linux-gnu-gcc}. All of the above transformations happen
766 before the target alias is prepended to the name---so, specifying
767 @option{--program-prefix=foo-} and @option{program-suffix=-3.1}, the
768 resulting binary would be installed as
769 @file{/usr/local/bin/i686-pc-linux-gnu-foo-gcc-3.1}.
771 As a last shortcoming, none of the installed Ada programs are
772 transformed yet, which will be fixed in some time.
774 @item --with-local-prefix=@var{dirname}
776 installation directory for local include files. The default is
777 @file{/usr/local}. Specify this option if you want the compiler to
778 search directory @file{@var{dirname}/include} for locally installed
779 header files @emph{instead} of @file{/usr/local/include}.
781 You should specify @option{--with-local-prefix} @strong{only} if your
782 site has a different convention (not @file{/usr/local}) for where to put
785 The default value for @option{--with-local-prefix} is @file{/usr/local}
786 regardless of the value of @option{--prefix}. Specifying
787 @option{--prefix} has no effect on which directory GCC searches for
788 local header files. This may seem counterintuitive, but actually it is
791 The purpose of @option{--prefix} is to specify where to @emph{install
792 GCC}. The local header files in @file{/usr/local/include}---if you put
793 any in that directory---are not part of GCC@. They are part of other
794 programs---perhaps many others. (GCC installs its own header files in
795 another directory which is based on the @option{--prefix} value.)
797 Both the local-prefix include directory and the GCC-prefix include
798 directory are part of GCC's ``system include'' directories. Although these
799 two directories are not fixed, they need to be searched in the proper
800 order for the correct processing of the include_next directive. The
801 local-prefix include directory is searched before the GCC-prefix
802 include directory. Another characteristic of system include directories
803 is that pedantic warnings are turned off for headers in these directories.
805 Some autoconf macros add @option{-I @var{directory}} options to the
806 compiler command line, to ensure that directories containing installed
807 packages' headers are searched. When @var{directory} is one of GCC's
808 system include directories, GCC will ignore the option so that system
809 directories continue to be processed in the correct order. This
810 may result in a search order different from what was specified but the
811 directory will still be searched.
813 GCC automatically searches for ordinary libraries using
814 @env{GCC_EXEC_PREFIX}. Thus, when the same installation prefix is
815 used for both GCC and packages, GCC will automatically search for
816 both headers and libraries. This provides a configuration that is
817 easy to use. GCC behaves in a manner similar to that when it is
818 installed as a system compiler in @file{/usr}.
820 Sites that need to install multiple versions of GCC may not want to
821 use the above simple configuration. It is possible to use the
822 @option{--program-prefix}, @option{--program-suffix} and
823 @option{--program-transform-name} options to install multiple versions
824 into a single directory, but it may be simpler to use different prefixes
825 and the @option{--with-local-prefix} option to specify the location of the
826 site-specific files for each version. It will then be necessary for
827 users to specify explicitly the location of local site libraries
828 (e.g., with @env{LIBRARY_PATH}).
830 The same value can be used for both @option{--with-local-prefix} and
831 @option{--prefix} provided it is not @file{/usr}. This can be used
832 to avoid the default search of @file{/usr/local/include}.
834 @strong{Do not} specify @file{/usr} as the @option{--with-local-prefix}!
835 The directory you use for @option{--with-local-prefix} @strong{must not}
836 contain any of the system's standard header files. If it did contain
837 them, certain programs would be miscompiled (including GNU Emacs, on
838 certain targets), because this would override and nullify the header
839 file corrections made by the @command{fixincludes} script.
841 Indications are that people who use this option use it based on mistaken
842 ideas of what it is for. People use it as if it specified where to
843 install part of GCC@. Perhaps they make this assumption because
844 installing GCC creates the directory.
846 @item --enable-shared[=@var{package}[,@dots{}]]
847 Build shared versions of libraries, if shared libraries are supported on
848 the target platform. Unlike GCC 2.95.x and earlier, shared libraries
849 are enabled by default on all platforms that support shared libraries.
851 If a list of packages is given as an argument, build shared libraries
852 only for the listed packages. For other packages, only static libraries
853 will be built. Package names currently recognized in the GCC tree are
854 @samp{libgcc} (also known as @samp{gcc}), @samp{libstdc++} (not
855 @samp{libstdc++-v3}), @samp{libffi}, @samp{zlib}, @samp{boehm-gc},
856 @samp{ada}, @samp{libada}, @samp{libjava} and @samp{libobjc}.
857 Note @samp{libiberty} does not support shared libraries at all.
859 Use @option{--disable-shared} to build only static libraries. Note that
860 @option{--disable-shared} does not accept a list of package names as
861 argument, only @option{--enable-shared} does.
863 @item @anchor{with-gnu-as}--with-gnu-as
864 Specify that the compiler should assume that the
865 assembler it finds is the GNU assembler. However, this does not modify
866 the rules to find an assembler and will result in confusion if the
867 assembler found is not actually the GNU assembler. (Confusion may also
868 result if the compiler finds the GNU assembler but has not been
869 configured with @option{--with-gnu-as}.) If you have more than one
870 assembler installed on your system, you may want to use this option in
871 connection with @option{--with-as=@var{pathname}} or
872 @option{--with-build-time-tools=@var{pathname}}.
874 The following systems are the only ones where it makes a difference
875 whether you use the GNU assembler. On any other system,
876 @option{--with-gnu-as} has no effect.
879 @item @samp{hppa1.0-@var{any}-@var{any}}
880 @item @samp{hppa1.1-@var{any}-@var{any}}
881 @item @samp{sparc-sun-solaris2.@var{any}}
882 @item @samp{sparc64-@var{any}-solaris2.@var{any}}
885 @item @anchor{with-as}--with-as=@var{pathname}
886 Specify that the compiler should use the assembler pointed to by
887 @var{pathname}, rather than the one found by the standard rules to find
888 an assembler, which are:
891 Unless GCC is being built with a cross compiler, check the
892 @file{@var{libexec}/gcc/@var{target}/@var{version}} directory.
893 @var{libexec} defaults to @file{@var{exec-prefix}/libexec};
894 @var{exec-prefix} defaults to @var{prefix}, which
895 defaults to @file{/usr/local} unless overridden by the
896 @option{--prefix=@var{pathname}} switch described above. @var{target}
897 is the target system triple, such as @samp{sparc-sun-solaris2.7}, and
898 @var{version} denotes the GCC version, such as 3.0.
901 If the target system is the same that you are building on, check
902 operating system specific directories (e.g.@: @file{/usr/ccs/bin} on
906 Check in the @env{PATH} for a tool whose name is prefixed by the
907 target system triple.
910 Check in the @env{PATH} for a tool whose name is not prefixed by the
911 target system triple, if the host and target system triple are
912 the same (in other words, we use a host tool if it can be used for
916 You may want to use @option{--with-as} if no assembler
917 is installed in the directories listed above, or if you have multiple
918 assemblers installed and want to choose one that is not found by the
921 @item @anchor{with-gnu-ld}--with-gnu-ld
922 Same as @uref{#with-gnu-as,,@option{--with-gnu-as}}
925 @item --with-ld=@var{pathname}
926 Same as @uref{#with-as,,@option{--with-as}}
930 Specify that stabs debugging
931 information should be used instead of whatever format the host normally
932 uses. Normally GCC uses the same debug format as the host system.
934 On MIPS based systems and on Alphas, you must specify whether you want
935 GCC to create the normal ECOFF debugging format, or to use BSD-style
936 stabs passed through the ECOFF symbol table. The normal ECOFF debug
937 format cannot fully handle languages other than C@. BSD stabs format can
938 handle other languages, but it only works with the GNU debugger GDB@.
940 Normally, GCC uses the ECOFF debugging format by default; if you
941 prefer BSD stabs, specify @option{--with-stabs} when you configure GCC@.
943 No matter which default you choose when you configure GCC, the user
944 can use the @option{-gcoff} and @option{-gstabs+} options to specify explicitly
945 the debug format for a particular compilation.
947 @option{--with-stabs} is meaningful on the ISC system on the 386, also, if
948 @option{--with-gas} is used. It selects use of stabs debugging
949 information embedded in COFF output. This kind of debugging information
950 supports C++ well; ordinary COFF debugging information does not.
952 @option{--with-stabs} is also meaningful on 386 systems running SVR4. It
953 selects use of stabs debugging information embedded in ELF output. The
954 C++ compiler currently (2.6.0) does not support the DWARF debugging
955 information normally used on 386 SVR4 platforms; stabs provide a
956 workable alternative. This requires gas and gdb, as the normal SVR4
957 tools can not generate or interpret stabs.
959 @item --disable-multilib
960 Specify that multiple target
961 libraries to support different target variants, calling
962 conventions, etc.@: should not be built. The default is to build a
963 predefined set of them.
965 Some targets provide finer-grained control over which multilibs are built
966 (e.g., @option{--disable-softfloat}):
972 fpu, 26bit, underscore, interwork, biendian, nofmult.
975 softfloat, m68881, m68000, m68020.
978 single-float, biendian, softfloat.
980 @item powerpc*-*-*, rs6000*-*-*
981 aix64, pthread, softfloat, powercpu, powerpccpu, powerpcos, biendian,
986 @item --with-multilib-list=@var{list}
987 @itemx --without-multilib-list
988 Specify what multilibs to build.
989 Currently only implemented for sh*-*-*.
991 @var{list} is a comma separated list of CPU names. These must be of the
992 form @code{sh*} or @code{m*} (in which case they match the compiler option
993 for that processor). The list should not contain any endian options -
994 these are handled by @option{--with-endian}.
996 If @var{list} is empty, then there will be no multilibs for extra
997 processors. The multilib for the secondary endian remains enabled.
999 As a special case, if an entry in the list starts with a @code{!}
1000 (exclamation point), then it is added to the list of excluded multilibs.
1001 Entries of this sort should be compatible with @samp{MULTILIB_EXCLUDES}
1002 (once the leading @code{!} has been stripped).
1004 If @option{--with-multilib-list} is not given, then a default set of
1005 multilibs is selected based on the value of @option{--target}. This is
1006 usually the complete set of libraries, but some targets imply a more
1009 Example 1: to configure a compiler for SH4A only, but supporting both
1010 endians, with little endian being the default:
1012 --with-cpu=sh4a --with-endian=little,big --with-multilib-list=
1015 Example 2: to configure a compiler for both SH4A and SH4AL-DSP, but with
1016 only little endian SH4AL:
1018 --with-cpu=sh4a --with-endian=little,big --with-multilib-list=sh4al,!mb/m4al
1021 @item --with-endian=@var{endians}
1022 Specify what endians to use.
1023 Currently only implemented for sh*-*-*.
1025 @var{endians} may be one of the following:
1028 Use big endian exclusively.
1030 Use little endian exclusively.
1032 Use big endian by default. Provide a multilib for little endian.
1034 Use little endian by default. Provide a multilib for big endian.
1037 @item --enable-threads
1038 Specify that the target
1039 supports threads. This affects the Objective-C compiler and runtime
1040 library, and exception handling for other languages like C++ and Java.
1041 On some systems, this is the default.
1043 In general, the best (and, in many cases, the only known) threading
1044 model available will be configured for use. Beware that on some
1045 systems, GCC has not been taught what threading models are generally
1046 available for the system. In this case, @option{--enable-threads} is an
1047 alias for @option{--enable-threads=single}.
1049 @item --disable-threads
1050 Specify that threading support should be disabled for the system.
1051 This is an alias for @option{--enable-threads=single}.
1053 @item --enable-threads=@var{lib}
1055 @var{lib} is the thread support library. This affects the Objective-C
1056 compiler and runtime library, and exception handling for other languages
1057 like C++ and Java. The possibilities for @var{lib} are:
1065 Ada tasking support. For non-Ada programs, this setting is equivalent
1066 to @samp{single}. When used in conjunction with the Ada run time, it
1067 causes GCC to use the same thread primitives as Ada uses. This option
1068 is necessary when using both Ada and the back end exception handling,
1069 which is the default for most Ada targets.
1071 Generic MACH thread support, known to work on NeXTSTEP@. (Please note
1072 that the file needed to support this configuration, @file{gthr-mach.h}, is
1073 missing and thus this setting will cause a known bootstrap failure.)
1075 This is an alias for @samp{single}.
1077 Generic POSIX/Unix98 thread support.
1079 Generic POSIX/Unix95 thread support.
1081 RTEMS thread support.
1083 Disable thread support, should work for all platforms.
1085 Sun Solaris 2 thread support.
1087 VxWorks thread support.
1089 Microsoft Win32 API thread support.
1091 Novell Kernel Services thread support.
1095 Specify that the target supports TLS (Thread Local Storage). Usually
1096 configure can correctly determine if TLS is supported. In cases where
1097 it guesses incorrectly, TLS can be explicitly enabled or disabled with
1098 @option{--enable-tls} or @option{--disable-tls}. This can happen if
1099 the assembler supports TLS but the C library does not, or if the
1100 assumptions made by the configure test are incorrect.
1103 Specify that the target does not support TLS.
1104 This is an alias for @option{--enable-tls=no}.
1106 @item --with-cpu=@var{cpu}
1107 @itemx --with-cpu-32=@var{cpu}
1108 @itemx --with-cpu-64=@var{cpu}
1109 Specify which cpu variant the compiler should generate code for by default.
1110 @var{cpu} will be used as the default value of the @option{-mcpu=} switch.
1111 This option is only supported on some targets, including ARM, i386, M68k,
1112 PowerPC, and SPARC@. The @option{--with-cpu-32} and
1113 @option{--with-cpu-64} options specify separate default CPUs for
1114 32-bit and 64-bit modes; these options are only supported for i386 and
1117 @item --with-schedule=@var{cpu}
1118 @itemx --with-arch=@var{cpu}
1119 @itemx --with-arch-32=@var{cpu}
1120 @itemx --with-arch-64=@var{cpu}
1121 @itemx --with-tune=@var{cpu}
1122 @itemx --with-tune-32=@var{cpu}
1123 @itemx --with-tune-64=@var{cpu}
1124 @itemx --with-abi=@var{abi}
1125 @itemx --with-fpu=@var{type}
1126 @itemx --with-float=@var{type}
1127 These configure options provide default values for the @option{-mschedule=},
1128 @option{-march=}, @option{-mtune=}, @option{-mabi=}, and @option{-mfpu=}
1129 options and for @option{-mhard-float} or @option{-msoft-float}. As with
1130 @option{--with-cpu}, which switches will be accepted and acceptable values
1131 of the arguments depend on the target.
1133 @item --with-mode=@var{mode}
1134 Specify if the compiler should default to @option{-marm} or @option{-mthumb}.
1135 This option is only supported on ARM targets.
1137 @item --with-divide=@var{type}
1138 Specify how the compiler should generate code for checking for
1139 division by zero. This option is only supported on the MIPS target.
1140 The possibilities for @var{type} are:
1143 Division by zero checks use conditional traps (this is the default on
1144 systems that support conditional traps).
1146 Division by zero checks use the break instruction.
1149 @c If you make --with-llsc the default for additional targets,
1150 @c update the --with-llsc description in the MIPS section below.
1153 On MIPS targets, make @option{-mllsc} the default when no
1154 @option{-mno-lsc} option is passed. This is the default for
1155 Linux-based targets, as the kernel will emulate them if the ISA does
1158 @item --without-llsc
1159 On MIPS targets, make @option{-mno-llsc} the default when no
1160 @option{-mllsc} option is passed.
1162 @item --with-mips-plt
1163 On MIPS targets, make use of copy relocations and PLTs.
1164 These features are extensions to the traditional
1165 SVR4-based MIPS ABIs and require support from GNU binutils
1166 and the runtime C library.
1168 @item --enable-__cxa_atexit
1169 Define if you want to use __cxa_atexit, rather than atexit, to
1170 register C++ destructors for local statics and global objects.
1171 This is essential for fully standards-compliant handling of
1172 destructors, but requires __cxa_atexit in libc. This option is currently
1173 only available on systems with GNU libc. When enabled, this will cause
1174 @option{-fuse-cxa-atexit} to be passed by default.
1176 @item --enable-target-optspace
1178 libraries should be optimized for code space instead of code speed.
1179 This is the default for the m32r platform.
1182 Specify that a user visible @command{cpp} program should not be installed.
1184 @item --with-cpp-install-dir=@var{dirname}
1185 Specify that the user visible @command{cpp} program should be installed
1186 in @file{@var{prefix}/@var{dirname}/cpp}, in addition to @var{bindir}.
1188 @item --enable-initfini-array
1189 Force the use of sections @code{.init_array} and @code{.fini_array}
1190 (instead of @code{.init} and @code{.fini}) for constructors and
1191 destructors. Option @option{--disable-initfini-array} has the
1192 opposite effect. If neither option is specified, the configure script
1193 will try to guess whether the @code{.init_array} and
1194 @code{.fini_array} sections are supported and, if they are, use them.
1196 @item --enable-maintainer-mode
1197 The build rules that
1198 regenerate the GCC master message catalog @file{gcc.pot} are normally
1199 disabled. This is because it can only be rebuilt if the complete source
1200 tree is present. If you have changed the sources and want to rebuild the
1201 catalog, configuring with @option{--enable-maintainer-mode} will enable
1202 this. Note that you need a recent version of the @code{gettext} tools
1205 @item --disable-bootstrap
1206 For a native build, the default configuration is to perform
1207 a 3-stage bootstrap of the compiler when @samp{make} is invoked,
1208 testing that GCC can compile itself correctly. If you want to disable
1209 this process, you can configure with @option{--disable-bootstrap}.
1211 @item --enable-bootstrap
1212 In special cases, you may want to perform a 3-stage build
1213 even if the target and host triplets are different.
1214 This could happen when the host can run code compiled for
1215 the target (e.g.@: host is i686-linux, target is i486-linux).
1216 Starting from GCC 4.2, to do this you have to configure explicitly
1217 with @option{--enable-bootstrap}.
1219 @item --enable-generated-files-in-srcdir
1220 Neither the .c and .h files that are generated from Bison and flex nor the
1221 info manuals and man pages that are built from the .texi files are present
1222 in the SVN development tree. When building GCC from that development tree,
1223 or from one of our snapshots, those generated files are placed in your
1224 build directory, which allows for the source to be in a readonly
1227 If you configure with @option{--enable-generated-files-in-srcdir} then those
1228 generated files will go into the source directory. This is mainly intended
1229 for generating release or prerelease tarballs of the GCC sources, since it
1230 is not a requirement that the users of source releases to have flex, Bison,
1233 @item --enable-version-specific-runtime-libs
1235 that runtime libraries should be installed in the compiler specific
1236 subdirectory (@file{@var{libdir}/gcc}) rather than the usual places. In
1237 addition, @samp{libstdc++}'s include files will be installed into
1238 @file{@var{libdir}} unless you overruled it by using
1239 @option{--with-gxx-include-dir=@var{dirname}}. Using this option is
1240 particularly useful if you intend to use several versions of GCC in
1241 parallel. This is currently supported by @samp{libgfortran},
1242 @samp{libjava}, @samp{libmudflap}, @samp{libstdc++}, and @samp{libobjc}.
1244 @item --enable-languages=@var{lang1},@var{lang2},@dots{}
1245 Specify that only a particular subset of compilers and
1246 their runtime libraries should be built. For a list of valid values for
1247 @var{langN} you can issue the following command in the
1248 @file{gcc} directory of your GCC source tree:@*
1250 grep language= */config-lang.in
1252 Currently, you can use any of the following:
1253 @code{all}, @code{ada}, @code{c}, @code{c++}, @code{fortran}, @code{java},
1254 @code{objc}, @code{obj-c++}.
1255 Building the Ada compiler has special requirements, see below.
1256 If you do not pass this flag, or specify the option @code{all}, then all
1257 default languages available in the @file{gcc} sub-tree will be configured.
1258 Ada and Objective-C++ are not default languages; the rest are.
1259 Re-defining @code{LANGUAGES} when calling @samp{make} @strong{does not}
1260 work anymore, as those language sub-directories might not have been
1263 @item --enable-stage1-languages=@var{lang1},@var{lang2},@dots{}
1264 Specify that a particular subset of compilers and their runtime
1265 libraries should be built with the system C compiler during stage 1 of
1266 the bootstrap process, rather than only in later stages with the
1267 bootstrapped C compiler. The list of valid values is the same as for
1268 @option{--enable-languages}, and the option @code{all} will select all
1269 of the languages enabled by @option{--enable-languages}. This option is
1270 primarily useful for GCC development; for instance, when a development
1271 version of the compiler cannot bootstrap due to compiler bugs, or when
1272 one is debugging front ends other than the C front end. When this
1273 option is used, one can then build the target libraries for the
1274 specified languages with the stage-1 compiler by using @command{make
1275 stage1-bubble all-target}, or run the testsuite on the stage-1 compiler
1276 for the specified languages using @command{make stage1-start check-gcc}.
1278 @item --disable-libada
1279 Specify that the run-time libraries and tools used by GNAT should not
1280 be built. This can be useful for debugging, or for compatibility with
1281 previous Ada build procedures, when it was required to explicitly
1282 do a @samp{make -C gcc gnatlib_and_tools}.
1284 @item --disable-libssp
1285 Specify that the run-time libraries for stack smashing protection
1286 should not be built.
1288 @item --disable-libgomp
1289 Specify that the run-time libraries used by GOMP should not be built.
1292 Specify that the compiler should
1293 use DWARF 2 debugging information as the default.
1295 @item --enable-targets=all
1296 @itemx --enable-targets=@var{target_list}
1297 Some GCC targets, e.g.@: powerpc64-linux, build bi-arch compilers.
1298 These are compilers that are able to generate either 64-bit or 32-bit
1299 code. Typically, the corresponding 32-bit target, e.g.@:
1300 powerpc-linux for powerpc64-linux, only generates 32-bit code. This
1301 option enables the 32-bit target to be a bi-arch compiler, which is
1302 useful when you want a bi-arch compiler that defaults to 32-bit, and
1303 you are building a bi-arch or multi-arch binutils in a combined tree.
1304 Currently, this option only affects sparc-linux, powerpc-linux and
1307 @item --enable-secureplt
1308 This option enables @option{-msecure-plt} by default for powerpc-linux.
1310 @xref{RS/6000 and PowerPC Options,, RS/6000 and PowerPC Options, gcc,
1311 Using the GNU Compiler Collection (GCC)},
1314 See ``RS/6000 and PowerPC Options'' in the main manual
1318 This option enables @option{-mcld} by default for 32-bit x86 targets.
1320 @xref{i386 and x86-64 Options,, i386 and x86-64 Options, gcc,
1321 Using the GNU Compiler Collection (GCC)},
1324 See ``i386 and x86-64 Options'' in the main manual
1327 @item --enable-win32-registry
1328 @itemx --enable-win32-registry=@var{key}
1329 @itemx --disable-win32-registry
1330 The @option{--enable-win32-registry} option enables Microsoft Windows-hosted GCC
1331 to look up installations paths in the registry using the following key:
1334 @code{HKEY_LOCAL_MACHINE\SOFTWARE\Free Software Foundation\@var{key}}
1337 @var{key} defaults to GCC version number, and can be overridden by the
1338 @option{--enable-win32-registry=@var{key}} option. Vendors and distributors
1339 who use custom installers are encouraged to provide a different key,
1340 perhaps one comprised of vendor name and GCC version number, to
1341 avoid conflict with existing installations. This feature is enabled
1342 by default, and can be disabled by @option{--disable-win32-registry}
1343 option. This option has no effect on the other hosts.
1346 Specify that the machine does not have a floating point unit. This
1347 option only applies to @samp{m68k-sun-sunos@var{n}}. On any other
1348 system, @option{--nfp} has no effect.
1350 @item --enable-werror
1351 @itemx --disable-werror
1352 @itemx --enable-werror=yes
1353 @itemx --enable-werror=no
1354 When you specify this option, it controls whether certain files in the
1355 compiler are built with @option{-Werror} in bootstrap stage2 and later.
1356 If you don't specify it, @option{-Werror} is turned on for the main
1357 development trunk. However it defaults to off for release branches and
1358 final releases. The specific files which get @option{-Werror} are
1359 controlled by the Makefiles.
1361 @item --enable-checking
1362 @itemx --enable-checking=@var{list}
1363 When you specify this option, the compiler is built to perform internal
1364 consistency checks of the requested complexity. This does not change the
1365 generated code, but adds error checking within the compiler. This will
1366 slow down the compiler and may only work properly if you are building
1367 the compiler with GCC@. This is @samp{yes} by default when building
1368 from SVN or snapshots, but @samp{release} for releases. The default
1369 for building the stage1 compiler is @samp{yes}. More control
1370 over the checks may be had by specifying @var{list}. The categories of
1371 checks available are @samp{yes} (most common checks
1372 @samp{assert,misc,tree,gc,rtlflag,runtime}), @samp{no} (no checks at
1373 all), @samp{all} (all but @samp{valgrind}), @samp{release} (cheapest
1374 checks @samp{assert,runtime}) or @samp{none} (same as @samp{no}).
1375 Individual checks can be enabled with these flags @samp{assert},
1376 @samp{df}, @samp{fold}, @samp{gc}, @samp{gcac} @samp{misc}, @samp{rtl},
1377 @samp{rtlflag}, @samp{runtime}, @samp{tree}, and @samp{valgrind}.
1379 The @samp{valgrind} check requires the external @command{valgrind}
1380 simulator, available from @uref{http://valgrind.org/}. The
1381 @samp{df}, @samp{rtl}, @samp{gcac} and @samp{valgrind} checks are very expensive.
1382 To disable all checking, @samp{--disable-checking} or
1383 @samp{--enable-checking=none} must be explicitly requested. Disabling
1384 assertions will make the compiler and runtime slightly faster but
1385 increase the risk of undetected internal errors causing wrong code to be
1388 @item --disable-stage1-checking
1389 @item --enable-stage1-checking
1390 @itemx --enable-stage1-checking=@var{list}
1391 If no @option{--enable-checking} option is specified the stage1
1392 compiler will be built with @samp{yes} checking enabled, otherwise
1393 the stage1 checking flags are the same as specified by
1394 @option{--enable-checking}. To build the stage1 compiler with
1395 different checking options use @option{--enable-stage1-checking}.
1396 The list of checking options is the same as for @option{--enable-checking}.
1397 If your system is too slow or too small to bootstrap a released compiler
1398 with checking for stage1 enabled, you can use @samp{--disable-stage1-checking}
1399 to disable checking for the stage1 compiler.
1401 @item --enable-coverage
1402 @itemx --enable-coverage=@var{level}
1403 With this option, the compiler is built to collect self coverage
1404 information, every time it is run. This is for internal development
1405 purposes, and only works when the compiler is being built with gcc. The
1406 @var{level} argument controls whether the compiler is built optimized or
1407 not, values are @samp{opt} and @samp{noopt}. For coverage analysis you
1408 want to disable optimization, for performance analysis you want to
1409 enable optimization. When coverage is enabled, the default level is
1410 without optimization.
1412 @item --enable-gather-detailed-mem-stats
1413 When this option is specified more detailed information on memory
1414 allocation is gathered. This information is printed when using
1415 @option{-fmem-report}.
1418 @itemx --with-gc=@var{choice}
1419 With this option you can specify the garbage collector implementation
1420 used during the compilation process. @var{choice} can be one of
1421 @samp{page} and @samp{zone}, where @samp{page} is the default.
1424 @itemx --disable-nls
1425 The @option{--enable-nls} option enables Native Language Support (NLS),
1426 which lets GCC output diagnostics in languages other than American
1427 English. Native Language Support is enabled by default if not doing a
1428 canadian cross build. The @option{--disable-nls} option disables NLS@.
1430 @item --with-included-gettext
1431 If NLS is enabled, the @option{--with-included-gettext} option causes the build
1432 procedure to prefer its copy of GNU @command{gettext}.
1434 @item --with-catgets
1435 If NLS is enabled, and if the host lacks @code{gettext} but has the
1436 inferior @code{catgets} interface, the GCC build procedure normally
1437 ignores @code{catgets} and instead uses GCC's copy of the GNU
1438 @code{gettext} library. The @option{--with-catgets} option causes the
1439 build procedure to use the host's @code{catgets} in this situation.
1441 @item --with-libiconv-prefix=@var{dir}
1442 Search for libiconv header files in @file{@var{dir}/include} and
1443 libiconv library files in @file{@var{dir}/lib}.
1445 @item --enable-obsolete
1446 Enable configuration for an obsoleted system. If you attempt to
1447 configure GCC for a system (build, host, or target) which has been
1448 obsoleted, and you do not specify this flag, configure will halt with an
1451 All support for systems which have been obsoleted in one release of GCC
1452 is removed entirely in the next major release, unless someone steps
1453 forward to maintain the port.
1455 @item --enable-decimal-float
1456 @itemx --enable-decimal-float=yes
1457 @itemx --enable-decimal-float=no
1458 @itemx --enable-decimal-float=bid
1459 @itemx --enable-decimal-float=dpd
1460 @itemx --disable-decimal-float
1461 Enable (or disable) support for the C decimal floating point extension
1462 that is in the IEEE 754-2008 standard. This is enabled by default only
1463 on PowerPC, i386, and x86_64 GNU/Linux systems. Other systems may also
1464 support it, but require the user to specifically enable it. You can
1465 optionally control which decimal floating point format is used (either
1466 @samp{bid} or @samp{dpd}). The @samp{bid} (binary integer decimal)
1467 format is default on i386 and x86_64 systems, and the @samp{dpd}
1468 (densely packed decimal) format is default on PowerPC systems.
1470 @item --enable-fixed-point
1471 @itemx --disable-fixed-point
1472 Enable (or disable) support for C fixed-point arithmetic.
1473 This option is enabled by default for some targets (such as MIPS) which
1474 have hardware-support for fixed-point operations. On other targets, you
1475 may enable this option manually.
1477 @item --with-long-double-128
1478 Specify if @code{long double} type should be 128-bit by default on selected
1479 GNU/Linux architectures. If using @code{--without-long-double-128},
1480 @code{long double} will be by default 64-bit, the same as @code{double} type.
1481 When neither of these configure options are used, the default will be
1482 128-bit @code{long double} when built against GNU C Library 2.4 and later,
1483 64-bit @code{long double} otherwise.
1485 @item --with-gmp=@var{pathname}
1486 @itemx --with-gmp-include=@var{pathname}
1487 @itemx --with-gmp-lib=@var{pathname}
1488 @itemx --with-mpfr=@var{pathname}
1489 @itemx --with-mpfr-include=@var{pathname}
1490 @itemx --with-mpfr-lib=@var{pathname}
1491 If you do not have GMP (the GNU Multiple Precision library) and the
1492 MPFR Libraries installed in a standard location and you want to build
1493 GCC, you can explicitly specify the directory where they are installed
1494 (@samp{--with-gmp=@var{gmpinstalldir}},
1495 @samp{--with-mpfr=@var{mpfrinstalldir}}). The
1496 @option{--with-gmp=@var{gmpinstalldir}} option is shorthand for
1497 @option{--with-gmp-lib=@var{gmpinstalldir}/lib} and
1498 @option{--with-gmp-include=@var{gmpinstalldir}/include}. Likewise the
1499 @option{--with-mpfr=@var{mpfrinstalldir}} option is shorthand for
1500 @option{--with-mpfr-lib=@var{mpfrinstalldir}/lib} and
1501 @option{--with-mpfr-include=@var{mpfrinstalldir}/include}. If these
1502 shorthand assumptions are not correct, you can use the explicit
1503 include and lib options directly.
1505 @item --with-ppl=@var{pathname}
1506 @itemx --with-ppl-include=@var{pathname}
1507 @itemx --with-ppl-lib=@var{pathname}
1508 @itemx --with-cloog=@var{pathname}
1509 @itemx --with-cloog-include=@var{pathname}
1510 @itemx --with-cloog-lib=@var{pathname}
1511 If you do not have PPL (the Parma Polyhedra Library) and the CLooG
1512 libraries installed in a standard location and you want to build GCC,
1513 you can explicitly specify the directory where they are installed
1514 (@samp{--with-ppl=@var{pplinstalldir}},
1515 @samp{--with-cloog=@var{clooginstalldir}}). The
1516 @option{--with-ppl=@var{pplinstalldir}} option is shorthand for
1517 @option{--with-ppl-lib=@var{pplinstalldir}/lib} and
1518 @option{--with-ppl-include=@var{pplinstalldir}/include}. Likewise the
1519 @option{--with-cloog=@var{clooginstalldir}} option is shorthand for
1520 @option{--with-cloog-lib=@var{clooginstalldir}/lib} and
1521 @option{--with-cloog-include=@var{clooginstalldir}/include}. If these
1522 shorthand assumptions are not correct, you can use the explicit
1523 include and lib options directly.
1525 @item --with-host-libstdcxx=@var{linker-args}
1526 If you are linking with a static copy of PPL, you can use this option
1527 to specify how the linker should find the standard C++ library used
1528 internally by PPL. Typical values of @var{linker-args} might be
1529 @samp{-lstdc++} or @samp{-Wl,-Bstatic,-lstdc++,-Bdynamic -lm}. If you are
1530 linking with a shared copy of PPL, you probably do not need this
1531 option; shared library dependencies will cause the linker to search
1532 for the standard C++ library automatically.
1534 @item --with-debug-prefix-map=@var{map}
1535 Convert source directory names using @option{-fdebug-prefix-map} when
1536 building runtime libraries. @samp{@var{map}} is a space-separated
1537 list of maps of the form @samp{@var{old}=@var{new}}.
1541 @subheading Cross-Compiler-Specific Options
1542 The following options only apply to building cross compilers.
1544 @item --with-sysroot
1545 @itemx --with-sysroot=@var{dir}
1546 Tells GCC to consider @var{dir} as the root of a tree that contains a
1547 (subset of) the root filesystem of the target operating system.
1548 Target system headers, libraries and run-time object files will be
1549 searched in there. The specified directory is not copied into the
1550 install tree, unlike the options @option{--with-headers} and
1551 @option{--with-libs} that this option obsoletes. The default value,
1552 in case @option{--with-sysroot} is not given an argument, is
1553 @option{$@{gcc_tooldir@}/sys-root}. If the specified directory is a
1554 subdirectory of @option{$@{exec_prefix@}}, then it will be found relative to
1555 the GCC binaries if the installation tree is moved.
1557 @item --with-build-sysroot
1558 @itemx --with-build-sysroot=@var{dir}
1559 Tells GCC to consider @var{dir} as the system root (see
1560 @option{--with-sysroot}) while building target libraries, instead of
1561 the directory specified with @option{--with-sysroot}. This option is
1562 only useful when you are already using @option{--with-sysroot}. You
1563 can use @option{--with-build-sysroot} when you are configuring with
1564 @option{--prefix} set to a directory that is different from the one in
1565 which you are installing GCC and your target libraries.
1567 This option affects the system root for the compiler used to build
1568 target libraries (which runs on the build system); it does not affect
1569 the compiler which is used to build GCC itself.
1571 @item --with-headers
1572 @itemx --with-headers=@var{dir}
1573 Deprecated in favor of @option{--with-sysroot}.
1574 Specifies that target headers are available when building a cross compiler.
1575 The @var{dir} argument specifies a directory which has the target include
1576 files. These include files will be copied into the @file{gcc} install
1577 directory. @emph{This option with the @var{dir} argument is required} when
1578 building a cross compiler, if @file{@var{prefix}/@var{target}/sys-include}
1579 doesn't pre-exist. If @file{@var{prefix}/@var{target}/sys-include} does
1580 pre-exist, the @var{dir} argument may be omitted. @command{fixincludes}
1581 will be run on these files to make them compatible with GCC@.
1583 @item --without-headers
1584 Tells GCC not use any target headers from a libc when building a cross
1585 compiler. When crossing to GNU/Linux, you need the headers so GCC
1586 can build the exception handling for libgcc.
1589 @itemx --with-libs=``@var{dir1} @var{dir2} @dots{} @var{dirN}''
1590 Deprecated in favor of @option{--with-sysroot}.
1591 Specifies a list of directories which contain the target runtime
1592 libraries. These libraries will be copied into the @file{gcc} install
1593 directory. If the directory list is omitted, this option has no
1597 Specifies that @samp{newlib} is
1598 being used as the target C library. This causes @code{__eprintf} to be
1599 omitted from @file{libgcc.a} on the assumption that it will be provided by
1602 @item --with-build-time-tools=@var{dir}
1603 Specifies where to find the set of target tools (assembler, linker, etc.)
1604 that will be used while building GCC itself. This option can be useful
1605 if the directory layouts are different between the system you are building
1606 GCC on, and the system where you will deploy it.
1608 For example, on an @samp{ia64-hp-hpux} system, you may have the GNU
1609 assembler and linker in @file{/usr/bin}, and the native tools in a
1610 different path, and build a toolchain that expects to find the
1611 native tools in @file{/usr/bin}.
1613 When you use this option, you should ensure that @var{dir} includes
1614 @command{ar}, @command{as}, @command{ld}, @command{nm},
1615 @command{ranlib} and @command{strip} if necessary, and possibly
1616 @command{objdump}. Otherwise, GCC may use an inconsistent set of
1620 @subheading Java-Specific Options
1622 The following option applies to the build of the Java front end.
1625 @item --disable-libgcj
1626 Specify that the run-time libraries
1627 used by GCJ should not be built. This is useful in case you intend
1628 to use GCJ with some other run-time, or you're going to install it
1629 separately, or it just happens not to build on your particular
1630 machine. In general, if the Java front end is enabled, the GCJ
1631 libraries will be enabled too, unless they're known to not work on
1632 the target platform. If GCJ is enabled but @samp{libgcj} isn't built, you
1633 may need to port it; in this case, before modifying the top-level
1634 @file{configure.in} so that @samp{libgcj} is enabled by default on this platform,
1635 you may use @option{--enable-libgcj} to override the default.
1639 The following options apply to building @samp{libgcj}.
1641 @subsubheading General Options
1644 @item --enable-java-maintainer-mode
1645 By default the @samp{libjava} build will not attempt to compile the
1646 @file{.java} source files to @file{.class}. Instead, it will use the
1647 @file{.class} files from the source tree. If you use this option you
1648 must have executables named @command{ecj1} and @command{gjavah} in your path
1649 for use by the build. You must use this option if you intend to
1650 modify any @file{.java} files in @file{libjava}.
1652 @item --with-java-home=@var{dirname}
1653 This @samp{libjava} option overrides the default value of the
1654 @samp{java.home} system property. It is also used to set
1655 @samp{sun.boot.class.path} to @file{@var{dirname}/lib/rt.jar}. By
1656 default @samp{java.home} is set to @file{@var{prefix}} and
1657 @samp{sun.boot.class.path} to
1658 @file{@var{datadir}/java/libgcj-@var{version}.jar}.
1660 @item --with-ecj-jar=@var{filename}
1661 This option can be used to specify the location of an external jar
1662 file containing the Eclipse Java compiler. A specially modified
1663 version of this compiler is used by @command{gcj} to parse
1664 @file{.java} source files. If this option is given, the
1665 @samp{libjava} build will create and install an @file{ecj1} executable
1666 which uses this jar file at runtime.
1668 If this option is not given, but an @file{ecj.jar} file is found in
1669 the topmost source tree at configure time, then the @samp{libgcj}
1670 build will create and install @file{ecj1}, and will also install the
1671 discovered @file{ecj.jar} into a suitable place in the install tree.
1673 If @file{ecj1} is not installed, then the user will have to supply one
1674 on his path in order for @command{gcj} to properly parse @file{.java}
1675 source files. A suitable jar is available from
1676 @uref{ftp://sourceware.org/pub/java/}.
1678 @item --disable-getenv-properties
1679 Don't set system properties from @env{GCJ_PROPERTIES}.
1681 @item --enable-hash-synchronization
1682 Use a global hash table for monitor locks. Ordinarily,
1683 @samp{libgcj}'s @samp{configure} script automatically makes
1684 the correct choice for this option for your platform. Only use
1685 this if you know you need the library to be configured differently.
1687 @item --enable-interpreter
1688 Enable the Java interpreter. The interpreter is automatically
1689 enabled by default on all platforms that support it. This option
1690 is really only useful if you want to disable the interpreter
1691 (using @option{--disable-interpreter}).
1693 @item --disable-java-net
1694 Disable java.net. This disables the native part of java.net only,
1695 using non-functional stubs for native method implementations.
1697 @item --disable-jvmpi
1698 Disable JVMPI support.
1700 @item --disable-libgcj-bc
1701 Disable BC ABI compilation of certain parts of libgcj. By default,
1702 some portions of libgcj are compiled with @option{-findirect-dispatch}
1703 and @option{-fno-indirect-classes}, allowing them to be overridden at
1706 If @option{--disable-libgcj-bc} is specified, libgcj is built without
1707 these options. This allows the compile-time linker to resolve
1708 dependencies when statically linking to libgcj. However it makes it
1709 impossible to override the affected portions of libgcj at run-time.
1711 @item --enable-reduced-reflection
1712 Build most of libgcj with @option{-freduced-reflection}. This reduces
1713 the size of libgcj at the expense of not being able to do accurate
1714 reflection on the classes it contains. This option is safe if you
1715 know that code using libgcj will never use reflection on the standard
1716 runtime classes in libgcj (including using serialization, RMI or CORBA).
1719 Enable runtime eCos target support.
1721 @item --without-libffi
1722 Don't use @samp{libffi}. This will disable the interpreter and JNI
1723 support as well, as these require @samp{libffi} to work.
1725 @item --enable-libgcj-debug
1726 Enable runtime debugging code.
1728 @item --enable-libgcj-multifile
1729 If specified, causes all @file{.java} source files to be
1730 compiled into @file{.class} files in one invocation of
1731 @samp{gcj}. This can speed up build time, but is more
1732 resource-intensive. If this option is unspecified or
1733 disabled, @samp{gcj} is invoked once for each @file{.java}
1734 file to compile into a @file{.class} file.
1736 @item --with-libiconv-prefix=DIR
1737 Search for libiconv in @file{DIR/include} and @file{DIR/lib}.
1739 @item --enable-sjlj-exceptions
1740 Force use of the @code{setjmp}/@code{longjmp}-based scheme for exceptions.
1741 @samp{configure} ordinarily picks the correct value based on the platform.
1742 Only use this option if you are sure you need a different setting.
1744 @item --with-system-zlib
1745 Use installed @samp{zlib} rather than that included with GCC@.
1747 @item --with-win32-nlsapi=ansi, unicows or unicode
1748 Indicates how MinGW @samp{libgcj} translates between UNICODE
1749 characters and the Win32 API@.
1751 @item --enable-java-home
1752 If enabled, this creates a JPackage compatible SDK environment during install.
1753 Note that if --enable-java-home is used, --with-arch-directory=ARCH must also
1756 @item --with-arch-directory=ARCH
1757 Specifies the name to use for the @file{jre/lib/ARCH} directory in the SDK
1758 environment created when --enable-java-home is passed. Typical names for this
1759 directory include i386, amd64, ia64, etc.
1761 @item --with-os-directory=DIR
1762 Specifies the OS directory for the SDK include directory. This is set to auto
1763 detect, and is typically 'linux'.
1765 @item --with-origin-name=NAME
1766 Specifies the JPackage origin name. This defaults to the 'gcj' in
1769 @item --with-arch-suffix=SUFFIX
1770 Specifies the suffix for the sdk directory. Defaults to the empty string.
1771 Examples include '.x86_64' in 'java-1.5.0-gcj-1.5.0.0.x86_64'.
1773 @item --with-jvm-root-dir=DIR
1774 Specifies where to install the SDK. Default is $(prefix)/lib/jvm.
1776 @item --with-jvm-jar-dir=DIR
1777 Specifies where to install jars. Default is $(prefix)/lib/jvm-exports.
1779 @item --with-python-dir=DIR
1780 Specifies where to install the Python modules used for aot-compile. DIR should
1781 not include the prefix used in installation. For example, if the Python modules
1782 are to be installed in /usr/lib/python2.5/site-packages, then
1783 --with-python-dir=/lib/python2.5/site-packages should be passed. If this is
1784 not specified, then the Python modules are installed in $(prefix)/share/python.
1786 @item --enable-aot-compile-rpm
1787 Adds aot-compile-rpm to the list of installed scripts.
1791 Use the single-byte @code{char} and the Win32 A functions natively,
1792 translating to and from UNICODE when using these functions. If
1793 unspecified, this is the default.
1796 Use the @code{WCHAR} and Win32 W functions natively. Adds
1797 @code{-lunicows} to @file{libgcj.spec} to link with @samp{libunicows}.
1798 @file{unicows.dll} needs to be deployed on Microsoft Windows 9X machines
1799 running built executables. @file{libunicows.a}, an open-source
1800 import library around Microsoft's @code{unicows.dll}, is obtained from
1801 @uref{http://libunicows.sourceforge.net/}, which also gives details
1802 on getting @file{unicows.dll} from Microsoft.
1805 Use the @code{WCHAR} and Win32 W functions natively. Does @emph{not}
1806 add @code{-lunicows} to @file{libgcj.spec}. The built executables will
1807 only run on Microsoft Windows NT and above.
1811 @subsubheading AWT-Specific Options
1815 Use the X Window System.
1817 @item --enable-java-awt=PEER(S)
1818 Specifies the AWT peer library or libraries to build alongside
1819 @samp{libgcj}. If this option is unspecified or disabled, AWT
1820 will be non-functional. Current valid values are @option{gtk} and
1821 @option{xlib}. Multiple libraries should be separated by a
1822 comma (i.e.@: @option{--enable-java-awt=gtk,xlib}).
1824 @item --enable-gtk-cairo
1825 Build the cairo Graphics2D implementation on GTK@.
1827 @item --enable-java-gc=TYPE
1828 Choose garbage collector. Defaults to @option{boehm} if unspecified.
1830 @item --disable-gtktest
1831 Do not try to compile and run a test GTK+ program.
1833 @item --disable-glibtest
1834 Do not try to compile and run a test GLIB program.
1836 @item --with-libart-prefix=PFX
1837 Prefix where libart is installed (optional).
1839 @item --with-libart-exec-prefix=PFX
1840 Exec prefix where libart is installed (optional).
1842 @item --disable-libarttest
1843 Do not try to compile and run a test libart program.
1852 @uref{./index.html,,Return to the GCC Installation page}
1856 @c ***Building****************************************************************
1858 @comment node-name, next, previous, up
1859 @node Building, Testing, Configuration, Installing GCC
1865 @cindex Installing GCC: Building
1867 Now that GCC is configured, you are ready to build the compiler and
1870 Some commands executed when making the compiler may fail (return a
1871 nonzero status) and be ignored by @command{make}. These failures, which
1872 are often due to files that were not found, are expected, and can safely
1875 It is normal to have compiler warnings when compiling certain files.
1876 Unless you are a GCC developer, you can generally ignore these warnings
1877 unless they cause compilation to fail. Developers should attempt to fix
1878 any warnings encountered, however they can temporarily continue past
1879 warnings-as-errors by specifying the configure flag
1880 @option{--disable-werror}.
1882 On certain old systems, defining certain environment variables such as
1883 @env{CC} can interfere with the functioning of @command{make}.
1885 If you encounter seemingly strange errors when trying to build the
1886 compiler in a directory other than the source directory, it could be
1887 because you have previously configured the compiler in the source
1888 directory. Make sure you have done all the necessary preparations.
1890 If you build GCC on a BSD system using a directory stored in an old System
1891 V file system, problems may occur in running @command{fixincludes} if the
1892 System V file system doesn't support symbolic links. These problems
1893 result in a failure to fix the declaration of @code{size_t} in
1894 @file{sys/types.h}. If you find that @code{size_t} is a signed type and
1895 that type mismatches occur, this could be the cause.
1897 The solution is not to use such a directory for building GCC@.
1899 Similarly, when building from SVN or snapshots, or if you modify
1900 @file{*.l} files, you need the Flex lexical analyzer generator
1901 installed. If you do not modify @file{*.l} files, releases contain
1902 the Flex-generated files and you do not need Flex installed to build
1903 them. There is still one Flex-based lexical analyzer (part of the
1904 build machinery, not of GCC itself) that is used even if you only
1905 build the C front end.
1907 When building from SVN or snapshots, or if you modify Texinfo
1908 documentation, you need version 4.7 or later of Texinfo installed if you
1909 want Info documentation to be regenerated. Releases contain Info
1910 documentation pre-built for the unmodified documentation in the release.
1912 @section Building a native compiler
1914 For a native build, the default configuration is to perform
1915 a 3-stage bootstrap of the compiler when @samp{make} is invoked.
1916 This will build the entire GCC system and ensure that it compiles
1917 itself correctly. It can be disabled with the @option{--disable-bootstrap}
1918 parameter to @samp{configure}, but bootstrapping is suggested because
1919 the compiler will be tested more completely and could also have
1922 The bootstrapping process will complete the following steps:
1926 Build tools necessary to build the compiler.
1929 Perform a 3-stage bootstrap of the compiler. This includes building
1930 three times the target tools for use by the compiler such as binutils
1931 (bfd, binutils, gas, gprof, ld, and opcodes) if they have been
1932 individually linked or moved into the top level GCC source tree before
1936 Perform a comparison test of the stage2 and stage3 compilers.
1939 Build runtime libraries using the stage3 compiler from the previous step.
1943 If you are short on disk space you might consider @samp{make
1944 bootstrap-lean} instead. The sequence of compilation is the
1945 same described above, but object files from the stage1 and
1946 stage2 of the 3-stage bootstrap of the compiler are deleted as
1947 soon as they are no longer needed.
1949 If you wish to use non-default GCC flags when compiling the stage2
1950 and stage3 compilers, set @code{BOOT_CFLAGS} on the command line when
1951 doing @samp{make}. For example, if you want to save additional space
1952 during the bootstrap and in the final installation as well, you can
1953 build the compiler binaries without debugging information as in the
1954 following example. This will save roughly 40% of disk space both for
1955 the bootstrap and the final installation. (Libraries will still contain
1956 debugging information.)
1959 make BOOT_CFLAGS='-O' bootstrap
1962 You can place non-default optimization flags into @code{BOOT_CFLAGS}; they
1963 are less well tested here than the default of @samp{-g -O2}, but should
1964 still work. In a few cases, you may find that you need to specify special
1965 flags such as @option{-msoft-float} here to complete the bootstrap; or,
1966 if the native compiler miscompiles the stage1 compiler, you may need
1967 to work around this, by choosing @code{BOOT_CFLAGS} to avoid the parts
1968 of the stage1 compiler that were miscompiled, or by using @samp{make
1969 bootstrap4} to increase the number of stages of bootstrap.
1971 @code{BOOT_CFLAGS} does not apply to bootstrapped target libraries.
1972 Since these are always compiled with the compiler currently being
1973 bootstrapped, you can use @code{CFLAGS_FOR_TARGET} to modify their
1974 compilation flags, as for non-bootstrapped target libraries.
1975 Again, if the native compiler miscompiles the stage1 compiler, you may
1976 need to work around this by avoiding non-working parts of the stage1
1977 compiler. Use @code{STAGE1_LIBCFLAGS} to this end.
1979 If you used the flag @option{--enable-languages=@dots{}} to restrict
1980 the compilers to be built, only those you've actually enabled will be
1981 built. This will of course only build those runtime libraries, for
1982 which the particular compiler has been built. Please note,
1983 that re-defining @env{LANGUAGES} when calling @samp{make}
1984 @strong{does not} work anymore!
1986 If the comparison of stage2 and stage3 fails, this normally indicates
1987 that the stage2 compiler has compiled GCC incorrectly, and is therefore
1988 a potentially serious bug which you should investigate and report. (On
1989 a few systems, meaningful comparison of object files is impossible; they
1990 always appear ``different''. If you encounter this problem, you will
1991 need to disable comparison in the @file{Makefile}.)
1993 If you do not want to bootstrap your compiler, you can configure with
1994 @option{--disable-bootstrap}. In particular cases, you may want to
1995 bootstrap your compiler even if the target system is not the same as
1996 the one you are building on: for example, you could build a
1997 @code{powerpc-unknown-linux-gnu} toolchain on a
1998 @code{powerpc64-unknown-linux-gnu} host. In this case, pass
1999 @option{--enable-bootstrap} to the configure script.
2002 @section Building a cross compiler
2004 When building a cross compiler, it is not generally possible to do a
2005 3-stage bootstrap of the compiler. This makes for an interesting problem
2006 as parts of GCC can only be built with GCC@.
2008 To build a cross compiler, we first recommend building and installing a
2009 native compiler. You can then use the native GCC compiler to build the
2010 cross compiler. The installed native compiler needs to be GCC version
2013 If the cross compiler is to be built with support for the Java
2014 programming language and the ability to compile .java source files is
2015 desired, the installed native compiler used to build the cross
2016 compiler needs to be the same GCC version as the cross compiler. In
2017 addition the cross compiler needs to be configured with
2018 @option{--with-ecj-jar=@dots{}}.
2020 Assuming you have already installed a native copy of GCC and configured
2021 your cross compiler, issue the command @command{make}, which performs the
2026 Build host tools necessary to build the compiler.
2029 Build target tools for use by the compiler such as binutils (bfd,
2030 binutils, gas, gprof, ld, and opcodes)
2031 if they have been individually linked or moved into the top level GCC source
2032 tree before configuring.
2035 Build the compiler (single stage only).
2038 Build runtime libraries using the compiler from the previous step.
2041 Note that if an error occurs in any step the make process will exit.
2043 If you are not building GNU binutils in the same source tree as GCC,
2044 you will need a cross-assembler and cross-linker installed before
2045 configuring GCC@. Put them in the directory
2046 @file{@var{prefix}/@var{target}/bin}. Here is a table of the tools
2047 you should put in this directory:
2051 This should be the cross-assembler.
2054 This should be the cross-linker.
2057 This should be the cross-archiver: a program which can manipulate
2058 archive files (linker libraries) in the target machine's format.
2061 This should be a program to construct a symbol table in an archive file.
2064 The installation of GCC will find these programs in that directory,
2065 and copy or link them to the proper place to for the cross-compiler to
2066 find them when run later.
2068 The easiest way to provide these files is to build the Binutils package.
2069 Configure it with the same @option{--host} and @option{--target}
2070 options that you use for configuring GCC, then build and install
2071 them. They install their executables automatically into the proper
2072 directory. Alas, they do not support all the targets that GCC
2075 If you are not building a C library in the same source tree as GCC,
2076 you should also provide the target libraries and headers before
2077 configuring GCC, specifying the directories with
2078 @option{--with-sysroot} or @option{--with-headers} and
2079 @option{--with-libs}. Many targets also require ``start files'' such
2080 as @file{crt0.o} and
2081 @file{crtn.o} which are linked into each executable. There may be several
2082 alternatives for @file{crt0.o}, for use with profiling or other
2083 compilation options. Check your target's definition of
2084 @code{STARTFILE_SPEC} to find out what start files it uses.
2086 @section Building in parallel
2088 GNU Make 3.79 and above, which is necessary to build GCC, support
2089 building in parallel. To activate this, you can use @samp{make -j 2}
2090 instead of @samp{make}. You can also specify a bigger number, and
2091 in most cases using a value greater than the number of processors in
2092 your machine will result in fewer and shorter I/O latency hits, thus
2093 improving overall throughput; this is especially true for slow drives
2094 and network filesystems.
2096 @section Building the Ada compiler
2098 In order to build GNAT, the Ada compiler, you need a working GNAT
2099 compiler (GCC version 3.4 or later).
2100 This includes GNAT tools such as @command{gnatmake} and
2101 @command{gnatlink}, since the Ada front end is written in Ada and
2102 uses some GNAT-specific extensions.
2104 In order to build a cross compiler, it is suggested to install
2105 the new compiler as native first, and then use it to build the cross
2108 @command{configure} does not test whether the GNAT installation works
2109 and has a sufficiently recent version; if too old a GNAT version is
2110 installed, the build will fail unless @option{--enable-languages} is
2111 used to disable building the Ada front end.
2113 @env{ADA_INCLUDE_PATH} and @env{ADA_OBJECT_PATH} environment variables
2114 must not be set when building the Ada compiler, the Ada tools, or the
2115 Ada runtime libraries. You can check that your build environment is clean
2116 by verifying that @samp{gnatls -v} lists only one explicit path in each
2119 @section Building with profile feedback
2121 It is possible to use profile feedback to optimize the compiler itself. This
2122 should result in a faster compiler binary. Experiments done on x86 using gcc
2123 3.3 showed approximately 7 percent speedup on compiling C programs. To
2124 bootstrap the compiler with profile feedback, use @code{make profiledbootstrap}.
2126 When @samp{make profiledbootstrap} is run, it will first build a @code{stage1}
2127 compiler. This compiler is used to build a @code{stageprofile} compiler
2128 instrumented to collect execution counts of instruction and branch
2129 probabilities. Then runtime libraries are compiled with profile collected.
2130 Finally a @code{stagefeedback} compiler is built using the information collected.
2132 Unlike standard bootstrap, several additional restrictions apply. The
2133 compiler used to build @code{stage1} needs to support a 64-bit integral type.
2134 It is recommended to only use GCC for this. Also parallel make is currently
2135 not supported since collisions in profile collecting may occur.
2142 @uref{./index.html,,Return to the GCC Installation page}
2146 @c ***Testing*****************************************************************
2148 @comment node-name, next, previous, up
2149 @node Testing, Final install, Building, Installing GCC
2153 @chapter Installing GCC: Testing
2156 @cindex Installing GCC: Testing
2159 Before you install GCC, we encourage you to run the testsuites and to
2160 compare your results with results from a similar configuration that have
2161 been submitted to the
2162 @uref{http://gcc.gnu.org/ml/gcc-testresults/,,gcc-testresults mailing list}.
2163 Some of these archived results are linked from the build status lists
2164 at @uref{http://gcc.gnu.org/buildstat.html}, although not everyone who
2165 reports a successful build runs the testsuites and submits the results.
2166 This step is optional and may require you to download additional software,
2167 but it can give you confidence in your new GCC installation or point out
2168 problems before you install and start using your new GCC@.
2170 First, you must have @uref{download.html,,downloaded the testsuites}.
2171 These are part of the full distribution, but if you downloaded the
2172 ``core'' compiler plus any front ends, you must download the testsuites
2175 Second, you must have the testing tools installed. This includes
2176 @uref{http://www.gnu.org/software/dejagnu/,,DejaGnu}, Tcl, and Expect;
2177 the DejaGnu site has links to these.
2179 If the directories where @command{runtest} and @command{expect} were
2180 installed are not in the @env{PATH}, you may need to set the following
2181 environment variables appropriately, as in the following example (which
2182 assumes that DejaGnu has been installed under @file{/usr/local}):
2185 TCL_LIBRARY = /usr/local/share/tcl8.0
2186 DEJAGNULIBS = /usr/local/share/dejagnu
2189 (On systems such as Cygwin, these paths are required to be actual
2190 paths, not mounts or links; presumably this is due to some lack of
2191 portability in the DejaGnu code.)
2194 Finally, you can run the testsuite (which may take a long time):
2196 cd @var{objdir}; make -k check
2199 This will test various components of GCC, such as compiler
2200 front ends and runtime libraries. While running the testsuite, DejaGnu
2201 might emit some harmless messages resembling
2202 @samp{WARNING: Couldn't find the global config file.} or
2203 @samp{WARNING: Couldn't find tool init file} that can be ignored.
2205 If you are testing a cross-compiler, you may want to run the testsuite
2206 on a simulator as described at @uref{http://gcc.gnu.org/simtest-howto.html}.
2208 @section How can you run the testsuite on selected tests?
2210 In order to run sets of tests selectively, there are targets
2211 @samp{make check-gcc} and @samp{make check-g++}
2212 in the @file{gcc} subdirectory of the object directory. You can also
2213 just run @samp{make check} in a subdirectory of the object directory.
2216 A more selective way to just run all @command{gcc} execute tests in the
2220 make check-gcc RUNTESTFLAGS="execute.exp @var{other-options}"
2223 Likewise, in order to run only the @command{g++} ``old-deja'' tests in
2224 the testsuite with filenames matching @samp{9805*}, you would use
2227 make check-g++ RUNTESTFLAGS="old-deja.exp=9805* @var{other-options}"
2230 The @file{*.exp} files are located in the testsuite directories of the GCC
2231 source, the most important ones being @file{compile.exp},
2232 @file{execute.exp}, @file{dg.exp} and @file{old-deja.exp}.
2233 To get a list of the possible @file{*.exp} files, pipe the
2234 output of @samp{make check} into a file and look at the
2235 @samp{Running @dots{} .exp} lines.
2237 @section Passing options and running multiple testsuites
2239 You can pass multiple options to the testsuite using the
2240 @samp{--target_board} option of DejaGNU, either passed as part of
2241 @samp{RUNTESTFLAGS}, or directly to @command{runtest} if you prefer to
2242 work outside the makefiles. For example,
2245 make check-g++ RUNTESTFLAGS="--target_board=unix/-O3/-fmerge-constants"
2248 will run the standard @command{g++} testsuites (``unix'' is the target name
2249 for a standard native testsuite situation), passing
2250 @samp{-O3 -fmerge-constants} to the compiler on every test, i.e.,
2251 slashes separate options.
2253 You can run the testsuites multiple times using combinations of options
2254 with a syntax similar to the brace expansion of popular shells:
2257 @dots{}"--target_board=arm-sim\@{-mhard-float,-msoft-float\@}\@{-O1,-O2,-O3,\@}"
2260 (Note the empty option caused by the trailing comma in the final group.)
2261 The following will run each testsuite eight times using the @samp{arm-sim}
2262 target, as if you had specified all possible combinations yourself:
2265 --target_board=arm-sim/-mhard-float/-O1
2266 --target_board=arm-sim/-mhard-float/-O2
2267 --target_board=arm-sim/-mhard-float/-O3
2268 --target_board=arm-sim/-mhard-float
2269 --target_board=arm-sim/-msoft-float/-O1
2270 --target_board=arm-sim/-msoft-float/-O2
2271 --target_board=arm-sim/-msoft-float/-O3
2272 --target_board=arm-sim/-msoft-float
2275 They can be combined as many times as you wish, in arbitrary ways. This
2279 @dots{}"--target_board=unix/-Wextra\@{-O3,-fno-strength\@}\@{-fomit-frame,\@}"
2282 will generate four combinations, all involving @samp{-Wextra}.
2284 The disadvantage to this method is that the testsuites are run in serial,
2285 which is a waste on multiprocessor systems. For users with GNU Make and
2286 a shell which performs brace expansion, you can run the testsuites in
2287 parallel by having the shell perform the combinations and @command{make}
2288 do the parallel runs. Instead of using @samp{--target_board}, use a
2289 special makefile target:
2292 make -j@var{N} check-@var{testsuite}//@var{test-target}/@var{option1}/@var{option2}/@dots{}
2298 make -j3 check-gcc//sh-hms-sim/@{-m1,-m2,-m3,-m3e,-m4@}/@{,-nofpu@}
2301 will run three concurrent ``make-gcc'' testsuites, eventually testing all
2302 ten combinations as described above. Note that this is currently only
2303 supported in the @file{gcc} subdirectory. (To see how this works, try
2304 typing @command{echo} before the example given here.)
2307 @section Additional testing for Java Class Libraries
2309 The Java runtime tests can be executed via @samp{make check}
2310 in the @file{@var{target}/libjava/testsuite} directory in
2313 The @uref{http://sourceware.org/mauve/,,Mauve Project} provides
2314 a suite of tests for the Java Class Libraries. This suite can be run
2315 as part of libgcj testing by placing the Mauve tree within the libjava
2316 testsuite at @file{libjava/testsuite/libjava.mauve/mauve}, or by
2317 specifying the location of that tree when invoking @samp{make}, as in
2318 @samp{make MAUVEDIR=~/mauve check}.
2320 @section How to interpret test results
2322 The result of running the testsuite are various @file{*.sum} and @file{*.log}
2323 files in the testsuite subdirectories. The @file{*.log} files contain a
2324 detailed log of the compiler invocations and the corresponding
2325 results, the @file{*.sum} files summarize the results. These summaries
2326 contain status codes for all tests:
2330 PASS: the test passed as expected
2332 XPASS: the test unexpectedly passed
2334 FAIL: the test unexpectedly failed
2336 XFAIL: the test failed as expected
2338 UNSUPPORTED: the test is not supported on this platform
2340 ERROR: the testsuite detected an error
2342 WARNING: the testsuite detected a possible problem
2345 It is normal for some tests to report unexpected failures. At the
2346 current time the testing harness does not allow fine grained control
2347 over whether or not a test is expected to fail. This problem should
2348 be fixed in future releases.
2351 @section Submitting test results
2353 If you want to report the results to the GCC project, use the
2354 @file{contrib/test_summary} shell script. Start it in the @var{objdir} with
2357 @var{srcdir}/contrib/test_summary -p your_commentary.txt \
2358 -m gcc-testresults@@gcc.gnu.org |sh
2361 This script uses the @command{Mail} program to send the results, so
2362 make sure it is in your @env{PATH}. The file @file{your_commentary.txt} is
2363 prepended to the testsuite summary and should contain any special
2364 remarks you have on your results or your build environment. Please
2365 do not edit the testsuite result block or the subject line, as these
2366 messages may be automatically processed.
2373 @uref{./index.html,,Return to the GCC Installation page}
2377 @c ***Final install***********************************************************
2379 @comment node-name, next, previous, up
2380 @node Final install, , Testing, Installing GCC
2382 @ifset finalinstallhtml
2384 @chapter Installing GCC: Final installation
2387 Now that GCC has been built (and optionally tested), you can install it with
2389 cd @var{objdir}; make install
2392 We strongly recommend to install into a target directory where there is
2393 no previous version of GCC present. Also, the GNAT runtime should not
2394 be stripped, as this would break certain features of the debugger that
2395 depend on this debugging information (catching Ada exceptions for
2398 That step completes the installation of GCC; user level binaries can
2399 be found in @file{@var{prefix}/bin} where @var{prefix} is the value
2400 you specified with the @option{--prefix} to configure (or
2401 @file{/usr/local} by default). (If you specified @option{--bindir},
2402 that directory will be used instead; otherwise, if you specified
2403 @option{--exec-prefix}, @file{@var{exec-prefix}/bin} will be used.)
2404 Headers for the C++ and Java libraries are installed in
2405 @file{@var{prefix}/include}; libraries in @file{@var{libdir}}
2406 (normally @file{@var{prefix}/lib}); internal parts of the compiler in
2407 @file{@var{libdir}/gcc} and @file{@var{libexecdir}/gcc}; documentation
2408 in info format in @file{@var{infodir}} (normally
2409 @file{@var{prefix}/info}).
2411 When installing cross-compilers, GCC's executables
2412 are not only installed into @file{@var{bindir}}, that
2413 is, @file{@var{exec-prefix}/bin}, but additionally into
2414 @file{@var{exec-prefix}/@var{target-alias}/bin}, if that directory
2415 exists. Typically, such @dfn{tooldirs} hold target-specific
2416 binutils, including assembler and linker.
2418 Installation into a temporary staging area or into a @command{chroot}
2419 jail can be achieved with the command
2422 make DESTDIR=@var{path-to-rootdir} install
2425 @noindent where @var{path-to-rootdir} is the absolute path of
2426 a directory relative to which all installation paths will be
2427 interpreted. Note that the directory specified by @code{DESTDIR}
2428 need not exist yet; it will be created if necessary.
2430 There is a subtle point with tooldirs and @code{DESTDIR}:
2431 If you relocate a cross-compiler installation with
2432 e.g.@: @samp{DESTDIR=@var{rootdir}}, then the directory
2433 @file{@var{rootdir}/@var{exec-prefix}/@var{target-alias}/bin} will
2434 be filled with duplicated GCC executables only if it already exists,
2435 it will not be created otherwise. This is regarded as a feature,
2436 not as a bug, because it gives slightly more control to the packagers
2437 using the @code{DESTDIR} feature.
2439 If you are bootstrapping a released version of GCC then please
2440 quickly review the build status page for your release, available from
2441 @uref{http://gcc.gnu.org/buildstat.html}.
2442 If your system is not listed for the version of GCC that you built,
2444 @email{gcc@@gcc.gnu.org} indicating
2445 that you successfully built and installed GCC@.
2446 Include the following information:
2450 Output from running @file{@var{srcdir}/config.guess}. Do not send
2451 that file itself, just the one-line output from running it.
2454 The output of @samp{gcc -v} for your newly installed @command{gcc}.
2455 This tells us which version of GCC you built and the options you passed to
2459 Whether you enabled all languages or a subset of them. If you used a
2460 full distribution then this information is part of the configure
2461 options in the output of @samp{gcc -v}, but if you downloaded the
2462 ``core'' compiler plus additional front ends then it isn't apparent
2463 which ones you built unless you tell us about it.
2466 If the build was for GNU/Linux, also include:
2469 The distribution name and version (e.g., Red Hat 7.1 or Debian 2.2.3);
2470 this information should be available from @file{/etc/issue}.
2473 The version of the Linux kernel, available from @samp{uname --version}
2477 The version of glibc you used; for RPM-based systems like Red Hat,
2478 Mandrake, and SuSE type @samp{rpm -q glibc} to get the glibc version,
2479 and on systems like Debian and Progeny use @samp{dpkg -l libc6}.
2481 For other systems, you can include similar information if you think it is
2485 Any other information that you think would be useful to people building
2486 GCC on the same configuration. The new entry in the build status list
2487 will include a link to the archived copy of your message.
2490 We'd also like to know if the
2492 @ref{Specific, host/target specific installation notes}
2495 @uref{specific.html,,host/target specific installation notes}
2497 didn't include your host/target information or if that information is
2498 incomplete or out of date. Send a note to
2499 @email{gcc@@gcc.gnu.org} detailing how the information should be changed.
2501 If you find a bug, please report it following the
2502 @uref{../bugs.html,,bug reporting guidelines}.
2504 If you want to print the GCC manuals, do @samp{cd @var{objdir}; make
2505 dvi}. You will need to have @command{texi2dvi} (version at least 4.7)
2506 and @TeX{} installed. This creates a number of @file{.dvi} files in
2507 subdirectories of @file{@var{objdir}}; these may be converted for
2508 printing with programs such as @command{dvips}. Alternately, by using
2509 @samp{make pdf} in place of @samp{make dvi}, you can create documentation
2510 in the form of @file{.pdf} files; this requires @command{texi2pdf}, which
2511 is included with Texinfo version 4.8 and later. You can also
2512 @uref{http://www.gnu.org/order/order.html,,buy printed manuals from the
2513 Free Software Foundation}, though such manuals may not be for the most
2514 recent version of GCC@.
2516 If you would like to generate online HTML documentation, do @samp{cd
2517 @var{objdir}; make html} and HTML will be generated for the gcc manuals in
2518 @file{@var{objdir}/gcc/HTML}.
2525 @uref{./index.html,,Return to the GCC Installation page}
2529 @c ***Binaries****************************************************************
2531 @comment node-name, next, previous, up
2532 @node Binaries, Specific, Installing GCC, Top
2536 @chapter Installing GCC: Binaries
2539 @cindex Installing GCC: Binaries
2541 We are often asked about pre-compiled versions of GCC@. While we cannot
2542 provide these for all platforms, below you'll find links to binaries for
2543 various platforms where creating them by yourself is not easy due to various
2546 Please note that we did not create these binaries, nor do we
2547 support them. If you have any problems installing them, please
2548 contact their makers.
2555 @uref{http://www.bullfreeware.com,,Bull's Freeware and Shareware Archive for AIX};
2558 @uref{http://pware.hvcc.edu,,Hudson Valley Community College Open Source Software for IBM System p};
2561 @uref{http://www.perzl.org/aix,,AIX 5L and 6 Open Source Packages}.
2565 DOS---@uref{http://www.delorie.com/djgpp/,,DJGPP}.
2568 Renesas H8/300[HS]---@uref{http://h8300-hms.sourceforge.net/,,GNU
2569 Development Tools for the Renesas H8/300[HS] Series}.
2575 @uref{http://hpux.cs.utah.edu/,,HP-UX Porting Center};
2578 @uref{ftp://sunsite.informatik.rwth-aachen.de/pub/packages/gcc_hpux/,,Binaries for HP-UX 11.00 at Aachen University of Technology}.
2582 Motorola 68HC11/68HC12---@uref{http://www.gnu-m68hc11.org,,GNU
2583 Development Tools for the Motorola 68HC11/68HC12}.
2586 @uref{http://www.sco.com/skunkware/devtools/index.html#gcc,,SCO
2587 OpenServer/Unixware}.
2590 Solaris 2 (SPARC, Intel)---@uref{http://www.sunfreeware.com/,,Sunfreeware}.
2593 SGI---@uref{http://freeware.sgi.com/,,SGI Freeware}.
2599 The @uref{http://sourceware.org/cygwin/,,Cygwin} project;
2601 The @uref{http://www.mingw.org/,,MinGW} project.
2605 @uref{ftp://ftp.thewrittenword.com/packages/by-name/,,The
2606 Written Word} offers binaries for
2607 AIX 4.3.3, 5.1 and 5.2,
2609 Tru64 UNIX 4.0D and 5.1,
2611 HP-UX 10.20, 11.00, and 11.11, and
2612 Solaris/SPARC 2.5.1, 2.6, 7, 8, 9 and 10.
2615 @uref{http://www.openpkg.org/,,OpenPKG} offers binaries for quite a
2616 number of platforms.
2619 The @uref{http://gcc.gnu.org/wiki/GFortranBinaries,,GFortran Wiki} has
2620 links to GNU Fortran binaries for several platforms.
2623 In addition to those specific offerings, you can get a binary
2624 distribution CD-ROM from the
2625 @uref{http://www.gnu.org/order/order.html,,Free Software Foundation}.
2626 It contains binaries for a number of platforms, and
2627 includes not only GCC, but other stuff as well. The current CD does
2628 not contain the latest version of GCC, but it should allow
2629 bootstrapping the compiler. An updated version of that disk is in the
2637 @uref{./index.html,,Return to the GCC Installation page}
2641 @c ***Specific****************************************************************
2643 @comment node-name, next, previous, up
2644 @node Specific, Old, Binaries, Top
2648 @chapter Host/target specific installation notes for GCC
2651 @cindex Specific installation notes
2652 @cindex Target specific installation
2653 @cindex Host specific installation
2654 @cindex Target specific installation notes
2656 Please read this document carefully @emph{before} installing the
2657 GNU Compiler Collection on your machine.
2659 Note that this list of install notes is @emph{not} a list of supported
2660 hosts or targets. Not all supported hosts and targets are listed
2661 here, only the ones that require host-specific or target-specific
2667 @uref{#alpha-x-x,,alpha*-*-*}
2669 @uref{#alpha-dec-osf,,alpha*-dec-osf*}
2671 @uref{#arc-x-elf,,arc-*-elf}
2673 @uref{#arm-x-elf,,arm-*-elf}
2677 @uref{#bfin,,Blackfin}
2681 @uref{#x-x-freebsd,,*-*-freebsd*}
2683 @uref{#h8300-hms,,h8300-hms}
2685 @uref{#hppa-hp-hpux,,hppa*-hp-hpux*}
2687 @uref{#hppa-hp-hpux10,,hppa*-hp-hpux10}
2689 @uref{#hppa-hp-hpux11,,hppa*-hp-hpux11}
2691 @uref{#x-x-linux-gnu,,*-*-linux-gnu}
2693 @uref{#ix86-x-linux,,i?86-*-linux*}
2695 @uref{#ix86-x-solaris210,,i?86-*-solaris2.10}
2697 @uref{#ia64-x-linux,,ia64-*-linux}
2699 @uref{#ia64-x-hpux,,ia64-*-hpux*}
2701 @uref{#x-ibm-aix,,*-ibm-aix*}
2703 @uref{#iq2000-x-elf,,iq2000-*-elf}
2705 @uref{#m32c-x-elf,,m32c-*-elf}
2707 @uref{#m32r-x-elf,,m32r-*-elf}
2709 @uref{#m6811-elf,,m6811-elf}
2711 @uref{#m6812-elf,,m6812-elf}
2713 @uref{#m68k-x-x,,m68k-*-*}
2715 @uref{#m68k-uclinux,,m68k-uclinux}
2717 @uref{#mips-x-x,,mips-*-*}
2719 @uref{#mips-sgi-irix5,,mips-sgi-irix5}
2721 @uref{#mips-sgi-irix6,,mips-sgi-irix6}
2723 @uref{#powerpc-x-x,,powerpc*-*-*}
2725 @uref{#powerpc-x-darwin,,powerpc-*-darwin*}
2727 @uref{#powerpc-x-elf,,powerpc-*-elf}
2729 @uref{#powerpc-x-linux-gnu,,powerpc*-*-linux-gnu*}
2731 @uref{#powerpc-x-netbsd,,powerpc-*-netbsd*}
2733 @uref{#powerpc-x-eabisim,,powerpc-*-eabisim}
2735 @uref{#powerpc-x-eabi,,powerpc-*-eabi}
2737 @uref{#powerpcle-x-elf,,powerpcle-*-elf}
2739 @uref{#powerpcle-x-eabisim,,powerpcle-*-eabisim}
2741 @uref{#powerpcle-x-eabi,,powerpcle-*-eabi}
2743 @uref{#s390-x-linux,,s390-*-linux*}
2745 @uref{#s390x-x-linux,,s390x-*-linux*}
2747 @uref{#s390x-ibm-tpf,,s390x-ibm-tpf*}
2749 @uref{#x-x-solaris2,,*-*-solaris2*}
2751 @uref{#sparc-sun-solaris2,,sparc-sun-solaris2*}
2753 @uref{#sparc-sun-solaris27,,sparc-sun-solaris2.7}
2755 @uref{#sparc-x-linux,,sparc-*-linux*}
2757 @uref{#sparc64-x-solaris2,,sparc64-*-solaris2*}
2759 @uref{#sparcv9-x-solaris2,,sparcv9-*-solaris2*}
2761 @uref{#x-x-vxworks,,*-*-vxworks*}
2763 @uref{#x86-64-x-x,,x86_64-*-*, amd64-*-*}
2765 @uref{#xtensa-x-elf,,xtensa*-*-elf}
2767 @uref{#xtensa-x-linux,,xtensa*-*-linux*}
2769 @uref{#windows,,Microsoft Windows}
2771 @uref{#x-x-cygwin,,*-*-cygwin}
2773 @uref{#x-x-interix,,*-*-interix}
2775 @uref{#x-x-mingw32,,*-*-mingw32}
2779 @uref{#older,,Older systems}
2784 @uref{#elf,,all ELF targets} (SVR4, Solaris 2, etc.)
2790 <!-- -------- host/target specific issues start here ---------------- -->
2793 @heading @anchor{alpha-x-x}alpha*-*-*
2795 This section contains general configuration information for all
2796 alpha-based platforms using ELF (in particular, ignore this section for
2797 DEC OSF/1, Digital UNIX and Tru64 UNIX)@. In addition to reading this
2798 section, please read all other sections that match your target.
2800 We require binutils 2.11.2 or newer.
2801 Previous binutils releases had a number of problems with DWARF 2
2802 debugging information, not the least of which is incorrect linking of
2808 @heading @anchor{alpha-dec-osf}alpha*-dec-osf*
2809 Systems using processors that implement the DEC Alpha architecture and
2810 are running the DEC/Compaq Unix (DEC OSF/1, Digital UNIX, or Compaq
2811 Tru64 UNIX) operating system, for example the DEC Alpha AXP systems.
2813 As of GCC 3.2, versions before @code{alpha*-dec-osf4} are no longer
2814 supported. (These are the versions which identify themselves as DEC
2817 In Digital Unix V4.0, virtual memory exhausted bootstrap failures
2818 may be fixed by configuring with @option{--with-gc=simple},
2819 reconfiguring Kernel Virtual Memory and Swap parameters
2820 per the @command{/usr/sbin/sys_check} Tuning Suggestions,
2821 or applying the patch in
2822 @uref{http://gcc.gnu.org/ml/gcc/2002-08/msg00822.html}.
2824 In Tru64 UNIX V5.1, Compaq introduced a new assembler that does not
2825 currently (2001-06-13) work with @command{mips-tfile}. As a workaround,
2826 we need to use the old assembler, invoked via the barely documented
2827 @option{-oldas} option. To bootstrap GCC, you either need to use the
2831 % CC=cc @var{srcdir}/configure [@var{options}] [@var{target}]
2834 or you can use a copy of GCC 2.95.3 or higher built on Tru64 UNIX V4.0:
2837 % CC=gcc -Wa,-oldas @var{srcdir}/configure [@var{options}] [@var{target}]
2840 As of GNU binutils 2.11.2, neither GNU @command{as} nor GNU @command{ld}
2841 are supported on Tru64 UNIX, so you must not configure GCC with
2842 @option{--with-gnu-as} or @option{--with-gnu-ld}.
2844 GCC writes a @samp{.verstamp} directive to the assembler output file
2845 unless it is built as a cross-compiler. It gets the version to use from
2846 the system header file @file{/usr/include/stamp.h}. If you install a
2847 new version of DEC Unix, you should rebuild GCC to pick up the new version
2850 @samp{make compare} may fail on old versions of DEC Unix unless you add
2851 @option{-save-temps} to @code{BOOT_CFLAGS}. On these systems, the name
2852 of the assembler input file is stored in the object file, and that makes
2853 comparison fail if it differs between the @code{stage1} and
2854 @code{stage2} compilations. The option @option{-save-temps} forces a
2855 fixed name to be used for the assembler input file, instead of a
2856 randomly chosen name in @file{/tmp}. Do not add @option{-save-temps}
2857 unless the comparisons fail without that option. If you add
2858 @option{-save-temps}, you will have to manually delete the @samp{.i} and
2859 @samp{.s} files after each series of compilations.
2861 GCC now supports both the native (ECOFF) debugging format used by DBX
2862 and GDB and an encapsulated STABS format for use only with GDB@. See the
2863 discussion of the @option{--with-stabs} option of @file{configure} above
2864 for more information on these formats and how to select them.
2866 There is a bug in DEC's assembler that produces incorrect line numbers
2867 for ECOFF format when the @samp{.align} directive is used. To work
2868 around this problem, GCC will not emit such alignment directives
2869 while writing ECOFF format debugging information even if optimization is
2870 being performed. Unfortunately, this has the very undesirable
2871 side-effect that code addresses when @option{-O} is specified are
2872 different depending on whether or not @option{-g} is also specified.
2874 To avoid this behavior, specify @option{-gstabs+} and use GDB instead of
2875 DBX@. DEC is now aware of this problem with the assembler and hopes to
2876 provide a fix shortly.
2881 @heading @anchor{arc-x-elf}arc-*-elf
2882 Argonaut ARC processor.
2883 This configuration is intended for embedded systems.
2888 @heading @anchor{arm-x-elf}arm-*-elf
2889 ARM-family processors. Subtargets that use the ELF object format
2890 require GNU binutils 2.13 or newer. Such subtargets include:
2891 @code{arm-*-freebsd}, @code{arm-*-netbsdelf}, @code{arm-*-*linux}
2892 and @code{arm-*-rtems}.
2897 @heading @anchor{avr}avr
2899 ATMEL AVR-family micro controllers. These are used in embedded
2900 applications. There are no standard Unix configurations.
2902 @xref{AVR Options,, AVR Options, gcc, Using the GNU Compiler
2906 See ``AVR Options'' in the main manual
2908 for the list of supported MCU types.
2910 Use @samp{configure --target=avr --enable-languages="c"} to configure GCC@.
2912 Further installation notes and other useful information about AVR tools
2913 can also be obtained from:
2917 @uref{http://www.nongnu.org/avr/,,http://www.nongnu.org/avr/}
2919 @uref{http://home.overta.ru/users/denisc/,,http://home.overta.ru/users/denisc/}
2921 @uref{http://www.amelek.gda.pl/avr/,,http://www.amelek.gda.pl/avr/}
2924 We @emph{strongly} recommend using binutils 2.13 or newer.
2926 The following error:
2928 Error: register required
2931 indicates that you should upgrade to a newer version of the binutils.
2936 @heading @anchor{bfin}Blackfin
2938 The Blackfin processor, an Analog Devices DSP.
2940 @xref{Blackfin Options,, Blackfin Options, gcc, Using the GNU Compiler
2944 See ``Blackfin Options'' in the main manual
2947 More information, and a version of binutils with support for this processor,
2948 is available at @uref{http://blackfin.uclinux.org}
2953 @heading @anchor{cris}CRIS
2955 CRIS is the CPU architecture in Axis Communications ETRAX system-on-a-chip
2956 series. These are used in embedded applications.
2959 @xref{CRIS Options,, CRIS Options, gcc, Using the GNU Compiler
2963 See ``CRIS Options'' in the main manual
2965 for a list of CRIS-specific options.
2967 There are a few different CRIS targets:
2970 Mainly for monolithic embedded systems. Includes a multilib for the
2971 @samp{v10} core used in @samp{ETRAX 100 LX}.
2972 @item cris-axis-linux-gnu
2973 A GNU/Linux port for the CRIS architecture, currently targeting
2974 @samp{ETRAX 100 LX} by default.
2977 For @code{cris-axis-elf} you need binutils 2.11
2978 or newer. For @code{cris-axis-linux-gnu} you need binutils 2.12 or newer.
2980 Pre-packaged tools can be obtained from
2981 @uref{ftp://ftp.axis.com/pub/axis/tools/cris/compiler-kit/}. More
2982 information about this platform is available at
2983 @uref{http://developer.axis.com/}.
2988 @heading @anchor{crx}CRX
2990 The CRX CompactRISC architecture is a low-power 32-bit architecture with
2991 fast context switching and architectural extensibility features.
2994 @xref{CRX Options,, CRX Options, gcc, Using and Porting the GNU Compiler
2999 See ``CRX Options'' in the main manual for a list of CRX-specific options.
3002 Use @samp{configure --target=crx-elf --enable-languages=c,c++} to configure
3003 GCC@ for building a CRX cross-compiler. The option @samp{--target=crx-elf}
3004 is also used to build the @samp{newlib} C library for CRX.
3006 It is also possible to build libstdc++-v3 for the CRX architecture. This
3007 needs to be done in a separate step with the following configure settings:
3008 @samp{gcc/libstdc++-v3/configure --host=crx-elf --with-newlib
3009 --enable-sjlj-exceptions --enable-cxx-flags='-fexceptions -frtti'}
3014 @heading @anchor{dos}DOS
3016 Please have a look at the @uref{binaries.html,,binaries page}.
3018 You cannot install GCC by itself on MSDOS; it will not compile under
3019 any MSDOS compiler except itself. You need to get the complete
3020 compilation package DJGPP, which includes binaries as well as sources,
3021 and includes all the necessary compilation tools and libraries.
3026 @heading @anchor{x-x-freebsd}*-*-freebsd*
3028 The version of binutils installed in @file{/usr/bin} probably works with
3029 this release of GCC@. However, on FreeBSD 4, bootstrapping against the
3030 latest FSF binutils is known to improve overall testsuite results; and,
3031 on FreeBSD/alpha, using binutils 2.14 or later is required to build libjava.
3033 Support for FreeBSD 1 was discontinued in GCC 3.2.
3035 Support for FreeBSD 2 will be discontinued after GCC 3.4. The
3036 following was true for GCC 3.1 but the current status is unknown.
3037 For FreeBSD 2 or any mutant a.out versions of FreeBSD 3: All
3038 configuration support and files as shipped with GCC 2.95 are still in
3039 place. FreeBSD 2.2.7 has been known to bootstrap completely; however,
3040 it is unknown which version of binutils was used (it is assumed that it
3041 was the system copy in @file{/usr/bin}) and C++ EH failures were noted.
3043 For FreeBSD using the ELF file format: DWARF 2 debugging is now the
3044 default for all CPU architectures. It had been the default on
3045 FreeBSD/alpha since its inception. You may use @option{-gstabs} instead
3046 of @option{-g}, if you really want the old debugging format. There are
3047 no known issues with mixing object files and libraries with different
3048 debugging formats. Otherwise, this release of GCC should now match more
3049 of the configuration used in the stock FreeBSD configuration of GCC@. In
3050 particular, @option{--enable-threads} is now configured by default.
3051 However, as a general user, do not attempt to replace the system
3052 compiler with this release. Known to bootstrap and check with good
3053 results on FreeBSD 4.9-STABLE and 5-CURRENT@. In the past, known to
3054 bootstrap and check with good results on FreeBSD 3.0, 3.4, 4.0, 4.2,
3055 4.3, 4.4, 4.5, 4.8-STABLE@.
3057 In principle, @option{--enable-threads} is now compatible with
3058 @option{--enable-libgcj} on FreeBSD@. However, it has only been built
3059 and tested on @samp{i386-*-freebsd[45]} and @samp{alpha-*-freebsd[45]}.
3061 library may be incorrectly built (symbols are missing at link time).
3062 There is a rare timing-based startup hang (probably involves an
3063 assumption about the thread library). Multi-threaded boehm-gc (required for
3064 libjava) exposes severe threaded signal-handling bugs on FreeBSD before
3065 4.5-RELEASE@. Other CPU architectures
3066 supported by FreeBSD will require additional configuration tuning in, at
3067 the very least, both boehm-gc and libffi.
3069 Shared @file{libgcc_s.so} is now built and installed by default.
3074 @heading @anchor{h8300-hms}h8300-hms
3075 Renesas H8/300 series of processors.
3077 Please have a look at the @uref{binaries.html,,binaries page}.
3079 The calling convention and structure layout has changed in release 2.6.
3080 All code must be recompiled. The calling convention now passes the
3081 first three arguments in function calls in registers. Structures are no
3082 longer a multiple of 2 bytes.
3087 @heading @anchor{hppa-hp-hpux}hppa*-hp-hpux*
3088 Support for HP-UX version 9 and older was discontinued in GCC 3.4.
3090 We require using gas/binutils on all hppa platforms. Version 2.19 or
3091 later is recommended.
3093 It may be helpful to configure GCC with the
3094 @uref{./configure.html#with-gnu-as,,@option{--with-gnu-as}} and
3095 @option{--with-as=@dots{}} options to ensure that GCC can find GAS@.
3097 The HP assembler should not be used with GCC. It is rarely tested and may
3098 not work. It shouldn't be used with any languages other than C due to its
3101 Specifically, @option{-g} does not work (HP-UX uses a peculiar debugging
3102 format which GCC does not know about). It also inserts timestamps
3103 into each object file it creates, causing the 3-stage comparison test to
3104 fail during a bootstrap. You should be able to continue by saying
3105 @samp{make all-host all-target} after getting the failure from @samp{make}.
3107 Various GCC features are not supported. For example, it does not support weak
3108 symbols or alias definitions. As a result, explicit template instantiations
3109 are required when using C++. This makes it difficult if not impossible to
3110 build many C++ applications.
3112 There are two default scheduling models for instructions. These are
3113 PROCESSOR_7100LC and PROCESSOR_8000. They are selected from the pa-risc
3114 architecture specified for the target machine when configuring.
3115 PROCESSOR_8000 is the default. PROCESSOR_7100LC is selected when
3116 the target is a @samp{hppa1*} machine.
3118 The PROCESSOR_8000 model is not well suited to older processors. Thus,
3119 it is important to completely specify the machine architecture when
3120 configuring if you want a model other than PROCESSOR_8000. The macro
3121 TARGET_SCHED_DEFAULT can be defined in BOOT_CFLAGS if a different
3122 default scheduling model is desired.
3124 As of GCC 4.0, GCC uses the UNIX 95 namespace for HP-UX 10.10
3125 through 11.00, and the UNIX 98 namespace for HP-UX 11.11 and later.
3126 This namespace change might cause problems when bootstrapping with
3127 an earlier version of GCC or the HP compiler as essentially the same
3128 namespace is required for an entire build. This problem can be avoided
3129 in a number of ways. With HP cc, @env{UNIX_STD} can be set to @samp{95}
3130 or @samp{98}. Another way is to add an appropriate set of predefines
3131 to @env{CC}. The description for the @option{munix=} option contains
3132 a list of the predefines used with each standard.
3134 More specific information to @samp{hppa*-hp-hpux*} targets follows.
3139 @heading @anchor{hppa-hp-hpux10}hppa*-hp-hpux10
3141 For hpux10.20, we @emph{highly} recommend you pick up the latest sed patch
3142 @code{PHCO_19798} from HP@. HP has two sites which provide patches free of
3148 <a href="http://us.itrc.hp.com/service/home/home.do">US, Canada, Asia-Pacific, and
3152 @uref{http://us.itrc.hp.com/service/home/home.do,,} US, Canada, Asia-Pacific,
3156 @uref{http://europe.itrc.hp.com/service/home/home.do,,} Europe.
3159 The C++ ABI has changed incompatibly in GCC 4.0. COMDAT subspaces are
3160 used for one-only code and data. This resolves many of the previous
3161 problems in using C++ on this target. However, the ABI is not compatible
3162 with the one implemented under HP-UX 11 using secondary definitions.
3167 @heading @anchor{hppa-hp-hpux11}hppa*-hp-hpux11
3169 GCC 3.0 and up support HP-UX 11. GCC 2.95.x is not supported and cannot
3170 be used to compile GCC 3.0 and up.
3172 The libffi and libjava libraries haven't been ported to 64-bit HP-UX@
3175 Refer to @uref{binaries.html,,binaries} for information about obtaining
3176 precompiled GCC binaries for HP-UX@. Precompiled binaries must be obtained
3177 to build the Ada language as it can't be bootstrapped using C@. Ada is
3178 only available for the 32-bit PA-RISC runtime.
3180 Starting with GCC 3.4 an ISO C compiler is required to bootstrap. The
3181 bundled compiler supports only traditional C; you will need either HP's
3182 unbundled compiler, or a binary distribution of GCC@.
3184 It is possible to build GCC 3.3 starting with the bundled HP compiler,
3185 but the process requires several steps. GCC 3.3 can then be used to
3186 build later versions. The fastjar program contains ISO C code and
3187 can't be built with the HP bundled compiler. This problem can be
3188 avoided by not building the Java language. For example, use the
3189 @option{--enable-languages="c,c++,f77,objc"} option in your configure
3192 There are several possible approaches to building the distribution.
3193 Binutils can be built first using the HP tools. Then, the GCC
3194 distribution can be built. The second approach is to build GCC
3195 first using the HP tools, then build binutils, then rebuild GCC@.
3196 There have been problems with various binary distributions, so it
3197 is best not to start from a binary distribution.
3199 On 64-bit capable systems, there are two distinct targets. Different
3200 installation prefixes must be used if both are to be installed on
3201 the same system. The @samp{hppa[1-2]*-hp-hpux11*} target generates code
3202 for the 32-bit PA-RISC runtime architecture and uses the HP linker.
3203 The @samp{hppa64-hp-hpux11*} target generates 64-bit code for the
3204 PA-RISC 2.0 architecture.
3206 The script config.guess now selects the target type based on the compiler
3207 detected during configuration. You must define @env{PATH} or @env{CC} so
3208 that configure finds an appropriate compiler for the initial bootstrap.
3209 When @env{CC} is used, the definition should contain the options that are
3210 needed whenever @env{CC} is used.
3212 Specifically, options that determine the runtime architecture must be
3213 in @env{CC} to correctly select the target for the build. It is also
3214 convenient to place many other compiler options in @env{CC}. For example,
3215 @env{CC="cc -Ac +DA2.0W -Wp,-H16376 -D_CLASSIC_TYPES -D_HPUX_SOURCE"}
3216 can be used to bootstrap the GCC 3.3 branch with the HP compiler in
3217 64-bit K&R/bundled mode. The @option{+DA2.0W} option will result in
3218 the automatic selection of the @samp{hppa64-hp-hpux11*} target. The
3219 macro definition table of cpp needs to be increased for a successful
3220 build with the HP compiler. _CLASSIC_TYPES and _HPUX_SOURCE need to
3221 be defined when building with the bundled compiler, or when using the
3222 @option{-Ac} option. These defines aren't necessary with @option{-Ae}.
3224 It is best to explicitly configure the @samp{hppa64-hp-hpux11*} target
3225 with the @option{--with-ld=@dots{}} option. This overrides the standard
3226 search for ld. The two linkers supported on this target require different
3227 commands. The default linker is determined during configuration. As a
3228 result, it's not possible to switch linkers in the middle of a GCC build.
3229 This has been reported to sometimes occur in unified builds of binutils
3232 A recent linker patch must be installed for the correct operation of
3233 GCC 3.3 and later. @code{PHSS_26559} and @code{PHSS_24304} are the
3234 oldest linker patches that are known to work. They are for HP-UX
3235 11.00 and 11.11, respectively. @code{PHSS_24303}, the companion to
3236 @code{PHSS_24304}, might be usable but it hasn't been tested. These
3237 patches have been superseded. Consult the HP patch database to obtain
3238 the currently recommended linker patch for your system.
3240 The patches are necessary for the support of weak symbols on the
3241 32-bit port, and for the running of initializers and finalizers. Weak
3242 symbols are implemented using SOM secondary definition symbols. Prior
3243 to HP-UX 11, there are bugs in the linker support for secondary symbols.
3244 The patches correct a problem of linker core dumps creating shared
3245 libraries containing secondary symbols, as well as various other
3246 linking issues involving secondary symbols.
3248 GCC 3.3 uses the ELF DT_INIT_ARRAY and DT_FINI_ARRAY capabilities to
3249 run initializers and finalizers on the 64-bit port. The 32-bit port
3250 uses the linker @option{+init} and @option{+fini} options for the same
3251 purpose. The patches correct various problems with the +init/+fini
3252 options, including program core dumps. Binutils 2.14 corrects a
3253 problem on the 64-bit port resulting from HP's non-standard use of
3254 the .init and .fini sections for array initializers and finalizers.
3256 Although the HP and GNU linkers are both supported for the
3257 @samp{hppa64-hp-hpux11*} target, it is strongly recommended that the
3258 HP linker be used for link editing on this target.
3260 At this time, the GNU linker does not support the creation of long
3261 branch stubs. As a result, it can't successfully link binaries
3262 containing branch offsets larger than 8 megabytes. In addition,
3263 there are problems linking shared libraries, linking executables
3264 with @option{-static}, and with dwarf2 unwind and exception support.
3265 It also doesn't provide stubs for internal calls to global functions
3266 in shared libraries, so these calls can't be overloaded.
3268 The HP dynamic loader does not support GNU symbol versioning, so symbol
3269 versioning is not supported. It may be necessary to disable symbol
3270 versioning with @option{--disable-symvers} when using GNU ld.
3272 POSIX threads are the default. The optional DCE thread library is not
3273 supported, so @option{--enable-threads=dce} does not work.
3278 @heading @anchor{x-x-linux-gnu}*-*-linux-gnu
3280 Versions of libstdc++-v3 starting with 3.2.1 require bug fixes present
3281 in glibc 2.2.5 and later. More information is available in the
3282 libstdc++-v3 documentation.
3287 @heading @anchor{ix86-x-linux}i?86-*-linux*
3289 As of GCC 3.3, binutils 2.13.1 or later is required for this platform.
3290 See @uref{http://gcc.gnu.org/PR10877,,bug 10877} for more information.
3292 If you receive Signal 11 errors when building on GNU/Linux, then it is
3293 possible you have a hardware problem. Further information on this can be
3294 found on @uref{http://www.bitwizard.nl/sig11/,,www.bitwizard.nl}.
3299 @heading @anchor{ix86-x-solaris210}i?86-*-solaris2.10
3300 Use this for Solaris 10 or later on x86 and x86-64 systems. This
3301 configuration is supported by GCC 4.0 and later versions only.
3303 It is recommended that you configure GCC to use the GNU assembler in
3304 @file{/usr/sfw/bin/gas} but the Sun linker, using the options
3305 @option{--with-gnu-as --with-as=/usr/sfw/bin/gas --without-gnu-ld
3306 --with-ld=/usr/ccs/bin/ld}.
3311 @heading @anchor{ia64-x-linux}ia64-*-linux
3312 IA-64 processor (also known as IPF, or Itanium Processor Family)
3315 If you are using the installed system libunwind library with
3316 @option{--with-system-libunwind}, then you must use libunwind 0.98 or
3319 None of the following versions of GCC has an ABI that is compatible
3320 with any of the other versions in this list, with the exception that
3321 Red Hat 2.96 and Trillian 000171 are compatible with each other:
3322 3.1, 3.0.2, 3.0.1, 3.0, Red Hat 2.96, and Trillian 000717.
3323 This primarily affects C++ programs and programs that create shared libraries.
3324 GCC 3.1 or later is recommended for compiling linux, the kernel.
3325 As of version 3.1 GCC is believed to be fully ABI compliant, and hence no
3326 more major ABI changes are expected.
3331 @heading @anchor{ia64-x-hpux}ia64-*-hpux*
3332 Building GCC on this target requires the GNU Assembler. The bundled HP
3333 assembler will not work. To prevent GCC from using the wrong assembler,
3334 the option @option{--with-gnu-as} may be necessary.
3336 The GCC libunwind library has not been ported to HPUX@. This means that for
3337 GCC versions 3.2.3 and earlier, @option{--enable-libunwind-exceptions}
3338 is required to build GCC@. For GCC 3.3 and later, this is the default.
3339 For gcc 3.4.3 and later, @option{--enable-libunwind-exceptions} is
3340 removed and the system libunwind library will always be used.
3344 <!-- rs6000-ibm-aix*, powerpc-ibm-aix* -->
3346 @heading @anchor{x-ibm-aix}*-ibm-aix*
3347 Support for AIX version 3 and older was discontinued in GCC 3.4.
3348 Support for AIX version 4.2 and older was discontinued in GCC 4.5.
3350 ``out of memory'' bootstrap failures may indicate a problem with
3351 process resource limits (ulimit). Hard limits are configured in the
3352 @file{/etc/security/limits} system configuration file.
3354 To speed up the configuration phases of bootstrapping and installing GCC,
3355 one may use GNU Bash instead of AIX @command{/bin/sh}, e.g.,
3358 % CONFIG_SHELL=/opt/freeware/bin/bash
3359 % export CONFIG_SHELL
3362 and then proceed as described in @uref{build.html,,the build
3363 instructions}, where we strongly recommend specifying an absolute path
3364 to invoke @var{srcdir}/configure.
3366 Because GCC on AIX is built as a 32-bit executable by default,
3367 (although it can generate 64-bit programs) the GMP and MPFR libraries
3368 required by gfortran must be 32-bit libraries. Building GMP and MPFR
3369 as static archive libraries works better than shared libraries.
3371 Errors involving @code{alloca} when building GCC generally are due
3372 to an incorrect definition of @code{CC} in the Makefile or mixing files
3373 compiled with the native C compiler and GCC@. During the stage1 phase of
3374 the build, the native AIX compiler @strong{must} be invoked as @command{cc}
3375 (not @command{xlc}). Once @command{configure} has been informed of
3376 @command{xlc}, one needs to use @samp{make distclean} to remove the
3377 configure cache files and ensure that @env{CC} environment variable
3378 does not provide a definition that will confuse @command{configure}.
3379 If this error occurs during stage2 or later, then the problem most likely
3380 is the version of Make (see above).
3382 The native @command{as} and @command{ld} are recommended for bootstrapping
3383 on AIX 4 and required for bootstrapping on AIX 5L@. The GNU Assembler
3384 reports that it supports WEAK symbols on AIX 4, which causes GCC to try to
3385 utilize weak symbol functionality although it is not supported. The GNU
3386 Assembler and Linker do not support AIX 5L sufficiently to bootstrap GCC@.
3387 The native AIX tools do interoperate with GCC@.
3389 Building @file{libstdc++.a} requires a fix for an AIX Assembler bug
3390 APAR IY26685 (AIX 4.3) or APAR IY25528 (AIX 5.1). It also requires a
3391 fix for another AIX Assembler bug and a co-dependent AIX Archiver fix
3392 referenced as APAR IY53606 (AIX 5.2) or as APAR IY54774 (AIX 5.1)
3394 @samp{libstdc++} in GCC 3.4 increments the major version number of the
3395 shared object and GCC installation places the @file{libstdc++.a}
3396 shared library in a common location which will overwrite the and GCC
3397 3.3 version of the shared library. Applications either need to be
3398 re-linked against the new shared library or the GCC 3.1 and GCC 3.3
3399 versions of the @samp{libstdc++} shared object needs to be available
3400 to the AIX runtime loader. The GCC 3.1 @samp{libstdc++.so.4}, if
3401 present, and GCC 3.3 @samp{libstdc++.so.5} shared objects can be
3402 installed for runtime dynamic loading using the following steps to set
3403 the @samp{F_LOADONLY} flag in the shared object for @emph{each}
3404 multilib @file{libstdc++.a} installed:
3406 Extract the shared objects from the currently installed
3407 @file{libstdc++.a} archive:
3409 % ar -x libstdc++.a libstdc++.so.4 libstdc++.so.5
3412 Enable the @samp{F_LOADONLY} flag so that the shared object will be
3413 available for runtime dynamic loading, but not linking:
3415 % strip -e libstdc++.so.4 libstdc++.so.5
3418 Archive the runtime-only shared object in the GCC 3.4
3419 @file{libstdc++.a} archive:
3421 % ar -q libstdc++.a libstdc++.so.4 libstdc++.so.5
3424 Linking executables and shared libraries may produce warnings of
3425 duplicate symbols. The assembly files generated by GCC for AIX always
3426 have included multiple symbol definitions for certain global variable
3427 and function declarations in the original program. The warnings should
3428 not prevent the linker from producing a correct library or runnable
3431 AIX 4.3 utilizes a ``large format'' archive to support both 32-bit and
3432 64-bit object modules. The routines provided in AIX 4.3.0 and AIX 4.3.1
3433 to parse archive libraries did not handle the new format correctly.
3434 These routines are used by GCC and result in error messages during
3435 linking such as ``not a COFF file''. The version of the routines shipped
3436 with AIX 4.3.1 should work for a 32-bit environment. The @option{-g}
3437 option of the archive command may be used to create archives of 32-bit
3438 objects using the original ``small format''. A correct version of the
3439 routines is shipped with AIX 4.3.2 and above.
3441 Some versions of the AIX binder (linker) can fail with a relocation
3442 overflow severe error when the @option{-bbigtoc} option is used to link
3443 GCC-produced object files into an executable that overflows the TOC@. A fix
3444 for APAR IX75823 (OVERFLOW DURING LINK WHEN USING GCC AND -BBIGTOC) is
3445 available from IBM Customer Support and from its
3446 @uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com}
3447 website as PTF U455193.
3449 The AIX 4.3.2.1 linker (bos.rte.bind_cmds Level 4.3.2.1) will dump core
3450 with a segmentation fault when invoked by any version of GCC@. A fix for
3451 APAR IX87327 is available from IBM Customer Support and from its
3452 @uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com}
3453 website as PTF U461879. This fix is incorporated in AIX 4.3.3 and above.
3455 The initial assembler shipped with AIX 4.3.0 generates incorrect object
3456 files. A fix for APAR IX74254 (64BIT DISASSEMBLED OUTPUT FROM COMPILER FAILS
3457 TO ASSEMBLE/BIND) is available from IBM Customer Support and from its
3458 @uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com}
3459 website as PTF U453956. This fix is incorporated in AIX 4.3.1 and above.
3461 AIX provides National Language Support (NLS)@. Compilers and assemblers
3462 use NLS to support locale-specific representations of various data
3463 formats including floating-point numbers (e.g., @samp{.} vs @samp{,} for
3464 separating decimal fractions). There have been problems reported where
3465 GCC does not produce the same floating-point formats that the assembler
3466 expects. If one encounters this problem, set the @env{LANG}
3467 environment variable to @samp{C} or @samp{En_US}.
3469 A default can be specified with the @option{-mcpu=@var{cpu_type}}
3470 switch and using the configure option @option{--with-cpu-@var{cpu_type}}.
3475 @heading @anchor{iq2000-x-elf}iq2000-*-elf
3476 Vitesse IQ2000 processors. These are used in embedded
3477 applications. There are no standard Unix configurations.
3482 @heading @anchor{m32c-x-elf}m32c-*-elf
3483 Renesas M32C processor.
3484 This configuration is intended for embedded systems.
3489 @heading @anchor{m32r-x-elf}m32r-*-elf
3490 Renesas M32R processor.
3491 This configuration is intended for embedded systems.
3496 @heading @anchor{m6811-elf}m6811-elf
3497 Motorola 68HC11 family micro controllers. These are used in embedded
3498 applications. There are no standard Unix configurations.
3503 @heading @anchor{m6812-elf}m6812-elf
3504 Motorola 68HC12 family micro controllers. These are used in embedded
3505 applications. There are no standard Unix configurations.
3510 @heading @anchor{m68k-x-x}m68k-*-*
3512 @samp{m68k-*-elf*}, @samp{m68k-*-rtems}, @samp{m68k-*-uclinux} and
3514 build libraries for both M680x0 and ColdFire processors. If you only
3515 need the M680x0 libraries, you can omit the ColdFire ones by passing
3516 @option{--with-arch=m68k} to @command{configure}. Alternatively, you
3517 can omit the M680x0 libraries by passing @option{--with-arch=cf} to
3518 @command{configure}. These targets default to 5206 or 5475 code as
3519 appropriate for the target system when
3520 configured with @option{--with-arch=cf} and 68020 code otherwise.
3522 The @samp{m68k-*-netbsd} and
3523 @samp{m68k-*-openbsd} targets also support the @option{--with-arch}
3524 option. They will generate ColdFire CFV4e code when configured with
3525 @option{--with-arch=cf} and 68020 code otherwise.
3527 You can override the default processors listed above by configuring
3528 with @option{--with-cpu=@var{target}}. This @var{target} can either
3529 be a @option{-mcpu} argument or one of the following values:
3530 @samp{m68000}, @samp{m68010}, @samp{m68020}, @samp{m68030},
3531 @samp{m68040}, @samp{m68060}, @samp{m68020-40} and @samp{m68020-60}.
3536 @heading @anchor{m68k-x-uclinux}m68k-*-uclinux
3537 GCC 4.3 changed the uClinux configuration so that it uses the
3538 @samp{m68k-linux-gnu} ABI rather than the @samp{m68k-elf} ABI.
3539 It also added improved support for C++ and flat shared libraries,
3540 both of which were ABI changes. However, you can still use the
3541 original ABI by configuring for @samp{m68k-uclinuxoldabi} or
3542 @samp{m68k-@var{vendor}-uclinuxoldabi}.
3547 @heading @anchor{mips-x-x}mips-*-*
3548 If on a MIPS system you get an error message saying ``does not have gp
3549 sections for all it's [sic] sectons [sic]'', don't worry about it. This
3550 happens whenever you use GAS with the MIPS linker, but there is not
3551 really anything wrong, and it is okay to use the output file. You can
3552 stop such warnings by installing the GNU linker.
3554 It would be nice to extend GAS to produce the gp tables, but they are
3555 optional, and there should not be a warning about their absence.
3557 The libstdc++ atomic locking routines for MIPS targets requires MIPS II
3558 and later. A patch went in just after the GCC 3.3 release to
3559 make @samp{mips*-*-*} use the generic implementation instead. You can also
3560 configure for @samp{mipsel-elf} as a workaround. The
3561 @samp{mips*-*-linux*} target continues to use the MIPS II routines. More
3562 work on this is expected in future releases.
3564 @c If you make --with-llsc the default for another target, please also
3565 @c update the description of the --with-llsc option.
3567 The built-in @code{__sync_*} functions are available on MIPS II and
3568 later systems and others that support the @samp{ll}, @samp{sc} and
3569 @samp{sync} instructions. This can be overridden by passing
3570 @option{--with-llsc} or @option{--without-llsc} when configuring GCC.
3571 Since the Linux kernel emulates these instructions if they are
3572 missing, the default for @samp{mips*-*-linux*} targets is
3573 @option{--with-llsc}. The @option{--with-llsc} and
3574 @option{--without-llsc} configure options may be overridden at compile
3575 time by passing the @option{-mllsc} or @option{-mno-llsc} options to
3578 MIPS systems check for division by zero (unless
3579 @option{-mno-check-zero-division} is passed to the compiler) by
3580 generating either a conditional trap or a break instruction. Using
3581 trap results in smaller code, but is only supported on MIPS II and
3582 later. Also, some versions of the Linux kernel have a bug that
3583 prevents trap from generating the proper signal (@code{SIGFPE}). To enable
3584 the use of break, use the @option{--with-divide=breaks}
3585 @command{configure} option when configuring GCC@. The default is to
3586 use traps on systems that support them.
3588 Cross-compilers for the MIPS as target using the MIPS assembler
3589 currently do not work, because the auxiliary programs
3590 @file{mips-tdump.c} and @file{mips-tfile.c} can't be compiled on
3591 anything but a MIPS@. It does work to cross compile for a MIPS
3592 if you use the GNU assembler and linker.
3594 The assembler from GNU binutils 2.17 and earlier has a bug in the way
3595 it sorts relocations for REL targets (o32, o64, EABI). This can cause
3596 bad code to be generated for simple C++ programs. Also the linker
3597 from GNU binutils versions prior to 2.17 has a bug which causes the
3598 runtime linker stubs in very large programs, like @file{libgcj.so}, to
3599 be incorrectly generated. GNU Binutils 2.18 and later (and snapshots
3600 made after Nov. 9, 2006) should be free from both of these problems.
3605 @heading @anchor{mips-sgi-irix5}mips-sgi-irix5
3607 In order to compile GCC on an SGI running IRIX 5, the @samp{compiler_dev.hdr}
3608 subsystem must be installed from the IDO CD-ROM supplied by SGI@.
3609 It is also available for download from
3610 @uref{ftp://ftp.sgi.com/sgi/IRIX5.3/iris-development-option-5.3.tardist}.
3612 If you use the MIPS C compiler to bootstrap, it may be necessary
3613 to increase its table size for switch statements with the
3614 @option{-Wf,-XNg1500} option. If you use the @option{-O2}
3615 optimization option, you also need to use @option{-Olimit 3000}.
3617 To enable debugging under IRIX 5, you must use GNU binutils 2.15 or
3618 later, and use the @option{--with-gnu-ld} @command{configure} option
3619 when configuring GCC@. You need to use GNU @command{ar} and @command{nm},
3620 also distributed with GNU binutils.
3622 Some users have reported that @command{/bin/sh} will hang during bootstrap.
3623 This problem can be avoided by running the commands:
3626 % CONFIG_SHELL=/bin/ksh
3627 % export CONFIG_SHELL
3630 before starting the build.
3635 @heading @anchor{mips-sgi-irix6}mips-sgi-irix6
3637 If you are using SGI's MIPSpro @command{cc} as your bootstrap compiler, you must
3638 ensure that the N32 ABI is in use. To test this, compile a simple C
3639 file with @command{cc} and then run @command{file} on the
3640 resulting object file. The output should look like:
3643 test.o: ELF N32 MSB @dots{}
3649 test.o: ELF 32-bit MSB @dots{}
3655 test.o: ELF 64-bit MSB @dots{}
3658 then your version of @command{cc} uses the O32 or N64 ABI by default. You
3659 should set the environment variable @env{CC} to @samp{cc -n32}
3660 before configuring GCC@.
3662 If you want the resulting @command{gcc} to run on old 32-bit systems
3663 with the MIPS R4400 CPU, you need to ensure that only code for the @samp{mips3}
3664 instruction set architecture (ISA) is generated. While GCC 3.x does
3665 this correctly, both GCC 2.95 and SGI's MIPSpro @command{cc} may change
3666 the ISA depending on the machine where GCC is built. Using one of them
3667 as the bootstrap compiler may result in @samp{mips4} code, which won't run at
3668 all on @samp{mips3}-only systems. For the test program above, you should see:
3671 test.o: ELF N32 MSB mips-3 @dots{}
3677 test.o: ELF N32 MSB mips-4 @dots{}
3680 instead, you should set the environment variable @env{CC} to @samp{cc
3681 -n32 -mips3} or @samp{gcc -mips3} respectively before configuring GCC@.
3683 MIPSpro C 7.4 may cause bootstrap failures, due to a bug when inlining
3684 @code{memcmp}. Either add @code{-U__INLINE_INTRINSICS} to the @env{CC}
3685 environment variable as a workaround or upgrade to MIPSpro C 7.4.1m.
3687 GCC on IRIX 6 is usually built to support the N32, O32 and N64 ABIs. If
3688 you build GCC on a system that doesn't have the N64 libraries installed
3689 or cannot run 64-bit binaries,
3690 you need to configure with @option{--disable-multilib} so GCC doesn't
3691 try to use them. This will disable building the O32 libraries, too.
3692 Look for @file{/usr/lib64/libc.so.1} to see if you
3693 have the 64-bit libraries installed.
3695 To enable debugging for the O32 ABI, you must use GNU @command{as} from
3696 GNU binutils 2.15 or later. You may also use GNU @command{ld}, but
3697 this is not required and currently causes some problems with Ada.
3699 The @option{--enable-libgcj}
3700 option is disabled by default: IRIX 6 uses a very low default limit
3701 (20480) for the command line length. Although @command{libtool} contains a
3702 workaround for this problem, at least the N64 @samp{libgcj} is known not
3703 to build despite this, running into an internal error of the native
3704 @command{ld}. A sure fix is to increase this limit (@samp{ncargs}) to
3705 its maximum of 262144 bytes. If you have root access, you can use the
3706 @command{systune} command to do this.
3708 @code{wchar_t} support in @samp{libstdc++} is not available for old
3709 IRIX 6.5.x releases, @math{x < 19}. The problem cannot be autodetected
3710 and in order to build GCC for such targets you need to configure with
3711 @option{--disable-wchar_t}.
3713 See @uref{http://freeware.sgi.com/} for more
3714 information about using GCC on IRIX platforms.
3719 @heading @anchor{powerpc-x-x}powerpc-*-*
3721 You can specify a default version for the @option{-mcpu=@var{cpu_type}}
3722 switch by using the configure option @option{--with-cpu-@var{cpu_type}}.
3725 @uref{ftp://ftp.kernel.org/pub/linux/devel/binutils,,binutils 2.15}
3726 or newer for a working GCC@.
3731 @heading @anchor{powerpc-x-darwin}powerpc-*-darwin*
3732 PowerPC running Darwin (Mac OS X kernel).
3734 Pre-installed versions of Mac OS X may not include any developer tools,
3735 meaning that you will not be able to build GCC from source. Tool
3736 binaries are available at
3737 @uref{http://developer.apple.com/darwin/projects/compiler/} (free
3738 registration required).
3740 This version of GCC requires at least cctools-590.36. The
3741 cctools-590.36 package referenced from
3742 @uref{http://gcc.gnu.org/ml/gcc/2006-03/msg00507.html} will not work
3743 on systems older than 10.3.9 (aka darwin7.9.0).
3748 @heading @anchor{powerpc-x-elf}powerpc-*-elf
3749 PowerPC system in big endian mode, running System V.4.
3754 @heading @anchor{powerpc-x-linux-gnu}powerpc*-*-linux-gnu*
3756 PowerPC system in big endian mode running Linux.
3761 @heading @anchor{powerpc-x-netbsd}powerpc-*-netbsd*
3762 PowerPC system in big endian mode running NetBSD@.
3767 @heading @anchor{powerpc-x-eabisim}powerpc-*-eabisim
3768 Embedded PowerPC system in big endian mode for use in running under the
3774 @heading @anchor{powerpc-x-eabi}powerpc-*-eabi
3775 Embedded PowerPC system in big endian mode.
3780 @heading @anchor{powerpcle-x-elf}powerpcle-*-elf
3781 PowerPC system in little endian mode, running System V.4.
3786 @heading @anchor{powerpcle-x-eabisim}powerpcle-*-eabisim
3787 Embedded PowerPC system in little endian mode for use in running under
3793 @heading @anchor{powerpcle-x-eabi}powerpcle-*-eabi
3794 Embedded PowerPC system in little endian mode.
3799 @heading @anchor{s390-x-linux}s390-*-linux*
3800 S/390 system running GNU/Linux for S/390@.
3805 @heading @anchor{s390x-x-linux}s390x-*-linux*
3806 zSeries system (64-bit) running GNU/Linux for zSeries@.
3811 @heading @anchor{s390x-ibm-tpf}s390x-ibm-tpf*
3812 zSeries system (64-bit) running TPF@. This platform is
3813 supported as cross-compilation target only.
3818 @c Please use Solaris 2 to refer to all release of Solaris, starting
3819 @c with 2.0 until 2.6, 7, 8, etc. Solaris 1 was a marketing name for
3820 @c SunOS 4 releases which we don't use to avoid confusion. Solaris
3821 @c alone is too unspecific and must be avoided.
3822 @heading @anchor{x-x-solaris2}*-*-solaris2*
3824 Sun does not ship a C compiler with Solaris 2. To bootstrap and install
3825 GCC you first have to install a pre-built compiler, see the
3826 @uref{binaries.html,,binaries page} for details.
3828 The Solaris 2 @command{/bin/sh} will often fail to configure
3829 @file{libstdc++-v3}, @file{boehm-gc} or @file{libjava}. We therefore
3830 recommend using the following initial sequence of commands
3833 % CONFIG_SHELL=/bin/ksh
3834 % export CONFIG_SHELL
3837 and proceed as described in @uref{configure.html,,the configure instructions}.
3838 In addition we strongly recommend specifying an absolute path to invoke
3839 @var{srcdir}/configure.
3841 Solaris 2 comes with a number of optional OS packages. Some of these
3842 are needed to use GCC fully, namely @code{SUNWarc},
3843 @code{SUNWbtool}, @code{SUNWesu}, @code{SUNWhea}, @code{SUNWlibm},
3844 @code{SUNWsprot}, and @code{SUNWtoo}. If you did not install all
3845 optional packages when installing Solaris 2, you will need to verify that
3846 the packages that GCC needs are installed.
3848 To check whether an optional package is installed, use
3849 the @command{pkginfo} command. To add an optional package, use the
3850 @command{pkgadd} command. For further details, see the Solaris 2
3853 Trying to use the linker and other tools in
3854 @file{/usr/ucb} to install GCC has been observed to cause trouble.
3855 For example, the linker may hang indefinitely. The fix is to remove
3856 @file{/usr/ucb} from your @env{PATH}.
3858 The build process works more smoothly with the legacy Sun tools so, if you
3859 have @file{/usr/xpg4/bin} in your @env{PATH}, we recommend that you place
3860 @file{/usr/bin} before @file{/usr/xpg4/bin} for the duration of the build.
3862 We recommend the use of GNU binutils 2.14 or later, or the vendor tools
3863 (Sun @command{as}, Sun @command{ld}). Note that your mileage may vary
3864 if you use a combination of the GNU tools and the Sun tools: while the
3865 combination GNU @command{as} + Sun @command{ld} should reasonably work,
3866 the reverse combination Sun @command{as} + GNU @command{ld} is known to
3867 cause memory corruption at runtime in some cases for C++ programs.
3869 The stock GNU binutils 2.15 release is broken on this platform because of a
3870 single bug. It has been fixed on the 2.15 branch in the CVS repository.
3871 You can obtain a working version by checking out the binutils-2_15-branch
3872 from the CVS repository or applying the patch
3873 @uref{http://sourceware.org/ml/binutils-cvs/2004-09/msg00036.html} to the
3876 We recommend the use of GNU binutils 2.16 or later in conjunction with GCC
3877 4.x, or the vendor tools (Sun @command{as}, Sun @command{ld}). However,
3878 for Solaris 10 and above, an additional patch is required in order for the
3879 GNU linker to be able to cope with a new flavor of shared libraries. You
3880 can obtain a working version by checking out the binutils-2_16-branch from
3881 the CVS repository or applying the patch
3882 @uref{http://sourceware.org/ml/binutils-cvs/2005-07/msg00122.html} to the
3885 Sun bug 4296832 turns up when compiling X11 headers with GCC 2.95 or
3886 newer: @command{g++} will complain that types are missing. These headers
3887 assume that omitting the type means @code{int}; this assumption worked for
3888 C89 but is wrong for C++, and is now wrong for C99 also.
3890 @command{g++} accepts such (invalid) constructs with the option
3891 @option{-fpermissive}; it will assume that any missing type is @code{int}
3892 (as defined by C89).
3894 There are patches for Solaris 7 (108376-21 or newer for SPARC,
3895 108377-20 for Intel), and Solaris 8 (108652-24 or newer for SPARC,
3896 108653-22 for Intel) that fix this bug.
3898 Sun bug 4927647 sometimes causes random spurious testsuite failures
3899 related to missing diagnostic output. This bug doesn't affect GCC
3900 itself, rather it is a kernel bug triggered by the @command{expect}
3901 program which is used only by the GCC testsuite driver. When the bug
3902 causes the @command{expect} program to miss anticipated output, extra
3903 testsuite failures appear.
3905 There are patches for Solaris 8 (117350-12 or newer for SPARC,
3906 117351-12 or newer for Intel) and Solaris 9 (117171-11 or newer for
3907 SPARC, 117172-11 or newer for Intel) that address this problem.
3912 @heading @anchor{sparc-sun-solaris2}sparc-sun-solaris2*
3914 When GCC is configured to use binutils 2.14 or later the binaries
3915 produced are smaller than the ones produced using Sun's native tools;
3916 this difference is quite significant for binaries containing debugging
3919 Starting with Solaris 7, the operating system is capable of executing
3920 64-bit SPARC V9 binaries. GCC 3.1 and later properly supports
3921 this; the @option{-m64} option enables 64-bit code generation.
3922 However, if all you want is code tuned for the UltraSPARC CPU, you
3923 should try the @option{-mtune=ultrasparc} option instead, which produces
3924 code that, unlike full 64-bit code, can still run on non-UltraSPARC
3927 When configuring on a Solaris 7 or later system that is running a kernel
3928 that supports only 32-bit binaries, one must configure with
3929 @option{--disable-multilib}, since we will not be able to build the
3930 64-bit target libraries.
3932 GCC 3.3 and GCC 3.4 trigger code generation bugs in earlier versions of
3933 the GNU compiler (especially GCC 3.0.x versions), which lead to the
3934 miscompilation of the stage1 compiler and the subsequent failure of the
3935 bootstrap process. A workaround is to use GCC 3.2.3 as an intermediary
3936 stage, i.e.@: to bootstrap that compiler with the base compiler and then
3937 use it to bootstrap the final compiler.
3939 GCC 3.4 triggers a code generation bug in versions 5.4 (Sun ONE Studio 7)
3940 and 5.5 (Sun ONE Studio 8) of the Sun compiler, which causes a bootstrap
3941 failure in form of a miscompilation of the stage1 compiler by the Sun
3942 compiler. This is Sun bug 4974440. This is fixed with patch 112760-07.
3944 GCC 3.4 changed the default debugging format from STABS to DWARF-2 for
3945 32-bit code on Solaris 7 and later. If you use the Sun assembler, this
3946 change apparently runs afoul of Sun bug 4910101 (which is referenced as
3947 an x86-only problem by Sun, probably because they do not use DWARF-2).
3948 A symptom of the problem is that you cannot compile C++ programs like
3949 @command{groff} 1.19.1 without getting messages similar to the following:
3952 ld: warning: relocation error: R_SPARC_UA32: @dots{}
3953 external symbolic relocation against non-allocatable section
3954 .debug_info cannot be processed at runtime: relocation ignored.
3957 To work around this problem, compile with @option{-gstabs+} instead of
3960 When configuring the GNU Multiple Precision Library (GMP) or the MPFR
3961 library on a Solaris 7 or later system, the canonical target triplet
3962 must be specified as the @command{build} parameter on the configure
3963 line. This triplet can be obtained by invoking ./config.guess in
3964 the toplevel source directory of GCC (and not that of GMP or MPFR).
3965 For example on a Solaris 7 system:
3968 % ./configure --build=sparc-sun-solaris2.7 --prefix=xxx
3974 @heading @anchor{sparc-sun-solaris27}sparc-sun-solaris2.7
3976 Sun patch 107058-01 (1999-01-13) for Solaris 7/SPARC triggers a bug in
3977 the dynamic linker. This problem (Sun bug 4210064) affects GCC 2.8
3978 and later, including all EGCS releases. Sun formerly recommended
3979 107058-01 for all Solaris 7 users, but around 1999-09-01 it started to
3980 recommend it only for people who use Sun's compilers.
3982 Here are some workarounds to this problem:
3985 Do not install Sun patch 107058-01 until after Sun releases a
3986 complete patch for bug 4210064. This is the simplest course to take,
3987 unless you must also use Sun's C compiler. Unfortunately 107058-01
3988 is preinstalled on some new Solaris 7-based hosts, so you may have to
3992 Copy the original, unpatched Solaris 7
3993 @command{/usr/ccs/bin/as} into
3994 @command{/usr/local/libexec/gcc/sparc-sun-solaris2.7/3.4/as},
3995 adjusting the latter name to fit your local conventions and software
3999 Install Sun patch 106950-03 (1999-05-25) or later. Nobody with
4000 both 107058-01 and 106950-03 installed has reported the bug with GCC
4001 and Sun's dynamic linker. This last course of action is riskiest,
4002 for two reasons. First, you must install 106950 on all hosts that
4003 run code generated by GCC; it doesn't suffice to install it only on
4004 the hosts that run GCC itself. Second, Sun says that 106950-03 is
4005 only a partial fix for bug 4210064, but Sun doesn't know whether the
4006 partial fix is adequate for GCC@. Revision -08 or later should fix
4007 the bug. The current (as of 2004-05-23) revision is -24, and is included in
4008 the Solaris 7 Recommended Patch Cluster.
4011 GCC 3.3 triggers a bug in version 5.0 Alpha 03/27/98 of the Sun assembler,
4012 which causes a bootstrap failure when linking the 64-bit shared version of
4013 libgcc. A typical error message is:
4016 ld: fatal: relocation error: R_SPARC_32: file libgcc/sparcv9/_muldi3.o:
4017 symbol <unknown>: offset 0xffffffff7ec133e7 is non-aligned.
4020 This bug has been fixed in the final 5.0 version of the assembler.
4022 A similar problem was reported for version Sun WorkShop 6 99/08/18 of the
4023 Sun assembler, which causes a bootstrap failure with GCC 4.0.0:
4026 ld: fatal: relocation error: R_SPARC_DISP32:
4027 file .libs/libstdc++.lax/libsupc++convenience.a/vterminate.o:
4028 symbol <unknown>: offset 0xfccd33ad is non-aligned
4031 This bug has been fixed in more recent revisions of the assembler.
4036 @heading @anchor{sparc-x-linux}sparc-*-linux*
4038 GCC versions 3.0 and higher require binutils 2.11.2 and glibc 2.2.4
4039 or newer on this platform. All earlier binutils and glibc
4040 releases mishandled unaligned relocations on @code{sparc-*-*} targets.
4046 @heading @anchor{sparc64-x-solaris2}sparc64-*-solaris2*
4048 When configuring the GNU Multiple Precision Library (GMP) or the
4049 MPFR library, the canonical target triplet must be specified as
4050 the @command{build} parameter on the configure line. For example
4051 on a Solaris 7 system:
4054 % ./configure --build=sparc64-sun-solaris2.7 --prefix=xxx
4057 The following compiler flags must be specified in the configure
4058 step in order to bootstrap this target with the Sun compiler:
4061 % CC="cc -xarch=v9 -xildoff" @var{srcdir}/configure [@var{options}] [@var{target}]
4064 @option{-xarch=v9} specifies the SPARC-V9 architecture to the Sun toolchain
4065 and @option{-xildoff} turns off the incremental linker.
4070 @heading @anchor{sparcv9-x-solaris2}sparcv9-*-solaris2*
4072 This is a synonym for sparc64-*-solaris2*.
4077 @heading @anchor{x-x-vxworks}*-*-vxworks*
4078 Support for VxWorks is in flux. At present GCC supports @emph{only} the
4079 very recent VxWorks 5.5 (aka Tornado 2.2) release, and only on PowerPC@.
4080 We welcome patches for other architectures supported by VxWorks 5.5.
4081 Support for VxWorks AE would also be welcome; we believe this is merely
4082 a matter of writing an appropriate ``configlette'' (see below). We are
4083 not interested in supporting older, a.out or COFF-based, versions of
4086 VxWorks comes with an older version of GCC installed in
4087 @file{@var{$WIND_BASE}/host}; we recommend you do not overwrite it.
4088 Choose an installation @var{prefix} entirely outside @var{$WIND_BASE}.
4089 Before running @command{configure}, create the directories @file{@var{prefix}}
4090 and @file{@var{prefix}/bin}. Link or copy the appropriate assembler,
4091 linker, etc.@: into @file{@var{prefix}/bin}, and set your @var{PATH} to
4092 include that directory while running both @command{configure} and
4095 You must give @command{configure} the
4096 @option{--with-headers=@var{$WIND_BASE}/target/h} switch so that it can
4097 find the VxWorks system headers. Since VxWorks is a cross compilation
4098 target only, you must also specify @option{--target=@var{target}}.
4099 @command{configure} will attempt to create the directory
4100 @file{@var{prefix}/@var{target}/sys-include} and copy files into it;
4101 make sure the user running @command{configure} has sufficient privilege
4104 GCC's exception handling runtime requires a special ``configlette''
4105 module, @file{contrib/gthr_supp_vxw_5x.c}. Follow the instructions in
4106 that file to add the module to your kernel build. (Future versions of
4107 VxWorks will incorporate this module.)
4112 @heading @anchor{x86-64-x-x}x86_64-*-*, amd64-*-*
4114 GCC supports the x86-64 architecture implemented by the AMD64 processor
4115 (amd64-*-* is an alias for x86_64-*-*) on GNU/Linux, FreeBSD and NetBSD@.
4116 On GNU/Linux the default is a bi-arch compiler which is able to generate
4117 both 64-bit x86-64 and 32-bit x86 code (via the @option{-m32} switch).
4122 @heading @anchor{xtensa-x-elf}xtensa*-*-elf
4124 This target is intended for embedded Xtensa systems using the
4125 @samp{newlib} C library. It uses ELF but does not support shared
4126 objects. Designed-defined instructions specified via the
4127 Tensilica Instruction Extension (TIE) language are only supported
4128 through inline assembly.
4130 The Xtensa configuration information must be specified prior to
4131 building GCC@. The @file{include/xtensa-config.h} header
4132 file contains the configuration information. If you created your
4133 own Xtensa configuration with the Xtensa Processor Generator, the
4134 downloaded files include a customized copy of this header file,
4135 which you can use to replace the default header file.
4140 @heading @anchor{xtensa-x-linux}xtensa*-*-linux*
4142 This target is for Xtensa systems running GNU/Linux. It supports ELF
4143 shared objects and the GNU C library (glibc). It also generates
4144 position-independent code (PIC) regardless of whether the
4145 @option{-fpic} or @option{-fPIC} options are used. In other
4146 respects, this target is the same as the
4147 @uref{#xtensa*-*-elf,,@samp{xtensa*-*-elf}} target.
4152 @heading @anchor{windows}Microsoft Windows
4154 @subheading Intel 16-bit versions
4155 The 16-bit versions of Microsoft Windows, such as Windows 3.1, are not
4158 However, the 32-bit port has limited support for Microsoft
4159 Windows 3.11 in the Win32s environment, as a target only. See below.
4161 @subheading Intel 32-bit versions
4163 The 32-bit versions of Windows, including Windows 95, Windows NT, Windows
4164 XP, and Windows Vista, are supported by several different target
4165 platforms. These targets differ in which Windows subsystem they target
4166 and which C libraries are used.
4169 @item Cygwin @uref{#x-x-cygwin,,*-*-cygwin}: Cygwin provides a user-space
4170 Linux API emulation layer in the Win32 subsystem.
4171 @item Interix @uref{#x-x-interix,,*-*-interix}: The Interix subsystem
4172 provides native support for POSIX.
4173 @item MinGW @uref{#x-x-mingw32,,*-*-mingw32}: MinGW is a native GCC port for
4174 the Win32 subsystem that provides a subset of POSIX.
4175 @item MKS i386-pc-mks: NuTCracker from MKS. See
4176 @uref{http://www.mkssoftware.com/} for more information.
4179 @subheading Intel 64-bit versions
4181 GCC contains support for x86-64 using the mingw-w64
4182 runtime library, available from @uref{http://mingw-w64.sourceforge.net/}.
4183 This library should be used with the target triple x86_64-pc-mingw32.
4185 Presently Windows for Itanium is not supported.
4187 @subheading Windows CE
4189 Windows CE is supported as a target only on ARM (arm-wince-pe), Hitachi
4190 SuperH (sh-wince-pe), and MIPS (mips-wince-pe).
4192 @subheading Other Windows Platforms
4194 GCC no longer supports Windows NT on the Alpha or PowerPC.
4196 GCC no longer supports the Windows POSIX subsystem. However, it does
4197 support the Interix subsystem. See above.
4199 Old target names including *-*-winnt and *-*-windowsnt are no longer used.
4201 PW32 (i386-pc-pw32) support was never completed, and the project seems to
4202 be inactive. See @uref{http://pw32.sourceforge.net/} for more information.
4204 UWIN support has been removed due to a lack of maintenance.
4209 @heading @anchor{x-x-cygwin}*-*-cygwin
4211 Ports of GCC are included with the
4212 @uref{http://www.cygwin.com/,,Cygwin environment}.
4214 GCC will build under Cygwin without modification; it does not build
4215 with Microsoft's C++ compiler and there are no plans to make it do so.
4217 Cygwin can be compiled with i?86-pc-cygwin.
4222 @heading @anchor{x-x-interix}*-*-interix
4224 The Interix target is used by OpenNT, Interix, Services For UNIX (SFU),
4225 and Subsystem for UNIX-based Applications (SUA). Applications compiled
4226 with this target run in the Interix subsystem, which is separate from
4227 the Win32 subsystem. This target was last known to work in GCC 3.3.
4229 For more information, see @uref{http://www.interix.com/}.
4234 @heading @anchor{x-x-mingw32}*-*-mingw32
4236 GCC will build with and support only MinGW runtime 3.12 and later.
4237 Earlier versions of headers are incompatible with the new default semantics
4238 of @code{extern inline} in @code{-std=c99} and @code{-std=gnu99} modes.
4243 @heading @anchor{os2}OS/2
4245 GCC does not currently support OS/2. However, Andrew Zabolotny has been
4246 working on a generic OS/2 port with pgcc. The current code can be found
4247 at @uref{http://www.goof.com/pcg/os2/,,http://www.goof.com/pcg/os2/}.
4252 @heading @anchor{older}Older systems
4254 GCC contains support files for many older (1980s and early
4255 1990s) Unix variants. For the most part, support for these systems
4256 has not been deliberately removed, but it has not been maintained for
4257 several years and may suffer from bitrot.
4259 Starting with GCC 3.1, each release has a list of ``obsoleted'' systems.
4260 Support for these systems is still present in that release, but
4261 @command{configure} will fail unless the @option{--enable-obsolete}
4262 option is given. Unless a maintainer steps forward, support for these
4263 systems will be removed from the next release of GCC@.
4265 Support for old systems as hosts for GCC can cause problems if the
4266 workarounds for compiler, library and operating system bugs affect the
4267 cleanliness or maintainability of the rest of GCC@. In some cases, to
4268 bring GCC up on such a system, if still possible with current GCC, may
4269 require first installing an old version of GCC which did work on that
4270 system, and using it to compile a more recent GCC, to avoid bugs in the
4271 vendor compiler. Old releases of GCC 1 and GCC 2 are available in the
4272 @file{old-releases} directory on the @uref{../mirrors.html,,GCC mirror
4273 sites}. Header bugs may generally be avoided using
4274 @command{fixincludes}, but bugs or deficiencies in libraries and the
4275 operating system may still cause problems.
4277 Support for older systems as targets for cross-compilation is less
4278 problematic than support for them as hosts for GCC; if an enthusiast
4279 wishes to make such a target work again (including resurrecting any of
4280 the targets that never worked with GCC 2, starting from the last
4281 version before they were removed), patches
4282 @uref{../contribute.html,,following the usual requirements} would be
4283 likely to be accepted, since they should not affect the support for more
4286 For some systems, old versions of GNU binutils may also be useful,
4287 and are available from @file{pub/binutils/old-releases} on
4288 @uref{http://sourceware.org/mirrors.html,,sourceware.org mirror sites}.
4290 Some of the information on specific systems above relates to
4291 such older systems, but much of the information
4292 about GCC on such systems (which may no longer be applicable to
4293 current GCC) is to be found in the GCC texinfo manual.
4298 @heading @anchor{elf}all ELF targets (SVR4, Solaris 2, etc.)
4300 C++ support is significantly better on ELF targets if you use the
4301 @uref{./configure.html#with-gnu-ld,,GNU linker}; duplicate copies of
4302 inlines, vtables and template instantiations will be discarded
4311 @uref{./index.html,,Return to the GCC Installation page}
4315 @c ***Old documentation******************************************************
4317 @include install-old.texi
4323 @uref{./index.html,,Return to the GCC Installation page}
4327 @c ***GFDL********************************************************************
4335 @uref{./index.html,,Return to the GCC Installation page}
4339 @c ***************************************************************************
4340 @c Part 6 The End of the Document
4342 @comment node-name, next, previous, up
4343 @node Concept Index, , GNU Free Documentation License, Top
4347 @unnumbered Concept Index