1 \input texinfo.tex @c -*-texinfo-*-
4 @setfilename gccinstall.info
5 @settitle Installing GCC
10 @c Specify title for specific html page
12 @settitle Installing GCC
15 @settitle Host/Target specific installation notes for GCC
17 @ifset prerequisiteshtml
18 @settitle Prerequisites for GCC
21 @settitle Downloading GCC
24 @settitle Installing GCC: Configuration
27 @settitle Installing GCC: Building
30 @settitle Installing GCC: Testing
32 @ifset finalinstallhtml
33 @settitle Installing GCC: Final installation
36 @settitle Installing GCC: Binaries
39 @settitle Installing GCC: Old documentation
42 @settitle Installing GCC: GNU Free Documentation License
45 @c Copyright (C) 1988, 1989, 1992, 1993, 1994, 1995, 1996, 1997, 1998,
46 @c 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
47 @c *** Converted to texinfo by Dean Wakerley, dean@wakerley.com
49 @c IMPORTANT: whenever you modify this file, run `install.texi2html' to
50 @c test the generation of HTML documents for the gcc.gnu.org web pages.
52 @c Do not use @footnote{} in this file as it breaks install.texi2html!
54 @c Include everything if we're not making html
58 @set prerequisiteshtml
69 @c Part 2 Summary Description and Copyright
71 Copyright @copyright{} 1988, 1989, 1992, 1993, 1994, 1995, 1996, 1997, 1998,
72 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006 Free Software Foundation, Inc.
74 Permission is granted to copy, distribute and/or modify this document
75 under the terms of the GNU Free Documentation License, Version 1.2 or
76 any later version published by the Free Software Foundation; with no
77 Invariant Sections, the Front-Cover texts being (a) (see below), and
78 with the Back-Cover Texts being (b) (see below). A copy of the
79 license is included in the section entitled ``@uref{./gfdl.html,,GNU
80 Free Documentation License}''.
82 (a) The FSF's Front-Cover Text is:
86 (b) The FSF's Back-Cover Text is:
88 You have freedom to copy and modify this GNU Manual, like GNU
89 software. Copies published by the Free Software Foundation raise
90 funds for GNU development.
95 @dircategory Software development
97 * gccinstall: (gccinstall). Installing the GNU Compiler Collection.
100 @c Part 3 Titlepage and Copyright
103 @comment The title is printed in a large font.
104 @center @titlefont{Installing GCC}
106 @c The following two commands start the copyright page.
108 @vskip 0pt plus 1filll
112 @c Part 4 Top node and Master Menu
115 @comment node-name, next, Previous, up
118 * Installing GCC:: This document describes the generic installation
119 procedure for GCC as well as detailing some target
120 specific installation instructions.
122 * Specific:: Host/target specific installation notes for GCC.
123 * Binaries:: Where to get pre-compiled binaries.
125 * Old:: Old installation documentation.
127 * GNU Free Documentation License:: How you can copy and share this manual.
128 * Concept Index:: This index has two entries.
132 @c Part 5 The Body of the Document
133 @c ***Installing GCC**********************************************************
135 @comment node-name, next, previous, up
136 @node Installing GCC, Binaries, , Top
140 @chapter Installing GCC
143 The latest version of this document is always available at
144 @uref{http://gcc.gnu.org/install/,,http://gcc.gnu.org/install/}.
146 This document describes the generic installation procedure for GCC as well
147 as detailing some target specific installation instructions.
149 GCC includes several components that previously were separate distributions
150 with their own installation instructions. This document supersedes all
151 package specific installation instructions.
153 @emph{Before} starting the build/install procedure please check the
155 @ref{Specific, host/target specific installation notes}.
158 @uref{specific.html,,host/target specific installation notes}.
160 We recommend you browse the entire generic installation instructions before
163 Lists of successful builds for released versions of GCC are
164 available at @uref{http://gcc.gnu.org/buildstat.html}.
165 These lists are updated as new information becomes available.
167 The installation procedure itself is broken into five steps.
172 * Downloading the source::
175 * Testing:: (optional)
182 @uref{prerequisites.html,,Prerequisites}
184 @uref{download.html,,Downloading the source}
186 @uref{configure.html,,Configuration}
188 @uref{build.html,,Building}
190 @uref{test.html,,Testing} (optional)
192 @uref{finalinstall.html,,Final install}
196 Please note that GCC does not support @samp{make uninstall} and probably
197 won't do so in the near future as this would open a can of worms. Instead,
198 we suggest that you install GCC into a directory of its own and simply
199 remove that directory when you do not need that specific version of GCC
200 any longer, and, if shared libraries are installed there as well, no
201 more binaries exist that use them.
204 There are also some @uref{old.html,,old installation instructions},
205 which are mostly obsolete but still contain some information which has
206 not yet been merged into the main part of this manual.
214 @uref{./index.html,,Return to the GCC Installation page}
220 @c ***Prerequisites**************************************************
222 @comment node-name, next, previous, up
223 @node Prerequisites, Downloading the source, , Installing GCC
225 @ifset prerequisiteshtml
227 @chapter Prerequisites
229 @cindex Prerequisites
231 GCC requires that various tools and packages be available for use in the
232 build procedure. Modifying GCC sources requires additional tools
235 @heading Tools/packages necessary for building GCC
237 @item ISO C90 compiler
238 Necessary to bootstrap GCC, although versions of GCC prior
239 to 3.4 also allow bootstrapping with a traditional (K&R) C compiler.
241 To build all languages in a cross-compiler or other configuration where
242 3-stage bootstrap is not performed, you need to start with an existing
243 GCC binary (version 2.95 or later) because source code for language
244 frontends other than C might use GCC extensions.
248 In order to build the Ada compiler (GNAT) you must already have GNAT
249 installed because portions of the Ada frontend are written in Ada (with
250 GNAT extensions.) Refer to the Ada installation instructions for more
251 specific information.
253 @item A ``working'' POSIX compatible shell, or GNU bash
255 Necessary when running @command{configure} because some
256 @command{/bin/sh} shells have bugs and may crash when configuring the
257 target libraries. In other cases, @command{/bin/sh} or @command{ksh}
258 have disastrous corner-case performance problems. This
259 can cause target @command{configure} runs to literally take days to
260 complete in some cases.
262 So on some platforms @command{/bin/ksh} is sufficient, on others it
263 isn't. See the host/target specific instructions for your platform, or
264 use @command{bash} to be sure. Then set @env{CONFIG_SHELL} in your
265 environment to your ``good'' shell prior to running
266 @command{configure}/@command{make}.
268 @command{zsh} is not a fully compliant POSIX shell and will not
269 work when configuring GCC@.
273 Necessary in some circumstances, optional in others. See the
274 host/target specific instructions for your platform for the exact
277 @item gzip version 1.2.4 (or later) or
278 @itemx bzip2 version 1.0.2 (or later)
280 Necessary to uncompress GCC @command{tar} files when source code is
281 obtained via FTP mirror sites.
283 @item GNU make version 3.79.1 (or later)
285 You must have GNU make installed to build GCC@.
287 @item GNU tar version 1.14 (or later)
289 Necessary (only on some platforms) to untar the source code. Many
290 systems' @command{tar} programs will also work, only try GNU
291 @command{tar} if you have problems.
293 @item GNU Multiple Precision Library (GMP) version 4.1 (or later)
295 Necessary to build GCC. If you do not have it installed in your
296 library search path, you will have to configure with the
297 @option{--with-gmp} configure option. See also
298 @option{--with-gmp-lib} and @option{--with-gmp-include}.
300 @item MPFR Library version 2.2.1 (or later)
302 Necessary to build GCC. It can be downloaded from
303 @uref{http://www.mpfr.org/}. The version of MPFR that is bundled with
304 GMP 4.1.x contains numerous bugs. Although GCC may appear to function
305 with the buggy versions of MPFR, there are a few bugs that will not be
306 fixed when using this version. It is strongly recommended to upgrade
307 to the recommended version of MPFR.
309 The @option{--with-mpfr} configure option should be used if your MPFR
310 Library is not installed in your default library search path. See
311 also @option{--with-mpfr-lib} and @option{--with-mpfr-include}.
313 @item @command{jar}, or InfoZIP (@command{zip} and @command{unzip})
315 Necessary to build libgcj, the GCJ runtime.
320 @heading Tools/packages necessary for modifying GCC
322 @item autoconf versions 2.13 and 2.59
323 @itemx GNU m4 version 1.4 (or later)
325 Necessary when modifying @file{configure.ac}, @file{aclocal.m4}, etc.@:
326 to regenerate @file{configure} and @file{config.in} files. Most
327 directories require autoconf 2.59 (exactly), but the toplevel
328 still requires autoconf 2.13 (exactly).
330 @item automake version 1.9.6
332 Necessary when modifying a @file{Makefile.am} file to regenerate its
333 associated @file{Makefile.in}.
335 Much of GCC does not use automake, so directly edit the @file{Makefile.in}
336 file. Specifically this applies to the @file{gcc}, @file{intl},
337 @file{libcpp}, @file{libiberty}, @file{libobjc} directories as well
338 as any of their subdirectories.
340 For directories that use automake, GCC requires the latest release in
341 the 1.9.x series, which is currently 1.9.6. When regenerating a directory
342 to a newer version, please update all the directories using an older 1.9.x
343 to the latest released version.
345 @item gettext version 0.14.5 (or later)
347 Needed to regenerate @file{gcc.pot}.
349 @item gperf version 2.7.2 (or later)
351 Necessary when modifying @command{gperf} input files, e.g.@:
352 @file{gcc/cp/cfns.gperf} to regenerate its associated header file, e.g.@:
353 @file{gcc/cp/cfns.h}.
359 Necessary to run the GCC testsuite; see the section on testing for details.
361 @item autogen version 5.5.4 (or later) and
362 @itemx guile version 1.4.1 (or later)
364 Necessary to regenerate @file{fixinc/fixincl.x} from
365 @file{fixinc/inclhack.def} and @file{fixinc/*.tpl}.
367 Necessary to run @samp{make check} for @file{fixinc}.
369 Necessary to regenerate the top level @file{Makefile.in} file from
370 @file{Makefile.tpl} and @file{Makefile.def}.
372 @item GNU Bison version 1.28 (or later)
373 Berkeley @command{yacc} (@command{byacc}) is also reported to work other
376 Necessary when modifying @file{*.y} files.
378 Necessary to build GCC during development because the generated output
379 files are not included in the SVN repository. They are included in
382 @item Flex version 2.5.4 (or later)
384 Necessary when modifying @file{*.l} files.
386 Necessary to build GCC during development because the generated output
387 files are not included in the SVN repository. They are included in
390 @item Texinfo version 4.4 (or later)
392 Necessary for running @command{makeinfo} when modifying @file{*.texi}
393 files to test your changes.
395 Necessary for running @command{make dvi} or @command{make pdf} to
396 create printable documentation in DVI or PDF format. Texinfo version
397 4.8 or later is required for @command{make pdf}.
399 Necessary to build GCC documentation during development because the
400 generated output files are not included in the SVN repository. They are
401 included in releases.
403 @item @TeX{} (any working version)
405 Necessary for running @command{texi2dvi} and @command{texi2pdf}, which
406 are used when running @command{make dvi} or @command{make pdf} to create
407 DVI or PDF files, respectively.
409 @item SVN (any version)
410 @itemx SSH (any version)
412 Necessary to access the SVN repository. Public releases and weekly
413 snapshots of the development sources are also available via FTP@.
415 @item Perl version 5.6.1 (or later)
417 Necessary when regenerating @file{Makefile} dependencies in libiberty.
418 Necessary when regenerating @file{libiberty/functions.texi}.
419 Necessary when generating manpages from Texinfo manuals.
420 Necessary when targetting Darwin, building libstdc++,
421 and not using @option{--disable-symvers}.
422 Used by various scripts to generate some files included in SVN (mainly
423 Unicode-related and rarely changing) from source tables.
425 @item GNU diffutils version 2.7 (or later)
427 Useful when submitting patches for the GCC source code.
429 @item patch version 2.5.4 (or later)
431 Necessary when applying patches, created with @command{diff}, to one's
437 If you wish to modify @file{.java} files in libjava, you will need to
438 configure with @option{--enable-java-maintainer-mode}, and you will need
439 to have executables named @command{ecj1} and @command{gjavah} in your path.
440 The @command{ecj1} executable should run the Eclipse Java compiler via
441 the GCC-specific entry point. You can download a suitable jar from
442 @uref{ftp://sourceware.org/pub/java/}, or by running the script
443 @command{contrib/download_ecj}.
452 @uref{./index.html,,Return to the GCC Installation page}
456 @c ***Downloading the source**************************************************
458 @comment node-name, next, previous, up
459 @node Downloading the source, Configuration, Prerequisites, Installing GCC
463 @chapter Downloading GCC
465 @cindex Downloading GCC
466 @cindex Downloading the Source
468 GCC is distributed via @uref{http://gcc.gnu.org/svn.html,,SVN} and FTP
469 tarballs compressed with @command{gzip} or
470 @command{bzip2}. It is possible to download a full distribution or specific
473 Please refer to the @uref{http://gcc.gnu.org/releases.html,,releases web page}
474 for information on how to obtain GCC@.
476 The full distribution includes the C, C++, Objective-C, Fortran, Java,
477 and Ada (in the case of GCC 3.1 and later) compilers. The full
478 distribution also includes runtime libraries for C++, Objective-C,
479 Fortran, and Java. In GCC 3.0 and later versions, the GNU compiler
480 testsuites are also included in the full distribution.
482 If you choose to download specific components, you must download the core
483 GCC distribution plus any language specific distributions you wish to
484 use. The core distribution includes the C language front end as well as the
485 shared components. Each language has a tarball which includes the language
486 front end as well as the language runtime (when appropriate).
488 Unpack the core distribution as well as any language specific
489 distributions in the same directory.
491 If you also intend to build binutils (either to upgrade an existing
492 installation or for use in place of the corresponding tools of your
493 OS), unpack the binutils distribution either in the same directory or
494 a separate one. In the latter case, add symbolic links to any
495 components of the binutils you intend to build alongside the compiler
496 (@file{bfd}, @file{binutils}, @file{gas}, @file{gprof}, @file{ld},
497 @file{opcodes}, @dots{}) to the directory containing the GCC sources.
504 @uref{./index.html,,Return to the GCC Installation page}
508 @c ***Configuration***********************************************************
510 @comment node-name, next, previous, up
511 @node Configuration, Building, Downloading the source, Installing GCC
515 @chapter Installing GCC: Configuration
517 @cindex Configuration
518 @cindex Installing GCC: Configuration
520 Like most GNU software, GCC must be configured before it can be built.
521 This document describes the recommended configuration procedure
522 for both native and cross targets.
524 We use @var{srcdir} to refer to the toplevel source directory for
525 GCC; we use @var{objdir} to refer to the toplevel build/object directory.
527 If you obtained the sources via SVN, @var{srcdir} must refer to the top
528 @file{gcc} directory, the one where the @file{MAINTAINERS} can be found,
529 and not its @file{gcc} subdirectory, otherwise the build will fail.
531 If either @var{srcdir} or @var{objdir} is located on an automounted NFS
532 file system, the shell's built-in @command{pwd} command will return
533 temporary pathnames. Using these can lead to various sorts of build
534 problems. To avoid this issue, set the @env{PWDCMD} environment
535 variable to an automounter-aware @command{pwd} command, e.g.,
536 @command{pawd} or @samp{amq -w}, during the configuration and build
539 First, we @strong{highly} recommend that GCC be built into a
540 separate directory than the sources which does @strong{not} reside
541 within the source tree. This is how we generally build GCC; building
542 where @var{srcdir} == @var{objdir} should still work, but doesn't
543 get extensive testing; building where @var{objdir} is a subdirectory
544 of @var{srcdir} is unsupported.
546 If you have previously built GCC in the same directory for a
547 different target machine, do @samp{make distclean} to delete all files
548 that might be invalid. One of the files this deletes is @file{Makefile};
549 if @samp{make distclean} complains that @file{Makefile} does not exist
550 or issues a message like ``don't know how to make distclean'' it probably
551 means that the directory is already suitably clean. However, with the
552 recommended method of building in a separate @var{objdir}, you should
553 simply use a different @var{objdir} for each target.
555 Second, when configuring a native system, either @command{cc} or
556 @command{gcc} must be in your path or you must set @env{CC} in
557 your environment before running configure. Otherwise the configuration
561 Note that the bootstrap compiler and the resulting GCC must be link
562 compatible, else the bootstrap will fail with linker errors about
563 incompatible object file formats. Several multilibed targets are
564 affected by this requirement, see
566 @ref{Specific, host/target specific installation notes}.
569 @uref{specific.html,,host/target specific installation notes}.
578 % @var{srcdir}/configure [@var{options}] [@var{target}]
582 @heading Target specification
585 GCC has code to correctly determine the correct value for @var{target}
586 for nearly all native systems. Therefore, we highly recommend you not
587 provide a configure target when configuring a native compiler.
590 @var{target} must be specified as @option{--target=@var{target}}
591 when configuring a cross compiler; examples of valid targets would be
592 m68k-coff, sh-elf, etc.
595 Specifying just @var{target} instead of @option{--target=@var{target}}
596 implies that the host defaults to @var{target}.
600 @heading Options specification
602 Use @var{options} to override several configure time options for
603 GCC@. A list of supported @var{options} follows; @samp{configure
604 --help} may list other options, but those not listed below may not
605 work and should not normally be used.
607 Note that each @option{--enable} option has a corresponding
608 @option{--disable} option and that each @option{--with} option has a
609 corresponding @option{--without} option.
612 @item --prefix=@var{dirname}
613 Specify the toplevel installation
614 directory. This is the recommended way to install the tools into a directory
615 other than the default. The toplevel installation directory defaults to
618 We @strong{highly} recommend against @var{dirname} being the same or a
619 subdirectory of @var{objdir} or vice versa. If specifying a directory
620 beneath a user's home directory tree, some shells will not expand
621 @var{dirname} correctly if it contains the @samp{~} metacharacter; use
624 The following standard @command{autoconf} options are supported. Normally you
625 should not need to use these options.
627 @item --exec-prefix=@var{dirname}
628 Specify the toplevel installation directory for architecture-dependent
629 files. The default is @file{@var{prefix}}.
631 @item --bindir=@var{dirname}
632 Specify the installation directory for the executables called by users
633 (such as @command{gcc} and @command{g++}). The default is
634 @file{@var{exec-prefix}/bin}.
636 @item --libdir=@var{dirname}
637 Specify the installation directory for object code libraries and
638 internal data files of GCC@. The default is @file{@var{exec-prefix}/lib}.
640 @item --libexecdir=@var{dirname}
641 Specify the installation directory for internal executables of GCC@.
642 The default is @file{@var{exec-prefix}/libexec}.
644 @item --with-slibdir=@var{dirname}
645 Specify the installation directory for the shared libgcc library. The
646 default is @file{@var{libdir}}.
648 @item --infodir=@var{dirname}
649 Specify the installation directory for documentation in info format.
650 The default is @file{@var{prefix}/info}.
652 @item --datadir=@var{dirname}
653 Specify the installation directory for some architecture-independent
654 data files referenced by GCC@. The default is @file{@var{prefix}/share}.
656 @item --mandir=@var{dirname}
657 Specify the installation directory for manual pages. The default is
658 @file{@var{prefix}/man}. (Note that the manual pages are only extracts from
659 the full GCC manuals, which are provided in Texinfo format. The manpages
660 are derived by an automatic conversion process from parts of the full
663 @item --with-gxx-include-dir=@var{dirname}
665 the installation directory for G++ header files. The default is
666 @file{@var{prefix}/include/c++/@var{version}}.
670 @item --program-prefix=@var{prefix}
671 GCC supports some transformations of the names of its programs when
672 installing them. This option prepends @var{prefix} to the names of
673 programs to install in @var{bindir} (see above). For example, specifying
674 @option{--program-prefix=foo-} would result in @samp{gcc}
675 being installed as @file{/usr/local/bin/foo-gcc}.
677 @item --program-suffix=@var{suffix}
678 Appends @var{suffix} to the names of programs to install in @var{bindir}
679 (see above). For example, specifying @option{--program-suffix=-3.1}
680 would result in @samp{gcc} being installed as
681 @file{/usr/local/bin/gcc-3.1}.
683 @item --program-transform-name=@var{pattern}
684 Applies the @samp{sed} script @var{pattern} to be applied to the names
685 of programs to install in @var{bindir} (see above). @var{pattern} has to
686 consist of one or more basic @samp{sed} editing commands, separated by
687 semicolons. For example, if you want the @samp{gcc} program name to be
688 transformed to the installed program @file{/usr/local/bin/myowngcc} and
689 the @samp{g++} program name to be transformed to
690 @file{/usr/local/bin/gspecial++} without changing other program names,
691 you could use the pattern
692 @option{--program-transform-name='s/^gcc$/myowngcc/; s/^g++$/gspecial++/'}
693 to achieve this effect.
695 All three options can be combined and used together, resulting in more
696 complex conversion patterns. As a basic rule, @var{prefix} (and
697 @var{suffix}) are prepended (appended) before further transformations
698 can happen with a special transformation script @var{pattern}.
700 As currently implemented, this option only takes effect for native
701 builds; cross compiler binaries' names are not transformed even when a
702 transformation is explicitly asked for by one of these options.
704 For native builds, some of the installed programs are also installed
705 with the target alias in front of their name, as in
706 @samp{i686-pc-linux-gnu-gcc}. All of the above transformations happen
707 before the target alias is prepended to the name---so, specifying
708 @option{--program-prefix=foo-} and @option{program-suffix=-3.1}, the
709 resulting binary would be installed as
710 @file{/usr/local/bin/i686-pc-linux-gnu-foo-gcc-3.1}.
712 As a last shortcoming, none of the installed Ada programs are
713 transformed yet, which will be fixed in some time.
715 @item --with-local-prefix=@var{dirname}
717 installation directory for local include files. The default is
718 @file{/usr/local}. Specify this option if you want the compiler to
719 search directory @file{@var{dirname}/include} for locally installed
720 header files @emph{instead} of @file{/usr/local/include}.
722 You should specify @option{--with-local-prefix} @strong{only} if your
723 site has a different convention (not @file{/usr/local}) for where to put
726 The default value for @option{--with-local-prefix} is @file{/usr/local}
727 regardless of the value of @option{--prefix}. Specifying
728 @option{--prefix} has no effect on which directory GCC searches for
729 local header files. This may seem counterintuitive, but actually it is
732 The purpose of @option{--prefix} is to specify where to @emph{install
733 GCC}. The local header files in @file{/usr/local/include}---if you put
734 any in that directory---are not part of GCC@. They are part of other
735 programs---perhaps many others. (GCC installs its own header files in
736 another directory which is based on the @option{--prefix} value.)
738 Both the local-prefix include directory and the GCC-prefix include
739 directory are part of GCC's ``system include'' directories. Although these
740 two directories are not fixed, they need to be searched in the proper
741 order for the correct processing of the include_next directive. The
742 local-prefix include directory is searched before the GCC-prefix
743 include directory. Another characteristic of system include directories
744 is that pedantic warnings are turned off for headers in these directories.
746 Some autoconf macros add @option{-I @var{directory}} options to the
747 compiler command line, to ensure that directories containing installed
748 packages' headers are searched. When @var{directory} is one of GCC's
749 system include directories, GCC will ignore the option so that system
750 directories continue to be processed in the correct order. This
751 may result in a search order different from what was specified but the
752 directory will still be searched.
754 GCC automatically searches for ordinary libraries using
755 @env{GCC_EXEC_PREFIX}. Thus, when the same installation prefix is
756 used for both GCC and packages, GCC will automatically search for
757 both headers and libraries. This provides a configuration that is
758 easy to use. GCC behaves in a manner similar to that when it is
759 installed as a system compiler in @file{/usr}.
761 Sites that need to install multiple versions of GCC may not want to
762 use the above simple configuration. It is possible to use the
763 @option{--program-prefix}, @option{--program-suffix} and
764 @option{--program-transform-name} options to install multiple versions
765 into a single directory, but it may be simpler to use different prefixes
766 and the @option{--with-local-prefix} option to specify the location of the
767 site-specific files for each version. It will then be necessary for
768 users to specify explicitly the location of local site libraries
769 (e.g., with @env{LIBRARY_PATH}).
771 The same value can be used for both @option{--with-local-prefix} and
772 @option{--prefix} provided it is not @file{/usr}. This can be used
773 to avoid the default search of @file{/usr/local/include}.
775 @strong{Do not} specify @file{/usr} as the @option{--with-local-prefix}!
776 The directory you use for @option{--with-local-prefix} @strong{must not}
777 contain any of the system's standard header files. If it did contain
778 them, certain programs would be miscompiled (including GNU Emacs, on
779 certain targets), because this would override and nullify the header
780 file corrections made by the @command{fixincludes} script.
782 Indications are that people who use this option use it based on mistaken
783 ideas of what it is for. People use it as if it specified where to
784 install part of GCC@. Perhaps they make this assumption because
785 installing GCC creates the directory.
787 @item --enable-shared[=@var{package}[,@dots{}]]
788 Build shared versions of libraries, if shared libraries are supported on
789 the target platform. Unlike GCC 2.95.x and earlier, shared libraries
790 are enabled by default on all platforms that support shared libraries.
792 If a list of packages is given as an argument, build shared libraries
793 only for the listed packages. For other packages, only static libraries
794 will be built. Package names currently recognized in the GCC tree are
795 @samp{libgcc} (also known as @samp{gcc}), @samp{libstdc++} (not
796 @samp{libstdc++-v3}), @samp{libffi}, @samp{zlib}, @samp{boehm-gc},
797 @samp{ada}, @samp{libada}, @samp{libjava} and @samp{libobjc}.
798 Note @samp{libiberty} does not support shared libraries at all.
800 Use @option{--disable-shared} to build only static libraries. Note that
801 @option{--disable-shared} does not accept a list of package names as
802 argument, only @option{--enable-shared} does.
804 @item @anchor{with-gnu-as}--with-gnu-as
805 Specify that the compiler should assume that the
806 assembler it finds is the GNU assembler. However, this does not modify
807 the rules to find an assembler and will result in confusion if the
808 assembler found is not actually the GNU assembler. (Confusion may also
809 result if the compiler finds the GNU assembler but has not been
810 configured with @option{--with-gnu-as}.) If you have more than one
811 assembler installed on your system, you may want to use this option in
812 connection with @option{--with-as=@var{pathname}} or
813 @option{--with-build-time-tools=@var{pathname}}.
815 The following systems are the only ones where it makes a difference
816 whether you use the GNU assembler. On any other system,
817 @option{--with-gnu-as} has no effect.
820 @item @samp{hppa1.0-@var{any}-@var{any}}
821 @item @samp{hppa1.1-@var{any}-@var{any}}
822 @item @samp{i386-@var{any}-sysv}
823 @item @samp{m68k-bull-sysv}
824 @item @samp{m68k-hp-hpux}
825 @item @samp{m68000-hp-hpux}
826 @item @samp{m68000-att-sysv}
827 @item @samp{sparc-sun-solaris2.@var{any}}
828 @item @samp{sparc64-@var{any}-solaris2.@var{any}}
831 On the systems listed above (except for the HP-PA, the SPARC, for ISC on
832 the 386, if you use the GNU assembler, you should also use the GNU linker
833 (and specify @option{--with-gnu-ld}).
835 @item @anchor{with-as}--with-as=@var{pathname}
836 Specify that the compiler should use the assembler pointed to by
837 @var{pathname}, rather than the one found by the standard rules to find
838 an assembler, which are:
841 Unless GCC is being built with a cross compiler, check the
842 @file{@var{libexec}/gcc/@var{target}/@var{version}} directory.
843 @var{libexec} defaults to @file{@var{exec-prefix}/libexec};
844 @var{exec-prefix} defaults to @var{prefix}, which
845 defaults to @file{/usr/local} unless overridden by the
846 @option{--prefix=@var{pathname}} switch described above. @var{target}
847 is the target system triple, such as @samp{sparc-sun-solaris2.7}, and
848 @var{version} denotes the GCC version, such as 3.0.
851 If the target system is the same that you are building on, check
852 operating system specific directories (e.g.@: @file{/usr/ccs/bin} on
856 Check in the @env{PATH} for a tool whose name is prefixed by the
857 target system triple.
860 Check in the @env{PATH} for a tool whose name is not prefixed by the
861 target system triple, if the host and target system triple are
862 the same (in other words, we use a host tool if it can be used for
866 You may want to use @option{--with-as} if no assembler
867 is installed in the directories listed above, or if you have multiple
868 assemblers installed and want to choose one that is not found by the
871 @item @anchor{with-gnu-ld}--with-gnu-ld
872 Same as @uref{#with-gnu-as,,@option{--with-gnu-as}}
875 @item --with-ld=@var{pathname}
876 Same as @uref{#with-as,,@option{--with-as}}
880 Specify that stabs debugging
881 information should be used instead of whatever format the host normally
882 uses. Normally GCC uses the same debug format as the host system.
884 On MIPS based systems and on Alphas, you must specify whether you want
885 GCC to create the normal ECOFF debugging format, or to use BSD-style
886 stabs passed through the ECOFF symbol table. The normal ECOFF debug
887 format cannot fully handle languages other than C@. BSD stabs format can
888 handle other languages, but it only works with the GNU debugger GDB@.
890 Normally, GCC uses the ECOFF debugging format by default; if you
891 prefer BSD stabs, specify @option{--with-stabs} when you configure GCC@.
893 No matter which default you choose when you configure GCC, the user
894 can use the @option{-gcoff} and @option{-gstabs+} options to specify explicitly
895 the debug format for a particular compilation.
897 @option{--with-stabs} is meaningful on the ISC system on the 386, also, if
898 @option{--with-gas} is used. It selects use of stabs debugging
899 information embedded in COFF output. This kind of debugging information
900 supports C++ well; ordinary COFF debugging information does not.
902 @option{--with-stabs} is also meaningful on 386 systems running SVR4. It
903 selects use of stabs debugging information embedded in ELF output. The
904 C++ compiler currently (2.6.0) does not support the DWARF debugging
905 information normally used on 386 SVR4 platforms; stabs provide a
906 workable alternative. This requires gas and gdb, as the normal SVR4
907 tools can not generate or interpret stabs.
909 @item --disable-multilib
910 Specify that multiple target
911 libraries to support different target variants, calling
912 conventions, etc.@: should not be built. The default is to build a
913 predefined set of them.
915 Some targets provide finer-grained control over which multilibs are built
916 (e.g., @option{--disable-softfloat}):
922 fpu, 26bit, underscore, interwork, biendian, nofmult.
925 softfloat, m68881, m68000, m68020.
928 single-float, biendian, softfloat.
930 @item powerpc*-*-*, rs6000*-*-*
931 aix64, pthread, softfloat, powercpu, powerpccpu, powerpcos, biendian,
936 @item --enable-threads
937 Specify that the target
938 supports threads. This affects the Objective-C compiler and runtime
939 library, and exception handling for other languages like C++ and Java.
940 On some systems, this is the default.
942 In general, the best (and, in many cases, the only known) threading
943 model available will be configured for use. Beware that on some
944 systems, GCC has not been taught what threading models are generally
945 available for the system. In this case, @option{--enable-threads} is an
946 alias for @option{--enable-threads=single}.
948 @item --disable-threads
949 Specify that threading support should be disabled for the system.
950 This is an alias for @option{--enable-threads=single}.
952 @item --enable-threads=@var{lib}
954 @var{lib} is the thread support library. This affects the Objective-C
955 compiler and runtime library, and exception handling for other languages
956 like C++ and Java. The possibilities for @var{lib} are:
964 Ada tasking support. For non-Ada programs, this setting is equivalent
965 to @samp{single}. When used in conjunction with the Ada run time, it
966 causes GCC to use the same thread primitives as Ada uses. This option
967 is necessary when using both Ada and the back end exception handling,
968 which is the default for most Ada targets.
970 Generic MACH thread support, known to work on NeXTSTEP@. (Please note
971 that the file needed to support this configuration, @file{gthr-mach.h}, is
972 missing and thus this setting will cause a known bootstrap failure.)
974 This is an alias for @samp{single}.
976 Generic POSIX/Unix98 thread support.
978 Generic POSIX/Unix95 thread support.
980 RTEMS thread support.
982 Disable thread support, should work for all platforms.
984 Sun Solaris 2 thread support.
986 VxWorks thread support.
988 Microsoft Win32 API thread support.
990 Novell Kernel Services thread support.
994 Specify that the target supports TLS (Thread Local Storage). Usually
995 configure can correctly determine if TLS is supported. In cases where
996 it guesses incorrectly, TLS can be explicitly enabled or disabled with
997 @option{--enable-tls} or @option{--disable-tls}. This can happen if
998 the assembler supports TLS but the C library does not, or if the
999 assumptions made by the configure test are incorrect.
1002 Specify that the target does not support TLS.
1003 This is an alias for @option{--enable-tls=no}.
1005 @item --with-cpu=@var{cpu}
1006 Specify which cpu variant the compiler should generate code for by default.
1007 @var{cpu} will be used as the default value of the @option{-mcpu=} switch.
1008 This option is only supported on some targets, including ARM, i386, M68k,
1009 PowerPC, and SPARC@.
1011 @item --with-schedule=@var{cpu}
1012 @itemx --with-arch=@var{cpu}
1013 @itemx --with-tune=@var{cpu}
1014 @itemx --with-abi=@var{abi}
1015 @itemx --with-fpu=@var{type}
1016 @itemx --with-float=@var{type}
1017 These configure options provide default values for the @option{-mschedule=},
1018 @option{-march=}, @option{-mtune=}, @option{-mabi=}, and @option{-mfpu=}
1019 options and for @option{-mhard-float} or @option{-msoft-float}. As with
1020 @option{--with-cpu}, which switches will be accepted and acceptable values
1021 of the arguments depend on the target.
1023 @item --with-mode=@var{mode}
1024 Specify if the compiler should default to @option{-marm} or @option{-mthumb}.
1025 This option is only supported on ARM targets.
1027 @item --with-divide=@var{type}
1028 Specify how the compiler should generate code for checking for
1029 division by zero. This option is only supported on the MIPS target.
1030 The possibilities for @var{type} are:
1033 Division by zero checks use conditional traps (this is the default on
1034 systems that support conditional traps).
1036 Division by zero checks use the break instruction.
1039 @item --enable-__cxa_atexit
1040 Define if you want to use __cxa_atexit, rather than atexit, to
1041 register C++ destructors for local statics and global objects.
1042 This is essential for fully standards-compliant handling of
1043 destructors, but requires __cxa_atexit in libc. This option is currently
1044 only available on systems with GNU libc. When enabled, this will cause
1045 @option{-fuse-cxa-exit} to be passed by default.
1047 @item --enable-target-optspace
1049 libraries should be optimized for code space instead of code speed.
1050 This is the default for the m32r platform.
1053 Specify that a user visible @command{cpp} program should not be installed.
1055 @item --with-cpp-install-dir=@var{dirname}
1056 Specify that the user visible @command{cpp} program should be installed
1057 in @file{@var{prefix}/@var{dirname}/cpp}, in addition to @var{bindir}.
1059 @item --enable-initfini-array
1060 Force the use of sections @code{.init_array} and @code{.fini_array}
1061 (instead of @code{.init} and @code{.fini}) for constructors and
1062 destructors. Option @option{--disable-initfini-array} has the
1063 opposite effect. If neither option is specified, the configure script
1064 will try to guess whether the @code{.init_array} and
1065 @code{.fini_array} sections are supported and, if they are, use them.
1067 @item --enable-maintainer-mode
1068 The build rules that
1069 regenerate the GCC master message catalog @file{gcc.pot} are normally
1070 disabled. This is because it can only be rebuilt if the complete source
1071 tree is present. If you have changed the sources and want to rebuild the
1072 catalog, configuring with @option{--enable-maintainer-mode} will enable
1073 this. Note that you need a recent version of the @code{gettext} tools
1076 @item --disable-bootstrap
1077 For a native build, the default configuration is to perform
1078 a 3-stage bootstrap of the compiler when @samp{make} is invoked,
1079 testing that GCC can compile itself correctly. If you want to disable
1080 this process, you can configure with @option{--disable-bootstrap}.
1082 @item --enable-bootstrap
1083 In special cases, you may want to perform a 3-stage build
1084 even if the target and host triplets are different.
1085 This could happen when the host can run code compiled for
1086 the target (e.g.@: host is i686-linux, target is i486-linux).
1087 Starting from GCC 4.2, to do this you have to configure explicitly
1088 with @option{--enable-bootstrap}.
1090 @item --enable-generated-files-in-srcdir
1091 Neither the .c and .h files that are generated from Bison and flex nor the
1092 info manuals and man pages that are built from the .texi files are present
1093 in the SVN development tree. When building GCC from that development tree,
1094 or from one of our snapshots, those generated files are placed in your
1095 build directory, which allows for the source to be in a readonly
1098 If you configure with @option{--enable-generated-files-in-srcdir} then those
1099 generated files will go into the source directory. This is mainly intended
1100 for generating release or prerelease tarballs of the GCC sources, since it
1101 is not a requirement that the users of source releases to have flex, Bison,
1104 @item --enable-version-specific-runtime-libs
1106 that runtime libraries should be installed in the compiler specific
1107 subdirectory (@file{@var{libdir}/gcc}) rather than the usual places. In
1108 addition, @samp{libstdc++}'s include files will be installed into
1109 @file{@var{libdir}} unless you overruled it by using
1110 @option{--with-gxx-include-dir=@var{dirname}}. Using this option is
1111 particularly useful if you intend to use several versions of GCC in
1112 parallel. This is currently supported by @samp{libgfortran},
1113 @samp{libjava}, @samp{libmudflap}, @samp{libstdc++}, and @samp{libobjc}.
1115 @item --enable-languages=@var{lang1},@var{lang2},@dots{}
1116 Specify that only a particular subset of compilers and
1117 their runtime libraries should be built. For a list of valid values for
1118 @var{langN} you can issue the following command in the
1119 @file{gcc} directory of your GCC source tree:@*
1121 grep language= */config-lang.in
1123 Currently, you can use any of the following:
1124 @code{all}, @code{ada}, @code{c}, @code{c++}, @code{fortran}, @code{java},
1125 @code{objc}, @code{obj-c++}, @code{treelang}.
1126 Building the Ada compiler has special requirements, see below.
1127 If you do not pass this flag, or specify the option @code{all}, then all
1128 default languages available in the @file{gcc} sub-tree will be configured.
1129 Ada, Objective-C++, and treelang are not default languages; the rest are.
1130 Re-defining @code{LANGUAGES} when calling @samp{make} @strong{does not}
1131 work anymore, as those language sub-directories might not have been
1134 @item --disable-libada
1135 Specify that the run-time libraries and tools used by GNAT should not
1136 be built. This can be useful for debugging, or for compatibility with
1137 previous Ada build procedures, when it was required to explicitly
1138 do a @samp{make -C gcc gnatlib_and_tools}.
1140 @item --disable-libssp
1141 Specify that the run-time libraries for stack smashing protection
1142 should not be built.
1144 @item --disable-libgomp
1145 Specify that the run-time libraries used by GOMP should not be built.
1148 Specify that the compiler should
1149 use DWARF 2 debugging information as the default.
1151 @item --enable-targets=all
1152 @itemx --enable-targets=@var{target_list}
1153 Some GCC targets, e.g.@: powerpc64-linux, build bi-arch compilers.
1154 These are compilers that are able to generate either 64-bit or 32-bit
1155 code. Typically, the corresponding 32-bit target, e.g.@:
1156 powerpc-linux for powerpc64-linux, only generates 32-bit code. This
1157 option enables the 32-bit target to be a bi-arch compiler, which is
1158 useful when you want a bi-arch compiler that defaults to 32-bit, and
1159 you are building a bi-arch or multi-arch binutils in a combined tree.
1160 Currently, this option only affects powerpc-linux and x86-linux.
1162 @item --enable-secureplt
1163 This option enables @option{-msecure-plt} by default for powerpc-linux.
1165 @xref{RS/6000 and PowerPC Options,, RS/6000 and PowerPC Options, gcc,
1166 Using the GNU Compiler Collection (GCC)},
1169 See ``RS/6000 and PowerPC Options'' in the main manual
1172 @item --enable-win32-registry
1173 @itemx --enable-win32-registry=@var{key}
1174 @itemx --disable-win32-registry
1175 The @option{--enable-win32-registry} option enables Microsoft Windows-hosted GCC
1176 to look up installations paths in the registry using the following key:
1179 @code{HKEY_LOCAL_MACHINE\SOFTWARE\Free Software Foundation\@var{key}}
1182 @var{key} defaults to GCC version number, and can be overridden by the
1183 @option{--enable-win32-registry=@var{key}} option. Vendors and distributors
1184 who use custom installers are encouraged to provide a different key,
1185 perhaps one comprised of vendor name and GCC version number, to
1186 avoid conflict with existing installations. This feature is enabled
1187 by default, and can be disabled by @option{--disable-win32-registry}
1188 option. This option has no effect on the other hosts.
1191 Specify that the machine does not have a floating point unit. This
1192 option only applies to @samp{m68k-sun-sunos@var{n}}. On any other
1193 system, @option{--nfp} has no effect.
1195 @item --enable-werror
1196 @itemx --disable-werror
1197 @itemx --enable-werror=yes
1198 @itemx --enable-werror=no
1199 When you specify this option, it controls whether certain files in the
1200 compiler are built with @option{-Werror} in bootstrap stage2 and later.
1201 If you don't specify it, @option{-Werror} is turned on for the main
1202 development trunk. However it defaults to off for release branches and
1203 final releases. The specific files which get @option{-Werror} are
1204 controlled by the Makefiles.
1206 @item --enable-checking
1207 @itemx --enable-checking=@var{list}
1208 When you specify this option, the compiler is built to perform internal
1209 consistency checks of the requested complexity. This does not change the
1210 generated code, but adds error checking within the compiler. This will
1211 slow down the compiler and may only work properly if you are building
1212 the compiler with GCC@. This is @samp{yes} by default when building
1213 from SVN or snapshots, but @samp{release} for releases. More control
1214 over the checks may be had by specifying @var{list}. The categories of
1215 checks available are @samp{yes} (most common checks
1216 @samp{assert,misc,tree,gc,rtlflag,runtime}), @samp{no} (no checks at
1217 all), @samp{all} (all but @samp{valgrind}), @samp{release} (cheapest
1218 checks @samp{assert,runtime}) or @samp{none} (same as @samp{no}).
1219 Individual checks can be enabled with these flags @samp{assert},
1220 @samp{fold}, @samp{gc}, @samp{gcac} @samp{misc}, @samp{rtl},
1221 @samp{rtlflag}, @samp{runtime}, @samp{tree}, and @samp{valgrind}.
1223 The @samp{valgrind} check requires the external @command{valgrind}
1224 simulator, available from @uref{http://valgrind.org/}. The
1225 @samp{rtl}, @samp{gcac} and @samp{valgrind} checks are very expensive.
1226 To disable all checking, @samp{--disable-checking} or
1227 @samp{--enable-checking=none} must be explicitly requested. Disabling
1228 assertions will make the compiler and runtime slightly faster but
1229 increase the risk of undetected internal errors causing wrong code to be
1232 @item --enable-coverage
1233 @itemx --enable-coverage=@var{level}
1234 With this option, the compiler is built to collect self coverage
1235 information, every time it is run. This is for internal development
1236 purposes, and only works when the compiler is being built with gcc. The
1237 @var{level} argument controls whether the compiler is built optimized or
1238 not, values are @samp{opt} and @samp{noopt}. For coverage analysis you
1239 want to disable optimization, for performance analysis you want to
1240 enable optimization. When coverage is enabled, the default level is
1241 without optimization.
1243 @item --enable-gather-detailed-mem-stats
1244 When this option is specified more detailed information on memory
1245 allocation is gathered. This information is printed when using
1246 @option{-fmem-report}.
1249 @itemx --with-gc=@var{choice}
1250 With this option you can specify the garbage collector implementation
1251 used during the compilation process. @var{choice} can be one of
1252 @samp{page} and @samp{zone}, where @samp{page} is the default.
1255 @itemx --disable-nls
1256 The @option{--enable-nls} option enables Native Language Support (NLS),
1257 which lets GCC output diagnostics in languages other than American
1258 English. Native Language Support is enabled by default if not doing a
1259 canadian cross build. The @option{--disable-nls} option disables NLS@.
1261 @item --with-included-gettext
1262 If NLS is enabled, the @option{--with-included-gettext} option causes the build
1263 procedure to prefer its copy of GNU @command{gettext}.
1265 @item --with-catgets
1266 If NLS is enabled, and if the host lacks @code{gettext} but has the
1267 inferior @code{catgets} interface, the GCC build procedure normally
1268 ignores @code{catgets} and instead uses GCC's copy of the GNU
1269 @code{gettext} library. The @option{--with-catgets} option causes the
1270 build procedure to use the host's @code{catgets} in this situation.
1272 @item --with-libiconv-prefix=@var{dir}
1273 Search for libiconv header files in @file{@var{dir}/include} and
1274 libiconv library files in @file{@var{dir}/lib}.
1276 @item --enable-obsolete
1277 Enable configuration for an obsoleted system. If you attempt to
1278 configure GCC for a system (build, host, or target) which has been
1279 obsoleted, and you do not specify this flag, configure will halt with an
1282 All support for systems which have been obsoleted in one release of GCC
1283 is removed entirely in the next major release, unless someone steps
1284 forward to maintain the port.
1286 @item --enable-decimal-float
1287 @itemx --disable-decimal-float
1288 Enable (or disable) support for the C decimal floating point
1289 extension. This is enabled by default only on PowerPC GNU/Linux
1290 systems. Other systems may also support it, but require the user to
1291 specifically enable it.
1293 @item --with-long-double-128
1294 Specify if @code{long double} type should be 128-bit by default on selected
1295 GNU/Linux architectures. If using @code{--without-long-double-128},
1296 @code{long double} will be by default 64-bit, the same as @code{double} type.
1297 When neither of these configure options are used, the default will be
1298 128-bit @code{long double} when built against GNU C Library 2.4 and later,
1299 64-bit @code{long double} otherwise.
1301 @item --with-gmp=@var{pathname}
1302 @itemx --with-gmp-include=@var{pathname}
1303 @itemx --with-gmp-lib=@var{pathname}
1304 @itemx --with-mpfr=@var{pathname}
1305 @itemx --with-mpfr-include=@var{pathname}
1306 @itemx --with-mpfr-lib=@var{pathname}
1307 If you do not have GMP (the GNU Multiple Precision library) and the
1308 MPFR Libraries installed in a standard location and you want to build
1309 GCC, you can explicitly specify the directory where they are installed
1310 (@samp{--with-gmp=@var{gmpinstalldir}},
1311 @samp{--with-mpfr=@var{mpfrinstalldir}}). The
1312 @option{--with-gmp=@var{gmpinstalldir}} option is shorthand for
1313 @option{--with-gmp-lib=@var{gmpinstalldir}/lib} and
1314 @option{--with-gmp-include=@var{gmpinstalldir}/include}. Likewise the
1315 @option{--with-mpfr=@var{mpfrinstalldir}} option is shorthand for
1316 @option{--with-mpfr-lib=@var{mpfrinstalldir}/lib} and
1317 @option{--with-mpfr-include=@var{mpfrinstalldir}/include}. If these
1318 shorthand assumptions are not correct, you can use the explicit
1319 include and lib options directly.
1323 @subheading Cross-Compiler-Specific Options
1324 The following options only apply to building cross compilers.
1326 @item --with-sysroot
1327 @itemx --with-sysroot=@var{dir}
1328 Tells GCC to consider @var{dir} as the root of a tree that contains a
1329 (subset of) the root filesystem of the target operating system.
1330 Target system headers, libraries and run-time object files will be
1331 searched in there. The specified directory is not copied into the
1332 install tree, unlike the options @option{--with-headers} and
1333 @option{--with-libs} that this option obsoletes. The default value,
1334 in case @option{--with-sysroot} is not given an argument, is
1335 @option{$@{gcc_tooldir@}/sys-root}. If the specified directory is a
1336 subdirectory of @option{$@{exec_prefix@}}, then it will be found relative to
1337 the GCC binaries if the installation tree is moved.
1339 @item --with-build-sysroot
1340 @itemx --with-build-sysroot=@var{dir}
1341 Tells GCC to consider @var{dir} as the system root (see
1342 @option{--with-sysroot}) while building target libraries, instead of
1343 the directory specified with @option{--with-sysroot}. This option is
1344 only useful when you are already using @option{--with-sysroot}. You
1345 can use @option{--with-build-sysroot} when you are configuring with
1346 @option{--prefix} set to a directory that is different from the one in
1347 which you are installing GCC and your target libraries.
1349 This option affects the system root for the compiler used to build
1350 target libraries (which runs on the build system); it does not affect
1351 the compiler which is used to build GCC itself.
1353 @item --with-headers
1354 @itemx --with-headers=@var{dir}
1355 Deprecated in favor of @option{--with-sysroot}.
1356 Specifies that target headers are available when building a cross compiler.
1357 The @var{dir} argument specifies a directory which has the target include
1358 files. These include files will be copied into the @file{gcc} install
1359 directory. @emph{This option with the @var{dir} argument is required} when
1360 building a cross compiler, if @file{@var{prefix}/@var{target}/sys-include}
1361 doesn't pre-exist. If @file{@var{prefix}/@var{target}/sys-include} does
1362 pre-exist, the @var{dir} argument may be omitted. @command{fixincludes}
1363 will be run on these files to make them compatible with GCC@.
1365 @item --without-headers
1366 Tells GCC not use any target headers from a libc when building a cross
1367 compiler. When crossing to GNU/Linux, you need the headers so GCC
1368 can build the exception handling for libgcc.
1371 @itemx --with-libs=``@var{dir1} @var{dir2} @dots{} @var{dirN}''
1372 Deprecated in favor of @option{--with-sysroot}.
1373 Specifies a list of directories which contain the target runtime
1374 libraries. These libraries will be copied into the @file{gcc} install
1375 directory. If the directory list is omitted, this option has no
1379 Specifies that @samp{newlib} is
1380 being used as the target C library. This causes @code{__eprintf} to be
1381 omitted from @file{libgcc.a} on the assumption that it will be provided by
1384 @item --with-build-time-tools=@var{dir}
1385 Specifies where to find the set of target tools (assembler, linker, etc.)
1386 that will be used while building GCC itself. This option can be useful
1387 if the directory layouts are different between the system you are building
1388 GCC on, and the system where you will deploy it.
1390 For example, on a @option{ia64-hp-hpux} system, you may have the GNU
1391 assembler and linker in @file{/usr/bin}, and the native tools in a
1392 different path, and build a toolchain that expects to find the
1393 native tools in @file{/usr/bin}.
1395 When you use this option, you should ensure that @var{dir} includes
1396 @command{ar}, @command{as}, @command{ld}, @command{nm},
1397 @command{ranlib} and @command{strip} if necessary, and possibly
1398 @command{objdump}. Otherwise, GCC may use an inconsistent set of
1402 @subheading Java-Specific Options
1404 The following option applies to the build of the Java front end.
1407 @item --disable-libgcj
1408 Specify that the run-time libraries
1409 used by GCJ should not be built. This is useful in case you intend
1410 to use GCJ with some other run-time, or you're going to install it
1411 separately, or it just happens not to build on your particular
1412 machine. In general, if the Java front end is enabled, the GCJ
1413 libraries will be enabled too, unless they're known to not work on
1414 the target platform. If GCJ is enabled but @samp{libgcj} isn't built, you
1415 may need to port it; in this case, before modifying the top-level
1416 @file{configure.in} so that @samp{libgcj} is enabled by default on this platform,
1417 you may use @option{--enable-libgcj} to override the default.
1421 The following options apply to building @samp{libgcj}.
1423 @subsubheading General Options
1426 @item --enable-java-maintainer-mode
1427 By default the @samp{libjava} build will not attempt to compile the
1428 @file{.java} source files to @file{.class}. Instead, it will use the
1429 @file{.class} files from the source tree. If you use this option you
1430 must have executables named @command{ecj1} and @command{gjavah} in your path
1431 for use by the build. You must use this option if you intend to
1432 modify any @file{.java} files in @file{libjava}.
1434 @item --with-java-home=@var{dirname}
1435 This @samp{libjava} option overrides the default value of the
1436 @samp{java.home} system property. It is also used to set
1437 @samp{sun.boot.class.path} to @file{@var{dirname}/lib/rt.jar}. By
1438 default @samp{java.home} is set to @file{@var{prefix}} and
1439 @samp{sun.boot.class.path} to
1440 @file{@var{datadir}/java/libgcj-@var{version}.jar}.
1442 @item --with-ecj-jar=@var{filename}
1443 This option can be used to specify the location of an external jar
1444 file containing the Eclipse Java compiler. A specially modified
1445 version of this compiler is used by @command{gcj} to parse
1446 @file{.java} source files. If this option is given, the
1447 @samp{libjava} build will create and install an @file{ecj1} executable
1448 which uses this jar file at runtime.
1450 If this option is not given, but an @file{ecj.jar} file is found in
1451 the topmost source tree at configure time, then the @samp{libgcj}
1452 build will create and install @file{ecj1}, and will also install the
1453 discovered @file{ecj.jar} into a suitable place in the install tree.
1455 If @file{ecj1} is not installed, then the user will have to supply one
1456 on his path in order for @command{gcj} to properly parse @file{.java}
1457 source files. A suitable jar is available from
1458 @uref{ftp://sourceware.org/pub/java/}.
1460 @item --disable-getenv-properties
1461 Don't set system properties from @env{GCJ_PROPERTIES}.
1463 @item --enable-hash-synchronization
1464 Use a global hash table for monitor locks. Ordinarily,
1465 @samp{libgcj}'s @samp{configure} script automatically makes
1466 the correct choice for this option for your platform. Only use
1467 this if you know you need the library to be configured differently.
1469 @item --enable-interpreter
1470 Enable the Java interpreter. The interpreter is automatically
1471 enabled by default on all platforms that support it. This option
1472 is really only useful if you want to disable the interpreter
1473 (using @option{--disable-interpreter}).
1475 @item --disable-java-net
1476 Disable java.net. This disables the native part of java.net only,
1477 using non-functional stubs for native method implementations.
1479 @item --disable-jvmpi
1480 Disable JVMPI support.
1483 Enable runtime eCos target support.
1485 @item --without-libffi
1486 Don't use @samp{libffi}. This will disable the interpreter and JNI
1487 support as well, as these require @samp{libffi} to work.
1489 @item --enable-libgcj-debug
1490 Enable runtime debugging code.
1492 @item --enable-libgcj-multifile
1493 If specified, causes all @file{.java} source files to be
1494 compiled into @file{.class} files in one invocation of
1495 @samp{gcj}. This can speed up build time, but is more
1496 resource-intensive. If this option is unspecified or
1497 disabled, @samp{gcj} is invoked once for each @file{.java}
1498 file to compile into a @file{.class} file.
1500 @item --with-libiconv-prefix=DIR
1501 Search for libiconv in @file{DIR/include} and @file{DIR/lib}.
1503 @item --enable-sjlj-exceptions
1504 Force use of the @code{setjmp}/@code{longjmp}-based scheme for exceptions.
1505 @samp{configure} ordinarily picks the correct value based on the platform.
1506 Only use this option if you are sure you need a different setting.
1508 @item --with-system-zlib
1509 Use installed @samp{zlib} rather than that included with GCC@.
1511 @item --with-win32-nlsapi=ansi, unicows or unicode
1512 Indicates how MinGW @samp{libgcj} translates between UNICODE
1513 characters and the Win32 API@.
1516 Use the single-byte @code{char} and the Win32 A functions natively,
1517 translating to and from UNICODE when using these functions. If
1518 unspecified, this is the default.
1521 Use the @code{WCHAR} and Win32 W functions natively. Adds
1522 @code{-lunicows} to @file{libgcj.spec} to link with @samp{libunicows}.
1523 @file{unicows.dll} needs to be deployed on Microsoft Windows 9X machines
1524 running built executables. @file{libunicows.a}, an open-source
1525 import library around Microsoft's @code{unicows.dll}, is obtained from
1526 @uref{http://libunicows.sourceforge.net/}, which also gives details
1527 on getting @file{unicows.dll} from Microsoft.
1530 Use the @code{WCHAR} and Win32 W functions natively. Does @emph{not}
1531 add @code{-lunicows} to @file{libgcj.spec}. The built executables will
1532 only run on Microsoft Windows NT and above.
1536 @subsubheading AWT-Specific Options
1540 Use the X Window System.
1542 @item --enable-java-awt=PEER(S)
1543 Specifies the AWT peer library or libraries to build alongside
1544 @samp{libgcj}. If this option is unspecified or disabled, AWT
1545 will be non-functional. Current valid values are @option{gtk} and
1546 @option{xlib}. Multiple libraries should be separated by a
1547 comma (i.e.@: @option{--enable-java-awt=gtk,xlib}).
1549 @item --enable-gtk-cairo
1550 Build the cairo Graphics2D implementation on GTK@.
1552 @item --enable-java-gc=TYPE
1553 Choose garbage collector. Defaults to @option{boehm} if unspecified.
1555 @item --disable-gtktest
1556 Do not try to compile and run a test GTK+ program.
1558 @item --disable-glibtest
1559 Do not try to compile and run a test GLIB program.
1561 @item --with-libart-prefix=PFX
1562 Prefix where libart is installed (optional).
1564 @item --with-libart-exec-prefix=PFX
1565 Exec prefix where libart is installed (optional).
1567 @item --disable-libarttest
1568 Do not try to compile and run a test libart program.
1577 @uref{./index.html,,Return to the GCC Installation page}
1581 @c ***Building****************************************************************
1583 @comment node-name, next, previous, up
1584 @node Building, Testing, Configuration, Installing GCC
1590 @cindex Installing GCC: Building
1592 Now that GCC is configured, you are ready to build the compiler and
1595 Some commands executed when making the compiler may fail (return a
1596 nonzero status) and be ignored by @command{make}. These failures, which
1597 are often due to files that were not found, are expected, and can safely
1600 It is normal to have compiler warnings when compiling certain files.
1601 Unless you are a GCC developer, you can generally ignore these warnings
1602 unless they cause compilation to fail. Developers should attempt to fix
1603 any warnings encountered, however they can temporarily continue past
1604 warnings-as-errors by specifying the configure flag
1605 @option{--disable-werror}.
1607 On certain old systems, defining certain environment variables such as
1608 @env{CC} can interfere with the functioning of @command{make}.
1610 If you encounter seemingly strange errors when trying to build the
1611 compiler in a directory other than the source directory, it could be
1612 because you have previously configured the compiler in the source
1613 directory. Make sure you have done all the necessary preparations.
1615 If you build GCC on a BSD system using a directory stored in an old System
1616 V file system, problems may occur in running @command{fixincludes} if the
1617 System V file system doesn't support symbolic links. These problems
1618 result in a failure to fix the declaration of @code{size_t} in
1619 @file{sys/types.h}. If you find that @code{size_t} is a signed type and
1620 that type mismatches occur, this could be the cause.
1622 The solution is not to use such a directory for building GCC@.
1624 When building from SVN or snapshots, or if you modify parser sources,
1625 you need the Bison parser generator installed. If you do not modify
1626 parser sources, releases contain the Bison-generated files and you do
1627 not need Bison installed to build them.
1629 When building from SVN or snapshots, or if you modify Texinfo
1630 documentation, you need version 4.4 or later of Texinfo installed if you
1631 want Info documentation to be regenerated. Releases contain Info
1632 documentation pre-built for the unmodified documentation in the release.
1634 @section Building a native compiler
1636 For a native build, the default configuration is to perform
1637 a 3-stage bootstrap of the compiler when @samp{make} is invoked.
1638 This will build the entire GCC system and ensure that it compiles
1639 itself correctly. It can be disabled with the @option{--disable-bootstrap}
1640 parameter to @samp{configure}, but bootstrapping is suggested because
1641 the compiler will be tested more completely and could also have
1644 The bootstrapping process will complete the following steps:
1648 Build tools necessary to build the compiler.
1651 Perform a 3-stage bootstrap of the compiler. This includes building
1652 three times the target tools for use by the compiler such as binutils
1653 (bfd, binutils, gas, gprof, ld, and opcodes) if they have been
1654 individually linked or moved into the top level GCC source tree before
1658 Perform a comparison test of the stage2 and stage3 compilers.
1661 Build runtime libraries using the stage3 compiler from the previous step.
1665 If you are short on disk space you might consider @samp{make
1666 bootstrap-lean} instead. The sequence of compilation is the
1667 same described above, but object files from the stage1 and
1668 stage2 of the 3-stage bootstrap of the compiler are deleted as
1669 soon as they are no longer needed.
1671 If you want to save additional space during the bootstrap and in
1672 the final installation as well, you can build the compiler binaries
1673 without debugging information as in the following example. This will save
1674 roughly 40% of disk space both for the bootstrap and the final installation.
1675 (Libraries will still contain debugging information.)
1678 make CFLAGS='-O' LIBCFLAGS='-g -O2' \
1679 LIBCXXFLAGS='-g -O2 -fno-implicit-templates' bootstrap
1682 If you wish to use non-default GCC flags when compiling the stage2 and
1683 stage3 compilers, set @code{BOOT_CFLAGS} on the command line when doing
1684 @samp{make}. Non-default optimization flags are less well
1685 tested here than the default of @samp{-g -O2}, but should still work.
1686 In a few cases, you may find that you need to specify special flags such
1687 as @option{-msoft-float} here to complete the bootstrap; or, if the
1688 native compiler miscompiles the stage1 compiler, you may need to work
1689 around this, by choosing @code{BOOT_CFLAGS} to avoid the parts of the
1690 stage1 compiler that were miscompiled, or by using @samp{make
1691 bootstrap4} to increase the number of stages of bootstrap.
1693 Note that using non-standard @code{CFLAGS} can cause bootstrap to fail
1694 if these trigger a warning with the new compiler. For example using
1695 @samp{-O2 -g -mcpu=i686} on @code{i686-pc-linux-gnu} will cause bootstrap
1696 failure as @option{-mcpu=} is deprecated in 3.4.0 and above.
1699 If you used the flag @option{--enable-languages=@dots{}} to restrict
1700 the compilers to be built, only those you've actually enabled will be
1701 built. This will of course only build those runtime libraries, for
1702 which the particular compiler has been built. Please note,
1703 that re-defining @env{LANGUAGES} when calling @samp{make}
1704 @strong{does not} work anymore!
1706 If the comparison of stage2 and stage3 fails, this normally indicates
1707 that the stage2 compiler has compiled GCC incorrectly, and is therefore
1708 a potentially serious bug which you should investigate and report. (On
1709 a few systems, meaningful comparison of object files is impossible; they
1710 always appear ``different''. If you encounter this problem, you will
1711 need to disable comparison in the @file{Makefile}.)
1713 If you do not want to bootstrap your compiler, you can configure with
1714 @option{--disable-bootstrap}. In particular cases, you may want to
1715 bootstrap your compiler even if the target system is not the same as
1716 the one you are building on: for example, you could build a
1717 @code{powerpc-unknown-linux-gnu} toolchain on a
1718 @code{powerpc64-unknown-linux-gnu} host. In this case, pass
1719 @option{--enable-bootstrap} to the configure script.
1722 @section Building a cross compiler
1724 When building a cross compiler, it is not generally possible to do a
1725 3-stage bootstrap of the compiler. This makes for an interesting problem
1726 as parts of GCC can only be built with GCC@.
1728 To build a cross compiler, we first recommend building and installing a
1729 native compiler. You can then use the native GCC compiler to build the
1730 cross compiler. The installed native compiler needs to be GCC version
1733 Assuming you have already installed a native copy of GCC and configured
1734 your cross compiler, issue the command @command{make}, which performs the
1739 Build host tools necessary to build the compiler.
1742 Build target tools for use by the compiler such as binutils (bfd,
1743 binutils, gas, gprof, ld, and opcodes)
1744 if they have been individually linked or moved into the top level GCC source
1745 tree before configuring.
1748 Build the compiler (single stage only).
1751 Build runtime libraries using the compiler from the previous step.
1754 Note that if an error occurs in any step the make process will exit.
1756 If you are not building GNU binutils in the same source tree as GCC,
1757 you will need a cross-assembler and cross-linker installed before
1758 configuring GCC@. Put them in the directory
1759 @file{@var{prefix}/@var{target}/bin}. Here is a table of the tools
1760 you should put in this directory:
1764 This should be the cross-assembler.
1767 This should be the cross-linker.
1770 This should be the cross-archiver: a program which can manipulate
1771 archive files (linker libraries) in the target machine's format.
1774 This should be a program to construct a symbol table in an archive file.
1777 The installation of GCC will find these programs in that directory,
1778 and copy or link them to the proper place to for the cross-compiler to
1779 find them when run later.
1781 The easiest way to provide these files is to build the Binutils package.
1782 Configure it with the same @option{--host} and @option{--target}
1783 options that you use for configuring GCC, then build and install
1784 them. They install their executables automatically into the proper
1785 directory. Alas, they do not support all the targets that GCC
1788 If you are not building a C library in the same source tree as GCC,
1789 you should also provide the target libraries and headers before
1790 configuring GCC, specifying the directories with
1791 @option{--with-sysroot} or @option{--with-headers} and
1792 @option{--with-libs}. Many targets also require ``start files'' such
1793 as @file{crt0.o} and
1794 @file{crtn.o} which are linked into each executable. There may be several
1795 alternatives for @file{crt0.o}, for use with profiling or other
1796 compilation options. Check your target's definition of
1797 @code{STARTFILE_SPEC} to find out what start files it uses.
1799 @section Building in parallel
1801 GNU Make 3.79 and above, which is necessary to build GCC, support
1802 building in parallel. To activate this, you can use @samp{make -j 2}
1803 instead of @samp{make}. You can also specify a bigger number, and
1804 in most cases using a value greater than the number of processors in
1805 your machine will result in fewer and shorter I/O latency hits, thus
1806 improving overall throughput; this is especially true for slow drives
1807 and network filesystems.
1809 @section Building the Ada compiler
1811 In order to build GNAT, the Ada compiler, you need a working GNAT
1812 compiler (GNAT version 3.14 or later, or GCC version 3.1 or later).
1813 This includes GNAT tools such as @command{gnatmake} and
1814 @command{gnatlink}, since the Ada front end is written in Ada and
1815 uses some GNAT-specific extensions.
1817 In order to build a cross compiler, it is suggested to install
1818 the new compiler as native first, and then use it to build the cross
1821 @command{configure} does not test whether the GNAT installation works
1822 and has a sufficiently recent version; if too old a GNAT version is
1823 installed, the build will fail unless @option{--enable-languages} is
1824 used to disable building the Ada front end.
1826 @section Building with profile feedback
1828 It is possible to use profile feedback to optimize the compiler itself. This
1829 should result in a faster compiler binary. Experiments done on x86 using gcc
1830 3.3 showed approximately 7 percent speedup on compiling C programs. To
1831 bootstrap the compiler with profile feedback, use @code{make profiledbootstrap}.
1833 When @samp{make profiledbootstrap} is run, it will first build a @code{stage1}
1834 compiler. This compiler is used to build a @code{stageprofile} compiler
1835 instrumented to collect execution counts of instruction and branch
1836 probabilities. Then runtime libraries are compiled with profile collected.
1837 Finally a @code{stagefeedback} compiler is built using the information collected.
1839 Unlike standard bootstrap, several additional restrictions apply. The
1840 compiler used to build @code{stage1} needs to support a 64-bit integral type.
1841 It is recommended to only use GCC for this. Also parallel make is currently
1842 not supported since collisions in profile collecting may occur.
1849 @uref{./index.html,,Return to the GCC Installation page}
1853 @c ***Testing*****************************************************************
1855 @comment node-name, next, previous, up
1856 @node Testing, Final install, Building, Installing GCC
1860 @chapter Installing GCC: Testing
1863 @cindex Installing GCC: Testing
1866 Before you install GCC, we encourage you to run the testsuites and to
1867 compare your results with results from a similar configuration that have
1868 been submitted to the
1869 @uref{http://gcc.gnu.org/ml/gcc-testresults/,,gcc-testresults mailing list}.
1870 Some of these archived results are linked from the build status lists
1871 at @uref{http://gcc.gnu.org/buildstat.html}, although not everyone who
1872 reports a successful build runs the testsuites and submits the results.
1873 This step is optional and may require you to download additional software,
1874 but it can give you confidence in your new GCC installation or point out
1875 problems before you install and start using your new GCC@.
1877 First, you must have @uref{download.html,,downloaded the testsuites}.
1878 These are part of the full distribution, but if you downloaded the
1879 ``core'' compiler plus any front ends, you must download the testsuites
1882 Second, you must have the testing tools installed. This includes
1883 @uref{http://www.gnu.org/software/dejagnu/,,DejaGnu}, Tcl, and Expect;
1884 the DejaGnu site has links to these.
1886 If the directories where @command{runtest} and @command{expect} were
1887 installed are not in the @env{PATH}, you may need to set the following
1888 environment variables appropriately, as in the following example (which
1889 assumes that DejaGnu has been installed under @file{/usr/local}):
1892 TCL_LIBRARY = /usr/local/share/tcl8.0
1893 DEJAGNULIBS = /usr/local/share/dejagnu
1896 (On systems such as Cygwin, these paths are required to be actual
1897 paths, not mounts or links; presumably this is due to some lack of
1898 portability in the DejaGnu code.)
1901 Finally, you can run the testsuite (which may take a long time):
1903 cd @var{objdir}; make -k check
1906 This will test various components of GCC, such as compiler
1907 front ends and runtime libraries. While running the testsuite, DejaGnu
1908 might emit some harmless messages resembling
1909 @samp{WARNING: Couldn't find the global config file.} or
1910 @samp{WARNING: Couldn't find tool init file} that can be ignored.
1912 @section How can you run the testsuite on selected tests?
1914 In order to run sets of tests selectively, there are targets
1915 @samp{make check-gcc} and @samp{make check-g++}
1916 in the @file{gcc} subdirectory of the object directory. You can also
1917 just run @samp{make check} in a subdirectory of the object directory.
1920 A more selective way to just run all @command{gcc} execute tests in the
1924 make check-gcc RUNTESTFLAGS="execute.exp @var{other-options}"
1927 Likewise, in order to run only the @command{g++} ``old-deja'' tests in
1928 the testsuite with filenames matching @samp{9805*}, you would use
1931 make check-g++ RUNTESTFLAGS="old-deja.exp=9805* @var{other-options}"
1934 The @file{*.exp} files are located in the testsuite directories of the GCC
1935 source, the most important ones being @file{compile.exp},
1936 @file{execute.exp}, @file{dg.exp} and @file{old-deja.exp}.
1937 To get a list of the possible @file{*.exp} files, pipe the
1938 output of @samp{make check} into a file and look at the
1939 @samp{Running @dots{} .exp} lines.
1941 @section Passing options and running multiple testsuites
1943 You can pass multiple options to the testsuite using the
1944 @samp{--target_board} option of DejaGNU, either passed as part of
1945 @samp{RUNTESTFLAGS}, or directly to @command{runtest} if you prefer to
1946 work outside the makefiles. For example,
1949 make check-g++ RUNTESTFLAGS="--target_board=unix/-O3/-fmerge-constants"
1952 will run the standard @command{g++} testsuites (``unix'' is the target name
1953 for a standard native testsuite situation), passing
1954 @samp{-O3 -fmerge-constants} to the compiler on every test, i.e.,
1955 slashes separate options.
1957 You can run the testsuites multiple times using combinations of options
1958 with a syntax similar to the brace expansion of popular shells:
1961 @dots{}"--target_board=arm-sim/@{-mhard-float,-msoft-float@}@{-O1,-O2,-O3,@}"
1964 (Note the empty option caused by the trailing comma in the final group.)
1965 The following will run each testsuite eight times using the @samp{arm-sim}
1966 target, as if you had specified all possible combinations yourself:
1969 --target_board=arm-sim/-mhard-float/-O1
1970 --target_board=arm-sim/-mhard-float/-O2
1971 --target_board=arm-sim/-mhard-float/-O3
1972 --target_board=arm-sim/-mhard-float
1973 --target_board=arm-sim/-msoft-float/-O1
1974 --target_board=arm-sim/-msoft-float/-O2
1975 --target_board=arm-sim/-msoft-float/-O3
1976 --target_board=arm-sim/-msoft-float
1979 They can be combined as many times as you wish, in arbitrary ways. This
1983 @dots{}"--target_board=unix/-Wextra@{-O3,-fno-strength-reduce@}@{-fomit-frame-pointer,@}"
1986 will generate four combinations, all involving @samp{-Wextra}.
1988 The disadvantage to this method is that the testsuites are run in serial,
1989 which is a waste on multiprocessor systems. For users with GNU Make and
1990 a shell which performs brace expansion, you can run the testsuites in
1991 parallel by having the shell perform the combinations and @command{make}
1992 do the parallel runs. Instead of using @samp{--target_board}, use a
1993 special makefile target:
1996 make -j@var{N} check-@var{testsuite}//@var{test-target}/@var{option1}/@var{option2}/@dots{}
2002 make -j3 check-gcc//sh-hms-sim/@{-m1,-m2,-m3,-m3e,-m4@}/@{,-nofpu@}
2005 will run three concurrent ``make-gcc'' testsuites, eventually testing all
2006 ten combinations as described above. Note that this is currently only
2007 supported in the @file{gcc} subdirectory. (To see how this works, try
2008 typing @command{echo} before the example given here.)
2011 @section Additional testing for Java Class Libraries
2013 The Java runtime tests can be executed via @samp{make check}
2014 in the @file{@var{target}/libjava/testsuite} directory in
2017 The @uref{http://sourceware.org/mauve/,,Mauve Project} provides
2018 a suite of tests for the Java Class Libraries. This suite can be run
2019 as part of libgcj testing by placing the Mauve tree within the libjava
2020 testsuite at @file{libjava/testsuite/libjava.mauve/mauve}, or by
2021 specifying the location of that tree when invoking @samp{make}, as in
2022 @samp{make MAUVEDIR=~/mauve check}.
2024 @section How to interpret test results
2026 The result of running the testsuite are various @file{*.sum} and @file{*.log}
2027 files in the testsuite subdirectories. The @file{*.log} files contain a
2028 detailed log of the compiler invocations and the corresponding
2029 results, the @file{*.sum} files summarize the results. These summaries
2030 contain status codes for all tests:
2034 PASS: the test passed as expected
2036 XPASS: the test unexpectedly passed
2038 FAIL: the test unexpectedly failed
2040 XFAIL: the test failed as expected
2042 UNSUPPORTED: the test is not supported on this platform
2044 ERROR: the testsuite detected an error
2046 WARNING: the testsuite detected a possible problem
2049 It is normal for some tests to report unexpected failures. At the
2050 current time the testing harness does not allow fine grained control
2051 over whether or not a test is expected to fail. This problem should
2052 be fixed in future releases.
2055 @section Submitting test results
2057 If you want to report the results to the GCC project, use the
2058 @file{contrib/test_summary} shell script. Start it in the @var{objdir} with
2061 @var{srcdir}/contrib/test_summary -p your_commentary.txt \
2062 -m gcc-testresults@@gcc.gnu.org |sh
2065 This script uses the @command{Mail} program to send the results, so
2066 make sure it is in your @env{PATH}. The file @file{your_commentary.txt} is
2067 prepended to the testsuite summary and should contain any special
2068 remarks you have on your results or your build environment. Please
2069 do not edit the testsuite result block or the subject line, as these
2070 messages may be automatically processed.
2077 @uref{./index.html,,Return to the GCC Installation page}
2081 @c ***Final install***********************************************************
2083 @comment node-name, next, previous, up
2084 @node Final install, , Testing, Installing GCC
2086 @ifset finalinstallhtml
2088 @chapter Installing GCC: Final installation
2091 Now that GCC has been built (and optionally tested), you can install it with
2093 cd @var{objdir}; make install
2096 We strongly recommend to install into a target directory where there is
2097 no previous version of GCC present. Also, the GNAT runtime should not
2098 be stripped, as this would break certain features of the debugger that
2099 depend on this debugging information (catching Ada exceptions for
2102 That step completes the installation of GCC; user level binaries can
2103 be found in @file{@var{prefix}/bin} where @var{prefix} is the value
2104 you specified with the @option{--prefix} to configure (or
2105 @file{/usr/local} by default). (If you specified @option{--bindir},
2106 that directory will be used instead; otherwise, if you specified
2107 @option{--exec-prefix}, @file{@var{exec-prefix}/bin} will be used.)
2108 Headers for the C++ and Java libraries are installed in
2109 @file{@var{prefix}/include}; libraries in @file{@var{libdir}}
2110 (normally @file{@var{prefix}/lib}); internal parts of the compiler in
2111 @file{@var{libdir}/gcc} and @file{@var{libexecdir}/gcc}; documentation
2112 in info format in @file{@var{infodir}} (normally
2113 @file{@var{prefix}/info}).
2115 When installing cross-compilers, GCC's executables
2116 are not only installed into @file{@var{bindir}}, that
2117 is, @file{@var{exec-prefix}/bin}, but additionally into
2118 @file{@var{exec-prefix}/@var{target-alias}/bin}, if that directory
2119 exists. Typically, such @dfn{tooldirs} hold target-specific
2120 binutils, including assembler and linker.
2122 Installation into a temporary staging area or into a @command{chroot}
2123 jail can be achieved with the command
2126 make DESTDIR=@var{path-to-rootdir} install
2129 @noindent where @var{path-to-rootdir} is the absolute path of
2130 a directory relative to which all installation paths will be
2131 interpreted. Note that the directory specified by @code{DESTDIR}
2132 need not exist yet; it will be created if necessary.
2134 There is a subtle point with tooldirs and @code{DESTDIR}:
2135 If you relocate a cross-compiler installation with
2136 e.g.@: @samp{DESTDIR=@var{rootdir}}, then the directory
2137 @file{@var{rootdir}/@var{exec-prefix}/@var{target-alias}/bin} will
2138 be filled with duplicated GCC executables only if it already exists,
2139 it will not be created otherwise. This is regarded as a feature,
2140 not as a bug, because it gives slightly more control to the packagers
2141 using the @code{DESTDIR} feature.
2143 If you are bootstrapping a released version of GCC then please
2144 quickly review the build status page for your release, available from
2145 @uref{http://gcc.gnu.org/buildstat.html}.
2146 If your system is not listed for the version of GCC that you built,
2148 @email{gcc@@gcc.gnu.org} indicating
2149 that you successfully built and installed GCC@.
2150 Include the following information:
2154 Output from running @file{@var{srcdir}/config.guess}. Do not send
2155 that file itself, just the one-line output from running it.
2158 The output of @samp{gcc -v} for your newly installed @command{gcc}.
2159 This tells us which version of GCC you built and the options you passed to
2163 Whether you enabled all languages or a subset of them. If you used a
2164 full distribution then this information is part of the configure
2165 options in the output of @samp{gcc -v}, but if you downloaded the
2166 ``core'' compiler plus additional front ends then it isn't apparent
2167 which ones you built unless you tell us about it.
2170 If the build was for GNU/Linux, also include:
2173 The distribution name and version (e.g., Red Hat 7.1 or Debian 2.2.3);
2174 this information should be available from @file{/etc/issue}.
2177 The version of the Linux kernel, available from @samp{uname --version}
2181 The version of glibc you used; for RPM-based systems like Red Hat,
2182 Mandrake, and SuSE type @samp{rpm -q glibc} to get the glibc version,
2183 and on systems like Debian and Progeny use @samp{dpkg -l libc6}.
2185 For other systems, you can include similar information if you think it is
2189 Any other information that you think would be useful to people building
2190 GCC on the same configuration. The new entry in the build status list
2191 will include a link to the archived copy of your message.
2194 We'd also like to know if the
2196 @ref{Specific, host/target specific installation notes}
2199 @uref{specific.html,,host/target specific installation notes}
2201 didn't include your host/target information or if that information is
2202 incomplete or out of date. Send a note to
2203 @email{gcc@@gcc.gnu.org} detailing how the information should be changed.
2205 If you find a bug, please report it following the
2206 @uref{../bugs.html,,bug reporting guidelines}.
2208 If you want to print the GCC manuals, do @samp{cd @var{objdir}; make
2209 dvi}. You will need to have @command{texi2dvi} (version at least 4.4)
2210 and @TeX{} installed. This creates a number of @file{.dvi} files in
2211 subdirectories of @file{@var{objdir}}; these may be converted for
2212 printing with programs such as @command{dvips}. Alternately, by using
2213 @samp{make pdf} in place of @samp{make dvi}, you can create documentation
2214 in the form of @file{.pdf} files; this requires @command{texi2pdf}, which
2215 is included with Texinfo version 4.8 and later. You can also
2216 @uref{http://www.gnu.org/order/order.html,,buy printed manuals from the
2217 Free Software Foundation}, though such manuals may not be for the most
2218 recent version of GCC@.
2220 If you would like to generate online HTML documentation, do @samp{cd
2221 @var{objdir}; make html} and HTML will be generated for the gcc manuals in
2222 @file{@var{objdir}/gcc/HTML}.
2229 @uref{./index.html,,Return to the GCC Installation page}
2233 @c ***Binaries****************************************************************
2235 @comment node-name, next, previous, up
2236 @node Binaries, Specific, Installing GCC, Top
2240 @chapter Installing GCC: Binaries
2243 @cindex Installing GCC: Binaries
2245 We are often asked about pre-compiled versions of GCC@. While we cannot
2246 provide these for all platforms, below you'll find links to binaries for
2247 various platforms where creating them by yourself is not easy due to various
2250 Please note that we did not create these binaries, nor do we
2251 support them. If you have any problems installing them, please
2252 contact their makers.
2259 @uref{http://www.bullfreeware.com,,Bull's Freeware and Shareware Archive for AIX};
2262 @uref{http://aixpdslib.seas.ucla.edu,,UCLA Software Library for AIX}.
2266 DOS---@uref{http://www.delorie.com/djgpp/,,DJGPP}.
2269 Renesas H8/300[HS]---@uref{http://h8300-hms.sourceforge.net/,,GNU
2270 Development Tools for the Renesas H8/300[HS] Series}.
2276 @uref{http://hpux.cs.utah.edu/,,HP-UX Porting Center};
2279 @uref{ftp://sunsite.informatik.rwth-aachen.de/pub/packages/gcc_hpux/,,Binaries for HP-UX 11.00 at Aachen University of Technology}.
2283 Motorola 68HC11/68HC12---@uref{http://www.gnu-m68hc11.org,,GNU
2284 Development Tools for the Motorola 68HC11/68HC12}.
2287 @uref{http://www.sco.com/skunkware/devtools/index.html#gcc,,SCO
2288 OpenServer/Unixware}.
2291 Solaris 2 (SPARC, Intel)---@uref{http://www.sunfreeware.com/,,Sunfreeware}.
2294 SGI---@uref{http://freeware.sgi.com/,,SGI Freeware}.
2300 The @uref{http://sourceware.org/cygwin/,,Cygwin} project;
2302 The @uref{http://www.mingw.org/,,MinGW} project.
2306 @uref{ftp://ftp.thewrittenword.com/packages/by-name/,,The
2307 Written Word} offers binaries for
2310 Digital UNIX 4.0D and 5.1,
2312 HP-UX 10.20, 11.00, and 11.11, and
2313 Solaris/SPARC 2.5.1, 2.6, 7, 8, and 9.
2316 @uref{http://www.openpkg.org/,,OpenPKG} offers binaries for quite a
2317 number of platforms.
2320 The @uref{http://gcc.gnu.org/wiki/GFortranBinaries,,GFortran Wiki} has
2321 links to GNU Fortran binaries for several platforms.
2324 In addition to those specific offerings, you can get a binary
2325 distribution CD-ROM from the
2326 @uref{http://www.gnu.org/order/order.html,,Free Software Foundation}.
2327 It contains binaries for a number of platforms, and
2328 includes not only GCC, but other stuff as well. The current CD does
2329 not contain the latest version of GCC, but it should allow
2330 bootstrapping the compiler. An updated version of that disk is in the
2338 @uref{./index.html,,Return to the GCC Installation page}
2342 @c ***Specific****************************************************************
2344 @comment node-name, next, previous, up
2345 @node Specific, Old, Binaries, Top
2349 @chapter Host/target specific installation notes for GCC
2352 @cindex Specific installation notes
2353 @cindex Target specific installation
2354 @cindex Host specific installation
2355 @cindex Target specific installation notes
2357 Please read this document carefully @emph{before} installing the
2358 GNU Compiler Collection on your machine.
2360 Note that this list of install notes is @emph{not} a list of supported
2361 hosts or targets. Not all supported hosts and targets are listed
2362 here, only the ones that require host-specific or target-specific
2368 @uref{#alpha-x-x,,alpha*-*-*}
2370 @uref{#alpha-dec-osf,,alpha*-dec-osf*}
2372 @uref{#alphaev5-cray-unicosmk,,alphaev5-cray-unicosmk*}
2374 @uref{#arc-x-elf,,arc-*-elf}
2376 @uref{#arm-x-elf,,arm-*-elf}
2377 @uref{#arm-x-coff,,arm-*-coff}
2378 @uref{#arm-x-aout,,arm-*-aout}
2380 @uref{#xscale-x-x,,xscale-*-*}
2384 @uref{#bfin,,Blackfin}
2390 @uref{#x-x-freebsd,,*-*-freebsd*}
2392 @uref{#h8300-hms,,h8300-hms}
2394 @uref{#hppa-hp-hpux,,hppa*-hp-hpux*}
2396 @uref{#hppa-hp-hpux10,,hppa*-hp-hpux10}
2398 @uref{#hppa-hp-hpux11,,hppa*-hp-hpux11}
2400 @uref{#x-x-linux-gnu,,*-*-linux-gnu}
2402 @uref{#ix86-x-linuxaout,,i?86-*-linux*aout}
2404 @uref{#ix86-x-linux,,i?86-*-linux*}
2406 @uref{#ix86-x-sco32v5,,i?86-*-sco3.2v5*}
2408 @uref{#ix86-x-solaris210,,i?86-*-solaris2.10}
2410 @uref{#ix86-x-udk,,i?86-*-udk}
2412 @uref{#ia64-x-linux,,ia64-*-linux}
2414 @uref{#ia64-x-hpux,,ia64-*-hpux*}
2416 @uref{#x-ibm-aix,,*-ibm-aix*}
2418 @uref{#iq2000-x-elf,,iq2000-*-elf}
2420 @uref{#m32c-x-elf,,m32c-*-elf}
2422 @uref{#m32r-x-elf,,m32r-*-elf}
2424 @uref{#m6811-elf,,m6811-elf}
2426 @uref{#m6812-elf,,m6812-elf}
2428 @uref{#m68k-x-x,,m68k-*-*}
2430 @uref{#m68k-hp-hpux,,m68k-hp-hpux}
2432 @uref{#m68k-uclinux,,m68k-uclinux}
2434 @uref{#mips-x-x,,mips-*-*}
2436 @uref{#mips-sgi-irix5,,mips-sgi-irix5}
2438 @uref{#mips-sgi-irix6,,mips-sgi-irix6}
2440 @uref{#powerpc-x-x,,powerpc*-*-*, powerpc-*-sysv4}
2442 @uref{#powerpc-x-darwin,,powerpc-*-darwin*}
2444 @uref{#powerpc-x-elf,,powerpc-*-elf, powerpc-*-sysv4}
2446 @uref{#powerpc-x-linux-gnu,,powerpc*-*-linux-gnu*}
2448 @uref{#powerpc-x-netbsd,,powerpc-*-netbsd*}
2450 @uref{#powerpc-x-eabisim,,powerpc-*-eabisim}
2452 @uref{#powerpc-x-eabi,,powerpc-*-eabi}
2454 @uref{#powerpcle-x-elf,,powerpcle-*-elf, powerpcle-*-sysv4}
2456 @uref{#powerpcle-x-eabisim,,powerpcle-*-eabisim}
2458 @uref{#powerpcle-x-eabi,,powerpcle-*-eabi}
2460 @uref{#s390-x-linux,,s390-*-linux*}
2462 @uref{#s390x-x-linux,,s390x-*-linux*}
2464 @uref{#s390x-ibm-tpf,,s390x-ibm-tpf*}
2466 @uref{#x-x-solaris2,,*-*-solaris2*}
2468 @uref{#sparc-sun-solaris2,,sparc-sun-solaris2*}
2470 @uref{#sparc-sun-solaris27,,sparc-sun-solaris2.7}
2472 @uref{#sparc-x-linux,,sparc-*-linux*}
2474 @uref{#sparc64-x-solaris2,,sparc64-*-solaris2*}
2476 @uref{#sparcv9-x-solaris2,,sparcv9-*-solaris2*}
2478 @uref{#x-x-sysv,,*-*-sysv*}
2480 @uref{#vax-dec-ultrix,,vax-dec-ultrix}
2482 @uref{#x-x-vxworks,,*-*-vxworks*}
2484 @uref{#x86-64-x-x,,x86_64-*-*, amd64-*-*}
2486 @uref{#xtensa-x-elf,,xtensa-*-elf}
2488 @uref{#xtensa-x-linux,,xtensa-*-linux*}
2490 @uref{#windows,,Microsoft Windows}
2494 @uref{#older,,Older systems}
2499 @uref{#elf,,all ELF targets} (SVR4, Solaris 2, etc.)
2505 <!-- -------- host/target specific issues start here ---------------- -->
2508 @heading @anchor{alpha-x-x}alpha*-*-*
2510 This section contains general configuration information for all
2511 alpha-based platforms using ELF (in particular, ignore this section for
2512 DEC OSF/1, Digital UNIX and Tru64 UNIX)@. In addition to reading this
2513 section, please read all other sections that match your target.
2515 We require binutils 2.11.2 or newer.
2516 Previous binutils releases had a number of problems with DWARF 2
2517 debugging information, not the least of which is incorrect linking of
2523 @heading @anchor{alpha-dec-osf}alpha*-dec-osf*
2524 Systems using processors that implement the DEC Alpha architecture and
2525 are running the DEC/Compaq Unix (DEC OSF/1, Digital UNIX, or Compaq
2526 Tru64 UNIX) operating system, for example the DEC Alpha AXP systems.
2528 As of GCC 3.2, versions before @code{alpha*-dec-osf4} are no longer
2529 supported. (These are the versions which identify themselves as DEC
2532 In Digital Unix V4.0, virtual memory exhausted bootstrap failures
2533 may be fixed by configuring with @option{--with-gc=simple},
2534 reconfiguring Kernel Virtual Memory and Swap parameters
2535 per the @command{/usr/sbin/sys_check} Tuning Suggestions,
2536 or applying the patch in
2537 @uref{http://gcc.gnu.org/ml/gcc/2002-08/msg00822.html}.
2539 In Tru64 UNIX V5.1, Compaq introduced a new assembler that does not
2540 currently (2001-06-13) work with @command{mips-tfile}. As a workaround,
2541 we need to use the old assembler, invoked via the barely documented
2542 @option{-oldas} option. To bootstrap GCC, you either need to use the
2546 % CC=cc @var{srcdir}/configure [@var{options}] [@var{target}]
2549 or you can use a copy of GCC 2.95.3 or higher built on Tru64 UNIX V4.0:
2552 % CC=gcc -Wa,-oldas @var{srcdir}/configure [@var{options}] [@var{target}]
2555 As of GNU binutils 2.11.2, neither GNU @command{as} nor GNU @command{ld}
2556 are supported on Tru64 UNIX, so you must not configure GCC with
2557 @option{--with-gnu-as} or @option{--with-gnu-ld}.
2559 GCC writes a @samp{.verstamp} directive to the assembler output file
2560 unless it is built as a cross-compiler. It gets the version to use from
2561 the system header file @file{/usr/include/stamp.h}. If you install a
2562 new version of DEC Unix, you should rebuild GCC to pick up the new version
2565 Note that since the Alpha is a 64-bit architecture, cross-compilers from
2566 32-bit machines will not generate code as efficient as that generated
2567 when the compiler is running on a 64-bit machine because many
2568 optimizations that depend on being able to represent a word on the
2569 target in an integral value on the host cannot be performed. Building
2570 cross-compilers on the Alpha for 32-bit machines has only been tested in
2571 a few cases and may not work properly.
2573 @samp{make compare} may fail on old versions of DEC Unix unless you add
2574 @option{-save-temps} to @code{CFLAGS}. On these systems, the name of the
2575 assembler input file is stored in the object file, and that makes
2576 comparison fail if it differs between the @code{stage1} and
2577 @code{stage2} compilations. The option @option{-save-temps} forces a
2578 fixed name to be used for the assembler input file, instead of a
2579 randomly chosen name in @file{/tmp}. Do not add @option{-save-temps}
2580 unless the comparisons fail without that option. If you add
2581 @option{-save-temps}, you will have to manually delete the @samp{.i} and
2582 @samp{.s} files after each series of compilations.
2584 GCC now supports both the native (ECOFF) debugging format used by DBX
2585 and GDB and an encapsulated STABS format for use only with GDB@. See the
2586 discussion of the @option{--with-stabs} option of @file{configure} above
2587 for more information on these formats and how to select them.
2589 There is a bug in DEC's assembler that produces incorrect line numbers
2590 for ECOFF format when the @samp{.align} directive is used. To work
2591 around this problem, GCC will not emit such alignment directives
2592 while writing ECOFF format debugging information even if optimization is
2593 being performed. Unfortunately, this has the very undesirable
2594 side-effect that code addresses when @option{-O} is specified are
2595 different depending on whether or not @option{-g} is also specified.
2597 To avoid this behavior, specify @option{-gstabs+} and use GDB instead of
2598 DBX@. DEC is now aware of this problem with the assembler and hopes to
2599 provide a fix shortly.
2604 @heading @anchor{alphaev5-cray-unicosmk}alphaev5-cray-unicosmk*
2605 Cray T3E systems running Unicos/Mk.
2607 This port is incomplete and has many known bugs. We hope to improve the
2608 support for this target soon. Currently, only the C front end is supported,
2609 and it is not possible to build parallel applications. Cray modules are not
2610 supported; in particular, Craylibs are assumed to be in
2611 @file{/opt/ctl/craylibs/craylibs}.
2613 On this platform, you need to tell GCC where to find the assembler and
2614 the linker. The simplest way to do so is by providing @option{--with-as}
2615 and @option{--with-ld} to @file{configure}, e.g.@:
2618 configure --with-as=/opt/ctl/bin/cam --with-ld=/opt/ctl/bin/cld \
2619 --enable-languages=c
2622 The comparison test at the end of the bootstrapping process fails on Unicos/Mk
2623 because the assembler inserts timestamps into object files. You should
2624 be able to work around this by doing @samp{make all} after getting this
2630 @heading @anchor{arc-x-elf}arc-*-elf
2631 Argonaut ARC processor.
2632 This configuration is intended for embedded systems.
2637 @heading @anchor{arm-x-elf}arm-*-elf
2638 @heading @anchor{xscale-x-x}xscale-*-*
2639 ARM-family processors. Subtargets that use the ELF object format
2640 require GNU binutils 2.13 or newer. Such subtargets include:
2641 @code{arm-*-freebsd}, @code{arm-*-netbsdelf}, @code{arm-*-*linux},
2642 @code{arm-*-rtems} and @code{arm-*-kaos}.
2647 @heading @anchor{arm-x-coff}arm-*-coff
2648 ARM-family processors. Note that there are two different varieties
2649 of PE format subtarget supported: @code{arm-wince-pe} and
2650 @code{arm-pe} as well as a standard COFF target @code{arm-*-coff}.
2655 @heading @anchor{arm-x-aout}arm-*-aout
2656 ARM-family processors. These targets support the AOUT file format:
2657 @code{arm-*-aout}, @code{arm-*-netbsd}.
2662 @heading @anchor{avr}avr
2664 ATMEL AVR-family micro controllers. These are used in embedded
2665 applications. There are no standard Unix configurations.
2667 @xref{AVR Options,, AVR Options, gcc, Using the GNU Compiler
2671 See ``AVR Options'' in the main manual
2673 for the list of supported MCU types.
2675 Use @samp{configure --target=avr --enable-languages="c"} to configure GCC@.
2677 Further installation notes and other useful information about AVR tools
2678 can also be obtained from:
2682 @uref{http://www.nongnu.org/avr/,,http://www.nongnu.org/avr/}
2684 @uref{http://home.overta.ru/users/denisc/,,http://home.overta.ru/users/denisc/}
2686 @uref{http://www.amelek.gda.pl/avr/,,http://www.amelek.gda.pl/avr/}
2689 We @emph{strongly} recommend using binutils 2.13 or newer.
2691 The following error:
2693 Error: register required
2696 indicates that you should upgrade to a newer version of the binutils.
2701 @heading @anchor{bfin}Blackfin
2703 The Blackfin processor, an Analog Devices DSP.
2705 @xref{Blackfin Options,, Blackfin Options, gcc, Using the GNU Compiler
2709 See ``Blackfin Options'' in the main manual
2712 More information, and a version of binutils with support for this processor,
2713 is available at @uref{http://blackfin.uclinux.org}
2718 @heading @anchor{c4x}c4x
2720 Texas Instruments TMS320C3x and TMS320C4x Floating Point Digital Signal
2721 Processors. These are used in embedded applications. There are no
2722 standard Unix configurations.
2724 @xref{TMS320C3x/C4x Options,, TMS320C3x/C4x Options, gcc, Using the
2725 GNU Compiler Collection (GCC)},
2728 See ``TMS320C3x/C4x Options'' in the main manual
2730 for the list of supported MCU types.
2732 GCC can be configured as a cross compiler for both the C3x and C4x
2733 architectures on the same system. Use @samp{configure --target=c4x
2734 --enable-languages="c,c++"} to configure.
2737 Further installation notes and other useful information about C4x tools
2738 can also be obtained from:
2742 @uref{http://www.elec.canterbury.ac.nz/c4x/,,http://www.elec.canterbury.ac.nz/c4x/}
2748 @heading @anchor{cris}CRIS
2750 CRIS is the CPU architecture in Axis Communications ETRAX system-on-a-chip
2751 series. These are used in embedded applications.
2754 @xref{CRIS Options,, CRIS Options, gcc, Using the GNU Compiler
2758 See ``CRIS Options'' in the main manual
2760 for a list of CRIS-specific options.
2762 There are a few different CRIS targets:
2764 @item cris-axis-aout
2765 Old target. Includes a multilib for the @samp{elinux} a.out-based
2766 target. No multilibs for newer architecture variants.
2768 Mainly for monolithic embedded systems. Includes a multilib for the
2769 @samp{v10} core used in @samp{ETRAX 100 LX}.
2770 @item cris-axis-linux-gnu
2771 A GNU/Linux port for the CRIS architecture, currently targeting
2772 @samp{ETRAX 100 LX} by default.
2775 For @code{cris-axis-aout} and @code{cris-axis-elf} you need binutils 2.11
2776 or newer. For @code{cris-axis-linux-gnu} you need binutils 2.12 or newer.
2778 Pre-packaged tools can be obtained from
2779 @uref{ftp://ftp.axis.com/pub/axis/tools/cris/compiler-kit/}. More
2780 information about this platform is available at
2781 @uref{http://developer.axis.com/}.
2786 @heading @anchor{crx}CRX
2788 The CRX CompactRISC architecture is a low-power 32-bit architecture with
2789 fast context switching and architectural extensibility features.
2792 @xref{CRX Options,, CRX Options, gcc, Using and Porting the GNU Compiler
2797 See ``CRX Options'' in the main manual for a list of CRX-specific options.
2800 Use @samp{configure --target=crx-elf --enable-languages=c,c++} to configure
2801 GCC@ for building a CRX cross-compiler. The option @samp{--target=crx-elf}
2802 is also used to build the @samp{newlib} C library for CRX.
2804 It is also possible to build libstdc++-v3 for the CRX architecture. This
2805 needs to be done in a separate step with the following configure settings:
2806 @samp{gcc/libstdc++-v3/configure --host=crx-elf --with-newlib
2807 --enable-sjlj-exceptions --enable-cxx-flags='-fexceptions -frtti'}
2812 @heading @anchor{dos}DOS
2814 Please have a look at the @uref{binaries.html,,binaries page}.
2816 You cannot install GCC by itself on MSDOS; it will not compile under
2817 any MSDOS compiler except itself. You need to get the complete
2818 compilation package DJGPP, which includes binaries as well as sources,
2819 and includes all the necessary compilation tools and libraries.
2824 @heading @anchor{x-x-freebsd}*-*-freebsd*
2826 The version of binutils installed in @file{/usr/bin} probably works with
2827 this release of GCC@. However, on FreeBSD 4, bootstrapping against the
2828 latest FSF binutils is known to improve overall testsuite results; and,
2829 on FreeBSD/alpha, using binutils 2.14 or later is required to build libjava.
2831 Support for FreeBSD 1 was discontinued in GCC 3.2.
2833 Support for FreeBSD 2 will be discontinued after GCC 3.4. The
2834 following was true for GCC 3.1 but the current status is unknown.
2835 For FreeBSD 2 or any mutant a.out versions of FreeBSD 3: All
2836 configuration support and files as shipped with GCC 2.95 are still in
2837 place. FreeBSD 2.2.7 has been known to bootstrap completely; however,
2838 it is unknown which version of binutils was used (it is assumed that it
2839 was the system copy in @file{/usr/bin}) and C++ EH failures were noted.
2841 For FreeBSD using the ELF file format: DWARF 2 debugging is now the
2842 default for all CPU architectures. It had been the default on
2843 FreeBSD/alpha since its inception. You may use @option{-gstabs} instead
2844 of @option{-g}, if you really want the old debugging format. There are
2845 no known issues with mixing object files and libraries with different
2846 debugging formats. Otherwise, this release of GCC should now match more
2847 of the configuration used in the stock FreeBSD configuration of GCC@. In
2848 particular, @option{--enable-threads} is now configured by default.
2849 However, as a general user, do not attempt to replace the system
2850 compiler with this release. Known to bootstrap and check with good
2851 results on FreeBSD 4.9-STABLE and 5-CURRENT@. In the past, known to
2852 bootstrap and check with good results on FreeBSD 3.0, 3.4, 4.0, 4.2,
2853 4.3, 4.4, 4.5, 4.8-STABLE@.
2855 In principle, @option{--enable-threads} is now compatible with
2856 @option{--enable-libgcj} on FreeBSD@. However, it has only been built
2857 and tested on @samp{i386-*-freebsd[45]} and @samp{alpha-*-freebsd[45]}.
2859 library may be incorrectly built (symbols are missing at link time).
2860 There is a rare timing-based startup hang (probably involves an
2861 assumption about the thread library). Multi-threaded boehm-gc (required for
2862 libjava) exposes severe threaded signal-handling bugs on FreeBSD before
2863 4.5-RELEASE@. Other CPU architectures
2864 supported by FreeBSD will require additional configuration tuning in, at
2865 the very least, both boehm-gc and libffi.
2867 Shared @file{libgcc_s.so} is now built and installed by default.
2872 @heading @anchor{h8300-hms}h8300-hms
2873 Renesas H8/300 series of processors.
2875 Please have a look at the @uref{binaries.html,,binaries page}.
2877 The calling convention and structure layout has changed in release 2.6.
2878 All code must be recompiled. The calling convention now passes the
2879 first three arguments in function calls in registers. Structures are no
2880 longer a multiple of 2 bytes.
2885 @heading @anchor{hppa-hp-hpux}hppa*-hp-hpux*
2886 Support for HP-UX version 9 and older was discontinued in GCC 3.4.
2888 We require using gas/binutils on all hppa platforms;
2889 you may encounter a variety of problems if you try to use the HP assembler.
2891 Specifically, @option{-g} does not work on HP-UX (since that system
2892 uses a peculiar debugging format which GCC does not know about), unless
2893 you use GAS and GDB@. It may be helpful to configure GCC with the
2894 @uref{./configure.html#with-gnu-as,,@option{--with-gnu-as}} and
2895 @option{--with-as=@dots{}} options to ensure that GCC can find GAS@.
2897 If you wish to use the pa-risc 2.0 architecture support with a 32-bit
2898 runtime, you must use gas/binutils 2.11 or newer.
2900 There are two default scheduling models for instructions. These are
2901 PROCESSOR_7100LC and PROCESSOR_8000. They are selected from the pa-risc
2902 architecture specified for the target machine when configuring.
2903 PROCESSOR_8000 is the default. PROCESSOR_7100LC is selected when
2904 the target is a @samp{hppa1*} machine.
2906 The PROCESSOR_8000 model is not well suited to older processors. Thus,
2907 it is important to completely specify the machine architecture when
2908 configuring if you want a model other than PROCESSOR_8000. The macro
2909 TARGET_SCHED_DEFAULT can be defined in BOOT_CFLAGS if a different
2910 default scheduling model is desired.
2912 As of GCC 4.0, GCC uses the UNIX 95 namespace for HP-UX 10.10
2913 through 11.00, and the UNIX 98 namespace for HP-UX 11.11 and later.
2914 This namespace change might cause problems when bootstrapping with
2915 an earlier version of GCC or the HP compiler as essentially the same
2916 namespace is required for an entire build. This problem can be avoided
2917 in a number of ways. With HP cc, @env{UNIX_STD} can be set to @samp{95}
2918 or @samp{98}. Another way is to add an appropriate set of predefines
2919 to @env{CC}. The description for the @option{munix=} option contains
2920 a list of the predefines used with each standard.
2922 As of GCC 4.1, @env{DWARF2} exception handling is available on HP-UX.
2923 It is now the default. This exposed a bug in the handling of data
2924 relocations in the GAS assembler. The handling of 64-bit data relocations
2925 was seriously broken, affecting debugging and exception support on all
2926 @samp{hppa64-*-*} targets. Under some circumstances, 32-bit data relocations
2927 could also be handled incorrectly. This problem is fixed in GAS version
2930 GCC versions prior to 4.1 incorrectly passed and returned complex
2931 values. They are now passed in the same manner as aggregates.
2933 More specific information to @samp{hppa*-hp-hpux*} targets follows.
2938 @heading @anchor{hppa-hp-hpux10}hppa*-hp-hpux10
2940 For hpux10.20, we @emph{highly} recommend you pick up the latest sed patch
2941 @code{PHCO_19798} from HP@. HP has two sites which provide patches free of
2947 <a href="http://us.itrc.hp.com/service/home/home.do">US, Canada, Asia-Pacific, and
2951 @uref{http://us.itrc.hp.com/service/home/home.do,,} US, Canada, Asia-Pacific,
2955 @uref{http://europe.itrc.hp.com/service/home/home.do,,} Europe.
2958 The HP assembler on these systems has some problems. Most notably the
2959 assembler inserts timestamps into each object file it creates, causing
2960 the 3-stage comparison test to fail during a bootstrap.
2961 You should be able to continue by saying @samp{make all-host all-target}
2962 after getting the failure from @samp{make}.
2964 GCC 4.0 requires CVS binutils as of April 28, 2004 or later. Earlier
2965 versions require binutils 2.8 or later.
2967 The C++ ABI has changed incompatibly in GCC 4.0. COMDAT subspaces are
2968 used for one-only code and data. This resolves many of the previous
2969 problems in using C++ on this target. However, the ABI is not compatible
2970 with the one implemented under HP-UX 11 using secondary definitions.
2975 @heading @anchor{hppa-hp-hpux11}hppa*-hp-hpux11
2977 GCC 3.0 and up support HP-UX 11. GCC 2.95.x is not supported and cannot
2978 be used to compile GCC 3.0 and up.
2980 Refer to @uref{binaries.html,,binaries} for information about obtaining
2981 precompiled GCC binaries for HP-UX@. Precompiled binaries must be obtained
2982 to build the Ada language as it can't be bootstrapped using C@. Ada is
2983 only available for the 32-bit PA-RISC runtime. The libffi and libjava
2984 haven't been ported to HP-UX and don't build.
2986 Starting with GCC 3.4 an ISO C compiler is required to bootstrap. The
2987 bundled compiler supports only traditional C; you will need either HP's
2988 unbundled compiler, or a binary distribution of GCC@.
2990 It is possible to build GCC 3.3 starting with the bundled HP compiler,
2991 but the process requires several steps. GCC 3.3 can then be used to
2992 build later versions. The fastjar program contains ISO C code and
2993 can't be built with the HP bundled compiler. This problem can be
2994 avoided by not building the Java language. For example, use the
2995 @option{--enable-languages="c,c++,f77,objc"} option in your configure
2998 There are several possible approaches to building the distribution.
2999 Binutils can be built first using the HP tools. Then, the GCC
3000 distribution can be built. The second approach is to build GCC
3001 first using the HP tools, then build binutils, then rebuild GCC@.
3002 There have been problems with various binary distributions, so it
3003 is best not to start from a binary distribution.
3005 On 64-bit capable systems, there are two distinct targets. Different
3006 installation prefixes must be used if both are to be installed on
3007 the same system. The @samp{hppa[1-2]*-hp-hpux11*} target generates code
3008 for the 32-bit PA-RISC runtime architecture and uses the HP linker.
3009 The @samp{hppa64-hp-hpux11*} target generates 64-bit code for the
3010 PA-RISC 2.0 architecture. The HP and GNU linkers are both supported
3013 The script config.guess now selects the target type based on the compiler
3014 detected during configuration. You must define @env{PATH} or @env{CC} so
3015 that configure finds an appropriate compiler for the initial bootstrap.
3016 When @env{CC} is used, the definition should contain the options that are
3017 needed whenever @env{CC} is used.
3019 Specifically, options that determine the runtime architecture must be
3020 in @env{CC} to correctly select the target for the build. It is also
3021 convenient to place many other compiler options in @env{CC}. For example,
3022 @env{CC="cc -Ac +DA2.0W -Wp,-H16376 -D_CLASSIC_TYPES -D_HPUX_SOURCE"}
3023 can be used to bootstrap the GCC 3.3 branch with the HP compiler in
3024 64-bit K&R/bundled mode. The @option{+DA2.0W} option will result in
3025 the automatic selection of the @samp{hppa64-hp-hpux11*} target. The
3026 macro definition table of cpp needs to be increased for a successful
3027 build with the HP compiler. _CLASSIC_TYPES and _HPUX_SOURCE need to
3028 be defined when building with the bundled compiler, or when using the
3029 @option{-Ac} option. These defines aren't necessary with @option{-Ae}.
3031 It is best to explicitly configure the @samp{hppa64-hp-hpux11*} target
3032 with the @option{--with-ld=@dots{}} option. This overrides the standard
3033 search for ld. The two linkers supported on this target require different
3034 commands. The default linker is determined during configuration. As a
3035 result, it's not possible to switch linkers in the middle of a GCC build.
3036 This has been been reported to sometimes occur in unified builds of
3039 GCC 3.0 through 3.2 require binutils 2.11 or above. GCC 3.3 through
3040 GCC 4.0 require binutils 2.14 or later.
3042 Although the HP assembler can be used for an initial build, it shouldn't
3043 be used with any languages other than C and perhaps Fortran due to its
3044 many limitations. For example, it does not support weak symbols or alias
3045 definitions. As a result, explicit template instantiations are required
3046 when using C++. This makes it difficult if not impossible to build many
3047 C++ applications. You can't generate debugging information when using
3048 the HP assembler. Finally, bootstrapping fails in the final
3049 comparison of object modules due to the time stamps that it inserts into
3050 the modules. The bootstrap can be continued from this point with
3051 @samp{make all-host all-target}.
3053 A recent linker patch must be installed for the correct operation of
3054 GCC 3.3 and later. @code{PHSS_26559} and @code{PHSS_24304} are the
3055 oldest linker patches that are known to work. They are for HP-UX
3056 11.00 and 11.11, respectively. @code{PHSS_24303}, the companion to
3057 @code{PHSS_24304}, might be usable but it hasn't been tested. These
3058 patches have been superseded. Consult the HP patch database to obtain
3059 the currently recommended linker patch for your system.
3061 The patches are necessary for the support of weak symbols on the
3062 32-bit port, and for the running of initializers and finalizers. Weak
3063 symbols are implemented using SOM secondary definition symbols. Prior
3064 to HP-UX 11, there are bugs in the linker support for secondary symbols.
3065 The patches correct a problem of linker core dumps creating shared
3066 libraries containing secondary symbols, as well as various other
3067 linking issues involving secondary symbols.
3069 GCC 3.3 uses the ELF DT_INIT_ARRAY and DT_FINI_ARRAY capabilities to
3070 run initializers and finalizers on the 64-bit port. The 32-bit port
3071 uses the linker @option{+init} and @option{+fini} options for the same
3072 purpose. The patches correct various problems with the +init/+fini
3073 options, including program core dumps. Binutils 2.14 corrects a
3074 problem on the 64-bit port resulting from HP's non-standard use of
3075 the .init and .fini sections for array initializers and finalizers.
3077 There are a number of issues to consider in selecting which linker to
3078 use with the 64-bit port. The GNU 64-bit linker can only create dynamic
3079 binaries. The @option{-static} option causes linking with archive
3080 libraries but doesn't produce a truly static binary. Dynamic binaries
3081 still require final binding by the dynamic loader to resolve a set of
3082 dynamic-loader-defined symbols. The default behavior of the HP linker
3083 is the same as the GNU linker. However, it can generate true 64-bit
3084 static binaries using the @option{+compat} option.
3086 The HP 64-bit linker doesn't support linkonce semantics. As a
3087 result, C++ programs have many more sections than they should.
3089 The GNU 64-bit linker has some issues with shared library support
3090 and exceptions. As a result, we only support libgcc in archive
3091 format. For similar reasons, dwarf2 unwind and exception support
3092 are disabled. The GNU linker also has problems creating binaries
3093 with @option{-static}. It doesn't provide stubs for internal
3094 calls to global functions in shared libraries, so these calls
3095 can't be overloaded.
3097 Thread support is not implemented in GCC 3.0 through 3.2, so the
3098 @option{--enable-threads} configure option does not work. In 3.3
3099 and later, POSIX threads are supported. The optional DCE thread
3100 library is not supported.
3102 This port still is undergoing significant development.
3107 @heading @anchor{x-x-linux-gnu}*-*-linux-gnu
3109 Versions of libstdc++-v3 starting with 3.2.1 require bug fixes present
3110 in glibc 2.2.5 and later. More information is available in the
3111 libstdc++-v3 documentation.
3116 @heading @anchor{ix86-x-linuxaout}i?86-*-linux*aout
3117 Use this configuration to generate @file{a.out} binaries on Linux-based
3118 GNU systems. This configuration is being superseded.
3123 @heading @anchor{ix86-x-linux}i?86-*-linux*
3125 As of GCC 3.3, binutils 2.13.1 or later is required for this platform.
3126 See @uref{http://gcc.gnu.org/PR10877,,bug 10877} for more information.
3128 If you receive Signal 11 errors when building on GNU/Linux, then it is
3129 possible you have a hardware problem. Further information on this can be
3130 found on @uref{http://www.bitwizard.nl/sig11/,,www.bitwizard.nl}.
3135 @heading @anchor{ix86-x-sco32v5}i?86-*-sco3.2v5*
3136 Use this for the SCO OpenServer Release 5 family of operating systems.
3138 Unlike earlier versions of GCC, the ability to generate COFF with this
3139 target is no longer provided.
3141 Earlier versions of GCC emitted DWARF 1 when generating ELF to allow
3142 the system debugger to be used. That support was too burdensome to
3143 maintain. GCC now emits only DWARF 2 for this target. This means you
3144 may use either the UDK debugger or GDB to debug programs built by this
3147 GCC is now only supported on releases 5.0.4 and later, and requires that
3148 you install Support Level Supplement OSS646B or later, and Support Level
3149 Supplement OSS631C or later. If you are using release 5.0.7 of
3150 OpenServer, you must have at least the first maintenance pack installed
3151 (this includes the relevant portions of OSS646). OSS646, also known as
3152 the ``Execution Environment Update'', provides updated link editors and
3153 assemblers, as well as updated standard C and math libraries. The C
3154 startup modules are also updated to support the System V gABI draft, and
3155 GCC relies on that behavior. OSS631 provides a collection of commonly
3156 used open source libraries, some of which GCC depends on (such as GNU
3157 gettext and zlib). SCO OpenServer Release 5.0.7 has all of this built
3158 in by default, but OSS631C and later also apply to that release. Please
3160 @uref{ftp://ftp.sco.com/pub/openserver5,,ftp://ftp.sco.com/pub/openserver5}
3161 for the latest versions of these (and other potentially useful)
3164 Although there is support for using the native assembler, it is
3165 recommended that you configure GCC to use the GNU assembler. You do
3166 this by using the flags
3167 @uref{./configure.html#with-gnu-as,,@option{--with-gnu-as}}. You should
3168 use a modern version of GNU binutils. Version 2.13.2.1 was used for all
3169 testing. In general, only the @option{--with-gnu-as} option is tested.
3170 A modern bintuils (as well as a plethora of other development related
3171 GNU utilities) can be found in Support Level Supplement OSS658A, the
3172 ``GNU Development Tools'' package. See the SCO web and ftp sites for details.
3173 That package also contains the currently ``officially supported'' version of
3174 GCC, version 2.95.3. It is useful for bootstrapping this version.
3179 @heading @anchor{ix86-x-solaris210}i?86-*-solaris2.10
3180 Use this for Solaris 10 or later on x86 and x86-64 systems. This
3181 configuration is supported by GCC 4.0 and later versions only.
3183 It is recommended that you configure GCC to use the GNU assembler in
3184 @file{/usr/sfw/bin/gas} but the Sun linker, using the options
3185 @option{--with-gnu-as --with-as=/usr/sfw/bin/gas --without-gnu-ld
3186 --with-ld=/usr/ccs/bin/ld}.
3191 @heading @anchor{ix86-x-udk}i?86-*-udk
3193 This target emulates the SCO Universal Development Kit and requires that
3194 package be installed. (If it is installed, you will have a
3195 @file{/udk/usr/ccs/bin/cc} file present.) It's very much like the
3196 @samp{i?86-*-unixware7*} target
3197 but is meant to be used when hosting on a system where UDK isn't the
3198 default compiler such as OpenServer 5 or Unixware 2. This target will
3199 generate binaries that will run on OpenServer, Unixware 2, or Unixware 7,
3200 with the same warnings and caveats as the SCO UDK@.
3202 This target is a little tricky to build because we have to distinguish
3203 it from the native tools (so it gets headers, startups, and libraries
3204 from the right place) while making the tools not think we're actually
3205 building a cross compiler. The easiest way to do this is with a configure
3209 CC=/udk/usr/ccs/bin/cc @var{/your/path/to}/gcc/configure \
3210 --host=i686-pc-udk --target=i686-pc-udk --program-prefix=udk-
3213 @emph{You should substitute @samp{i686} in the above command with the appropriate
3214 processor for your host.}
3216 After the usual @samp{make} and
3217 @samp{make install}, you can then access the UDK-targeted GCC
3218 tools by adding @command{udk-} before the commonly known name. For
3219 example, to invoke the C compiler, you would use @command{udk-gcc}.
3220 They will coexist peacefully with any native-target GCC tools you may
3227 @heading @anchor{ia64-x-linux}ia64-*-linux
3228 IA-64 processor (also known as IPF, or Itanium Processor Family)
3231 If you are using the installed system libunwind library with
3232 @option{--with-system-libunwind}, then you must use libunwind 0.98 or
3235 None of the following versions of GCC has an ABI that is compatible
3236 with any of the other versions in this list, with the exception that
3237 Red Hat 2.96 and Trillian 000171 are compatible with each other:
3238 3.1, 3.0.2, 3.0.1, 3.0, Red Hat 2.96, and Trillian 000717.
3239 This primarily affects C++ programs and programs that create shared libraries.
3240 GCC 3.1 or later is recommended for compiling linux, the kernel.
3241 As of version 3.1 GCC is believed to be fully ABI compliant, and hence no
3242 more major ABI changes are expected.
3247 @heading @anchor{ia64-x-hpux}ia64-*-hpux*
3248 Building GCC on this target requires the GNU Assembler. The bundled HP
3249 assembler will not work. To prevent GCC from using the wrong assembler,
3250 the option @option{--with-gnu-as} may be necessary.
3252 The GCC libunwind library has not been ported to HPUX@. This means that for
3253 GCC versions 3.2.3 and earlier, @option{--enable-libunwind-exceptions}
3254 is required to build GCC@. For GCC 3.3 and later, this is the default.
3255 For gcc 3.4.3 and later, @option{--enable-libunwind-exceptions} is
3256 removed and the system libunwind library will always be used.
3260 <!-- rs6000-ibm-aix*, powerpc-ibm-aix* -->
3262 @heading @anchor{x-ibm-aix}*-ibm-aix*
3263 Support for AIX version 3 and older was discontinued in GCC 3.4.
3265 ``out of memory'' bootstrap failures may indicate a problem with
3266 process resource limits (ulimit). Hard limits are configured in the
3267 @file{/etc/security/limits} system configuration file.
3269 To speed up the configuration phases of bootstrapping and installing GCC,
3270 one may use GNU Bash instead of AIX @command{/bin/sh}, e.g.,
3273 % CONFIG_SHELL=/opt/freeware/bin/bash
3274 % export CONFIG_SHELL
3277 and then proceed as described in @uref{build.html,,the build
3278 instructions}, where we strongly recommend specifying an absolute path
3279 to invoke @var{srcdir}/configure.
3281 Because GCC on AIX is built as a 32-bit executable by default,
3282 (although it can generate 64-bit programs) the GMP and MPFR libraries
3283 required by gfortran must be 32-bit libraries. Building GMP and MPFR
3284 as static archive libraries works better than shared libraries.
3286 Errors involving @code{alloca} when building GCC generally are due
3287 to an incorrect definition of @code{CC} in the Makefile or mixing files
3288 compiled with the native C compiler and GCC@. During the stage1 phase of
3289 the build, the native AIX compiler @strong{must} be invoked as @command{cc}
3290 (not @command{xlc}). Once @command{configure} has been informed of
3291 @command{xlc}, one needs to use @samp{make distclean} to remove the
3292 configure cache files and ensure that @env{CC} environment variable
3293 does not provide a definition that will confuse @command{configure}.
3294 If this error occurs during stage2 or later, then the problem most likely
3295 is the version of Make (see above).
3297 The native @command{as} and @command{ld} are recommended for bootstrapping
3298 on AIX 4 and required for bootstrapping on AIX 5L@. The GNU Assembler
3299 reports that it supports WEAK symbols on AIX 4, which causes GCC to try to
3300 utilize weak symbol functionality although it is not supported. The GNU
3301 Assembler and Linker do not support AIX 5L sufficiently to bootstrap GCC@.
3302 The native AIX tools do interoperate with GCC@.
3304 Building @file{libstdc++.a} requires a fix for an AIX Assembler bug
3305 APAR IY26685 (AIX 4.3) or APAR IY25528 (AIX 5.1). It also requires a
3306 fix for another AIX Assembler bug and a co-dependent AIX Archiver fix
3307 referenced as APAR IY53606 (AIX 5.2) or a APAR IY54774 (AIX 5.1)
3309 @samp{libstdc++} in GCC 3.4 increments the major version number of the
3310 shared object and GCC installation places the @file{libstdc++.a}
3311 shared library in a common location which will overwrite the and GCC
3312 3.3 version of the shared library. Applications either need to be
3313 re-linked against the new shared library or the GCC 3.1 and GCC 3.3
3314 versions of the @samp{libstdc++} shared object needs to be available
3315 to the AIX runtime loader. The GCC 3.1 @samp{libstdc++.so.4}, if
3316 present, and GCC 3.3 @samp{libstdc++.so.5} shared objects can be
3317 installed for runtime dynamic loading using the following steps to set
3318 the @samp{F_LOADONLY} flag in the shared object for @emph{each}
3319 multilib @file{libstdc++.a} installed:
3321 Extract the shared objects from the currently installed
3322 @file{libstdc++.a} archive:
3324 % ar -x libstdc++.a libstdc++.so.4 libstdc++.so.5
3327 Enable the @samp{F_LOADONLY} flag so that the shared object will be
3328 available for runtime dynamic loading, but not linking:
3330 % strip -e libstdc++.so.4 libstdc++.so.5
3333 Archive the runtime-only shared object in the GCC 3.4
3334 @file{libstdc++.a} archive:
3336 % ar -q libstdc++.a libstdc++.so.4 libstdc++.so.5
3339 Linking executables and shared libraries may produce warnings of
3340 duplicate symbols. The assembly files generated by GCC for AIX always
3341 have included multiple symbol definitions for certain global variable
3342 and function declarations in the original program. The warnings should
3343 not prevent the linker from producing a correct library or runnable
3346 AIX 4.3 utilizes a ``large format'' archive to support both 32-bit and
3347 64-bit object modules. The routines provided in AIX 4.3.0 and AIX 4.3.1
3348 to parse archive libraries did not handle the new format correctly.
3349 These routines are used by GCC and result in error messages during
3350 linking such as ``not a COFF file''. The version of the routines shipped
3351 with AIX 4.3.1 should work for a 32-bit environment. The @option{-g}
3352 option of the archive command may be used to create archives of 32-bit
3353 objects using the original ``small format''. A correct version of the
3354 routines is shipped with AIX 4.3.2 and above.
3356 Some versions of the AIX binder (linker) can fail with a relocation
3357 overflow severe error when the @option{-bbigtoc} option is used to link
3358 GCC-produced object files into an executable that overflows the TOC@. A fix
3359 for APAR IX75823 (OVERFLOW DURING LINK WHEN USING GCC AND -BBIGTOC) is
3360 available from IBM Customer Support and from its
3361 @uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com}
3362 website as PTF U455193.
3364 The AIX 4.3.2.1 linker (bos.rte.bind_cmds Level 4.3.2.1) will dump core
3365 with a segmentation fault when invoked by any version of GCC@. A fix for
3366 APAR IX87327 is available from IBM Customer Support and from its
3367 @uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com}
3368 website as PTF U461879. This fix is incorporated in AIX 4.3.3 and above.
3370 The initial assembler shipped with AIX 4.3.0 generates incorrect object
3371 files. A fix for APAR IX74254 (64BIT DISASSEMBLED OUTPUT FROM COMPILER FAILS
3372 TO ASSEMBLE/BIND) is available from IBM Customer Support and from its
3373 @uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com}
3374 website as PTF U453956. This fix is incorporated in AIX 4.3.1 and above.
3376 AIX provides National Language Support (NLS)@. Compilers and assemblers
3377 use NLS to support locale-specific representations of various data
3378 formats including floating-point numbers (e.g., @samp{.} vs @samp{,} for
3379 separating decimal fractions). There have been problems reported where
3380 GCC does not produce the same floating-point formats that the assembler
3381 expects. If one encounters this problem, set the @env{LANG}
3382 environment variable to @samp{C} or @samp{En_US}.
3384 By default, GCC for AIX 4.1 and above produces code that can be used on
3385 both Power or PowerPC processors.
3387 A default can be specified with the @option{-mcpu=@var{cpu_type}}
3388 switch and using the configure option @option{--with-cpu-@var{cpu_type}}.
3393 @heading @anchor{iq2000-x-elf}iq2000-*-elf
3394 Vitesse IQ2000 processors. These are used in embedded
3395 applications. There are no standard Unix configurations.
3400 @heading @anchor{m32c-x-elf}m32c-*-elf
3401 Renesas M32C processor.
3402 This configuration is intended for embedded systems.
3407 @heading @anchor{m32r-x-elf}m32r-*-elf
3408 Renesas M32R processor.
3409 This configuration is intended for embedded systems.
3414 @heading @anchor{m6811-elf}m6811-elf
3415 Motorola 68HC11 family micro controllers. These are used in embedded
3416 applications. There are no standard Unix configurations.
3421 @heading @anchor{m6812-elf}m6812-elf
3422 Motorola 68HC12 family micro controllers. These are used in embedded
3423 applications. There are no standard Unix configurations.
3428 @heading @anchor{m68k-x-x}m68k-*-*
3429 You can specify a default target using @option{--with-cpu=@var{target}}.
3430 This @var{target} can either be a @option{-mcpu} argument or one of the
3431 following values: @samp{m68000}, @samp{m68010}, @samp{m68020}, @samp{m68030},
3432 @samp{m68040}, @samp{m68060}, @samp{m68020-40} and @samp{m68020-60}.
3437 @heading @anchor{m68k-hp-hpux}m68k-hp-hpux
3438 HP 9000 series 300 or 400 running HP-UX@. HP-UX version 8.0 has a bug in
3439 the assembler that prevents compilation of GCC@. This
3440 bug manifests itself during the first stage of compilation, while
3441 building @file{libgcc2.a}:
3445 cc1: warning: `-g' option not supported on this version of GCC
3446 cc1: warning: `-g1' option not supported on this version of GCC
3447 ./xgcc: Internal compiler error: program as got fatal signal 11
3450 A patched version of the assembler is available as the file
3451 @uref{ftp://altdorf.ai.mit.edu/archive/cph/hpux-8.0-assembler}. If you
3452 have HP software support, the patch can also be obtained directly from
3453 HP, as described in the following note:
3456 This is the patched assembler, to patch SR#1653-010439, where the
3457 assembler aborts on floating point constants.
3459 The bug is not really in the assembler, but in the shared library
3460 version of the function ``cvtnum(3c)''. The bug on ``cvtnum(3c)'' is
3461 SR#4701-078451. Anyway, the attached assembler uses the archive
3462 library version of ``cvtnum(3c)'' and thus does not exhibit the bug.
3465 This patch is also known as PHCO_4484.
3467 In addition gdb does not understand that native HP-UX format, so
3468 you must use gas if you wish to use gdb.
3470 On HP-UX version 8.05, but not on 8.07 or more recent versions, the
3471 @command{fixproto} shell script triggers a bug in the system shell. If you
3472 encounter this problem, upgrade your operating system or use BASH (the
3473 GNU shell) to run @command{fixproto}. This bug will cause the fixproto
3474 program to report an error of the form:
3477 ./fixproto: sh internal 1K buffer overflow
3480 To fix this, you can also change the first line of the fixproto script
3490 @heading @anchor{m68k-x-uclinux}m68k-*-uclinux
3491 GCC 4.3 changed the uClinux configuration so that it uses the
3492 @samp{m68k-linux-gnu} ABI rather than the @samp{m68k-elf} ABI.
3493 It also added improved support for C++ and flat shared libraries,
3494 both of which were ABI changes. However, you can still use the
3495 original ABI by configuring for @samp{m68k-uclinuxoldabi} or
3496 @samp{m68k-@var{vendor}-uclinuxoldabi}.
3501 @heading @anchor{mips-x-x}mips-*-*
3502 If on a MIPS system you get an error message saying ``does not have gp
3503 sections for all it's [sic] sectons [sic]'', don't worry about it. This
3504 happens whenever you use GAS with the MIPS linker, but there is not
3505 really anything wrong, and it is okay to use the output file. You can
3506 stop such warnings by installing the GNU linker.
3508 It would be nice to extend GAS to produce the gp tables, but they are
3509 optional, and there should not be a warning about their absence.
3511 The libstdc++ atomic locking routines for MIPS targets requires MIPS II
3512 and later. A patch went in just after the GCC 3.3 release to
3513 make @samp{mips*-*-*} use the generic implementation instead. You can also
3514 configure for @samp{mipsel-elf} as a workaround. The
3515 @samp{mips*-*-linux*} target continues to use the MIPS II routines. More
3516 work on this is expected in future releases.
3518 MIPS systems check for division by zero (unless
3519 @option{-mno-check-zero-division} is passed to the compiler) by
3520 generating either a conditional trap or a break instruction. Using
3521 trap results in smaller code, but is only supported on MIPS II and
3522 later. Also, some versions of the Linux kernel have a bug that
3523 prevents trap from generating the proper signal (@code{SIGFPE}). To enable
3524 the use of break, use the @option{--with-divide=breaks}
3525 @command{configure} option when configuring GCC@. The default is to
3526 use traps on systems that support them.
3528 Cross-compilers for the MIPS as target using the MIPS assembler
3529 currently do not work, because the auxiliary programs
3530 @file{mips-tdump.c} and @file{mips-tfile.c} can't be compiled on
3531 anything but a MIPS. It does work to cross compile for a MIPS
3532 if you use the GNU assembler and linker.
3534 The linker from GNU binutils versions prior to 2.17 has a bug which
3535 causes the runtime linker stubs in @file{libgcj.so} to be incorrectly
3536 generated. If you want to use libgcj, either use binutils 2.17 or
3537 later to build it or export @samp{LD_BIND_NOW=1} in your runtime environment.
3542 @heading @anchor{mips-sgi-irix5}mips-sgi-irix5
3544 In order to compile GCC on an SGI running IRIX 5, the @samp{compiler_dev.hdr}
3545 subsystem must be installed from the IDO CD-ROM supplied by SGI@.
3546 It is also available for download from
3547 @uref{ftp://ftp.sgi.com/sgi/IRIX5.3/iris-development-option-5.3.tardist}.
3549 If you use the MIPS C compiler to bootstrap, it may be necessary
3550 to increase its table size for switch statements with the
3551 @option{-Wf,-XNg1500} option. If you use the @option{-O2}
3552 optimization option, you also need to use @option{-Olimit 3000}.
3554 To enable debugging under IRIX 5, you must use GNU binutils 2.15 or
3555 later, and use the @option{--with-gnu-ld} @command{configure} option
3556 when configuring GCC@. You need to use GNU @command{ar} and @command{nm},
3557 also distributed with GNU binutils.
3559 Some users have reported that @command{/bin/sh} will hang during bootstrap.
3560 This problem can be avoided by running the commands:
3563 % CONFIG_SHELL=/bin/ksh
3564 % export CONFIG_SHELL
3567 before starting the build.
3572 @heading @anchor{mips-sgi-irix6}mips-sgi-irix6
3574 If you are using SGI's MIPSpro @command{cc} as your bootstrap compiler, you must
3575 ensure that the N32 ABI is in use. To test this, compile a simple C
3576 file with @command{cc} and then run @command{file} on the
3577 resulting object file. The output should look like:
3580 test.o: ELF N32 MSB @dots{}
3586 test.o: ELF 32-bit MSB @dots{}
3592 test.o: ELF 64-bit MSB @dots{}
3595 then your version of @command{cc} uses the O32 or N64 ABI by default. You
3596 should set the environment variable @env{CC} to @samp{cc -n32}
3597 before configuring GCC@.
3599 If you want the resulting @command{gcc} to run on old 32-bit systems
3600 with the MIPS R4400 CPU, you need to ensure that only code for the @samp{mips3}
3601 instruction set architecture (ISA) is generated. While GCC 3.x does
3602 this correctly, both GCC 2.95 and SGI's MIPSpro @command{cc} may change
3603 the ISA depending on the machine where GCC is built. Using one of them
3604 as the bootstrap compiler may result in @samp{mips4} code, which won't run at
3605 all on @samp{mips3}-only systems. For the test program above, you should see:
3608 test.o: ELF N32 MSB mips-3 @dots{}
3614 test.o: ELF N32 MSB mips-4 @dots{}
3617 instead, you should set the environment variable @env{CC} to @samp{cc
3618 -n32 -mips3} or @samp{gcc -mips3} respectively before configuring GCC@.
3620 MIPSpro C 7.4 may cause bootstrap failures, due to a bug when inlining
3621 @code{memcmp}. Either add @code{-U__INLINE_INTRINSICS} to the @env{CC}
3622 environment variable as a workaround or upgrade to MIPSpro C 7.4.1m.
3624 GCC on IRIX 6 is usually built to support the N32, O32 and N64 ABIs. If
3625 you build GCC on a system that doesn't have the N64 libraries installed
3626 or cannot run 64-bit binaries,
3627 you need to configure with @option{--disable-multilib} so GCC doesn't
3628 try to use them. This will disable building the O32 libraries, too.
3629 Look for @file{/usr/lib64/libc.so.1} to see if you
3630 have the 64-bit libraries installed.
3632 To enable debugging for the O32 ABI, you must use GNU @command{as} from
3633 GNU binutils 2.15 or later. You may also use GNU @command{ld}, but
3634 this is not required and currently causes some problems with Ada.
3636 The @option{--enable-threads} option doesn't currently work, a patch is
3637 in preparation for a future release. The @option{--enable-libgcj}
3638 option is disabled by default: IRIX 6 uses a very low default limit
3639 (20480) for the command line length. Although @command{libtool} contains a
3640 workaround for this problem, at least the N64 @samp{libgcj} is known not
3641 to build despite this, running into an internal error of the native
3642 @command{ld}. A sure fix is to increase this limit (@samp{ncargs}) to
3643 its maximum of 262144 bytes. If you have root access, you can use the
3644 @command{systune} command to do this.
3646 @code{wchar_t} support in @samp{libstdc++} is not available for old
3647 IRIX 6.5.x releases, @math{x < 19}. The problem cannot be autodetected
3648 and in order to build GCC for such targets you need to configure with
3649 @option{--disable-wchar_t}.
3651 See @uref{http://freeware.sgi.com/} for more
3652 information about using GCC on IRIX platforms.
3657 @heading @anchor{powerpc-x-x}powerpc-*-*
3659 You can specify a default version for the @option{-mcpu=@var{cpu_type}}
3660 switch by using the configure option @option{--with-cpu-@var{cpu_type}}.
3665 @heading @anchor{powerpc-x-darwin}powerpc-*-darwin*
3666 PowerPC running Darwin (Mac OS X kernel).
3668 Pre-installed versions of Mac OS X may not include any developer tools,
3669 meaning that you will not be able to build GCC from source. Tool
3670 binaries are available at
3671 @uref{http://developer.apple.com/darwin/projects/compiler/} (free
3672 registration required).
3674 This version of GCC requires at least cctools-590.36. The
3675 cctools-590.36 package referenced from
3676 @uref{http://gcc.gnu.org/ml/gcc/2006-03/msg00507.html} will not work
3677 on systems older than 10.3.9 (aka darwin7.9.0).
3682 @heading @anchor{powerpc-x-elf}powerpc-*-elf, powerpc-*-sysv4
3683 PowerPC system in big endian mode, running System V.4.
3688 @heading @anchor{powerpc-x-linux-gnu}powerpc*-*-linux-gnu*
3691 @uref{ftp://ftp.kernel.org/pub/linux/devel/binutils,,binutils 2.15}
3692 or newer for a working GCC@.
3697 @heading @anchor{powerpc-x-netbsd}powerpc-*-netbsd*
3698 PowerPC system in big endian mode running NetBSD@. To build the
3699 documentation you will need Texinfo version 4.4 (NetBSD 1.5.1 included
3700 Texinfo version 3.12).
3705 @heading @anchor{powerpc-x-eabisim}powerpc-*-eabisim
3706 Embedded PowerPC system in big endian mode for use in running under the
3712 @heading @anchor{powerpc-x-eabi}powerpc-*-eabi
3713 Embedded PowerPC system in big endian mode.
3718 @heading @anchor{powerpcle-x-elf}powerpcle-*-elf, powerpcle-*-sysv4
3719 PowerPC system in little endian mode, running System V.4.
3724 @heading @anchor{powerpcle-x-eabisim}powerpcle-*-eabisim
3725 Embedded PowerPC system in little endian mode for use in running under
3731 @heading @anchor{powerpcle-x-eabi}powerpcle-*-eabi
3732 Embedded PowerPC system in little endian mode.
3737 @heading @anchor{s390-x-linux}s390-*-linux*
3738 S/390 system running GNU/Linux for S/390@.
3743 @heading @anchor{s390x-x-linux}s390x-*-linux*
3744 zSeries system (64-bit) running GNU/Linux for zSeries@.
3749 @heading @anchor{s390x-ibm-tpf}s390x-ibm-tpf*
3750 zSeries system (64-bit) running TPF@. This platform is
3751 supported as cross-compilation target only.
3756 @c Please use Solaris 2 to refer to all release of Solaris, starting
3757 @c with 2.0 until 2.6, 7, 8, etc. Solaris 1 was a marketing name for
3758 @c SunOS 4 releases which we don't use to avoid confusion. Solaris
3759 @c alone is too unspecific and must be avoided.
3760 @heading @anchor{x-x-solaris2}*-*-solaris2*
3762 Sun does not ship a C compiler with Solaris 2. To bootstrap and install
3763 GCC you first have to install a pre-built compiler, see the
3764 @uref{binaries.html,,binaries page} for details.
3766 The Solaris 2 @command{/bin/sh} will often fail to configure
3767 @file{libstdc++-v3}, @file{boehm-gc} or @file{libjava}. We therefore
3768 recommend using the following initial sequence of commands
3771 % CONFIG_SHELL=/bin/ksh
3772 % export CONFIG_SHELL
3775 and proceed as described in @uref{configure.html,,the configure instructions}.
3776 In addition we strongly recommend specifying an absolute path to invoke
3777 @var{srcdir}/configure.
3779 Solaris 2 comes with a number of optional OS packages. Some of these
3780 are needed to use GCC fully, namely @code{SUNWarc},
3781 @code{SUNWbtool}, @code{SUNWesu}, @code{SUNWhea}, @code{SUNWlibm},
3782 @code{SUNWsprot}, and @code{SUNWtoo}. If you did not install all
3783 optional packages when installing Solaris 2, you will need to verify that
3784 the packages that GCC needs are installed.
3786 To check whether an optional package is installed, use
3787 the @command{pkginfo} command. To add an optional package, use the
3788 @command{pkgadd} command. For further details, see the Solaris 2
3791 Trying to use the linker and other tools in
3792 @file{/usr/ucb} to install GCC has been observed to cause trouble.
3793 For example, the linker may hang indefinitely. The fix is to remove
3794 @file{/usr/ucb} from your @env{PATH}.
3796 The build process works more smoothly with the legacy Sun tools so, if you
3797 have @file{/usr/xpg4/bin} in your @env{PATH}, we recommend that you place
3798 @file{/usr/bin} before @file{/usr/xpg4/bin} for the duration of the build.
3800 All releases of GNU binutils prior to 2.11.2 have known bugs on this
3801 platform. We recommend the use of GNU binutils 2.11.2 or later, or the
3802 vendor tools (Sun @command{as}, Sun @command{ld}). Note that your mileage
3803 may vary if you use a combination of the GNU tools and the Sun tools: while
3804 the combination GNU @command{as} + Sun @command{ld} should reasonably work,
3805 the reverse combination Sun @command{as} + GNU @command{ld} is known to
3806 cause memory corruption at runtime in some cases for C++ programs.
3808 The stock GNU binutils 2.15 release is broken on this platform because of a
3809 single bug. It has been fixed on the 2.15 branch in the CVS repository.
3810 You can obtain a working version by checking out the binutils-2_15-branch
3811 from the CVS repository or applying the patch
3812 @uref{http://sourceware.org/ml/binutils-cvs/2004-09/msg00036.html} to the
3815 We recommend using GNU binutils 2.16 or later in conjunction with GCC 4.x,
3816 or the vendor tools (Sun @command{as}, Sun @command{ld}). However, for
3817 Solaris 10 and above, an additional patch is required in order for the GNU
3818 linker to be able to cope with a new flavor of shared libraries. You
3819 can obtain a working version by checking out the binutils-2_16-branch from
3820 the CVS repository or applying the patch
3821 @uref{http://sourceware.org/ml/binutils-cvs/2005-07/msg00122.html} to the
3824 Sun bug 4296832 turns up when compiling X11 headers with GCC 2.95 or
3825 newer: @command{g++} will complain that types are missing. These headers assume
3826 that omitting the type means @code{int}; this assumption worked for C89 but
3827 is wrong for C++, and is now wrong for C99 also.
3829 @command{g++} accepts such (invalid) constructs with the option
3830 @option{-fpermissive}; it
3831 will assume that any missing type is @code{int} (as defined by C89).
3833 There are patches for Solaris 2.6 (105633-56 or newer for SPARC,
3834 106248-42 or newer for Intel), Solaris 7 (108376-21 or newer for SPARC,
3835 108377-20 for Intel), and Solaris 8 (108652-24 or newer for SPARC,
3836 108653-22 for Intel) that fix this bug.
3838 Sun bug 4927647 sometimes causes random spurious testsuite failures
3839 related to missing diagnostic output. This bug doesn't affect GCC
3840 itself, rather it is a kernel bug triggered by the @command{expect}
3841 program which is used only by the GCC testsuite driver. When the bug
3842 causes the @command{expect} program to miss anticipated output, extra
3843 testsuite failures appear.
3845 There are patches for Solaris 8 (117350-12 or newer for SPARC,
3846 117351-12 or newer for Intel) and Solaris 9 (117171-11 or newer for
3847 SPARC, 117172-11 or newer for Intel) that address this problem.
3852 @heading @anchor{sparc-sun-solaris2}sparc-sun-solaris2*
3854 When GCC is configured to use binutils 2.11.2 or later the binaries
3855 produced are smaller than the ones produced using Sun's native tools;
3856 this difference is quite significant for binaries containing debugging
3859 Sun @command{as} 4.x is broken in that it cannot cope with long symbol names.
3860 A typical error message might look similar to the following:
3863 /usr/ccs/bin/as: "/var/tmp/ccMsw135.s", line 11041: error:
3864 can't compute value of an expression involving an external symbol.
3867 This is Sun bug 4237974. This is fixed with patch 108908-02 for Solaris
3868 2.6 and has been fixed in later (5.x) versions of the assembler,
3869 starting with Solaris 7.
3871 Starting with Solaris 7, the operating system is capable of executing
3872 64-bit SPARC V9 binaries. GCC 3.1 and later properly supports
3873 this; the @option{-m64} option enables 64-bit code generation.
3874 However, if all you want is code tuned for the UltraSPARC CPU, you
3875 should try the @option{-mtune=ultrasparc} option instead, which produces
3876 code that, unlike full 64-bit code, can still run on non-UltraSPARC
3879 When configuring on a Solaris 7 or later system that is running a kernel
3880 that supports only 32-bit binaries, one must configure with
3881 @option{--disable-multilib}, since we will not be able to build the
3882 64-bit target libraries.
3884 GCC 3.3 and GCC 3.4 trigger code generation bugs in earlier versions of
3885 the GNU compiler (especially GCC 3.0.x versions), which lead to the
3886 miscompilation of the stage1 compiler and the subsequent failure of the
3887 bootstrap process. A workaround is to use GCC 3.2.3 as an intermediary
3888 stage, i.e.@: to bootstrap that compiler with the base compiler and then
3889 use it to bootstrap the final compiler.
3891 GCC 3.4 triggers a code generation bug in versions 5.4 (Sun ONE Studio 7)
3892 and 5.5 (Sun ONE Studio 8) of the Sun compiler, which causes a bootstrap
3893 failure in form of a miscompilation of the stage1 compiler by the Sun
3894 compiler. This is Sun bug 4974440. This is fixed with patch 112760-07.
3896 GCC 3.4 changed the default debugging format from STABS to DWARF-2 for
3897 32-bit code on Solaris 7 and later. If you use the Sun assembler, this
3898 change apparently runs afoul of Sun bug 4910101 (which is referenced as
3899 a x86-only problem by Sun, probably because they do not use DWARF-2).
3900 A symptom of the problem is that you cannot compile C++ programs like
3901 @command{groff} 1.19.1 without getting messages similar to the following:
3904 ld: warning: relocation error: R_SPARC_UA32: @dots{}
3905 external symbolic relocation against non-allocatable section
3906 .debug_info cannot be processed at runtime: relocation ignored.
3909 To work around this problem, compile with @option{-gstabs+} instead of
3912 When configuring the GNU Multiple Precision Library (GMP) or the MPFR
3913 library on a Solaris 7 or later system, the canonical target triplet
3914 must be specified as the @command{build} parameter on the configure
3915 line. This triplet can be obtained by invoking ./config.guess in
3916 the toplevel source directory of GCC (and not that of GMP or MPFR).
3917 For example on a Solaris 7 system:
3920 % ./configure --build=sparc-sun-solaris2.7 --prefix=xxx
3926 @heading @anchor{sparc-sun-solaris27}sparc-sun-solaris2.7
3928 Sun patch 107058-01 (1999-01-13) for Solaris 7/SPARC triggers a bug in
3929 the dynamic linker. This problem (Sun bug 4210064) affects GCC 2.8
3930 and later, including all EGCS releases. Sun formerly recommended
3931 107058-01 for all Solaris 7 users, but around 1999-09-01 it started to
3932 recommend it only for people who use Sun's compilers.
3934 Here are some workarounds to this problem:
3937 Do not install Sun patch 107058-01 until after Sun releases a
3938 complete patch for bug 4210064. This is the simplest course to take,
3939 unless you must also use Sun's C compiler. Unfortunately 107058-01
3940 is preinstalled on some new Solaris 7-based hosts, so you may have to
3944 Copy the original, unpatched Solaris 7
3945 @command{/usr/ccs/bin/as} into
3946 @command{/usr/local/libexec/gcc/sparc-sun-solaris2.7/3.4/as},
3947 adjusting the latter name to fit your local conventions and software
3951 Install Sun patch 106950-03 (1999-05-25) or later. Nobody with
3952 both 107058-01 and 106950-03 installed has reported the bug with GCC
3953 and Sun's dynamic linker. This last course of action is riskiest,
3954 for two reasons. First, you must install 106950 on all hosts that
3955 run code generated by GCC; it doesn't suffice to install it only on
3956 the hosts that run GCC itself. Second, Sun says that 106950-03 is
3957 only a partial fix for bug 4210064, but Sun doesn't know whether the
3958 partial fix is adequate for GCC@. Revision -08 or later should fix
3959 the bug. The current (as of 2004-05-23) revision is -24, and is included in
3960 the Solaris 7 Recommended Patch Cluster.
3963 GCC 3.3 triggers a bug in version 5.0 Alpha 03/27/98 of the Sun assembler,
3964 which causes a bootstrap failure when linking the 64-bit shared version of
3965 libgcc. A typical error message is:
3968 ld: fatal: relocation error: R_SPARC_32: file libgcc/sparcv9/_muldi3.o:
3969 symbol <unknown>: offset 0xffffffff7ec133e7 is non-aligned.
3972 This bug has been fixed in the final 5.0 version of the assembler.
3974 A similar problem was reported for version Sun WorkShop 6 99/08/18 of the
3975 Sun assembler, which causes a bootstrap failure with GCC 4.0.0:
3978 ld: fatal: relocation error: R_SPARC_DISP32:
3979 file .libs/libstdc++.lax/libsupc++convenience.a/vterminate.o:
3980 symbol <unknown>: offset 0xfccd33ad is non-aligned
3983 This bug has been fixed in more recent revisions of the assembler.
3988 @heading @anchor{sparc-x-linux}sparc-*-linux*
3990 GCC versions 3.0 and higher require binutils 2.11.2 and glibc 2.2.4
3991 or newer on this platform. All earlier binutils and glibc
3992 releases mishandled unaligned relocations on @code{sparc-*-*} targets.
3998 @heading @anchor{sparc64-x-solaris2}sparc64-*-solaris2*
4000 When configuring the GNU Multiple Precision Library (GMP) or the
4001 MPFR library, the canonical target triplet must be specified as
4002 the @command{build} parameter on the configure line. For example
4003 on a Solaris 7 system:
4006 % ./configure --build=sparc64-sun-solaris2.7 --prefix=xxx
4009 The following compiler flags must be specified in the configure
4010 step in order to bootstrap this target with the Sun compiler:
4013 % CC="cc -xarch=v9 -xildoff" @var{srcdir}/configure [@var{options}] [@var{target}]
4016 @option{-xarch=v9} specifies the SPARC-V9 architecture to the Sun toolchain
4017 and @option{-xildoff} turns off the incremental linker.
4022 @heading @anchor{sparcv9-x-solaris2}sparcv9-*-solaris2*
4024 This is a synonym for sparc64-*-solaris2*.
4029 @heading @anchor{x-x-sysv}*-*-sysv*
4030 On System V release 3, you may get this error message
4034 ld fatal: failed to write symbol name @var{something}
4035 in strings table for file @var{whatever}
4038 This probably indicates that the disk is full or your ulimit won't allow
4039 the file to be as large as it needs to be.
4041 This problem can also result because the kernel parameter @code{MAXUMEM}
4042 is too small. If so, you must regenerate the kernel and make the value
4043 much larger. The default value is reported to be 1024; a value of 32768
4044 is said to work. Smaller values may also work.
4046 On System V, if you get an error like this,
4049 /usr/local/lib/bison.simple: In function `yyparse':
4050 /usr/local/lib/bison.simple:625: virtual memory exhausted
4054 that too indicates a problem with disk space, ulimit, or @code{MAXUMEM}.
4056 On a System V release 4 system, make sure @file{/usr/bin} precedes
4057 @file{/usr/ucb} in @code{PATH}. The @command{cc} command in
4058 @file{/usr/ucb} uses libraries which have bugs.
4063 @heading @anchor{vax-dec-ultrix}vax-dec-ultrix
4064 Don't try compiling with VAX C (@command{vcc}). It produces incorrect code
4065 in some cases (for example, when @code{alloca} is used).
4070 @heading @anchor{x-x-vxworks}*-*-vxworks*
4071 Support for VxWorks is in flux. At present GCC supports @emph{only} the
4072 very recent VxWorks 5.5 (aka Tornado 2.2) release, and only on PowerPC@.
4073 We welcome patches for other architectures supported by VxWorks 5.5.
4074 Support for VxWorks AE would also be welcome; we believe this is merely
4075 a matter of writing an appropriate ``configlette'' (see below). We are
4076 not interested in supporting older, a.out or COFF-based, versions of
4079 VxWorks comes with an older version of GCC installed in
4080 @file{@var{$WIND_BASE}/host}; we recommend you do not overwrite it.
4081 Choose an installation @var{prefix} entirely outside @var{$WIND_BASE}.
4082 Before running @command{configure}, create the directories @file{@var{prefix}}
4083 and @file{@var{prefix}/bin}. Link or copy the appropriate assembler,
4084 linker, etc.@: into @file{@var{prefix}/bin}, and set your @var{PATH} to
4085 include that directory while running both @command{configure} and
4088 You must give @command{configure} the
4089 @option{--with-headers=@var{$WIND_BASE}/target/h} switch so that it can
4090 find the VxWorks system headers. Since VxWorks is a cross compilation
4091 target only, you must also specify @option{--target=@var{target}}.
4092 @command{configure} will attempt to create the directory
4093 @file{@var{prefix}/@var{target}/sys-include} and copy files into it;
4094 make sure the user running @command{configure} has sufficient privilege
4097 GCC's exception handling runtime requires a special ``configlette''
4098 module, @file{contrib/gthr_supp_vxw_5x.c}. Follow the instructions in
4099 that file to add the module to your kernel build. (Future versions of
4100 VxWorks will incorporate this module.)
4105 @heading @anchor{x86-64-x-x}x86_64-*-*, amd64-*-*
4107 GCC supports the x86-64 architecture implemented by the AMD64 processor
4108 (amd64-*-* is an alias for x86_64-*-*) on GNU/Linux, FreeBSD and NetBSD@.
4109 On GNU/Linux the default is a bi-arch compiler which is able to generate
4110 both 64-bit x86-64 and 32-bit x86 code (via the @option{-m32} switch).
4115 @heading @anchor{xtensa-x-elf}xtensa-*-elf
4117 This target is intended for embedded Xtensa systems using the
4118 @samp{newlib} C library. It uses ELF but does not support shared
4119 objects. Designed-defined instructions specified via the
4120 Tensilica Instruction Extension (TIE) language are only supported
4121 through inline assembly.
4123 The Xtensa configuration information must be specified prior to
4124 building GCC@. The @file{include/xtensa-config.h} header
4125 file contains the configuration information. If you created your
4126 own Xtensa configuration with the Xtensa Processor Generator, the
4127 downloaded files include a customized copy of this header file,
4128 which you can use to replace the default header file.
4133 @heading @anchor{xtensa-x-linux}xtensa-*-linux*
4135 This target is for Xtensa systems running GNU/Linux. It supports ELF
4136 shared objects and the GNU C library (glibc). It also generates
4137 position-independent code (PIC) regardless of whether the
4138 @option{-fpic} or @option{-fPIC} options are used. In other
4139 respects, this target is the same as the
4140 @uref{#xtensa-*-elf,,@samp{xtensa-*-elf}} target.
4145 @heading @anchor{windows}Microsoft Windows (32-bit)
4147 Ports of GCC are included with the
4148 @uref{http://www.cygwin.com/,,Cygwin environment}.
4150 GCC will build under Cygwin without modification; it does not build
4151 with Microsoft's C++ compiler and there are no plans to make it do so.
4156 @heading @anchor{os2}OS/2
4158 GCC does not currently support OS/2. However, Andrew Zabolotny has been
4159 working on a generic OS/2 port with pgcc. The current code can be found
4160 at @uref{http://www.goof.com/pcg/os2/,,http://www.goof.com/pcg/os2/}.
4165 @heading @anchor{older}Older systems
4167 GCC contains support files for many older (1980s and early
4168 1990s) Unix variants. For the most part, support for these systems
4169 has not been deliberately removed, but it has not been maintained for
4170 several years and may suffer from bitrot.
4172 Starting with GCC 3.1, each release has a list of ``obsoleted'' systems.
4173 Support for these systems is still present in that release, but
4174 @command{configure} will fail unless the @option{--enable-obsolete}
4175 option is given. Unless a maintainer steps forward, support for these
4176 systems will be removed from the next release of GCC@.
4178 Support for old systems as hosts for GCC can cause problems if the
4179 workarounds for compiler, library and operating system bugs affect the
4180 cleanliness or maintainability of the rest of GCC@. In some cases, to
4181 bring GCC up on such a system, if still possible with current GCC, may
4182 require first installing an old version of GCC which did work on that
4183 system, and using it to compile a more recent GCC, to avoid bugs in the
4184 vendor compiler. Old releases of GCC 1 and GCC 2 are available in the
4185 @file{old-releases} directory on the @uref{../mirrors.html,,GCC mirror
4186 sites}. Header bugs may generally be avoided using
4187 @command{fixincludes}, but bugs or deficiencies in libraries and the
4188 operating system may still cause problems.
4190 Support for older systems as targets for cross-compilation is less
4191 problematic than support for them as hosts for GCC; if an enthusiast
4192 wishes to make such a target work again (including resurrecting any of
4193 the targets that never worked with GCC 2, starting from the last
4194 version before they were removed), patches
4195 @uref{../contribute.html,,following the usual requirements} would be
4196 likely to be accepted, since they should not affect the support for more
4199 For some systems, old versions of GNU binutils may also be useful,
4200 and are available from @file{pub/binutils/old-releases} on
4201 @uref{http://sourceware.org/mirrors.html,,sourceware.org mirror sites}.
4203 Some of the information on specific systems above relates to
4204 such older systems, but much of the information
4205 about GCC on such systems (which may no longer be applicable to
4206 current GCC) is to be found in the GCC texinfo manual.
4211 @heading @anchor{elf}all ELF targets (SVR4, Solaris 2, etc.)
4213 C++ support is significantly better on ELF targets if you use the
4214 @uref{./configure.html#with-gnu-ld,,GNU linker}; duplicate copies of
4215 inlines, vtables and template instantiations will be discarded
4224 @uref{./index.html,,Return to the GCC Installation page}
4228 @c ***Old documentation******************************************************
4230 @include install-old.texi
4236 @uref{./index.html,,Return to the GCC Installation page}
4240 @c ***GFDL********************************************************************
4248 @uref{./index.html,,Return to the GCC Installation page}
4252 @c ***************************************************************************
4253 @c Part 6 The End of the Document
4255 @comment node-name, next, previous, up
4256 @node Concept Index, , GNU Free Documentation License, Top
4260 @unnumbered Concept Index