1 \input texinfo.tex @c -*-texinfo-*-
4 @setfilename gccinstall.info
5 @settitle Installing GCC
10 @include gcc-common.texi
12 @c Specify title for specific html page
14 @settitle Installing GCC
17 @settitle Host/Target specific installation notes for GCC
19 @ifset prerequisiteshtml
20 @settitle Prerequisites for GCC
23 @settitle Downloading GCC
26 @settitle Installing GCC: Configuration
29 @settitle Installing GCC: Building
32 @settitle Installing GCC: Testing
34 @ifset finalinstallhtml
35 @settitle Installing GCC: Final installation
38 @settitle Installing GCC: Binaries
41 @settitle Installing GCC: Old documentation
44 @settitle Installing GCC: GNU Free Documentation License
47 @c Copyright (C) 1988-2016 Free Software Foundation, Inc.
48 @c *** Converted to texinfo by Dean Wakerley, dean@wakerley.com
50 @c IMPORTANT: whenever you modify this file, run `install.texi2html' to
51 @c test the generation of HTML documents for the gcc.gnu.org web pages.
53 @c Do not use @footnote{} in this file as it breaks install.texi2html!
55 @c Include everything if we're not making html
59 @set prerequisiteshtml
70 @c Part 2 Summary Description and Copyright
72 Copyright @copyright{} 1988-2016 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 1.3 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
102 @title Installing GCC
105 @c The following two commands start the copyright page.
107 @vskip 0pt plus 1filll
111 @c Part 4 Top node, Master Menu, and/or Table of Contents
114 @comment node-name, next, Previous, up
117 * Installing GCC:: This document describes the generic installation
118 procedure for GCC as well as detailing some target
119 specific installation instructions.
121 * Specific:: Host/target specific installation notes for GCC.
122 * Binaries:: Where to get pre-compiled binaries.
124 * Old:: Old installation documentation.
126 * GNU Free Documentation License:: How you can copy and share this manual.
127 * Concept Index:: This index has two entries.
135 @c Part 5 The Body of the Document
136 @c ***Installing GCC**********************************************************
138 @comment node-name, next, previous, up
139 @node Installing GCC, Binaries, , Top
143 @chapter Installing GCC
146 The latest version of this document is always available at
147 @uref{http://gcc.gnu.org/install/,,http://gcc.gnu.org/install/}.
148 It refers to the current development sources, instructions for
149 specific released versions are included with the sources.
151 This document describes the generic installation procedure for GCC as well
152 as detailing some target specific installation instructions.
154 GCC includes several components that previously were separate distributions
155 with their own installation instructions. This document supersedes all
156 package-specific installation instructions.
158 @emph{Before} starting the build/install procedure please check the
160 @ref{Specific, host/target specific installation notes}.
163 @uref{specific.html,,host/target specific installation notes}.
165 We recommend you browse the entire generic installation instructions before
168 Lists of successful builds for released versions of GCC are
169 available at @uref{http://gcc.gnu.org/buildstat.html}.
170 These lists are updated as new information becomes available.
172 The installation procedure itself is broken into five steps.
177 * Downloading the source::
180 * Testing:: (optional)
187 @uref{prerequisites.html,,Prerequisites}
189 @uref{download.html,,Downloading the source}
191 @uref{configure.html,,Configuration}
193 @uref{build.html,,Building}
195 @uref{test.html,,Testing} (optional)
197 @uref{finalinstall.html,,Final install}
201 Please note that GCC does not support @samp{make uninstall} and probably
202 won't do so in the near future as this would open a can of worms. Instead,
203 we suggest that you install GCC into a directory of its own and simply
204 remove that directory when you do not need that specific version of GCC
205 any longer, and, if shared libraries are installed there as well, no
206 more binaries exist that use them.
209 There are also some @uref{old.html,,old installation instructions},
210 which are mostly obsolete but still contain some information which has
211 not yet been merged into the main part of this manual.
219 @uref{./index.html,,Return to the GCC Installation page}
225 @c ***Prerequisites**************************************************
227 @comment node-name, next, previous, up
228 @node Prerequisites, Downloading the source, , Installing GCC
230 @ifset prerequisiteshtml
232 @chapter Prerequisites
234 @cindex Prerequisites
236 GCC requires that various tools and packages be available for use in the
237 build procedure. Modifying GCC sources requires additional tools
240 @heading Tools/packages necessary for building GCC
242 @item ISO C++98 compiler
243 Necessary to bootstrap GCC, although versions of GCC prior
244 to 4.8 also allow bootstrapping with a ISO C89 compiler and versions
245 of GCC prior to 3.4 also allow bootstrapping with a traditional
248 To build all languages in a cross-compiler or other configuration where
249 3-stage bootstrap is not performed, you need to start with an existing
250 GCC binary (version 3.4 or later) because source code for language
251 frontends other than C might use GCC extensions.
253 Note that to bootstrap GCC with versions of GCC earlier than 3.4, you
254 may need to use @option{--disable-stage1-checking}, though
255 bootstrapping the compiler with such earlier compilers is strongly
258 @item C standard library and headers
260 In order to build GCC, the C standard library and headers must be present
261 for all target variants for which target libraries will be built (and not
262 only the variant of the host C++ compiler).
264 This affects the popular @samp{x86_64-unknown-linux-gnu} platform (among
265 other multilib targets), for which 64-bit (@samp{x86_64}) and 32-bit
266 (@samp{i386}) libc headers are usually packaged separately. If you do a
267 build of a native compiler on @samp{x86_64-unknown-linux-gnu}, make sure you
268 either have the 32-bit libc developer package properly installed (the exact
269 name of the package depends on your distro) or you must build GCC as a
270 64-bit only compiler by configuring with the option
271 @option{--disable-multilib}. Otherwise, you may encounter an error such as
272 @samp{fatal error: gnu/stubs-32.h: No such file}
276 In order to build the Ada compiler (GNAT) you must already have GNAT
277 installed because portions of the Ada frontend are written in Ada (with
278 GNAT extensions.) Refer to the Ada installation instructions for more
279 specific information.
281 @item A ``working'' POSIX compatible shell, or GNU bash
283 Necessary when running @command{configure} because some
284 @command{/bin/sh} shells have bugs and may crash when configuring the
285 target libraries. In other cases, @command{/bin/sh} or @command{ksh}
286 have disastrous corner-case performance problems. This
287 can cause target @command{configure} runs to literally take days to
288 complete in some cases.
290 So on some platforms @command{/bin/ksh} is sufficient, on others it
291 isn't. See the host/target specific instructions for your platform, or
292 use @command{bash} to be sure. Then set @env{CONFIG_SHELL} in your
293 environment to your ``good'' shell prior to running
294 @command{configure}/@command{make}.
296 @command{zsh} is not a fully compliant POSIX shell and will not
297 work when configuring GCC@.
299 @item A POSIX or SVR4 awk
301 Necessary for creating some of the generated source files for GCC@.
302 If in doubt, use a recent GNU awk version, as some of the older ones
303 are broken. GNU awk version 3.1.5 is known to work.
307 Necessary in some circumstances, optional in others. See the
308 host/target specific instructions for your platform for the exact
311 @item gzip version 1.2.4 (or later) or
312 @itemx bzip2 version 1.0.2 (or later)
314 Necessary to uncompress GCC @command{tar} files when source code is
315 obtained via FTP mirror sites.
317 @item GNU make version 3.80 (or later)
319 You must have GNU make installed to build GCC@.
321 @item GNU tar version 1.14 (or later)
323 Necessary (only on some platforms) to untar the source code. Many
324 systems' @command{tar} programs will also work, only try GNU
325 @command{tar} if you have problems.
327 @item Perl version 5.6.1 (or later)
329 Necessary when targeting Darwin, building @samp{libstdc++},
330 and not using @option{--disable-symvers}.
331 Necessary when targeting Solaris 2 with Sun @command{ld} and not using
332 @option{--disable-symvers}. The bundled @command{perl} in Solaris@tie{}8
335 Necessary when regenerating @file{Makefile} dependencies in libiberty.
336 Necessary when regenerating @file{libiberty/functions.texi}.
337 Necessary when generating manpages from Texinfo manuals.
338 Used by various scripts to generate some files included in SVN (mainly
339 Unicode-related and rarely changing) from source tables.
341 @item @command{jar}, or InfoZIP (@command{zip} and @command{unzip})
343 Necessary to build libgcj, the GCJ runtime.
347 Several support libraries are necessary to build GCC, some are required,
348 others optional. While any sufficiently new version of required tools
349 usually work, library requirements are generally stricter. Newer
350 versions may work in some cases, but it's safer to use the exact
351 versions documented. We appreciate bug reports about problems with
352 newer versions, though. If your OS vendor provides packages for the
353 support libraries then using those packages may be the simplest way to
354 install the libraries.
357 @item GNU Multiple Precision Library (GMP) version 4.3.2 (or later)
359 Necessary to build GCC@. If a GMP source distribution is found in a
360 subdirectory of your GCC sources named @file{gmp}, it will be built
361 together with GCC. Alternatively, if GMP is already installed but it
362 is not in your library search path, you will have to configure with the
363 @option{--with-gmp} configure option. See also @option{--with-gmp-lib}
364 and @option{--with-gmp-include}.
365 The in-tree build is only supported with the GMP version that
366 download_prerequisites installs.
368 @item MPFR Library version 2.4.2 (or later)
370 Necessary to build GCC@. It can be downloaded from
371 @uref{http://www.mpfr.org/}. If an MPFR source distribution is found
372 in a subdirectory of your GCC sources named @file{mpfr}, it will be
373 built together with GCC. Alternatively, if MPFR is already installed
374 but it is not in your default library search path, the
375 @option{--with-mpfr} configure option should be used. See also
376 @option{--with-mpfr-lib} and @option{--with-mpfr-include}.
377 The in-tree build is only supported with the MPFR version that
378 download_prerequisites installs.
380 @item MPC Library version 0.8.1 (or later)
382 Necessary to build GCC@. It can be downloaded from
383 @uref{http://www.multiprecision.org/}. If an MPC source distribution
384 is found in a subdirectory of your GCC sources named @file{mpc}, it
385 will be built together with GCC. Alternatively, if MPC is already
386 installed but it is not in your default library search path, the
387 @option{--with-mpc} configure option should be used. See also
388 @option{--with-mpc-lib} and @option{--with-mpc-include}.
389 The in-tree build is only supported with the MPC version that
390 download_prerequisites installs.
392 @item isl Library version 0.16, 0.15, or 0.14.
394 Necessary to build GCC with the Graphite loop optimizations.
395 It can be downloaded from @uref{ftp://gcc.gnu.org/pub/gcc/infrastructure/}.
396 If an isl source distribution is found
397 in a subdirectory of your GCC sources named @file{isl}, it will be
398 built together with GCC. Alternatively, the @option{--with-isl} configure
399 option should be used if isl is not installed in your default library
404 @heading Tools/packages necessary for modifying GCC
406 @item autoconf version 2.64
407 @itemx GNU m4 version 1.4.6 (or later)
409 Necessary when modifying @file{configure.ac}, @file{aclocal.m4}, etc.@:
410 to regenerate @file{configure} and @file{config.in} files.
412 @item automake version 1.11.6
414 Necessary when modifying a @file{Makefile.am} file to regenerate its
415 associated @file{Makefile.in}.
417 Much of GCC does not use automake, so directly edit the @file{Makefile.in}
418 file. Specifically this applies to the @file{gcc}, @file{intl},
419 @file{libcpp}, @file{libiberty}, @file{libobjc} directories as well
420 as any of their subdirectories.
422 For directories that use automake, GCC requires the latest release in
423 the 1.11 series, which is currently 1.11.6. When regenerating a directory
424 to a newer version, please update all the directories using an older 1.11
425 to the latest released version.
427 @item gettext version 0.14.5 (or later)
429 Needed to regenerate @file{gcc.pot}.
431 @item gperf version 2.7.2 (or later)
433 Necessary when modifying @command{gperf} input files, e.g.@:
434 @file{gcc/cp/cfns.gperf} to regenerate its associated header file, e.g.@:
435 @file{gcc/cp/cfns.h}.
441 Necessary to run the GCC testsuite; see the section on testing for
442 details. Tcl 8.6 has a known regression in RE pattern handling that
443 make parts of the testsuite fail. See
444 @uref{http://core.tcl.tk/tcl/tktview/267b7e2334ee2e9de34c4b00d6e72e2f1997085f}
445 for more information. This bug has been fixed in 8.6.1.
447 @item autogen version 5.5.4 (or later) and
448 @itemx guile version 1.4.1 (or later)
450 Necessary to regenerate @file{fixinc/fixincl.x} from
451 @file{fixinc/inclhack.def} and @file{fixinc/*.tpl}.
453 Necessary to run @samp{make check} for @file{fixinc}.
455 Necessary to regenerate the top level @file{Makefile.in} file from
456 @file{Makefile.tpl} and @file{Makefile.def}.
458 @item Flex version 2.5.4 (or later)
460 Necessary when modifying @file{*.l} files.
462 Necessary to build GCC during development because the generated output
463 files are not included in the SVN repository. They are included in
466 @item Texinfo version 4.7 (or later)
468 Necessary for running @command{makeinfo} when modifying @file{*.texi}
469 files to test your changes.
471 Necessary for running @command{make dvi} or @command{make pdf} to
472 create printable documentation in DVI or PDF format. Texinfo version
473 4.8 or later is required for @command{make pdf}.
475 Necessary to build GCC documentation during development because the
476 generated output files are not included in the SVN repository. They are
477 included in releases.
479 @item @TeX{} (any working version)
481 Necessary for running @command{texi2dvi} and @command{texi2pdf}, which
482 are used when running @command{make dvi} or @command{make pdf} to create
483 DVI or PDF files, respectively.
485 @item Sphinx version 1.0 (or later)
487 Necessary to regenerate @file{jit/docs/_build/texinfo} from the @file{.rst}
488 files in the directories below @file{jit/docs}.
490 @item SVN (any version)
491 @itemx SSH (any version)
493 Necessary to access the SVN repository. Public releases and weekly
494 snapshots of the development sources are also available via FTP@.
496 @item GNU diffutils version 2.7 (or later)
498 Useful when submitting patches for the GCC source code.
500 @item patch version 2.5.4 (or later)
502 Necessary when applying patches, created with @command{diff}, to one's
508 If you wish to modify @file{.java} files in libjava, you will need to
509 configure with @option{--enable-java-maintainer-mode}, and you will need
510 to have executables named @command{ecj1} and @command{gjavah} in your path.
511 The @command{ecj1} executable should run the Eclipse Java compiler via
512 the GCC-specific entry point. You can download a suitable jar from
513 @uref{ftp://sourceware.org/pub/java/}, or by running the script
514 @command{contrib/download_ecj}.
516 @item antlr.jar version 2.7.1 (or later)
519 If you wish to build the @command{gjdoc} binary in libjava, you will
520 need to have an @file{antlr.jar} library available. The library is
521 searched for in system locations but can be specified with
522 @option{--with-antlr-jar=} instead. When configuring with
523 @option{--enable-java-maintainer-mode}, you will need to have one of
524 the executables named @command{cantlr}, @command{runantlr} or
525 @command{antlr} in your path.
534 @uref{./index.html,,Return to the GCC Installation page}
538 @c ***Downloading the source**************************************************
540 @comment node-name, next, previous, up
541 @node Downloading the source, Configuration, Prerequisites, Installing GCC
545 @chapter Downloading GCC
547 @cindex Downloading GCC
548 @cindex Downloading the Source
550 GCC is distributed via @uref{http://gcc.gnu.org/svn.html,,SVN} and FTP
551 tarballs compressed with @command{gzip} or
554 Please refer to the @uref{http://gcc.gnu.org/releases.html,,releases web page}
555 for information on how to obtain GCC@.
557 The source distribution includes the C, C++, Objective-C, Fortran, Java,
558 and Ada (in the case of GCC 3.1 and later) compilers, as well as
559 runtime libraries for C++, Objective-C, Fortran, and Java.
560 For previous versions these were downloadable as separate components such
561 as the core GCC distribution, which included the C language front end and
562 shared components, and language-specific distributions including the
563 language front end and the language runtime (where appropriate).
565 If you also intend to build binutils (either to upgrade an existing
566 installation or for use in place of the corresponding tools of your
567 OS), unpack the binutils distribution either in the same directory or
568 a separate one. In the latter case, add symbolic links to any
569 components of the binutils you intend to build alongside the compiler
570 (@file{bfd}, @file{binutils}, @file{gas}, @file{gprof}, @file{ld},
571 @file{opcodes}, @dots{}) to the directory containing the GCC sources.
573 Likewise the GMP, MPFR and MPC libraries can be automatically built
574 together with GCC. You may simply run the
575 @command{contrib/download_prerequisites} script in the GCC source directory
576 to set up everything.
577 Otherwise unpack the GMP, MPFR and/or MPC source
578 distributions in the directory containing the GCC sources and rename
579 their directories to @file{gmp}, @file{mpfr} and @file{mpc},
580 respectively (or use symbolic links with the same name).
587 @uref{./index.html,,Return to the GCC Installation page}
591 @c ***Configuration***********************************************************
593 @comment node-name, next, previous, up
594 @node Configuration, Building, Downloading the source, Installing GCC
598 @chapter Installing GCC: Configuration
600 @cindex Configuration
601 @cindex Installing GCC: Configuration
603 Like most GNU software, GCC must be configured before it can be built.
604 This document describes the recommended configuration procedure
605 for both native and cross targets.
607 We use @var{srcdir} to refer to the toplevel source directory for
608 GCC; we use @var{objdir} to refer to the toplevel build/object directory.
610 If you obtained the sources via SVN, @var{srcdir} must refer to the top
611 @file{gcc} directory, the one where the @file{MAINTAINERS} file can be
612 found, and not its @file{gcc} subdirectory, otherwise the build will fail.
614 If either @var{srcdir} or @var{objdir} is located on an automounted NFS
615 file system, the shell's built-in @command{pwd} command will return
616 temporary pathnames. Using these can lead to various sorts of build
617 problems. To avoid this issue, set the @env{PWDCMD} environment
618 variable to an automounter-aware @command{pwd} command, e.g.,
619 @command{pawd} or @samp{amq -w}, during the configuration and build
622 First, we @strong{highly} recommend that GCC be built into a
623 separate directory from the sources which does @strong{not} reside
624 within the source tree. This is how we generally build GCC; building
625 where @var{srcdir} == @var{objdir} should still work, but doesn't
626 get extensive testing; building where @var{objdir} is a subdirectory
627 of @var{srcdir} is unsupported.
629 If you have previously built GCC in the same directory for a
630 different target machine, do @samp{make distclean} to delete all files
631 that might be invalid. One of the files this deletes is @file{Makefile};
632 if @samp{make distclean} complains that @file{Makefile} does not exist
633 or issues a message like ``don't know how to make distclean'' it probably
634 means that the directory is already suitably clean. However, with the
635 recommended method of building in a separate @var{objdir}, you should
636 simply use a different @var{objdir} for each target.
638 Second, when configuring a native system, either @command{cc} or
639 @command{gcc} must be in your path or you must set @env{CC} in
640 your environment before running configure. Otherwise the configuration
644 Note that the bootstrap compiler and the resulting GCC must be link
645 compatible, else the bootstrap will fail with linker errors about
646 incompatible object file formats. Several multilibed targets are
647 affected by this requirement, see
649 @ref{Specific, host/target specific installation notes}.
652 @uref{specific.html,,host/target specific installation notes}.
661 % @var{srcdir}/configure [@var{options}] [@var{target}]
664 @heading Distributor options
666 If you will be distributing binary versions of GCC, with modifications
667 to the source code, you should use the options described in this
668 section to make clear that your version contains modifications.
671 @item --with-pkgversion=@var{version}
672 Specify a string that identifies your package. You may wish
673 to include a build number or build date. This version string will be
674 included in the output of @command{gcc --version}. This suffix does
675 not replace the default version string, only the @samp{GCC} part.
677 The default value is @samp{GCC}.
679 @item --with-bugurl=@var{url}
680 Specify the URL that users should visit if they wish to report a bug.
681 You are of course welcome to forward bugs reported to you to the FSF,
682 if you determine that they are not bugs in your modifications.
684 The default value refers to the FSF's GCC bug tracker.
688 @heading Target specification
691 GCC has code to correctly determine the correct value for @var{target}
692 for nearly all native systems. Therefore, we highly recommend you do
693 not provide a configure target when configuring a native compiler.
696 @var{target} must be specified as @option{--target=@var{target}}
697 when configuring a cross compiler; examples of valid targets would be
698 m68k-elf, sh-elf, etc.
701 Specifying just @var{target} instead of @option{--target=@var{target}}
702 implies that the host defaults to @var{target}.
706 @heading Options specification
708 Use @var{options} to override several configure time options for
709 GCC@. A list of supported @var{options} follows; @samp{configure
710 --help} may list other options, but those not listed below may not
711 work and should not normally be used.
713 Note that each @option{--enable} option has a corresponding
714 @option{--disable} option and that each @option{--with} option has a
715 corresponding @option{--without} option.
718 @item --prefix=@var{dirname}
719 Specify the toplevel installation
720 directory. This is the recommended way to install the tools into a directory
721 other than the default. The toplevel installation directory defaults to
724 We @strong{highly} recommend against @var{dirname} being the same or a
725 subdirectory of @var{objdir} or vice versa. If specifying a directory
726 beneath a user's home directory tree, some shells will not expand
727 @var{dirname} correctly if it contains the @samp{~} metacharacter; use
730 The following standard @command{autoconf} options are supported. Normally you
731 should not need to use these options.
733 @item --exec-prefix=@var{dirname}
734 Specify the toplevel installation directory for architecture-dependent
735 files. The default is @file{@var{prefix}}.
737 @item --bindir=@var{dirname}
738 Specify the installation directory for the executables called by users
739 (such as @command{gcc} and @command{g++}). The default is
740 @file{@var{exec-prefix}/bin}.
742 @item --libdir=@var{dirname}
743 Specify the installation directory for object code libraries and
744 internal data files of GCC@. The default is @file{@var{exec-prefix}/lib}.
746 @item --libexecdir=@var{dirname}
747 Specify the installation directory for internal executables of GCC@.
748 The default is @file{@var{exec-prefix}/libexec}.
750 @item --with-slibdir=@var{dirname}
751 Specify the installation directory for the shared libgcc library. The
752 default is @file{@var{libdir}}.
754 @item --datarootdir=@var{dirname}
755 Specify the root of the directory tree for read-only architecture-independent
756 data files referenced by GCC@. The default is @file{@var{prefix}/share}.
758 @item --infodir=@var{dirname}
759 Specify the installation directory for documentation in info format.
760 The default is @file{@var{datarootdir}/info}.
762 @item --datadir=@var{dirname}
763 Specify the installation directory for some architecture-independent
764 data files referenced by GCC@. The default is @file{@var{datarootdir}}.
766 @item --docdir=@var{dirname}
767 Specify the installation directory for documentation files (other
768 than Info) for GCC@. The default is @file{@var{datarootdir}/doc}.
770 @item --htmldir=@var{dirname}
771 Specify the installation directory for HTML documentation files.
772 The default is @file{@var{docdir}}.
774 @item --pdfdir=@var{dirname}
775 Specify the installation directory for PDF documentation files.
776 The default is @file{@var{docdir}}.
778 @item --mandir=@var{dirname}
779 Specify the installation directory for manual pages. The default is
780 @file{@var{datarootdir}/man}. (Note that the manual pages are only extracts
781 from the full GCC manuals, which are provided in Texinfo format. The manpages
782 are derived by an automatic conversion process from parts of the full
785 @item --with-gxx-include-dir=@var{dirname}
787 the installation directory for G++ header files. The default depends
788 on other configuration options, and differs between cross and native
791 @item --with-specs=@var{specs}
792 Specify additional command line driver SPECS.
793 This can be useful if you need to turn on a non-standard feature by
794 default without modifying the compiler's source code, for instance
795 @option{--with-specs=%@{!fcommon:%@{!fno-common:-fno-common@}@}}.
797 @xref{Spec Files,, Specifying subprocesses and the switches to pass to them,
798 gcc, Using the GNU Compiler Collection (GCC)},
801 See ``Spec Files'' in the main manual
806 @item --program-prefix=@var{prefix}
807 GCC supports some transformations of the names of its programs when
808 installing them. This option prepends @var{prefix} to the names of
809 programs to install in @var{bindir} (see above). For example, specifying
810 @option{--program-prefix=foo-} would result in @samp{gcc}
811 being installed as @file{/usr/local/bin/foo-gcc}.
813 @item --program-suffix=@var{suffix}
814 Appends @var{suffix} to the names of programs to install in @var{bindir}
815 (see above). For example, specifying @option{--program-suffix=-3.1}
816 would result in @samp{gcc} being installed as
817 @file{/usr/local/bin/gcc-3.1}.
819 @item --program-transform-name=@var{pattern}
820 Applies the @samp{sed} script @var{pattern} to be applied to the names
821 of programs to install in @var{bindir} (see above). @var{pattern} has to
822 consist of one or more basic @samp{sed} editing commands, separated by
823 semicolons. For example, if you want the @samp{gcc} program name to be
824 transformed to the installed program @file{/usr/local/bin/myowngcc} and
825 the @samp{g++} program name to be transformed to
826 @file{/usr/local/bin/gspecial++} without changing other program names,
827 you could use the pattern
828 @option{--program-transform-name='s/^gcc$/myowngcc/; s/^g++$/gspecial++/'}
829 to achieve this effect.
831 All three options can be combined and used together, resulting in more
832 complex conversion patterns. As a basic rule, @var{prefix} (and
833 @var{suffix}) are prepended (appended) before further transformations
834 can happen with a special transformation script @var{pattern}.
836 As currently implemented, this option only takes effect for native
837 builds; cross compiler binaries' names are not transformed even when a
838 transformation is explicitly asked for by one of these options.
840 For native builds, some of the installed programs are also installed
841 with the target alias in front of their name, as in
842 @samp{i686-pc-linux-gnu-gcc}. All of the above transformations happen
843 before the target alias is prepended to the name---so, specifying
844 @option{--program-prefix=foo-} and @option{program-suffix=-3.1}, the
845 resulting binary would be installed as
846 @file{/usr/local/bin/i686-pc-linux-gnu-foo-gcc-3.1}.
848 As a last shortcoming, none of the installed Ada programs are
849 transformed yet, which will be fixed in some time.
851 @item --with-local-prefix=@var{dirname}
853 installation directory for local include files. The default is
854 @file{/usr/local}. Specify this option if you want the compiler to
855 search directory @file{@var{dirname}/include} for locally installed
856 header files @emph{instead} of @file{/usr/local/include}.
858 You should specify @option{--with-local-prefix} @strong{only} if your
859 site has a different convention (not @file{/usr/local}) for where to put
862 The default value for @option{--with-local-prefix} is @file{/usr/local}
863 regardless of the value of @option{--prefix}. Specifying
864 @option{--prefix} has no effect on which directory GCC searches for
865 local header files. This may seem counterintuitive, but actually it is
868 The purpose of @option{--prefix} is to specify where to @emph{install
869 GCC}. The local header files in @file{/usr/local/include}---if you put
870 any in that directory---are not part of GCC@. They are part of other
871 programs---perhaps many others. (GCC installs its own header files in
872 another directory which is based on the @option{--prefix} value.)
874 Both the local-prefix include directory and the GCC-prefix include
875 directory are part of GCC's ``system include'' directories. Although these
876 two directories are not fixed, they need to be searched in the proper
877 order for the correct processing of the include_next directive. The
878 local-prefix include directory is searched before the GCC-prefix
879 include directory. Another characteristic of system include directories
880 is that pedantic warnings are turned off for headers in these directories.
882 Some autoconf macros add @option{-I @var{directory}} options to the
883 compiler command line, to ensure that directories containing installed
884 packages' headers are searched. When @var{directory} is one of GCC's
885 system include directories, GCC will ignore the option so that system
886 directories continue to be processed in the correct order. This
887 may result in a search order different from what was specified but the
888 directory will still be searched.
890 GCC automatically searches for ordinary libraries using
891 @env{GCC_EXEC_PREFIX}. Thus, when the same installation prefix is
892 used for both GCC and packages, GCC will automatically search for
893 both headers and libraries. This provides a configuration that is
894 easy to use. GCC behaves in a manner similar to that when it is
895 installed as a system compiler in @file{/usr}.
897 Sites that need to install multiple versions of GCC may not want to
898 use the above simple configuration. It is possible to use the
899 @option{--program-prefix}, @option{--program-suffix} and
900 @option{--program-transform-name} options to install multiple versions
901 into a single directory, but it may be simpler to use different prefixes
902 and the @option{--with-local-prefix} option to specify the location of the
903 site-specific files for each version. It will then be necessary for
904 users to specify explicitly the location of local site libraries
905 (e.g., with @env{LIBRARY_PATH}).
907 The same value can be used for both @option{--with-local-prefix} and
908 @option{--prefix} provided it is not @file{/usr}. This can be used
909 to avoid the default search of @file{/usr/local/include}.
911 @strong{Do not} specify @file{/usr} as the @option{--with-local-prefix}!
912 The directory you use for @option{--with-local-prefix} @strong{must not}
913 contain any of the system's standard header files. If it did contain
914 them, certain programs would be miscompiled (including GNU Emacs, on
915 certain targets), because this would override and nullify the header
916 file corrections made by the @command{fixincludes} script.
918 Indications are that people who use this option use it based on mistaken
919 ideas of what it is for. People use it as if it specified where to
920 install part of GCC@. Perhaps they make this assumption because
921 installing GCC creates the directory.
923 @item --with-native-system-header-dir=@var{dirname}
924 Specifies that @var{dirname} is the directory that contains native system
925 header files, rather than @file{/usr/include}. This option is most useful
926 if you are creating a compiler that should be isolated from the system
927 as much as possible. It is most commonly used with the
928 @option{--with-sysroot} option and will cause GCC to search
929 @var{dirname} inside the system root specified by that option.
931 @item --enable-shared[=@var{package}[,@dots{}]]
932 Build shared versions of libraries, if shared libraries are supported on
933 the target platform. Unlike GCC 2.95.x and earlier, shared libraries
934 are enabled by default on all platforms that support shared libraries.
936 If a list of packages is given as an argument, build shared libraries
937 only for the listed packages. For other packages, only static libraries
938 will be built. Package names currently recognized in the GCC tree are
939 @samp{libgcc} (also known as @samp{gcc}), @samp{libstdc++} (not
940 @samp{libstdc++-v3}), @samp{libffi}, @samp{zlib}, @samp{boehm-gc},
941 @samp{ada}, @samp{libada}, @samp{libjava}, @samp{libgo}, and @samp{libobjc}.
942 Note @samp{libiberty} does not support shared libraries at all.
944 Use @option{--disable-shared} to build only static libraries. Note that
945 @option{--disable-shared} does not accept a list of package names as
946 argument, only @option{--enable-shared} does.
948 Contrast with @option{--enable-host-shared}, which affects @emph{host}
951 @item --enable-host-shared
952 Specify that the @emph{host} code should be built into position-independent
953 machine code (with -fPIC), allowing it to be used within shared libraries,
954 but yielding a slightly slower compiler.
956 This option is required when building the libgccjit.so library.
958 Contrast with @option{--enable-shared}, which affects @emph{target}
961 @item @anchor{with-gnu-as}--with-gnu-as
962 Specify that the compiler should assume that the
963 assembler it finds is the GNU assembler. However, this does not modify
964 the rules to find an assembler and will result in confusion if the
965 assembler found is not actually the GNU assembler. (Confusion may also
966 result if the compiler finds the GNU assembler but has not been
967 configured with @option{--with-gnu-as}.) If you have more than one
968 assembler installed on your system, you may want to use this option in
969 connection with @option{--with-as=@var{pathname}} or
970 @option{--with-build-time-tools=@var{pathname}}.
972 The following systems are the only ones where it makes a difference
973 whether you use the GNU assembler. On any other system,
974 @option{--with-gnu-as} has no effect.
977 @item @samp{hppa1.0-@var{any}-@var{any}}
978 @item @samp{hppa1.1-@var{any}-@var{any}}
979 @item @samp{sparc-sun-solaris2.@var{any}}
980 @item @samp{sparc64-@var{any}-solaris2.@var{any}}
983 @item @anchor{with-as}--with-as=@var{pathname}
984 Specify that the compiler should use the assembler pointed to by
985 @var{pathname}, rather than the one found by the standard rules to find
986 an assembler, which are:
989 Unless GCC is being built with a cross compiler, check the
990 @file{@var{libexec}/gcc/@var{target}/@var{version}} directory.
991 @var{libexec} defaults to @file{@var{exec-prefix}/libexec};
992 @var{exec-prefix} defaults to @var{prefix}, which
993 defaults to @file{/usr/local} unless overridden by the
994 @option{--prefix=@var{pathname}} switch described above. @var{target}
995 is the target system triple, such as @samp{sparc-sun-solaris2.7}, and
996 @var{version} denotes the GCC version, such as 3.0.
999 If the target system is the same that you are building on, check
1000 operating system specific directories (e.g.@: @file{/usr/ccs/bin} on
1004 Check in the @env{PATH} for a tool whose name is prefixed by the
1005 target system triple.
1008 Check in the @env{PATH} for a tool whose name is not prefixed by the
1009 target system triple, if the host and target system triple are
1010 the same (in other words, we use a host tool if it can be used for
1011 the target as well).
1014 You may want to use @option{--with-as} if no assembler
1015 is installed in the directories listed above, or if you have multiple
1016 assemblers installed and want to choose one that is not found by the
1019 @item @anchor{with-gnu-ld}--with-gnu-ld
1020 Same as @uref{#with-gnu-as,,@option{--with-gnu-as}}
1023 @item --with-ld=@var{pathname}
1024 Same as @uref{#with-as,,@option{--with-as}}
1028 Specify that stabs debugging
1029 information should be used instead of whatever format the host normally
1030 uses. Normally GCC uses the same debug format as the host system.
1032 On MIPS based systems and on Alphas, you must specify whether you want
1033 GCC to create the normal ECOFF debugging format, or to use BSD-style
1034 stabs passed through the ECOFF symbol table. The normal ECOFF debug
1035 format cannot fully handle languages other than C@. BSD stabs format can
1036 handle other languages, but it only works with the GNU debugger GDB@.
1038 Normally, GCC uses the ECOFF debugging format by default; if you
1039 prefer BSD stabs, specify @option{--with-stabs} when you configure GCC@.
1041 No matter which default you choose when you configure GCC, the user
1042 can use the @option{-gcoff} and @option{-gstabs+} options to specify explicitly
1043 the debug format for a particular compilation.
1045 @option{--with-stabs} is meaningful on the ISC system on the 386, also, if
1046 @option{--with-gas} is used. It selects use of stabs debugging
1047 information embedded in COFF output. This kind of debugging information
1048 supports C++ well; ordinary COFF debugging information does not.
1050 @option{--with-stabs} is also meaningful on 386 systems running SVR4. It
1051 selects use of stabs debugging information embedded in ELF output. The
1052 C++ compiler currently (2.6.0) does not support the DWARF debugging
1053 information normally used on 386 SVR4 platforms; stabs provide a
1054 workable alternative. This requires gas and gdb, as the normal SVR4
1055 tools can not generate or interpret stabs.
1057 @item --with-tls=@var{dialect}
1058 Specify the default TLS dialect, for systems were there is a choice.
1059 For ARM targets, possible values for @var{dialect} are @code{gnu} or
1060 @code{gnu2}, which select between the original GNU dialect and the GNU TLS
1061 descriptor-based dialect.
1063 @item --enable-multiarch
1064 Specify whether to enable or disable multiarch support. The default is
1065 to check for glibc start files in a multiarch location, and enable it
1066 if the files are found. The auto detection is enabled for native builds,
1067 and for cross builds configured with @option{--with-sysroot}, and without
1068 @option{--with-native-system-header-dir}.
1069 More documentation about multiarch can be found at
1070 @uref{https://wiki.debian.org/Multiarch}.
1072 @item --enable-vtable-verify
1073 Specify whether to enable or disable the vtable verification feature.
1074 Enabling this feature causes libstdc++ to be built with its virtual calls
1075 in verifiable mode. This means that, when linked with libvtv, every
1076 virtual call in libstdc++ will verify the vtable pointer through which the
1077 call will be made before actually making the call. If not linked with libvtv,
1078 the verifier will call stub functions (in libstdc++ itself) and do nothing.
1079 If vtable verification is disabled, then libstdc++ is not built with its
1080 virtual calls in verifiable mode at all. However the libvtv library will
1081 still be built (see @option{--disable-libvtv} to turn off building libvtv).
1082 @option{--disable-vtable-verify} is the default.
1084 @item --disable-multilib
1085 Specify that multiple target
1086 libraries to support different target variants, calling
1087 conventions, etc.@: should not be built. The default is to build a
1088 predefined set of them.
1090 Some targets provide finer-grained control over which multilibs are built
1091 (e.g., @option{--disable-softfloat}):
1094 fpu, 26bit, underscore, interwork, biendian, nofmult.
1097 softfloat, m68881, m68000, m68020.
1100 single-float, biendian, softfloat.
1102 @item powerpc*-*-*, rs6000*-*-*
1103 aix64, pthread, softfloat, powercpu, powerpccpu, powerpcos, biendian,
1108 @item --with-multilib-list=@var{list}
1109 @itemx --without-multilib-list
1110 Specify what multilibs to build.
1111 Currently only implemented for arm*-*-*, sh*-*-* and x86-64-*-linux*.
1115 @var{list} is either @code{default} or @code{aprofile}. Specifying
1116 @code{default} is equivalent to omitting this option while specifying
1117 @code{aprofile} builds multilibs for each combination of ISA (@code{-marm} or
1118 @code{-mthumb}), architecture (@code{-march=armv7-a}, @code{-march=armv7ve},
1119 or @code{-march=armv8-a}), FPU available (none, @code{-mfpu=vfpv3-d16},
1120 @code{-mfpu=neon}, @code{-mfpu=vfpv4-d16}, @code{-mfpu=neon-vfpv4} or
1121 @code{-mfpu=neon-fp-armv8} depending on architecture) and floating-point ABI
1122 (@code{-mfloat-abi=softfp} or @code{-mfloat-abi=hard}).
1125 @var{list} is a comma separated list of CPU names. These must be of the
1126 form @code{sh*} or @code{m*} (in which case they match the compiler option
1127 for that processor). The list should not contain any endian options -
1128 these are handled by @option{--with-endian}.
1130 If @var{list} is empty, then there will be no multilibs for extra
1131 processors. The multilib for the secondary endian remains enabled.
1133 As a special case, if an entry in the list starts with a @code{!}
1134 (exclamation point), then it is added to the list of excluded multilibs.
1135 Entries of this sort should be compatible with @samp{MULTILIB_EXCLUDES}
1136 (once the leading @code{!} has been stripped).
1138 If @option{--with-multilib-list} is not given, then a default set of
1139 multilibs is selected based on the value of @option{--target}. This is
1140 usually the complete set of libraries, but some targets imply a more
1143 Example 1: to configure a compiler for SH4A only, but supporting both
1144 endians, with little endian being the default:
1146 --with-cpu=sh4a --with-endian=little,big --with-multilib-list=
1149 Example 2: to configure a compiler for both SH4A and SH4AL-DSP, but with
1150 only little endian SH4AL:
1152 --with-cpu=sh4a --with-endian=little,big \
1153 --with-multilib-list=sh4al,!mb/m4al
1156 @item x86-64-*-linux*
1157 @var{list} is a comma separated list of @code{m32}, @code{m64} and
1158 @code{mx32} to enable 32-bit, 64-bit and x32 run-time libraries,
1159 respectively. If @var{list} is empty, then there will be no multilibs
1160 and only the default run-time library will be enabled.
1162 If @option{--with-multilib-list} is not given, then only 32-bit and
1163 64-bit run-time libraries will be enabled.
1166 @item --with-endian=@var{endians}
1167 Specify what endians to use.
1168 Currently only implemented for sh*-*-*.
1170 @var{endians} may be one of the following:
1173 Use big endian exclusively.
1175 Use little endian exclusively.
1177 Use big endian by default. Provide a multilib for little endian.
1179 Use little endian by default. Provide a multilib for big endian.
1182 @item --enable-threads
1183 Specify that the target
1184 supports threads. This affects the Objective-C compiler and runtime
1185 library, and exception handling for other languages like C++ and Java.
1186 On some systems, this is the default.
1188 In general, the best (and, in many cases, the only known) threading
1189 model available will be configured for use. Beware that on some
1190 systems, GCC has not been taught what threading models are generally
1191 available for the system. In this case, @option{--enable-threads} is an
1192 alias for @option{--enable-threads=single}.
1194 @item --disable-threads
1195 Specify that threading support should be disabled for the system.
1196 This is an alias for @option{--enable-threads=single}.
1198 @item --enable-threads=@var{lib}
1200 @var{lib} is the thread support library. This affects the Objective-C
1201 compiler and runtime library, and exception handling for other languages
1202 like C++ and Java. The possibilities for @var{lib} are:
1210 LynxOS thread support.
1212 MIPS SDE thread support.
1214 This is an alias for @samp{single}.
1216 Generic POSIX/Unix98 thread support.
1218 RTEMS thread support.
1220 Disable thread support, should work for all platforms.
1224 VxWorks thread support.
1226 Microsoft Win32 API thread support.
1230 Specify that the target supports TLS (Thread Local Storage). Usually
1231 configure can correctly determine if TLS is supported. In cases where
1232 it guesses incorrectly, TLS can be explicitly enabled or disabled with
1233 @option{--enable-tls} or @option{--disable-tls}. This can happen if
1234 the assembler supports TLS but the C library does not, or if the
1235 assumptions made by the configure test are incorrect.
1238 Specify that the target does not support TLS.
1239 This is an alias for @option{--enable-tls=no}.
1241 @item --with-cpu=@var{cpu}
1242 @itemx --with-cpu-32=@var{cpu}
1243 @itemx --with-cpu-64=@var{cpu}
1244 Specify which cpu variant the compiler should generate code for by default.
1245 @var{cpu} will be used as the default value of the @option{-mcpu=} switch.
1246 This option is only supported on some targets, including ARC, ARM, i386, M68k,
1247 PowerPC, and SPARC@. It is mandatory for ARC@. The @option{--with-cpu-32} and
1248 @option{--with-cpu-64} options specify separate default CPUs for
1249 32-bit and 64-bit modes; these options are only supported for i386,
1250 x86-64, PowerPC, and SPARC@.
1252 @item --with-schedule=@var{cpu}
1253 @itemx --with-arch=@var{cpu}
1254 @itemx --with-arch-32=@var{cpu}
1255 @itemx --with-arch-64=@var{cpu}
1256 @itemx --with-tune=@var{cpu}
1257 @itemx --with-tune-32=@var{cpu}
1258 @itemx --with-tune-64=@var{cpu}
1259 @itemx --with-abi=@var{abi}
1260 @itemx --with-fpu=@var{type}
1261 @itemx --with-float=@var{type}
1262 These configure options provide default values for the @option{-mschedule=},
1263 @option{-march=}, @option{-mtune=}, @option{-mabi=}, and @option{-mfpu=}
1264 options and for @option{-mhard-float} or @option{-msoft-float}. As with
1265 @option{--with-cpu}, which switches will be accepted and acceptable values
1266 of the arguments depend on the target.
1268 @item --with-mode=@var{mode}
1269 Specify if the compiler should default to @option{-marm} or @option{-mthumb}.
1270 This option is only supported on ARM targets.
1272 @item --with-stack-offset=@var{num}
1273 This option sets the default for the -mstack-offset=@var{num} option,
1274 and will thus generally also control the setting of this option for
1275 libraries. This option is only supported on Epiphany targets.
1277 @item --with-fpmath=@var{isa}
1278 This options sets @option{-mfpmath=sse} by default and specifies the default
1279 ISA for floating-point arithmetics. You can select either @samp{sse} which
1280 enables @option{-msse2} or @samp{avx} which enables @option{-mavx} by default.
1281 This option is only supported on i386 and x86-64 targets.
1283 @item --with-fp-32=@var{mode}
1284 On MIPS targets, set the default value for the @option{-mfp} option when using
1285 the o32 ABI. The possibilities for @var{mode} are:
1288 Use the o32 FP32 ABI extension, as with the @option{-mfp32} command-line
1291 Use the o32 FPXX ABI extension, as with the @option{-mfpxx} command-line
1294 Use the o32 FP64 ABI extension, as with the @option{-mfp64} command-line
1297 In the absence of this configuration option the default is to use the o32
1300 @item --with-odd-spreg-32
1301 On MIPS targets, set the @option{-modd-spreg} option by default when using
1304 @item --without-odd-spreg-32
1305 On MIPS targets, set the @option{-mno-odd-spreg} option by default when using
1306 the o32 ABI. This is normally used in conjunction with
1307 @option{--with-fp-32=64} in order to target the o32 FP64A ABI extension.
1309 @item --with-nan=@var{encoding}
1310 On MIPS targets, set the default encoding convention to use for the
1311 special not-a-number (NaN) IEEE 754 floating-point data. The
1312 possibilities for @var{encoding} are:
1315 Use the legacy encoding, as with the @option{-mnan=legacy} command-line
1318 Use the 754-2008 encoding, as with the @option{-mnan=2008} command-line
1321 To use this configuration option you must have an assembler version
1322 installed that supports the @option{-mnan=} command-line option too.
1323 In the absence of this configuration option the default convention is
1324 the legacy encoding, as when neither of the @option{-mnan=2008} and
1325 @option{-mnan=legacy} command-line options has been used.
1327 @item --with-divide=@var{type}
1328 Specify how the compiler should generate code for checking for
1329 division by zero. This option is only supported on the MIPS target.
1330 The possibilities for @var{type} are:
1333 Division by zero checks use conditional traps (this is the default on
1334 systems that support conditional traps).
1336 Division by zero checks use the break instruction.
1339 @c If you make --with-llsc the default for additional targets,
1340 @c update the --with-llsc description in the MIPS section below.
1343 On MIPS targets, make @option{-mllsc} the default when no
1344 @option{-mno-llsc} option is passed. This is the default for
1345 Linux-based targets, as the kernel will emulate them if the ISA does
1348 @item --without-llsc
1349 On MIPS targets, make @option{-mno-llsc} the default when no
1350 @option{-mllsc} option is passed.
1353 On MIPS targets, make @option{-msynci} the default when no
1354 @option{-mno-synci} option is passed.
1356 @item --without-synci
1357 On MIPS targets, make @option{-mno-synci} the default when no
1358 @option{-msynci} option is passed. This is the default.
1360 @item --with-mips-plt
1361 On MIPS targets, make use of copy relocations and PLTs.
1362 These features are extensions to the traditional
1363 SVR4-based MIPS ABIs and require support from GNU binutils
1364 and the runtime C library.
1366 @item --enable-__cxa_atexit
1367 Define if you want to use __cxa_atexit, rather than atexit, to
1368 register C++ destructors for local statics and global objects.
1369 This is essential for fully standards-compliant handling of
1370 destructors, but requires __cxa_atexit in libc. This option is currently
1371 only available on systems with GNU libc. When enabled, this will cause
1372 @option{-fuse-cxa-atexit} to be passed by default.
1374 @item --enable-gnu-indirect-function
1375 Define if you want to enable the @code{ifunc} attribute. This option is
1376 currently only available on systems with GNU libc on certain targets.
1378 @item --enable-target-optspace
1380 libraries should be optimized for code space instead of code speed.
1381 This is the default for the m32r platform.
1383 @item --with-cpp-install-dir=@var{dirname}
1384 Specify that the user visible @command{cpp} program should be installed
1385 in @file{@var{prefix}/@var{dirname}/cpp}, in addition to @var{bindir}.
1387 @item --enable-comdat
1388 Enable COMDAT group support. This is primarily used to override the
1389 automatically detected value.
1391 @item --enable-initfini-array
1392 Force the use of sections @code{.init_array} and @code{.fini_array}
1393 (instead of @code{.init} and @code{.fini}) for constructors and
1394 destructors. Option @option{--disable-initfini-array} has the
1395 opposite effect. If neither option is specified, the configure script
1396 will try to guess whether the @code{.init_array} and
1397 @code{.fini_array} sections are supported and, if they are, use them.
1399 @item --enable-link-mutex
1400 When building GCC, use a mutex to avoid linking the compilers for
1401 multiple languages at the same time, to avoid thrashing on build
1402 systems with limited free memory. The default is not to use such a mutex.
1404 @item --enable-maintainer-mode
1405 The build rules that regenerate the Autoconf and Automake output files as
1406 well as the GCC master message catalog @file{gcc.pot} are normally
1407 disabled. This is because it can only be rebuilt if the complete source
1408 tree is present. If you have changed the sources and want to rebuild the
1409 catalog, configuring with @option{--enable-maintainer-mode} will enable
1410 this. Note that you need a recent version of the @code{gettext} tools
1413 @item --disable-bootstrap
1414 For a native build, the default configuration is to perform
1415 a 3-stage bootstrap of the compiler when @samp{make} is invoked,
1416 testing that GCC can compile itself correctly. If you want to disable
1417 this process, you can configure with @option{--disable-bootstrap}.
1419 @item --enable-bootstrap
1420 In special cases, you may want to perform a 3-stage build
1421 even if the target and host triplets are different.
1422 This is possible when the host can run code compiled for
1423 the target (e.g.@: host is i686-linux, target is i486-linux).
1424 Starting from GCC 4.2, to do this you have to configure explicitly
1425 with @option{--enable-bootstrap}.
1427 @item --enable-generated-files-in-srcdir
1428 Neither the .c and .h files that are generated from Bison and flex nor the
1429 info manuals and man pages that are built from the .texi files are present
1430 in the SVN development tree. When building GCC from that development tree,
1431 or from one of our snapshots, those generated files are placed in your
1432 build directory, which allows for the source to be in a readonly
1435 If you configure with @option{--enable-generated-files-in-srcdir} then those
1436 generated files will go into the source directory. This is mainly intended
1437 for generating release or prerelease tarballs of the GCC sources, since it
1438 is not a requirement that the users of source releases to have flex, Bison,
1441 @item --enable-version-specific-runtime-libs
1443 that runtime libraries should be installed in the compiler specific
1444 subdirectory (@file{@var{libdir}/gcc}) rather than the usual places. In
1445 addition, @samp{libstdc++}'s include files will be installed into
1446 @file{@var{libdir}} unless you overruled it by using
1447 @option{--with-gxx-include-dir=@var{dirname}}. Using this option is
1448 particularly useful if you intend to use several versions of GCC in
1449 parallel. This is currently supported by @samp{libgfortran},
1450 @samp{libjava}, @samp{libstdc++}, and @samp{libobjc}.
1452 @item @anchor{WithAixSoname}--with-aix-soname=@samp{aix}, @samp{svr4} or @samp{both}
1453 Traditional AIX shared library versioning (versioned @code{Shared Object}
1454 files as members of unversioned @code{Archive Library} files named
1455 @samp{lib.a}) causes numerous headaches for package managers. However,
1456 @code{Import Files} as members of @code{Archive Library} files allow for
1457 @strong{filename-based versioning} of shared libraries as seen on Linux/SVR4,
1458 where this is called the "SONAME". But as they prevent static linking,
1459 @code{Import Files} may be used with @code{Runtime Linking} only, where the
1460 linker does search for @samp{libNAME.so} before @samp{libNAME.a} library
1461 filenames with the @samp{-lNAME} linker flag.
1463 @anchor{AixLdCommand}For detailed information please refer to the AIX
1464 @uref{http://www-01.ibm.com/support/knowledgecenter/search/%22the%20ld%20command%2C%20also%20called%20the%20linkage%20editor%20or%20binder%22,,ld
1467 As long as shared library creation is enabled, upon:
1469 @item --with-aix-soname=aix
1470 @item --with-aix-soname=both
1471 A (traditional AIX) @code{Shared Archive Library} file is created:
1473 @item using the @samp{libNAME.a} filename scheme
1474 @item with the @code{Shared Object} file as archive member named
1475 @samp{libNAME.so.V} (except for @samp{libgcc_s}, where the @code{Shared
1476 Object} file is named @samp{shr.o} for backwards compatibility), which
1478 @item is used for runtime loading from inside the @samp{libNAME.a} file
1479 @item is used for dynamic loading via
1480 @code{dlopen("libNAME.a(libNAME.so.V)", RTLD_MEMBER)}
1481 @item is used for shared linking
1482 @item is used for static linking, so no separate @code{Static Archive
1483 Library} file is needed
1486 @item --with-aix-soname=both
1487 @item --with-aix-soname=svr4
1488 A (second) @code{Shared Archive Library} file is created:
1490 @item using the @samp{libNAME.so.V} filename scheme
1491 @item with the @code{Shared Object} file as archive member named
1494 @item is created with the @code{-G linker flag}
1495 @item has the @code{F_LOADONLY} flag set
1496 @item is used for runtime loading from inside the @samp{libNAME.so.V} file
1497 @item is used for dynamic loading via @code{dlopen("libNAME.so.V(shr.o)",
1500 @item with the @code{Import File} as archive member named @samp{shr.imp},
1503 @item refers to @samp{libNAME.so.V(shr.o)} as the "SONAME", to be recorded
1504 in the @code{Loader Section} of subsequent binaries
1505 @item indicates whether @samp{libNAME.so.V(shr.o)} is 32 or 64 bit
1506 @item lists all the public symbols exported by @samp{lib.so.V(shr.o)},
1507 eventually decorated with the @code{@samp{weak} Keyword}
1508 @item is necessary for shared linking against @samp{lib.so.V(shr.o)}
1511 A symbolic link using the @samp{libNAME.so} filename scheme is created:
1513 @item pointing to the @samp{libNAME.so.V} @code{Shared Archive Library} file
1514 @item to permit the @code{ld Command} to find @samp{lib.so.V(shr.imp)} via
1515 the @samp{-lNAME} argument (requires @code{Runtime Linking} to be enabled)
1516 @item to permit dynamic loading of @samp{lib.so.V(shr.o)} without the need
1517 to specify the version number via @code{dlopen("libNAME.so(shr.o)",
1522 As long as static library creation is enabled, upon:
1524 @item --with-aix-soname=svr4
1525 A @code{Static Archive Library} is created:
1527 @item using the @samp{libNAME.a} filename scheme
1528 @item with all the @code{Static Object} files as archive members, which
1530 @item are used for static linking
1535 While the aix-soname=@samp{svr4} option does not create @code{Shared Object}
1536 files as members of unversioned @code{Archive Library} files any more, package
1537 managers still are responsible to
1538 @uref{./specific.html#TransferAixShobj,,transfer} @code{Shared Object} files
1539 found as member of a previously installed unversioned @code{Archive Library}
1540 file into the newly installed @code{Archive Library} file with the same
1543 @emph{WARNING:} Creating @code{Shared Object} files with @code{Runtime Linking}
1544 enabled may bloat the TOC, eventually leading to @code{TOC overflow} errors,
1545 requiring the use of either the @option{-Wl,-bbigtoc} linker flag (seen to
1546 break with the @code{GDB} debugger) or some of the TOC-related compiler flags,
1548 @xref{RS/6000 and PowerPC Options,, RS/6000 and PowerPC Options, gcc,
1549 Using the GNU Compiler Collection (GCC)}.
1552 see ``RS/6000 and PowerPC Options'' in the main manual.
1555 @option{--with-aix-soname} is currently supported by @samp{libgcc_s} only, so
1556 this option is still experimental and not for normal use yet.
1558 Default is the traditional behavior @option{--with-aix-soname=@samp{aix}}.
1560 @item --enable-languages=@var{lang1},@var{lang2},@dots{}
1561 Specify that only a particular subset of compilers and
1562 their runtime libraries should be built. For a list of valid values for
1563 @var{langN} you can issue the following command in the
1564 @file{gcc} directory of your GCC source tree:@*
1566 grep ^language= */config-lang.in
1568 Currently, you can use any of the following:
1569 @code{all}, @code{ada}, @code{c}, @code{c++}, @code{fortran},
1570 @code{go}, @code{java}, @code{jit}, @code{lto}, @code{objc}, @code{obj-c++}.
1571 Building the Ada compiler has special requirements, see below.
1572 If you do not pass this flag, or specify the option @code{all}, then all
1573 default languages available in the @file{gcc} sub-tree will be configured.
1574 Ada, Go, Jit, and Objective-C++ are not default languages. LTO is not a
1575 default language, but is built by default because @option{--enable-lto} is
1576 enabled by default. The other languages are default languages.
1578 @item --enable-stage1-languages=@var{lang1},@var{lang2},@dots{}
1579 Specify that a particular subset of compilers and their runtime
1580 libraries should be built with the system C compiler during stage 1 of
1581 the bootstrap process, rather than only in later stages with the
1582 bootstrapped C compiler. The list of valid values is the same as for
1583 @option{--enable-languages}, and the option @code{all} will select all
1584 of the languages enabled by @option{--enable-languages}. This option is
1585 primarily useful for GCC development; for instance, when a development
1586 version of the compiler cannot bootstrap due to compiler bugs, or when
1587 one is debugging front ends other than the C front end. When this
1588 option is used, one can then build the target libraries for the
1589 specified languages with the stage-1 compiler by using @command{make
1590 stage1-bubble all-target}, or run the testsuite on the stage-1 compiler
1591 for the specified languages using @command{make stage1-start check-gcc}.
1593 @item --disable-libada
1594 Specify that the run-time libraries and tools used by GNAT should not
1595 be built. This can be useful for debugging, or for compatibility with
1596 previous Ada build procedures, when it was required to explicitly
1597 do a @samp{make -C gcc gnatlib_and_tools}.
1599 @item --disable-libsanitizer
1600 Specify that the run-time libraries for the various sanitizers should
1603 @item --disable-libssp
1604 Specify that the run-time libraries for stack smashing protection
1605 should not be built.
1607 @item --disable-libquadmath
1608 Specify that the GCC quad-precision math library should not be built.
1609 On some systems, the library is required to be linkable when building
1610 the Fortran front end, unless @option{--disable-libquadmath-support}
1613 @item --disable-libquadmath-support
1614 Specify that the Fortran front end and @code{libgfortran} do not add
1615 support for @code{libquadmath} on systems supporting it.
1617 @item --disable-libgomp
1618 Specify that the GNU Offloading and Multi Processing Runtime Library
1619 should not be built.
1621 @item --disable-libvtv
1622 Specify that the run-time libraries used by vtable verification
1623 should not be built.
1626 Specify that the compiler should
1627 use DWARF 2 debugging information as the default.
1629 @item --with-advance-toolchain=@var{at}
1630 On 64-bit PowerPC Linux systems, configure the compiler to use the
1631 header files, library files, and the dynamic linker from the Advance
1632 Toolchain release @var{at} instead of the default versions that are
1633 provided by the Linux distribution. In general, this option is
1634 intended for the developers of GCC, and it is not intended for general
1637 @item --enable-targets=all
1638 @itemx --enable-targets=@var{target_list}
1639 Some GCC targets, e.g.@: powerpc64-linux, build bi-arch compilers.
1640 These are compilers that are able to generate either 64-bit or 32-bit
1641 code. Typically, the corresponding 32-bit target, e.g.@:
1642 powerpc-linux for powerpc64-linux, only generates 32-bit code. This
1643 option enables the 32-bit target to be a bi-arch compiler, which is
1644 useful when you want a bi-arch compiler that defaults to 32-bit, and
1645 you are building a bi-arch or multi-arch binutils in a combined tree.
1646 On mips-linux, this will build a tri-arch compiler (ABI o32/n32/64),
1648 Currently, this option only affects sparc-linux, powerpc-linux, x86-linux,
1649 mips-linux and s390-linux.
1651 @item --enable-default-pie
1652 Turn on @option{-fPIE} and @option{-pie} by default.
1654 @item --enable-secureplt
1655 This option enables @option{-msecure-plt} by default for powerpc-linux.
1657 @xref{RS/6000 and PowerPC Options,, RS/6000 and PowerPC Options, gcc,
1658 Using the GNU Compiler Collection (GCC)},
1661 See ``RS/6000 and PowerPC Options'' in the main manual
1664 @item --enable-default-ssp
1665 Turn on @option{-fstack-protector-strong} by default.
1668 This option enables @option{-mcld} by default for 32-bit x86 targets.
1670 @xref{i386 and x86-64 Options,, i386 and x86-64 Options, gcc,
1671 Using the GNU Compiler Collection (GCC)},
1674 See ``i386 and x86-64 Options'' in the main manual
1677 @item --enable-win32-registry
1678 @itemx --enable-win32-registry=@var{key}
1679 @itemx --disable-win32-registry
1680 The @option{--enable-win32-registry} option enables Microsoft Windows-hosted GCC
1681 to look up installations paths in the registry using the following key:
1684 @code{HKEY_LOCAL_MACHINE\SOFTWARE\Free Software Foundation\@var{key}}
1687 @var{key} defaults to GCC version number, and can be overridden by the
1688 @option{--enable-win32-registry=@var{key}} option. Vendors and distributors
1689 who use custom installers are encouraged to provide a different key,
1690 perhaps one comprised of vendor name and GCC version number, to
1691 avoid conflict with existing installations. This feature is enabled
1692 by default, and can be disabled by @option{--disable-win32-registry}
1693 option. This option has no effect on the other hosts.
1696 Specify that the machine does not have a floating point unit. This
1697 option only applies to @samp{m68k-sun-sunos@var{n}}. On any other
1698 system, @option{--nfp} has no effect.
1700 @item --enable-werror
1701 @itemx --disable-werror
1702 @itemx --enable-werror=yes
1703 @itemx --enable-werror=no
1704 When you specify this option, it controls whether certain files in the
1705 compiler are built with @option{-Werror} in bootstrap stage2 and later.
1706 If you don't specify it, @option{-Werror} is turned on for the main
1707 development trunk. However it defaults to off for release branches and
1708 final releases. The specific files which get @option{-Werror} are
1709 controlled by the Makefiles.
1711 @item --enable-checking
1712 @itemx --enable-checking=@var{list}
1713 When you specify this option, the compiler is built to perform internal
1714 consistency checks of the requested complexity. This does not change the
1715 generated code, but adds error checking within the compiler. This will
1716 slow down the compiler and may only work properly if you are building
1717 the compiler with GCC@. This is @samp{yes,extra} by default when building
1718 from SVN or snapshots, but @samp{release} for releases. The default
1719 for building the stage1 compiler is @samp{yes}. More control
1720 over the checks may be had by specifying @var{list}. The categories of
1721 checks available are @samp{yes} (most common checks
1722 @samp{assert,misc,tree,gc,rtlflag,runtime}), @samp{no} (no checks at
1723 all), @samp{all} (all but @samp{valgrind}), @samp{release} (cheapest
1724 checks @samp{assert,runtime}) or @samp{none} (same as @samp{no}).
1725 Individual checks can be enabled with these flags @samp{assert},
1726 @samp{df}, @samp{fold}, @samp{gc}, @samp{gcac}, @samp{misc}, @samp{rtl},
1727 @samp{rtlflag}, @samp{runtime}, @samp{tree}, @samp{extra} and @samp{valgrind}.
1728 @samp{extra} adds for @samp{misc} checking extra checks that might affect
1729 code generation and should therefore not differ between stage1 and later
1732 The @samp{valgrind} check requires the external @command{valgrind}
1733 simulator, available from @uref{http://valgrind.org/}. The
1734 @samp{df}, @samp{rtl}, @samp{gcac} and @samp{valgrind} checks are very expensive.
1735 To disable all checking, @samp{--disable-checking} or
1736 @samp{--enable-checking=none} must be explicitly requested. Disabling
1737 assertions will make the compiler and runtime slightly faster but
1738 increase the risk of undetected internal errors causing wrong code to be
1741 @item --disable-stage1-checking
1742 @itemx --enable-stage1-checking
1743 @itemx --enable-stage1-checking=@var{list}
1744 If no @option{--enable-checking} option is specified the stage1
1745 compiler will be built with @samp{yes} checking enabled, otherwise
1746 the stage1 checking flags are the same as specified by
1747 @option{--enable-checking}. To build the stage1 compiler with
1748 different checking options use @option{--enable-stage1-checking}.
1749 The list of checking options is the same as for @option{--enable-checking}.
1750 If your system is too slow or too small to bootstrap a released compiler
1751 with checking for stage1 enabled, you can use @samp{--disable-stage1-checking}
1752 to disable checking for the stage1 compiler.
1754 @item --enable-coverage
1755 @itemx --enable-coverage=@var{level}
1756 With this option, the compiler is built to collect self coverage
1757 information, every time it is run. This is for internal development
1758 purposes, and only works when the compiler is being built with gcc. The
1759 @var{level} argument controls whether the compiler is built optimized or
1760 not, values are @samp{opt} and @samp{noopt}. For coverage analysis you
1761 want to disable optimization, for performance analysis you want to
1762 enable optimization. When coverage is enabled, the default level is
1763 without optimization.
1765 @item --enable-gather-detailed-mem-stats
1766 When this option is specified more detailed information on memory
1767 allocation is gathered. This information is printed when using
1768 @option{-fmem-report}.
1770 @item --enable-valgrind-annotations
1771 Mark selected memory related operations in the compiler when run under
1772 valgrind to suppress false positives.
1775 @itemx --disable-nls
1776 The @option{--enable-nls} option enables Native Language Support (NLS),
1777 which lets GCC output diagnostics in languages other than American
1778 English. Native Language Support is enabled by default if not doing a
1779 canadian cross build. The @option{--disable-nls} option disables NLS@.
1781 @item --with-included-gettext
1782 If NLS is enabled, the @option{--with-included-gettext} option causes the build
1783 procedure to prefer its copy of GNU @command{gettext}.
1785 @item --with-catgets
1786 If NLS is enabled, and if the host lacks @code{gettext} but has the
1787 inferior @code{catgets} interface, the GCC build procedure normally
1788 ignores @code{catgets} and instead uses GCC's copy of the GNU
1789 @code{gettext} library. The @option{--with-catgets} option causes the
1790 build procedure to use the host's @code{catgets} in this situation.
1792 @item --with-libiconv-prefix=@var{dir}
1793 Search for libiconv header files in @file{@var{dir}/include} and
1794 libiconv library files in @file{@var{dir}/lib}.
1796 @item --enable-obsolete
1797 Enable configuration for an obsoleted system. If you attempt to
1798 configure GCC for a system (build, host, or target) which has been
1799 obsoleted, and you do not specify this flag, configure will halt with an
1802 All support for systems which have been obsoleted in one release of GCC
1803 is removed entirely in the next major release, unless someone steps
1804 forward to maintain the port.
1806 @item --enable-decimal-float
1807 @itemx --enable-decimal-float=yes
1808 @itemx --enable-decimal-float=no
1809 @itemx --enable-decimal-float=bid
1810 @itemx --enable-decimal-float=dpd
1811 @itemx --disable-decimal-float
1812 Enable (or disable) support for the C decimal floating point extension
1813 that is in the IEEE 754-2008 standard. This is enabled by default only
1814 on PowerPC, i386, and x86_64 GNU/Linux systems. Other systems may also
1815 support it, but require the user to specifically enable it. You can
1816 optionally control which decimal floating point format is used (either
1817 @samp{bid} or @samp{dpd}). The @samp{bid} (binary integer decimal)
1818 format is default on i386 and x86_64 systems, and the @samp{dpd}
1819 (densely packed decimal) format is default on PowerPC systems.
1821 @item --enable-fixed-point
1822 @itemx --disable-fixed-point
1823 Enable (or disable) support for C fixed-point arithmetic.
1824 This option is enabled by default for some targets (such as MIPS) which
1825 have hardware-support for fixed-point operations. On other targets, you
1826 may enable this option manually.
1828 @item --with-long-double-128
1829 Specify if @code{long double} type should be 128-bit by default on selected
1830 GNU/Linux architectures. If using @code{--without-long-double-128},
1831 @code{long double} will be by default 64-bit, the same as @code{double} type.
1832 When neither of these configure options are used, the default will be
1833 128-bit @code{long double} when built against GNU C Library 2.4 and later,
1834 64-bit @code{long double} otherwise.
1836 @item --enable-fdpic
1837 On SH Linux systems, generate ELF FDPIC code.
1839 @item --with-gmp=@var{pathname}
1840 @itemx --with-gmp-include=@var{pathname}
1841 @itemx --with-gmp-lib=@var{pathname}
1842 @itemx --with-mpfr=@var{pathname}
1843 @itemx --with-mpfr-include=@var{pathname}
1844 @itemx --with-mpfr-lib=@var{pathname}
1845 @itemx --with-mpc=@var{pathname}
1846 @itemx --with-mpc-include=@var{pathname}
1847 @itemx --with-mpc-lib=@var{pathname}
1848 If you want to build GCC but do not have the GMP library, the MPFR
1849 library and/or the MPC library installed in a standard location and
1850 do not have their sources present in the GCC source tree then you
1851 can explicitly specify the directory where they are installed
1852 (@samp{--with-gmp=@var{gmpinstalldir}},
1853 @samp{--with-mpfr=@/@var{mpfrinstalldir}},
1854 @samp{--with-mpc=@/@var{mpcinstalldir}}). The
1855 @option{--with-gmp=@/@var{gmpinstalldir}} option is shorthand for
1856 @option{--with-gmp-lib=@/@var{gmpinstalldir}/lib} and
1857 @option{--with-gmp-include=@/@var{gmpinstalldir}/include}. Likewise the
1858 @option{--with-mpfr=@/@var{mpfrinstalldir}} option is shorthand for
1859 @option{--with-mpfr-lib=@/@var{mpfrinstalldir}/lib} and
1860 @option{--with-mpfr-include=@/@var{mpfrinstalldir}/include}, also the
1861 @option{--with-mpc=@/@var{mpcinstalldir}} option is shorthand for
1862 @option{--with-mpc-lib=@/@var{mpcinstalldir}/lib} and
1863 @option{--with-mpc-include=@/@var{mpcinstalldir}/include}. If these
1864 shorthand assumptions are not correct, you can use the explicit
1865 include and lib options directly. You might also need to ensure the
1866 shared libraries can be found by the dynamic linker when building and
1867 using GCC, for example by setting the runtime shared library path
1868 variable (@env{LD_LIBRARY_PATH} on GNU/Linux and Solaris systems).
1870 These flags are applicable to the host platform only. When building
1871 a cross compiler, they will not be used to configure target libraries.
1873 @item --with-isl=@var{pathname}
1874 @itemx --with-isl-include=@var{pathname}
1875 @itemx --with-isl-lib=@var{pathname}
1876 If you do not have the isl library installed in a standard location and you
1877 want to build GCC, you can explicitly specify the directory where it is
1878 installed (@samp{--with-isl=@/@var{islinstalldir}}). The
1879 @option{--with-isl=@/@var{islinstalldir}} option is shorthand for
1880 @option{--with-isl-lib=@/@var{islinstalldir}/lib} and
1881 @option{--with-isl-include=@/@var{islinstalldir}/include}. If this
1882 shorthand assumption is not correct, you can use the explicit
1883 include and lib options directly.
1885 These flags are applicable to the host platform only. When building
1886 a cross compiler, they will not be used to configure target libraries.
1888 @item --with-stage1-ldflags=@var{flags}
1889 This option may be used to set linker flags to be used when linking
1890 stage 1 of GCC. These are also used when linking GCC if configured with
1891 @option{--disable-bootstrap}. If @option{--with-stage1-libs} is not set to a
1892 value, then the default is @samp{-static-libstdc++ -static-libgcc}, if
1895 @item --with-stage1-libs=@var{libs}
1896 This option may be used to set libraries to be used when linking stage 1
1897 of GCC. These are also used when linking GCC if configured with
1898 @option{--disable-bootstrap}.
1900 @item --with-boot-ldflags=@var{flags}
1901 This option may be used to set linker flags to be used when linking
1902 stage 2 and later when bootstrapping GCC. If --with-boot-libs
1903 is not is set to a value, then the default is
1904 @samp{-static-libstdc++ -static-libgcc}.
1906 @item --with-boot-libs=@var{libs}
1907 This option may be used to set libraries to be used when linking stage 2
1908 and later when bootstrapping GCC.
1910 @item --with-debug-prefix-map=@var{map}
1911 Convert source directory names using @option{-fdebug-prefix-map} when
1912 building runtime libraries. @samp{@var{map}} is a space-separated
1913 list of maps of the form @samp{@var{old}=@var{new}}.
1915 @item --enable-linker-build-id
1916 Tells GCC to pass @option{--build-id} option to the linker for all final
1917 links (links performed without the @option{-r} or @option{--relocatable}
1918 option), if the linker supports it. If you specify
1919 @option{--enable-linker-build-id}, but your linker does not
1920 support @option{--build-id} option, a warning is issued and the
1921 @option{--enable-linker-build-id} option is ignored. The default is off.
1923 @item --with-linker-hash-style=@var{choice}
1924 Tells GCC to pass @option{--hash-style=@var{choice}} option to the
1925 linker for all final links. @var{choice} can be one of
1926 @samp{sysv}, @samp{gnu}, and @samp{both} where @samp{sysv} is the default.
1928 @item --enable-gnu-unique-object
1929 @itemx --disable-gnu-unique-object
1930 Tells GCC to use the gnu_unique_object relocation for C++ template
1931 static data members and inline function local statics. Enabled by
1932 default for a toolchain with an assembler that accepts it and
1933 GLIBC 2.11 or above, otherwise disabled.
1935 @item --with-diagnostics-color=@var{choice}
1936 Tells GCC to use @var{choice} as the default for @option{-fdiagnostics-color=}
1937 option (if not used explicitly on the command line). @var{choice}
1938 can be one of @samp{never}, @samp{auto}, @samp{always}, and @samp{auto-if-env}
1939 where @samp{auto} is the default. @samp{auto-if-env} means that
1940 @option{-fdiagnostics-color=auto} will be the default if @code{GCC_COLORS}
1941 is present and non-empty in the environment, and
1942 @option{-fdiagnostics-color=never} otherwise.
1945 @itemx --disable-lto
1946 Enable support for link-time optimization (LTO). This is enabled by
1947 default, and may be disabled using @option{--disable-lto}.
1949 @item --enable-linker-plugin-configure-flags=FLAGS
1950 @itemx --enable-linker-plugin-flags=FLAGS
1951 By default, linker plugins (such as the LTO plugin) are built for the
1952 host system architecture. For the case that the linker has a
1953 different (but run-time compatible) architecture, these flags can be
1954 specified to build plugins that are compatible to the linker. For
1955 example, if you are building GCC for a 64-bit x86_64
1956 (@samp{x86_64-unknown-linux-gnu}) host system, but have a 32-bit x86
1957 GNU/Linux (@samp{i686-pc-linux-gnu}) linker executable (which is
1958 executable on the former system), you can configure GCC as follows for
1959 getting compatible linker plugins:
1962 % @var{srcdir}/configure \
1963 --host=x86_64-unknown-linux-gnu \
1964 --enable-linker-plugin-configure-flags=--host=i686-pc-linux-gnu \
1965 --enable-linker-plugin-flags='CC=gcc\ -m32\ -Wl,-rpath,[...]/i686-pc-linux-gnu/lib'
1968 @item --with-plugin-ld=@var{pathname}
1969 Enable an alternate linker to be used at link-time optimization (LTO)
1970 link time when @option{-fuse-linker-plugin} is enabled.
1971 This linker should have plugin support such as gold starting with
1972 version 2.20 or GNU ld starting with version 2.21.
1973 See @option{-fuse-linker-plugin} for details.
1975 @item --enable-canonical-system-headers
1976 @itemx --disable-canonical-system-headers
1977 Enable system header path canonicalization for @file{libcpp}. This can
1978 produce shorter header file paths in diagnostics and dependency output
1979 files, but these changed header paths may conflict with some compilation
1980 environments. Enabled by default, and may be disabled using
1981 @option{--disable-canonical-system-headers}.
1983 @item --with-glibc-version=@var{major}.@var{minor}
1984 Tell GCC that when the GNU C Library (glibc) is used on the target it
1985 will be version @var{major}.@var{minor} or later. Normally this can
1986 be detected from the C library's header files, but this option may be
1987 needed when bootstrapping a cross toolchain without the header files
1988 available for building the initial bootstrap compiler.
1990 If GCC is configured with some multilibs that use glibc and some that
1991 do not, this option applies only to the multilibs that use glibc.
1992 However, such configurations may not work well as not all the relevant
1993 configuration in GCC is on a per-multilib basis.
1995 @item --enable-as-accelerator-for=@var{target}
1996 Build as offload target compiler. Specify offload host triple by @var{target}.
1998 @item --enable-offload-targets=@var{target1}[=@var{path1}],@dots{},@var{targetN}[=@var{pathN}]
1999 Enable offloading to targets @var{target1}, @dots{}, @var{targetN}.
2000 Offload compilers are expected to be already installed. Default search
2001 path for them is @file{@var{exec-prefix}}, but it can be changed by
2002 specifying paths @var{path1}, @dots{}, @var{pathN}.
2005 % @var{srcdir}/configure \
2006 --enable-offload-target=i686-unknown-linux-gnu=/path/to/i686/compiler,x86_64-pc-linux-gnu
2009 If @samp{hsa} is specified as one of the targets, the compiler will be
2010 built with support for HSA GPU accelerators. Because the same
2011 compiler will emit the accelerator code, no path should be specified.
2013 @item --with-hsa-runtime=@var{pathname}
2014 @itemx --with-hsa-runtime-include=@var{pathname}
2015 @itemx --with-hsa-runtime-lib=@var{pathname}
2017 If you configure GCC with HSA offloading but do not have the HSA
2018 run-time library installed in a standard location then you can
2019 explicitly specify the directory where they are installed. The
2020 @option{--with-hsa-runtime=@/@var{hsainstalldir}} option is a
2022 @option{--with-hsa-runtime-lib=@/@var{hsainstalldir}/lib} and
2023 @option{--with-hsa-runtime-include=@/@var{hsainstalldir}/include}.
2025 @item --with-hsa-kmt-lib=@var{pathname}
2027 If you configure GCC with HSA offloading but do not have the HSA
2028 KMT library installed in a standard location then you can
2029 explicitly specify the directory where it resides.
2032 @subheading Cross-Compiler-Specific Options
2033 The following options only apply to building cross compilers.
2036 @item --with-sysroot
2037 @itemx --with-sysroot=@var{dir}
2038 Tells GCC to consider @var{dir} as the root of a tree that contains
2039 (a subset of) the root filesystem of the target operating system.
2040 Target system headers, libraries and run-time object files will be
2041 searched for in there. More specifically, this acts as if
2042 @option{--sysroot=@var{dir}} was added to the default options of the built
2043 compiler. The specified directory is not copied into the
2044 install tree, unlike the options @option{--with-headers} and
2045 @option{--with-libs} that this option obsoletes. The default value,
2046 in case @option{--with-sysroot} is not given an argument, is
2047 @option{$@{gcc_tooldir@}/sys-root}. If the specified directory is a
2048 subdirectory of @option{$@{exec_prefix@}}, then it will be found relative to
2049 the GCC binaries if the installation tree is moved.
2051 This option affects the system root for the compiler used to build
2052 target libraries (which runs on the build system) and the compiler newly
2053 installed with @code{make install}; it does not affect the compiler which is
2054 used to build GCC itself.
2056 If you specify the @option{--with-native-system-header-dir=@var{dirname}}
2057 option then the compiler will search that directory within @var{dirname} for
2058 native system headers rather than the default @file{/usr/include}.
2060 @item --with-build-sysroot
2061 @itemx --with-build-sysroot=@var{dir}
2062 Tells GCC to consider @var{dir} as the system root (see
2063 @option{--with-sysroot}) while building target libraries, instead of
2064 the directory specified with @option{--with-sysroot}. This option is
2065 only useful when you are already using @option{--with-sysroot}. You
2066 can use @option{--with-build-sysroot} when you are configuring with
2067 @option{--prefix} set to a directory that is different from the one in
2068 which you are installing GCC and your target libraries.
2070 This option affects the system root for the compiler used to build
2071 target libraries (which runs on the build system); it does not affect
2072 the compiler which is used to build GCC itself.
2074 If you specify the @option{--with-native-system-header-dir=@var{dirname}}
2075 option then the compiler will search that directory within @var{dirname} for
2076 native system headers rather than the default @file{/usr/include}.
2078 @item --with-headers
2079 @itemx --with-headers=@var{dir}
2080 Deprecated in favor of @option{--with-sysroot}.
2081 Specifies that target headers are available when building a cross compiler.
2082 The @var{dir} argument specifies a directory which has the target include
2083 files. These include files will be copied into the @file{gcc} install
2084 directory. @emph{This option with the @var{dir} argument is required} when
2085 building a cross compiler, if @file{@var{prefix}/@var{target}/sys-include}
2086 doesn't pre-exist. If @file{@var{prefix}/@var{target}/sys-include} does
2087 pre-exist, the @var{dir} argument may be omitted. @command{fixincludes}
2088 will be run on these files to make them compatible with GCC@.
2090 @item --without-headers
2091 Tells GCC not use any target headers from a libc when building a cross
2092 compiler. When crossing to GNU/Linux, you need the headers so GCC
2093 can build the exception handling for libgcc.
2096 @itemx --with-libs="@var{dir1} @var{dir2} @dots{} @var{dirN}"
2097 Deprecated in favor of @option{--with-sysroot}.
2098 Specifies a list of directories which contain the target runtime
2099 libraries. These libraries will be copied into the @file{gcc} install
2100 directory. If the directory list is omitted, this option has no
2104 Specifies that @samp{newlib} is
2105 being used as the target C library. This causes @code{__eprintf} to be
2106 omitted from @file{libgcc.a} on the assumption that it will be provided by
2109 @item --with-avrlibc
2110 Specifies that @samp{AVR-Libc} is
2111 being used as the target C library. This causes float support
2112 functions like @code{__addsf3} to be omitted from @file{libgcc.a} on
2113 the assumption that it will be provided by @file{libm.a}. For more
2114 technical details, cf. @uref{http://gcc.gnu.org/PR54461,,PR54461}.
2115 This option is only supported for the AVR target. It is not supported for
2116 RTEMS configurations, which currently use newlib. The option is
2117 supported since version 4.7.2 and is the default in 4.8.0 and newer.
2119 @item --with-nds32-lib=@var{library}
2120 Specifies that @var{library} setting is used for building @file{libgcc.a}.
2121 Currently, the valid @var{library} is @samp{newlib} or @samp{mculib}.
2122 This option is only supported for the NDS32 target.
2124 @item --with-build-time-tools=@var{dir}
2125 Specifies where to find the set of target tools (assembler, linker, etc.)
2126 that will be used while building GCC itself. This option can be useful
2127 if the directory layouts are different between the system you are building
2128 GCC on, and the system where you will deploy it.
2130 For example, on an @samp{ia64-hp-hpux} system, you may have the GNU
2131 assembler and linker in @file{/usr/bin}, and the native tools in a
2132 different path, and build a toolchain that expects to find the
2133 native tools in @file{/usr/bin}.
2135 When you use this option, you should ensure that @var{dir} includes
2136 @command{ar}, @command{as}, @command{ld}, @command{nm},
2137 @command{ranlib} and @command{strip} if necessary, and possibly
2138 @command{objdump}. Otherwise, GCC may use an inconsistent set of
2142 @subheading Java-Specific Options
2144 The following option applies to the build of the Java front end.
2147 @item --disable-libgcj
2148 Specify that the run-time libraries
2149 used by GCJ should not be built. This is useful in case you intend
2150 to use GCJ with some other run-time, or you're going to install it
2151 separately, or it just happens not to build on your particular
2152 machine. In general, if the Java front end is enabled, the GCJ
2153 libraries will be enabled too, unless they're known to not work on
2154 the target platform. If GCJ is enabled but @samp{libgcj} isn't built, you
2155 may need to port it; in this case, before modifying the top-level
2156 @file{configure.ac} so that @samp{libgcj} is enabled by default on this platform,
2157 you may use @option{--enable-libgcj} to override the default.
2161 The following options apply to building @samp{libgcj}.
2163 @subsubheading General Options
2166 @item --enable-java-maintainer-mode
2167 By default the @samp{libjava} build will not attempt to compile the
2168 @file{.java} source files to @file{.class}. Instead, it will use the
2169 @file{.class} files from the source tree. If you use this option you
2170 must have executables named @command{ecj1} and @command{gjavah} in your path
2171 for use by the build. You must use this option if you intend to
2172 modify any @file{.java} files in @file{libjava}.
2174 @item --with-java-home=@var{dirname}
2175 This @samp{libjava} option overrides the default value of the
2176 @samp{java.home} system property. It is also used to set
2177 @samp{sun.boot.class.path} to @file{@var{dirname}/lib/rt.jar}. By
2178 default @samp{java.home} is set to @file{@var{prefix}} and
2179 @samp{sun.boot.class.path} to
2180 @file{@var{datadir}/java/libgcj-@var{version}.jar}.
2182 @item --with-ecj-jar=@var{filename}
2183 This option can be used to specify the location of an external jar
2184 file containing the Eclipse Java compiler. A specially modified
2185 version of this compiler is used by @command{gcj} to parse
2186 @file{.java} source files. If this option is given, the
2187 @samp{libjava} build will create and install an @file{ecj1} executable
2188 which uses this jar file at runtime.
2190 If this option is not given, but an @file{ecj.jar} file is found in
2191 the topmost source tree at configure time, then the @samp{libgcj}
2192 build will create and install @file{ecj1}, and will also install the
2193 discovered @file{ecj.jar} into a suitable place in the install tree.
2195 If @file{ecj1} is not installed, then the user will have to supply one
2196 on his path in order for @command{gcj} to properly parse @file{.java}
2197 source files. A suitable jar is available from
2198 @uref{ftp://sourceware.org/pub/java/}.
2200 @item --disable-getenv-properties
2201 Don't set system properties from @env{GCJ_PROPERTIES}.
2203 @item --enable-hash-synchronization
2204 Use a global hash table for monitor locks. Ordinarily,
2205 @samp{libgcj}'s @samp{configure} script automatically makes
2206 the correct choice for this option for your platform. Only use
2207 this if you know you need the library to be configured differently.
2209 @item --enable-interpreter
2210 Enable the Java interpreter. The interpreter is automatically
2211 enabled by default on all platforms that support it. This option
2212 is really only useful if you want to disable the interpreter
2213 (using @option{--disable-interpreter}).
2215 @item --disable-java-net
2216 Disable java.net. This disables the native part of java.net only,
2217 using non-functional stubs for native method implementations.
2219 @item --disable-jvmpi
2220 Disable JVMPI support.
2222 @item --disable-libgcj-bc
2223 Disable BC ABI compilation of certain parts of libgcj. By default,
2224 some portions of libgcj are compiled with @option{-findirect-dispatch}
2225 and @option{-fno-indirect-classes}, allowing them to be overridden at
2228 If @option{--disable-libgcj-bc} is specified, libgcj is built without
2229 these options. This allows the compile-time linker to resolve
2230 dependencies when statically linking to libgcj. However it makes it
2231 impossible to override the affected portions of libgcj at run-time.
2233 @item --enable-reduced-reflection
2234 Build most of libgcj with @option{-freduced-reflection}. This reduces
2235 the size of libgcj at the expense of not being able to do accurate
2236 reflection on the classes it contains. This option is safe if you
2237 know that code using libgcj will never use reflection on the standard
2238 runtime classes in libgcj (including using serialization, RMI or CORBA).
2241 Enable runtime eCos target support.
2243 @item --without-libffi
2244 Don't use @samp{libffi}. This will disable the interpreter and JNI
2245 support as well, as these require @samp{libffi} to work.
2247 @item --enable-libgcj-debug
2248 Enable runtime debugging code.
2250 @item --enable-libgcj-multifile
2251 If specified, causes all @file{.java} source files to be
2252 compiled into @file{.class} files in one invocation of
2253 @samp{gcj}. This can speed up build time, but is more
2254 resource-intensive. If this option is unspecified or
2255 disabled, @samp{gcj} is invoked once for each @file{.java}
2256 file to compile into a @file{.class} file.
2258 @item --with-libiconv-prefix=DIR
2259 Search for libiconv in @file{DIR/include} and @file{DIR/lib}.
2261 @item --enable-sjlj-exceptions
2262 Force use of the @code{setjmp}/@code{longjmp}-based scheme for exceptions.
2263 @samp{configure} ordinarily picks the correct value based on the platform.
2264 Only use this option if you are sure you need a different setting.
2266 @item --with-system-zlib
2267 Use installed @samp{zlib} rather than that included with GCC@.
2269 @item --with-win32-nlsapi=ansi, unicows or unicode
2270 Indicates how MinGW @samp{libgcj} translates between UNICODE
2271 characters and the Win32 API@.
2273 @item --enable-java-home
2274 If enabled, this creates a JPackage compatible SDK environment during install.
2275 Note that if --enable-java-home is used, --with-arch-directory=ARCH must also
2278 @item --with-arch-directory=ARCH
2279 Specifies the name to use for the @file{jre/lib/ARCH} directory in the SDK
2280 environment created when --enable-java-home is passed. Typical names for this
2281 directory include i386, amd64, ia64, etc.
2283 @item --with-os-directory=DIR
2284 Specifies the OS directory for the SDK include directory. This is set to auto
2285 detect, and is typically 'linux'.
2287 @item --with-origin-name=NAME
2288 Specifies the JPackage origin name. This defaults to the 'gcj' in
2291 @item --with-arch-suffix=SUFFIX
2292 Specifies the suffix for the sdk directory. Defaults to the empty string.
2293 Examples include '.x86_64' in 'java-1.5.0-gcj-1.5.0.0.x86_64'.
2295 @item --with-jvm-root-dir=DIR
2296 Specifies where to install the SDK. Default is $(prefix)/lib/jvm.
2298 @item --with-jvm-jar-dir=DIR
2299 Specifies where to install jars. Default is $(prefix)/lib/jvm-exports.
2301 @item --with-python-dir=DIR
2302 Specifies where to install the Python modules used for aot-compile. DIR should
2303 not include the prefix used in installation. For example, if the Python modules
2304 are to be installed in /usr/lib/python2.5/site-packages, then
2305 --with-python-dir=/lib/python2.5/site-packages should be passed. If this is
2306 not specified, then the Python modules are installed in $(prefix)/share/python.
2308 @item --enable-aot-compile-rpm
2309 Adds aot-compile-rpm to the list of installed scripts.
2311 @item --enable-browser-plugin
2312 Build the gcjwebplugin web browser plugin.
2314 @item --enable-static-libjava
2315 Build static libraries in libjava. The default is to only build shared
2320 Use the single-byte @code{char} and the Win32 A functions natively,
2321 translating to and from UNICODE when using these functions. If
2322 unspecified, this is the default.
2325 Use the @code{WCHAR} and Win32 W functions natively. Adds
2326 @code{-lunicows} to @file{libgcj.spec} to link with @samp{libunicows}.
2327 @file{unicows.dll} needs to be deployed on Microsoft Windows 9X machines
2328 running built executables. @file{libunicows.a}, an open-source
2329 import library around Microsoft's @code{unicows.dll}, is obtained from
2330 @uref{http://libunicows.sourceforge.net/}, which also gives details
2331 on getting @file{unicows.dll} from Microsoft.
2334 Use the @code{WCHAR} and Win32 W functions natively. Does @emph{not}
2335 add @code{-lunicows} to @file{libgcj.spec}. The built executables will
2336 only run on Microsoft Windows NT and above.
2340 @subsubheading AWT-Specific Options
2344 Use the X Window System.
2346 @item --enable-java-awt=PEER(S)
2347 Specifies the AWT peer library or libraries to build alongside
2348 @samp{libgcj}. If this option is unspecified or disabled, AWT
2349 will be non-functional. Current valid values are @option{gtk} and
2350 @option{xlib}. Multiple libraries should be separated by a
2351 comma (i.e.@: @option{--enable-java-awt=gtk,xlib}).
2353 @item --enable-gtk-cairo
2354 Build the cairo Graphics2D implementation on GTK@.
2356 @item --enable-java-gc=TYPE
2357 Choose garbage collector. Defaults to @option{boehm} if unspecified.
2359 @item --disable-gtktest
2360 Do not try to compile and run a test GTK+ program.
2362 @item --disable-glibtest
2363 Do not try to compile and run a test GLIB program.
2365 @item --with-libart-prefix=PFX
2366 Prefix where libart is installed (optional).
2368 @item --with-libart-exec-prefix=PFX
2369 Exec prefix where libart is installed (optional).
2371 @item --disable-libarttest
2372 Do not try to compile and run a test libart program.
2376 @subsubheading Overriding @command{configure} test results
2378 Sometimes, it might be necessary to override the result of some
2379 @command{configure} test, for example in order to ease porting to a new
2380 system or work around a bug in a test. The toplevel @command{configure}
2381 script provides three variables for this:
2385 @item build_configargs
2386 @cindex @code{build_configargs}
2387 The contents of this variable is passed to all build @command{configure}
2390 @item host_configargs
2391 @cindex @code{host_configargs}
2392 The contents of this variable is passed to all host @command{configure}
2395 @item target_configargs
2396 @cindex @code{target_configargs}
2397 The contents of this variable is passed to all target @command{configure}
2402 In order to avoid shell and @command{make} quoting issues for complex
2403 overrides, you can pass a setting for @env{CONFIG_SITE} and set
2404 variables in the site file.
2411 @uref{./index.html,,Return to the GCC Installation page}
2415 @c ***Building****************************************************************
2417 @comment node-name, next, previous, up
2418 @node Building, Testing, Configuration, Installing GCC
2424 @cindex Installing GCC: Building
2426 Now that GCC is configured, you are ready to build the compiler and
2429 Some commands executed when making the compiler may fail (return a
2430 nonzero status) and be ignored by @command{make}. These failures, which
2431 are often due to files that were not found, are expected, and can safely
2434 It is normal to have compiler warnings when compiling certain files.
2435 Unless you are a GCC developer, you can generally ignore these warnings
2436 unless they cause compilation to fail. Developers should attempt to fix
2437 any warnings encountered, however they can temporarily continue past
2438 warnings-as-errors by specifying the configure flag
2439 @option{--disable-werror}.
2441 On certain old systems, defining certain environment variables such as
2442 @env{CC} can interfere with the functioning of @command{make}.
2444 If you encounter seemingly strange errors when trying to build the
2445 compiler in a directory other than the source directory, it could be
2446 because you have previously configured the compiler in the source
2447 directory. Make sure you have done all the necessary preparations.
2449 If you build GCC on a BSD system using a directory stored in an old System
2450 V file system, problems may occur in running @command{fixincludes} if the
2451 System V file system doesn't support symbolic links. These problems
2452 result in a failure to fix the declaration of @code{size_t} in
2453 @file{sys/types.h}. If you find that @code{size_t} is a signed type and
2454 that type mismatches occur, this could be the cause.
2456 The solution is not to use such a directory for building GCC@.
2458 Similarly, when building from SVN or snapshots, or if you modify
2459 @file{*.l} files, you need the Flex lexical analyzer generator
2460 installed. If you do not modify @file{*.l} files, releases contain
2461 the Flex-generated files and you do not need Flex installed to build
2462 them. There is still one Flex-based lexical analyzer (part of the
2463 build machinery, not of GCC itself) that is used even if you only
2464 build the C front end.
2466 When building from SVN or snapshots, or if you modify Texinfo
2467 documentation, you need version 4.7 or later of Texinfo installed if you
2468 want Info documentation to be regenerated. Releases contain Info
2469 documentation pre-built for the unmodified documentation in the release.
2471 @section Building a native compiler
2473 For a native build, the default configuration is to perform
2474 a 3-stage bootstrap of the compiler when @samp{make} is invoked.
2475 This will build the entire GCC system and ensure that it compiles
2476 itself correctly. It can be disabled with the @option{--disable-bootstrap}
2477 parameter to @samp{configure}, but bootstrapping is suggested because
2478 the compiler will be tested more completely and could also have
2481 The bootstrapping process will complete the following steps:
2485 Build tools necessary to build the compiler.
2488 Perform a 3-stage bootstrap of the compiler. This includes building
2489 three times the target tools for use by the compiler such as binutils
2490 (bfd, binutils, gas, gprof, ld, and opcodes) if they have been
2491 individually linked or moved into the top level GCC source tree before
2495 Perform a comparison test of the stage2 and stage3 compilers.
2498 Build runtime libraries using the stage3 compiler from the previous step.
2502 If you are short on disk space you might consider @samp{make
2503 bootstrap-lean} instead. The sequence of compilation is the
2504 same described above, but object files from the stage1 and
2505 stage2 of the 3-stage bootstrap of the compiler are deleted as
2506 soon as they are no longer needed.
2508 If you wish to use non-default GCC flags when compiling the stage2
2509 and stage3 compilers, set @code{BOOT_CFLAGS} on the command line when
2510 doing @samp{make}. For example, if you want to save additional space
2511 during the bootstrap and in the final installation as well, you can
2512 build the compiler binaries without debugging information as in the
2513 following example. This will save roughly 40% of disk space both for
2514 the bootstrap and the final installation. (Libraries will still contain
2515 debugging information.)
2518 make BOOT_CFLAGS='-O' bootstrap
2521 You can place non-default optimization flags into @code{BOOT_CFLAGS}; they
2522 are less well tested here than the default of @samp{-g -O2}, but should
2523 still work. In a few cases, you may find that you need to specify special
2524 flags such as @option{-msoft-float} here to complete the bootstrap; or,
2525 if the native compiler miscompiles the stage1 compiler, you may need
2526 to work around this, by choosing @code{BOOT_CFLAGS} to avoid the parts
2527 of the stage1 compiler that were miscompiled, or by using @samp{make
2528 bootstrap4} to increase the number of stages of bootstrap.
2530 @code{BOOT_CFLAGS} does not apply to bootstrapped target libraries.
2531 Since these are always compiled with the compiler currently being
2532 bootstrapped, you can use @code{CFLAGS_FOR_TARGET} to modify their
2533 compilation flags, as for non-bootstrapped target libraries.
2534 Again, if the native compiler miscompiles the stage1 compiler, you may
2535 need to work around this by avoiding non-working parts of the stage1
2536 compiler. Use @code{STAGE1_TFLAGS} to this end.
2538 If you used the flag @option{--enable-languages=@dots{}} to restrict
2539 the compilers to be built, only those you've actually enabled will be
2540 built. This will of course only build those runtime libraries, for
2541 which the particular compiler has been built. Please note,
2542 that re-defining @env{LANGUAGES} when calling @samp{make}
2543 @strong{does not} work anymore!
2545 If the comparison of stage2 and stage3 fails, this normally indicates
2546 that the stage2 compiler has compiled GCC incorrectly, and is therefore
2547 a potentially serious bug which you should investigate and report. (On
2548 a few systems, meaningful comparison of object files is impossible; they
2549 always appear ``different''. If you encounter this problem, you will
2550 need to disable comparison in the @file{Makefile}.)
2552 If you do not want to bootstrap your compiler, you can configure with
2553 @option{--disable-bootstrap}. In particular cases, you may want to
2554 bootstrap your compiler even if the target system is not the same as
2555 the one you are building on: for example, you could build a
2556 @code{powerpc-unknown-linux-gnu} toolchain on a
2557 @code{powerpc64-unknown-linux-gnu} host. In this case, pass
2558 @option{--enable-bootstrap} to the configure script.
2560 @code{BUILD_CONFIG} can be used to bring in additional customization
2561 to the build. It can be set to a whitespace-separated list of names.
2562 For each such @code{NAME}, top-level @file{config/@code{NAME}.mk} will
2563 be included by the top-level @file{Makefile}, bringing in any settings
2564 it contains. The default @code{BUILD_CONFIG} can be set using the
2565 configure option @option{--with-build-config=@code{NAME}...}. Some
2566 examples of supported build configurations are:
2569 @item @samp{bootstrap-O1}
2570 Removes any @option{-O}-started option from @code{BOOT_CFLAGS}, and adds
2571 @option{-O1} to it. @samp{BUILD_CONFIG=bootstrap-O1} is equivalent to
2572 @samp{BOOT_CFLAGS='-g -O1'}.
2574 @item @samp{bootstrap-O3}
2575 Analogous to @code{bootstrap-O1}.
2577 @item @samp{bootstrap-lto}
2578 Enables Link-Time Optimization for host tools during bootstrapping.
2579 @samp{BUILD_CONFIG=bootstrap-lto} is equivalent to adding
2580 @option{-flto} to @samp{BOOT_CFLAGS}. This option assumes that the host
2581 supports the linker plugin (e.g. GNU ld version 2.21 or later or GNU gold
2582 version 2.21 or later).
2584 @item @samp{bootstrap-lto-noplugin}
2585 This option is similar to @code{bootstrap-lto}, but is intended for
2586 hosts that do not support the linker plugin. Without the linker plugin
2587 static libraries are not compiled with link-time optimizations. Since
2588 the GCC middle end and back end are in @file{libbackend.a} this means
2589 that only the front end is actually LTO optimized.
2591 @item @samp{bootstrap-debug}
2592 Verifies that the compiler generates the same executable code, whether
2593 or not it is asked to emit debug information. To this end, this
2594 option builds stage2 host programs without debug information, and uses
2595 @file{contrib/compare-debug} to compare them with the stripped stage3
2596 object files. If @code{BOOT_CFLAGS} is overridden so as to not enable
2597 debug information, stage2 will have it, and stage3 won't. This option
2598 is enabled by default when GCC bootstrapping is enabled, if
2599 @code{strip} can turn object files compiled with and without debug
2600 info into identical object files. In addition to better test
2601 coverage, this option makes default bootstraps faster and leaner.
2603 @item @samp{bootstrap-debug-big}
2604 Rather than comparing stripped object files, as in
2605 @code{bootstrap-debug}, this option saves internal compiler dumps
2606 during stage2 and stage3 and compares them as well, which helps catch
2607 additional potential problems, but at a great cost in terms of disk
2608 space. It can be specified in addition to @samp{bootstrap-debug}.
2610 @item @samp{bootstrap-debug-lean}
2611 This option saves disk space compared with @code{bootstrap-debug-big},
2612 but at the expense of some recompilation. Instead of saving the dumps
2613 of stage2 and stage3 until the final compare, it uses
2614 @option{-fcompare-debug} to generate, compare and remove the dumps
2615 during stage3, repeating the compilation that already took place in
2616 stage2, whose dumps were not saved.
2618 @item @samp{bootstrap-debug-lib}
2619 This option tests executable code invariance over debug information
2620 generation on target libraries, just like @code{bootstrap-debug-lean}
2621 tests it on host programs. It builds stage3 libraries with
2622 @option{-fcompare-debug}, and it can be used along with any of the
2623 @code{bootstrap-debug} options above.
2625 There aren't @code{-lean} or @code{-big} counterparts to this option
2626 because most libraries are only build in stage3, so bootstrap compares
2627 would not get significant coverage. Moreover, the few libraries built
2628 in stage2 are used in stage3 host programs, so we wouldn't want to
2629 compile stage2 libraries with different options for comparison purposes.
2631 @item @samp{bootstrap-debug-ckovw}
2632 Arranges for error messages to be issued if the compiler built on any
2633 stage is run without the option @option{-fcompare-debug}. This is
2634 useful to verify the full @option{-fcompare-debug} testing coverage. It
2635 must be used along with @code{bootstrap-debug-lean} and
2636 @code{bootstrap-debug-lib}.
2638 @item @samp{bootstrap-time}
2639 Arranges for the run time of each program started by the GCC driver,
2640 built in any stage, to be logged to @file{time.log}, in the top level of
2645 @section Building a cross compiler
2647 When building a cross compiler, it is not generally possible to do a
2648 3-stage bootstrap of the compiler. This makes for an interesting problem
2649 as parts of GCC can only be built with GCC@.
2651 To build a cross compiler, we recommend first building and installing a
2652 native compiler. You can then use the native GCC compiler to build the
2653 cross compiler. The installed native compiler needs to be GCC version
2656 If the cross compiler is to be built with support for the Java
2657 programming language and the ability to compile .java source files is
2658 desired, the installed native compiler used to build the cross
2659 compiler needs to be the same GCC version as the cross compiler. In
2660 addition the cross compiler needs to be configured with
2661 @option{--with-ecj-jar=@dots{}}.
2663 Assuming you have already installed a native copy of GCC and configured
2664 your cross compiler, issue the command @command{make}, which performs the
2669 Build host tools necessary to build the compiler.
2672 Build target tools for use by the compiler such as binutils (bfd,
2673 binutils, gas, gprof, ld, and opcodes)
2674 if they have been individually linked or moved into the top level GCC source
2675 tree before configuring.
2678 Build the compiler (single stage only).
2681 Build runtime libraries using the compiler from the previous step.
2684 Note that if an error occurs in any step the make process will exit.
2686 If you are not building GNU binutils in the same source tree as GCC,
2687 you will need a cross-assembler and cross-linker installed before
2688 configuring GCC@. Put them in the directory
2689 @file{@var{prefix}/@var{target}/bin}. Here is a table of the tools
2690 you should put in this directory:
2694 This should be the cross-assembler.
2697 This should be the cross-linker.
2700 This should be the cross-archiver: a program which can manipulate
2701 archive files (linker libraries) in the target machine's format.
2704 This should be a program to construct a symbol table in an archive file.
2707 The installation of GCC will find these programs in that directory,
2708 and copy or link them to the proper place to for the cross-compiler to
2709 find them when run later.
2711 The easiest way to provide these files is to build the Binutils package.
2712 Configure it with the same @option{--host} and @option{--target}
2713 options that you use for configuring GCC, then build and install
2714 them. They install their executables automatically into the proper
2715 directory. Alas, they do not support all the targets that GCC
2718 If you are not building a C library in the same source tree as GCC,
2719 you should also provide the target libraries and headers before
2720 configuring GCC, specifying the directories with
2721 @option{--with-sysroot} or @option{--with-headers} and
2722 @option{--with-libs}. Many targets also require ``start files'' such
2723 as @file{crt0.o} and
2724 @file{crtn.o} which are linked into each executable. There may be several
2725 alternatives for @file{crt0.o}, for use with profiling or other
2726 compilation options. Check your target's definition of
2727 @code{STARTFILE_SPEC} to find out what start files it uses.
2729 @section Building in parallel
2731 GNU Make 3.80 and above, which is necessary to build GCC, support
2732 building in parallel. To activate this, you can use @samp{make -j 2}
2733 instead of @samp{make}. You can also specify a bigger number, and
2734 in most cases using a value greater than the number of processors in
2735 your machine will result in fewer and shorter I/O latency hits, thus
2736 improving overall throughput; this is especially true for slow drives
2737 and network filesystems.
2739 @section Building the Ada compiler
2741 In order to build GNAT, the Ada compiler, you need a working GNAT
2742 compiler (GCC version 4.0 or later).
2743 This includes GNAT tools such as @command{gnatmake} and
2744 @command{gnatlink}, since the Ada front end is written in Ada and
2745 uses some GNAT-specific extensions.
2747 In order to build a cross compiler, it is suggested to install
2748 the new compiler as native first, and then use it to build the cross
2751 @command{configure} does not test whether the GNAT installation works
2752 and has a sufficiently recent version; if too old a GNAT version is
2753 installed, the build will fail unless @option{--enable-languages} is
2754 used to disable building the Ada front end.
2756 @env{ADA_INCLUDE_PATH} and @env{ADA_OBJECT_PATH} environment variables
2757 must not be set when building the Ada compiler, the Ada tools, or the
2758 Ada runtime libraries. You can check that your build environment is clean
2759 by verifying that @samp{gnatls -v} lists only one explicit path in each
2762 @section Building with profile feedback
2764 It is possible to use profile feedback to optimize the compiler itself. This
2765 should result in a faster compiler binary. Experiments done on x86 using gcc
2766 3.3 showed approximately 7 percent speedup on compiling C programs. To
2767 bootstrap the compiler with profile feedback, use @code{make profiledbootstrap}.
2769 When @samp{make profiledbootstrap} is run, it will first build a @code{stage1}
2770 compiler. This compiler is used to build a @code{stageprofile} compiler
2771 instrumented to collect execution counts of instruction and branch
2772 probabilities. Then runtime libraries are compiled with profile collected.
2773 Finally a @code{stagefeedback} compiler is built using the information collected.
2775 Unlike standard bootstrap, several additional restrictions apply. The
2776 compiler used to build @code{stage1} needs to support a 64-bit integral type.
2777 It is recommended to only use GCC for this.
2779 On Linux/x86_64 hosts with some restrictions (no virtualization) it is
2780 also possible to do autofdo build with @samp{make
2781 autoprofiledback}. This uses Linux perf to sample branches in the
2782 binary and then rebuild it with feedback derived from the profile.
2783 Linux perf and the @code{autofdo} toolkit needs to be installed for
2786 Only the profile from the current build is used, so when an error
2787 occurs it is recommended to clean before restarting. Otherwise
2788 the code quality may be much worse.
2795 @uref{./index.html,,Return to the GCC Installation page}
2799 @c ***Testing*****************************************************************
2801 @comment node-name, next, previous, up
2802 @node Testing, Final install, Building, Installing GCC
2806 @chapter Installing GCC: Testing
2809 @cindex Installing GCC: Testing
2812 Before you install GCC, we encourage you to run the testsuites and to
2813 compare your results with results from a similar configuration that have
2814 been submitted to the
2815 @uref{http://gcc.gnu.org/ml/gcc-testresults/,,gcc-testresults mailing list}.
2816 Some of these archived results are linked from the build status lists
2817 at @uref{http://gcc.gnu.org/buildstat.html}, although not everyone who
2818 reports a successful build runs the testsuites and submits the results.
2819 This step is optional and may require you to download additional software,
2820 but it can give you confidence in your new GCC installation or point out
2821 problems before you install and start using your new GCC@.
2823 First, you must have @uref{download.html,,downloaded the testsuites}.
2824 These are part of the full distribution, but if you downloaded the
2825 ``core'' compiler plus any front ends, you must download the testsuites
2828 Second, you must have the testing tools installed. This includes
2829 @uref{http://www.gnu.org/software/dejagnu/,,DejaGnu}, Tcl, and Expect;
2830 the DejaGnu site has links to these.
2832 If the directories where @command{runtest} and @command{expect} were
2833 installed are not in the @env{PATH}, you may need to set the following
2834 environment variables appropriately, as in the following example (which
2835 assumes that DejaGnu has been installed under @file{/usr/local}):
2838 TCL_LIBRARY = /usr/local/share/tcl8.0
2839 DEJAGNULIBS = /usr/local/share/dejagnu
2842 (On systems such as Cygwin, these paths are required to be actual
2843 paths, not mounts or links; presumably this is due to some lack of
2844 portability in the DejaGnu code.)
2847 Finally, you can run the testsuite (which may take a long time):
2849 cd @var{objdir}; make -k check
2852 This will test various components of GCC, such as compiler
2853 front ends and runtime libraries. While running the testsuite, DejaGnu
2854 might emit some harmless messages resembling
2855 @samp{WARNING: Couldn't find the global config file.} or
2856 @samp{WARNING: Couldn't find tool init file} that can be ignored.
2858 If you are testing a cross-compiler, you may want to run the testsuite
2859 on a simulator as described at @uref{http://gcc.gnu.org/simtest-howto.html}.
2861 @section How can you run the testsuite on selected tests?
2863 In order to run sets of tests selectively, there are targets
2864 @samp{make check-gcc} and language specific @samp{make check-c},
2865 @samp{make check-c++}, @samp{make check-fortran}, @samp{make check-java},
2866 @samp{make check-ada}, @samp{make check-objc}, @samp{make check-obj-c++},
2867 @samp{make check-lto}
2868 in the @file{gcc} subdirectory of the object directory. You can also
2869 just run @samp{make check} in a subdirectory of the object directory.
2872 A more selective way to just run all @command{gcc} execute tests in the
2876 make check-gcc RUNTESTFLAGS="execute.exp @var{other-options}"
2879 Likewise, in order to run only the @command{g++} ``old-deja'' tests in
2880 the testsuite with filenames matching @samp{9805*}, you would use
2883 make check-g++ RUNTESTFLAGS="old-deja.exp=9805* @var{other-options}"
2886 The @file{*.exp} files are located in the testsuite directories of the GCC
2887 source, the most important ones being @file{compile.exp},
2888 @file{execute.exp}, @file{dg.exp} and @file{old-deja.exp}.
2889 To get a list of the possible @file{*.exp} files, pipe the
2890 output of @samp{make check} into a file and look at the
2891 @samp{Running @dots{} .exp} lines.
2893 @section Passing options and running multiple testsuites
2895 You can pass multiple options to the testsuite using the
2896 @samp{--target_board} option of DejaGNU, either passed as part of
2897 @samp{RUNTESTFLAGS}, or directly to @command{runtest} if you prefer to
2898 work outside the makefiles. For example,
2901 make check-g++ RUNTESTFLAGS="--target_board=unix/-O3/-fmerge-constants"
2904 will run the standard @command{g++} testsuites (``unix'' is the target name
2905 for a standard native testsuite situation), passing
2906 @samp{-O3 -fmerge-constants} to the compiler on every test, i.e.,
2907 slashes separate options.
2909 You can run the testsuites multiple times using combinations of options
2910 with a syntax similar to the brace expansion of popular shells:
2913 @dots{}"--target_board=arm-sim\@{-mhard-float,-msoft-float\@}\@{-O1,-O2,-O3,\@}"
2916 (Note the empty option caused by the trailing comma in the final group.)
2917 The following will run each testsuite eight times using the @samp{arm-sim}
2918 target, as if you had specified all possible combinations yourself:
2921 --target_board='arm-sim/-mhard-float/-O1 \
2922 arm-sim/-mhard-float/-O2 \
2923 arm-sim/-mhard-float/-O3 \
2924 arm-sim/-mhard-float \
2925 arm-sim/-msoft-float/-O1 \
2926 arm-sim/-msoft-float/-O2 \
2927 arm-sim/-msoft-float/-O3 \
2928 arm-sim/-msoft-float'
2931 They can be combined as many times as you wish, in arbitrary ways. This
2935 @dots{}"--target_board=unix/-Wextra\@{-O3,-fno-strength\@}\@{-fomit-frame,\@}"
2938 will generate four combinations, all involving @samp{-Wextra}.
2940 The disadvantage to this method is that the testsuites are run in serial,
2941 which is a waste on multiprocessor systems. For users with GNU Make and
2942 a shell which performs brace expansion, you can run the testsuites in
2943 parallel by having the shell perform the combinations and @command{make}
2944 do the parallel runs. Instead of using @samp{--target_board}, use a
2945 special makefile target:
2948 make -j@var{N} check-@var{testsuite}//@var{test-target}/@var{option1}/@var{option2}/@dots{}
2954 make -j3 check-gcc//sh-hms-sim/@{-m1,-m2,-m3,-m3e,-m4@}/@{,-nofpu@}
2957 will run three concurrent ``make-gcc'' testsuites, eventually testing all
2958 ten combinations as described above. Note that this is currently only
2959 supported in the @file{gcc} subdirectory. (To see how this works, try
2960 typing @command{echo} before the example given here.)
2963 @section Additional testing for Java Class Libraries
2965 The Java runtime tests can be executed via @samp{make check}
2966 in the @file{@var{target}/libjava/testsuite} directory in
2969 The @uref{http://sourceware.org/mauve/,,Mauve Project} provides
2970 a suite of tests for the Java Class Libraries. This suite can be run
2971 as part of libgcj testing by placing the Mauve tree within the libjava
2972 testsuite at @file{libjava/testsuite/libjava.mauve/mauve}, or by
2973 specifying the location of that tree when invoking @samp{make}, as in
2974 @samp{make MAUVEDIR=~/mauve check}.
2976 @section How to interpret test results
2978 The result of running the testsuite are various @file{*.sum} and @file{*.log}
2979 files in the testsuite subdirectories. The @file{*.log} files contain a
2980 detailed log of the compiler invocations and the corresponding
2981 results, the @file{*.sum} files summarize the results. These summaries
2982 contain status codes for all tests:
2986 PASS: the test passed as expected
2988 XPASS: the test unexpectedly passed
2990 FAIL: the test unexpectedly failed
2992 XFAIL: the test failed as expected
2994 UNSUPPORTED: the test is not supported on this platform
2996 ERROR: the testsuite detected an error
2998 WARNING: the testsuite detected a possible problem
3001 It is normal for some tests to report unexpected failures. At the
3002 current time the testing harness does not allow fine grained control
3003 over whether or not a test is expected to fail. This problem should
3004 be fixed in future releases.
3007 @section Submitting test results
3009 If you want to report the results to the GCC project, use the
3010 @file{contrib/test_summary} shell script. Start it in the @var{objdir} with
3013 @var{srcdir}/contrib/test_summary -p your_commentary.txt \
3014 -m gcc-testresults@@gcc.gnu.org |sh
3017 This script uses the @command{Mail} program to send the results, so
3018 make sure it is in your @env{PATH}. The file @file{your_commentary.txt} is
3019 prepended to the testsuite summary and should contain any special
3020 remarks you have on your results or your build environment. Please
3021 do not edit the testsuite result block or the subject line, as these
3022 messages may be automatically processed.
3029 @uref{./index.html,,Return to the GCC Installation page}
3033 @c ***Final install***********************************************************
3035 @comment node-name, next, previous, up
3036 @node Final install, , Testing, Installing GCC
3038 @ifset finalinstallhtml
3040 @chapter Installing GCC: Final installation
3043 Now that GCC has been built (and optionally tested), you can install it with
3045 cd @var{objdir} && make install
3048 We strongly recommend to install into a target directory where there is
3049 no previous version of GCC present. Also, the GNAT runtime should not
3050 be stripped, as this would break certain features of the debugger that
3051 depend on this debugging information (catching Ada exceptions for
3054 That step completes the installation of GCC; user level binaries can
3055 be found in @file{@var{prefix}/bin} where @var{prefix} is the value
3056 you specified with the @option{--prefix} to configure (or
3057 @file{/usr/local} by default). (If you specified @option{--bindir},
3058 that directory will be used instead; otherwise, if you specified
3059 @option{--exec-prefix}, @file{@var{exec-prefix}/bin} will be used.)
3060 Headers for the C++ and Java libraries are installed in
3061 @file{@var{prefix}/include}; libraries in @file{@var{libdir}}
3062 (normally @file{@var{prefix}/lib}); internal parts of the compiler in
3063 @file{@var{libdir}/gcc} and @file{@var{libexecdir}/gcc}; documentation
3064 in info format in @file{@var{infodir}} (normally
3065 @file{@var{prefix}/info}).
3067 When installing cross-compilers, GCC's executables
3068 are not only installed into @file{@var{bindir}}, that
3069 is, @file{@var{exec-prefix}/bin}, but additionally into
3070 @file{@var{exec-prefix}/@var{target-alias}/bin}, if that directory
3071 exists. Typically, such @dfn{tooldirs} hold target-specific
3072 binutils, including assembler and linker.
3074 Installation into a temporary staging area or into a @command{chroot}
3075 jail can be achieved with the command
3078 make DESTDIR=@var{path-to-rootdir} install
3082 where @var{path-to-rootdir} is the absolute path of
3083 a directory relative to which all installation paths will be
3084 interpreted. Note that the directory specified by @code{DESTDIR}
3085 need not exist yet; it will be created if necessary.
3087 There is a subtle point with tooldirs and @code{DESTDIR}:
3088 If you relocate a cross-compiler installation with
3089 e.g.@: @samp{DESTDIR=@var{rootdir}}, then the directory
3090 @file{@var{rootdir}/@var{exec-prefix}/@var{target-alias}/bin} will
3091 be filled with duplicated GCC executables only if it already exists,
3092 it will not be created otherwise. This is regarded as a feature,
3093 not as a bug, because it gives slightly more control to the packagers
3094 using the @code{DESTDIR} feature.
3096 You can install stripped programs and libraries with
3102 If you are bootstrapping a released version of GCC then please
3103 quickly review the build status page for your release, available from
3104 @uref{http://gcc.gnu.org/buildstat.html}.
3105 If your system is not listed for the version of GCC that you built,
3107 @email{gcc@@gcc.gnu.org} indicating
3108 that you successfully built and installed GCC@.
3109 Include the following information:
3113 Output from running @file{@var{srcdir}/config.guess}. Do not send
3114 that file itself, just the one-line output from running it.
3117 The output of @samp{gcc -v} for your newly installed @command{gcc}.
3118 This tells us which version of GCC you built and the options you passed to
3122 Whether you enabled all languages or a subset of them. If you used a
3123 full distribution then this information is part of the configure
3124 options in the output of @samp{gcc -v}, but if you downloaded the
3125 ``core'' compiler plus additional front ends then it isn't apparent
3126 which ones you built unless you tell us about it.
3129 If the build was for GNU/Linux, also include:
3132 The distribution name and version (e.g., Red Hat 7.1 or Debian 2.2.3);
3133 this information should be available from @file{/etc/issue}.
3136 The version of the Linux kernel, available from @samp{uname --version}
3140 The version of glibc you used; for RPM-based systems like Red Hat,
3141 Mandrake, and SuSE type @samp{rpm -q glibc} to get the glibc version,
3142 and on systems like Debian and Progeny use @samp{dpkg -l libc6}.
3144 For other systems, you can include similar information if you think it is
3148 Any other information that you think would be useful to people building
3149 GCC on the same configuration. The new entry in the build status list
3150 will include a link to the archived copy of your message.
3153 We'd also like to know if the
3155 @ref{Specific, host/target specific installation notes}
3158 @uref{specific.html,,host/target specific installation notes}
3160 didn't include your host/target information or if that information is
3161 incomplete or out of date. Send a note to
3162 @email{gcc@@gcc.gnu.org} detailing how the information should be changed.
3164 If you find a bug, please report it following the
3165 @uref{../bugs/,,bug reporting guidelines}.
3167 If you want to print the GCC manuals, do @samp{cd @var{objdir}; make
3168 dvi}. You will need to have @command{texi2dvi} (version at least 4.7)
3169 and @TeX{} installed. This creates a number of @file{.dvi} files in
3170 subdirectories of @file{@var{objdir}}; these may be converted for
3171 printing with programs such as @command{dvips}. Alternately, by using
3172 @samp{make pdf} in place of @samp{make dvi}, you can create documentation
3173 in the form of @file{.pdf} files; this requires @command{texi2pdf}, which
3174 is included with Texinfo version 4.8 and later. You can also
3175 @uref{https://shop.fsf.org/,,buy printed manuals from the
3176 Free Software Foundation}, though such manuals may not be for the most
3177 recent version of GCC@.
3179 If you would like to generate online HTML documentation, do @samp{cd
3180 @var{objdir}; make html} and HTML will be generated for the gcc manuals in
3181 @file{@var{objdir}/gcc/HTML}.
3188 @uref{./index.html,,Return to the GCC Installation page}
3192 @c ***Binaries****************************************************************
3194 @comment node-name, next, previous, up
3195 @node Binaries, Specific, Installing GCC, Top
3199 @chapter Installing GCC: Binaries
3202 @cindex Installing GCC: Binaries
3204 We are often asked about pre-compiled versions of GCC@. While we cannot
3205 provide these for all platforms, below you'll find links to binaries for
3206 various platforms where creating them by yourself is not easy due to various
3209 Please note that we did not create these binaries, nor do we
3210 support them. If you have any problems installing them, please
3211 contact their makers.
3218 @uref{http://www.bullfreeware.com,,Bull's Open Source Software Archive for
3219 for AIX 5L and AIX 6};
3222 @uref{http://www.perzl.org/aix/,,AIX Open Source Packages (AIX5L AIX 6.1
3227 DOS---@uref{http://www.delorie.com/djgpp/,,DJGPP}.
3233 @uref{http://hpux.connect.org.uk/,,HP-UX Porting Center};
3237 Solaris 2 (SPARC, Intel):
3240 @uref{http://www.opencsw.org/,,OpenCSW}
3243 @uref{http://jupiterrise.com/tgcware/,,TGCware}
3250 The @uref{http://sourceware.org/cygwin/,,Cygwin} project;
3252 The @uref{http://www.mingw.org/,,MinGW} and
3253 @uref{http://mingw-w64.org/,,mingw-w64} projects.
3257 @uref{http://www.openpkg.org/,,OpenPKG} offers binaries for quite a
3258 number of platforms.
3261 The @uref{http://gcc.gnu.org/wiki/GFortranBinaries,,GFortran Wiki} has
3262 links to GNU Fortran binaries for several platforms.
3270 @uref{./index.html,,Return to the GCC Installation page}
3274 @c ***Specific****************************************************************
3276 @comment node-name, next, previous, up
3277 @node Specific, Old, Binaries, Top
3281 @chapter Host/target specific installation notes for GCC
3284 @cindex Specific installation notes
3285 @cindex Target specific installation
3286 @cindex Host specific installation
3287 @cindex Target specific installation notes
3289 Please read this document carefully @emph{before} installing the
3290 GNU Compiler Collection on your machine.
3292 Note that this list of install notes is @emph{not} a list of supported
3293 hosts or targets. Not all supported hosts and targets are listed
3294 here, only the ones that require host-specific or target-specific
3295 information have to.
3300 @uref{#aarch64-x-x,,aarch64*-*-*}
3302 @uref{#alpha-x-x,,alpha*-*-*}
3304 @uref{#alpha-dec-osf51,,alpha*-dec-osf5.1}
3306 @uref{#amd64-x-solaris210,,amd64-*-solaris2.10}
3308 @uref{#arm-x-eabi,,arm-*-eabi}
3312 @uref{#bfin,,Blackfin}
3316 @uref{#x-x-freebsd,,*-*-freebsd*}
3318 @uref{#h8300-hms,,h8300-hms}
3320 @uref{#hppa-hp-hpux,,hppa*-hp-hpux*}
3322 @uref{#hppa-hp-hpux10,,hppa*-hp-hpux10}
3324 @uref{#hppa-hp-hpux11,,hppa*-hp-hpux11}
3326 @uref{#x-x-linux-gnu,,*-*-linux-gnu}
3328 @uref{#ix86-x-linux,,i?86-*-linux*}
3330 @uref{#ix86-x-solaris210,,i?86-*-solaris2.10}
3332 @uref{#ia64-x-linux,,ia64-*-linux}
3334 @uref{#ia64-x-hpux,,ia64-*-hpux*}
3336 @uref{#x-ibm-aix,,*-ibm-aix*}
3338 @uref{#iq2000-x-elf,,iq2000-*-elf}
3340 @uref{#lm32-x-elf,,lm32-*-elf}
3342 @uref{#lm32-x-uclinux,,lm32-*-uclinux}
3344 @uref{#m32c-x-elf,,m32c-*-elf}
3346 @uref{#m32r-x-elf,,m32r-*-elf}
3348 @uref{#m68k-x-x,,m68k-*-*}
3350 @uref{#m68k-uclinux,,m68k-uclinux}
3352 @uref{#microblaze-x-elf,,microblaze-*-elf}
3354 @uref{#mips-x-x,,mips-*-*}
3356 @uref{#mips-sgi-irix5,,mips-sgi-irix5}
3358 @uref{#mips-sgi-irix6,,mips-sgi-irix6}
3360 @uref{#nds32le-x-elf,,nds32le-*-elf}
3362 @uref{#nds32be-x-elf,,nds32be-*-elf}
3364 @uref{#nvptx-x-none,,nvptx-*-none}
3366 @uref{#powerpc-x-x,,powerpc*-*-*}
3368 @uref{#powerpc-x-darwin,,powerpc-*-darwin*}
3370 @uref{#powerpc-x-elf,,powerpc-*-elf}
3372 @uref{#powerpc-x-linux-gnu,,powerpc*-*-linux-gnu*}
3374 @uref{#powerpc-x-netbsd,,powerpc-*-netbsd*}
3376 @uref{#powerpc-x-eabisim,,powerpc-*-eabisim}
3378 @uref{#powerpc-x-eabi,,powerpc-*-eabi}
3380 @uref{#powerpcle-x-elf,,powerpcle-*-elf}
3382 @uref{#powerpcle-x-eabisim,,powerpcle-*-eabisim}
3384 @uref{#powerpcle-x-eabi,,powerpcle-*-eabi}
3386 @uref{#s390-x-linux,,s390-*-linux*}
3388 @uref{#s390x-x-linux,,s390x-*-linux*}
3390 @uref{#s390x-ibm-tpf,,s390x-ibm-tpf*}
3392 @uref{#x-x-solaris2,,*-*-solaris2*}
3394 @uref{#sparc-x-x,,sparc*-*-*}
3396 @uref{#sparc-sun-solaris2,,sparc-sun-solaris2*}
3398 @uref{#sparc-sun-solaris210,,sparc-sun-solaris2.10}
3400 @uref{#sparc-x-linux,,sparc-*-linux*}
3402 @uref{#sparc64-x-solaris2,,sparc64-*-solaris2*}
3404 @uref{#sparcv9-x-solaris2,,sparcv9-*-solaris2*}
3406 @uref{#c6x-x-x,,c6x-*-*}
3408 @uref{#tilegx-x-linux,,tilegx-*-linux*}
3410 @uref{#tilegxbe-x-linux,,tilegxbe-*-linux*}
3412 @uref{#tilepro-x-linux,,tilepro-*-linux*}
3414 @uref{#visium-x-elf, visium-*-elf}
3416 @uref{#x-x-vxworks,,*-*-vxworks*}
3418 @uref{#x86-64-x-x,,x86_64-*-*, amd64-*-*}
3420 @uref{#x86-64-x-solaris210,,x86_64-*-solaris2.1[0-9]*}
3422 @uref{#xtensa-x-elf,,xtensa*-*-elf}
3424 @uref{#xtensa-x-linux,,xtensa*-*-linux*}
3426 @uref{#windows,,Microsoft Windows}
3428 @uref{#x-x-cygwin,,*-*-cygwin}
3430 @uref{#x-x-mingw32,,*-*-mingw32}
3434 @uref{#older,,Older systems}
3439 @uref{#elf,,all ELF targets} (SVR4, Solaris 2, etc.)
3445 <!-- -------- host/target specific issues start here ---------------- -->
3448 @anchor{aarch64-x-x}
3449 @heading aarch64*-*-*
3450 Binutils pre 2.24 does not have support for selecting @option{-mabi} and
3451 does not support ILP32. If it is used to build GCC 4.9 or later, GCC will
3452 not support option @option{-mabi=ilp32}.
3454 To enable a workaround for the Cortex-A53 erratum number 835769 by default
3455 (for all CPUs regardless of -mcpu option given) at configure time use the
3456 @option{--enable-fix-cortex-a53-835769} option. This will enable the fix by
3457 default and can be explicitly disabled during compilation by passing the
3458 @option{-mno-fix-cortex-a53-835769} option. Conversely,
3459 @option{--disable-fix-cortex-a53-835769} will disable the workaround by
3460 default. The workaround is disabled by default if neither of
3461 @option{--enable-fix-cortex-a53-835769} or
3462 @option{--disable-fix-cortex-a53-835769} is given at configure time.
3464 To enable a workaround for the Cortex-A53 erratum number 843419 by default
3465 (for all CPUs regardless of -mcpu option given) at configure time use the
3466 @option{--enable-fix-cortex-a53-843419} option. This workaround is applied at
3467 link time. Enabling the workaround will cause GCC to pass the relevant option
3468 to the linker. It can be explicitly disabled during compilation by passing the
3469 @option{-mno-fix-cortex-a53-843419} option. Conversely,
3470 @option{--disable-fix-cortex-a53-843419} will disable the workaround by default.
3471 The workaround is disabled by default if neither of
3472 @option{--enable-fix-cortex-a53-843419} or
3473 @option{--disable-fix-cortex-a53-843419} is given at configure time.
3480 This section contains general configuration information for all
3481 alpha-based platforms using ELF (in particular, ignore this section for
3482 DEC OSF/1, Digital UNIX and Tru64 UNIX)@. In addition to reading this
3483 section, please read all other sections that match your target.
3485 We require binutils 2.11.2 or newer.
3486 Previous binutils releases had a number of problems with DWARF 2
3487 debugging information, not the least of which is incorrect linking of
3493 @anchor{alpha-dec-osf51}
3494 @heading alpha*-dec-osf5.1
3495 Systems using processors that implement the DEC Alpha architecture and
3496 are running the DEC/Compaq/HP Unix (DEC OSF/1, Digital UNIX, or Compaq/HP
3497 Tru64 UNIX) operating system, for example the DEC Alpha AXP systems.
3499 Support for Tru64 UNIX V5.1 has been removed in GCC 4.8. As of GCC 4.6,
3500 support for Tru64 UNIX V4.0 and V5.0 has been removed. As of GCC 3.2,
3501 versions before @code{alpha*-dec-osf4} are no longer supported. (These
3502 are the versions which identify themselves as DEC OSF/1.)
3507 @anchor{amd64-x-solaris210}
3508 @heading amd64-*-solaris2.1[0-9]*
3509 This is a synonym for @samp{x86_64-*-solaris2.1[0-9]*}.
3514 @anchor{arc-x-elf32}
3515 @heading arc-*-elf32
3517 Use @samp{configure --target=arc-elf32 --with-cpu=@var{cpu} --enable-languages="c,c++"}
3518 to configure GCC, with @var{cpu} being one of @samp{arc600}, @samp{arc601},
3524 @anchor{arc-linux-uclibc}
3525 @heading arc-linux-uclibc
3527 Use @samp{configure --target=arc-linux-uclibc --with-cpu=arc700 --enable-languages="c,c++"} to configure GCC@.
3534 ARM-family processors. Subtargets that use the ELF object format
3535 require GNU binutils 2.13 or newer. Such subtargets include:
3536 @code{arm-*-netbsdelf}, @code{arm-*-*linux-*}
3537 and @code{arm-*-rtemseabi}.
3539 Building the Ada frontend commonly fails (an infinite loop executing
3540 @code{xsinfo}) if the host compiler is GNAT 4.8. Host compilers built from the
3541 GNAT 4.6, 4.9 or 5 release branches are known to succeed.
3548 ATMEL AVR-family micro controllers. These are used in embedded
3549 applications. There are no standard Unix configurations.
3551 @xref{AVR Options,, AVR Options, gcc, Using the GNU Compiler
3555 See ``AVR Options'' in the main manual
3557 for the list of supported MCU types.
3559 Use @samp{configure --target=avr --enable-languages="c"} to configure GCC@.
3561 Further installation notes and other useful information about AVR tools
3562 can also be obtained from:
3566 @uref{http://www.nongnu.org/avr/,,http://www.nongnu.org/avr/}
3568 @uref{http://www.amelek.gda.pl/avr/,,http://www.amelek.gda.pl/avr/}
3571 We @emph{strongly} recommend using binutils 2.13 or newer.
3573 The following error:
3575 Error: register required
3578 indicates that you should upgrade to a newer version of the binutils.
3585 The Blackfin processor, an Analog Devices DSP.
3587 @xref{Blackfin Options,, Blackfin Options, gcc, Using the GNU Compiler
3591 See ``Blackfin Options'' in the main manual
3594 More information, and a version of binutils with support for this processor,
3595 is available at @uref{http://blackfin.uclinux.org}
3602 The CR16 CompactRISC architecture is a 16-bit architecture. This
3603 architecture is used in embedded applications.
3606 @xref{CR16 Options,, CR16 Options, gcc, Using and Porting the GNU Compiler
3611 See ``CR16 Options'' in the main manual for a list of CR16-specific options.
3614 Use @samp{configure --target=cr16-elf --enable-languages=c,c++} to configure
3615 GCC@ for building a CR16 elf cross-compiler.
3617 Use @samp{configure --target=cr16-uclinux --enable-languages=c,c++} to
3618 configure GCC@ for building a CR16 uclinux cross-compiler.
3625 CRIS is the CPU architecture in Axis Communications ETRAX system-on-a-chip
3626 series. These are used in embedded applications.
3629 @xref{CRIS Options,, CRIS Options, gcc, Using the GNU Compiler
3633 See ``CRIS Options'' in the main manual
3635 for a list of CRIS-specific options.
3637 There are a few different CRIS targets:
3640 Mainly for monolithic embedded systems. Includes a multilib for the
3641 @samp{v10} core used in @samp{ETRAX 100 LX}.
3642 @item cris-axis-linux-gnu
3643 A GNU/Linux port for the CRIS architecture, currently targeting
3644 @samp{ETRAX 100 LX} by default.
3647 For @code{cris-axis-elf} you need binutils 2.11
3648 or newer. For @code{cris-axis-linux-gnu} you need binutils 2.12 or newer.
3650 Pre-packaged tools can be obtained from
3651 @uref{ftp://ftp.axis.com/@/pub/@/axis/@/tools/@/cris/@/compiler-kit/}. More
3652 information about this platform is available at
3653 @uref{http://developer.axis.com/}.
3660 Please have a look at the @uref{binaries.html,,binaries page}.
3662 You cannot install GCC by itself on MSDOS; it will not compile under
3663 any MSDOS compiler except itself. You need to get the complete
3664 compilation package DJGPP, which includes binaries as well as sources,
3665 and includes all the necessary compilation tools and libraries.
3670 @anchor{epiphany-x-elf}
3671 @heading epiphany-*-elf
3673 This configuration is intended for embedded systems.
3678 @anchor{x-x-freebsd}
3679 @heading *-*-freebsd*
3680 Support for FreeBSD 1 was discontinued in GCC 3.2. Support for
3681 FreeBSD 2 (and any mutant a.out variants of FreeBSD 3) was
3682 discontinued in GCC 4.0.
3684 In order to better utilize FreeBSD base system functionality and match
3685 the configuration of the system compiler, GCC 4.5 and above as well as
3686 GCC 4.4 past 2010-06-20 leverage SSP support in libc (which is present
3687 on FreeBSD 7 or later) and the use of @code{__cxa_atexit} by default
3688 (on FreeBSD 6 or later). The use of @code{dl_iterate_phdr} inside
3689 @file{libgcc_s.so.1} and boehm-gc (on FreeBSD 7 or later) is enabled
3690 by GCC 4.5 and above.
3692 We support FreeBSD using the ELF file format with DWARF 2 debugging
3693 for all CPU architectures. You may use @option{-gstabs} instead of
3694 @option{-g}, if you really want the old debugging format. There are
3695 no known issues with mixing object files and libraries with different
3696 debugging formats. Otherwise, this release of GCC should now match
3697 more of the configuration used in the stock FreeBSD configuration of
3698 GCC@. In particular, @option{--enable-threads} is now configured by
3699 default. However, as a general user, do not attempt to replace the
3700 system compiler with this release. Known to bootstrap and check with
3701 good results on FreeBSD 7.2-STABLE@. In the past, known to bootstrap
3702 and check with good results on FreeBSD 3.0, 3.4, 4.0, 4.2, 4.3, 4.4,
3703 4.5, 4.8, 4.9 and 5-CURRENT@.
3705 The version of binutils installed in @file{/usr/bin} probably works
3706 with this release of GCC@. Bootstrapping against the latest GNU
3707 binutils and/or the version found in @file{/usr/ports/devel/binutils} has
3708 been known to enable additional features and improve overall testsuite
3709 results. However, it is currently known that boehm-gc (which itself
3710 is required for java) may not configure properly on FreeBSD prior to
3711 the FreeBSD 7.0 release with GNU binutils after 2.16.1.
3719 This configuration is intended for embedded systems.
3726 Renesas H8/300 series of processors.
3728 Please have a look at the @uref{binaries.html,,binaries page}.
3730 The calling convention and structure layout has changed in release 2.6.
3731 All code must be recompiled. The calling convention now passes the
3732 first three arguments in function calls in registers. Structures are no
3733 longer a multiple of 2 bytes.
3738 @anchor{hppa-hp-hpux}
3739 @heading hppa*-hp-hpux*
3740 Support for HP-UX version 9 and older was discontinued in GCC 3.4.
3742 We require using gas/binutils on all hppa platforms. Version 2.19 or
3743 later is recommended.
3745 It may be helpful to configure GCC with the
3746 @uref{./configure.html#with-gnu-as,,@option{--with-gnu-as}} and
3747 @option{--with-as=@dots{}} options to ensure that GCC can find GAS@.
3749 The HP assembler should not be used with GCC. It is rarely tested and may
3750 not work. It shouldn't be used with any languages other than C due to its
3753 Specifically, @option{-g} does not work (HP-UX uses a peculiar debugging
3754 format which GCC does not know about). It also inserts timestamps
3755 into each object file it creates, causing the 3-stage comparison test to
3756 fail during a bootstrap. You should be able to continue by saying
3757 @samp{make all-host all-target} after getting the failure from @samp{make}.
3759 Various GCC features are not supported. For example, it does not support weak
3760 symbols or alias definitions. As a result, explicit template instantiations
3761 are required when using C++. This makes it difficult if not impossible to
3762 build many C++ applications.
3764 There are two default scheduling models for instructions. These are
3765 PROCESSOR_7100LC and PROCESSOR_8000. They are selected from the pa-risc
3766 architecture specified for the target machine when configuring.
3767 PROCESSOR_8000 is the default. PROCESSOR_7100LC is selected when
3768 the target is a @samp{hppa1*} machine.
3770 The PROCESSOR_8000 model is not well suited to older processors. Thus,
3771 it is important to completely specify the machine architecture when
3772 configuring if you want a model other than PROCESSOR_8000. The macro
3773 TARGET_SCHED_DEFAULT can be defined in BOOT_CFLAGS if a different
3774 default scheduling model is desired.
3776 As of GCC 4.0, GCC uses the UNIX 95 namespace for HP-UX 10.10
3777 through 11.00, and the UNIX 98 namespace for HP-UX 11.11 and later.
3778 This namespace change might cause problems when bootstrapping with
3779 an earlier version of GCC or the HP compiler as essentially the same
3780 namespace is required for an entire build. This problem can be avoided
3781 in a number of ways. With HP cc, @env{UNIX_STD} can be set to @samp{95}
3782 or @samp{98}. Another way is to add an appropriate set of predefines
3783 to @env{CC}. The description for the @option{munix=} option contains
3784 a list of the predefines used with each standard.
3786 More specific information to @samp{hppa*-hp-hpux*} targets follows.
3791 @anchor{hppa-hp-hpux10}
3792 @heading hppa*-hp-hpux10
3793 For hpux10.20, we @emph{highly} recommend you pick up the latest sed patch
3794 @code{PHCO_19798} from HP@.
3796 The C++ ABI has changed incompatibly in GCC 4.0. COMDAT subspaces are
3797 used for one-only code and data. This resolves many of the previous
3798 problems in using C++ on this target. However, the ABI is not compatible
3799 with the one implemented under HP-UX 11 using secondary definitions.
3804 @anchor{hppa-hp-hpux11}
3805 @heading hppa*-hp-hpux11
3806 GCC 3.0 and up support HP-UX 11. GCC 2.95.x is not supported and cannot
3807 be used to compile GCC 3.0 and up.
3809 The libffi and libjava libraries haven't been ported to 64-bit HP-UX@
3812 Refer to @uref{binaries.html,,binaries} for information about obtaining
3813 precompiled GCC binaries for HP-UX@. Precompiled binaries must be obtained
3814 to build the Ada language as it can't be bootstrapped using C@. Ada is
3815 only available for the 32-bit PA-RISC runtime.
3817 Starting with GCC 3.4 an ISO C compiler is required to bootstrap. The
3818 bundled compiler supports only traditional C; you will need either HP's
3819 unbundled compiler, or a binary distribution of GCC@.
3821 It is possible to build GCC 3.3 starting with the bundled HP compiler,
3822 but the process requires several steps. GCC 3.3 can then be used to
3823 build later versions. The fastjar program contains ISO C code and
3824 can't be built with the HP bundled compiler. This problem can be
3825 avoided by not building the Java language. For example, use the
3826 @option{--enable-languages="c,c++,f77,objc"} option in your configure
3829 There are several possible approaches to building the distribution.
3830 Binutils can be built first using the HP tools. Then, the GCC
3831 distribution can be built. The second approach is to build GCC
3832 first using the HP tools, then build binutils, then rebuild GCC@.
3833 There have been problems with various binary distributions, so it
3834 is best not to start from a binary distribution.
3836 On 64-bit capable systems, there are two distinct targets. Different
3837 installation prefixes must be used if both are to be installed on
3838 the same system. The @samp{hppa[1-2]*-hp-hpux11*} target generates code
3839 for the 32-bit PA-RISC runtime architecture and uses the HP linker.
3840 The @samp{hppa64-hp-hpux11*} target generates 64-bit code for the
3841 PA-RISC 2.0 architecture.
3843 The script config.guess now selects the target type based on the compiler
3844 detected during configuration. You must define @env{PATH} or @env{CC} so
3845 that configure finds an appropriate compiler for the initial bootstrap.
3846 When @env{CC} is used, the definition should contain the options that are
3847 needed whenever @env{CC} is used.
3849 Specifically, options that determine the runtime architecture must be
3850 in @env{CC} to correctly select the target for the build. It is also
3851 convenient to place many other compiler options in @env{CC}. For example,
3852 @env{CC="cc -Ac +DA2.0W -Wp,-H16376 -D_CLASSIC_TYPES -D_HPUX_SOURCE"}
3853 can be used to bootstrap the GCC 3.3 branch with the HP compiler in
3854 64-bit K&R/bundled mode. The @option{+DA2.0W} option will result in
3855 the automatic selection of the @samp{hppa64-hp-hpux11*} target. The
3856 macro definition table of cpp needs to be increased for a successful
3857 build with the HP compiler. _CLASSIC_TYPES and _HPUX_SOURCE need to
3858 be defined when building with the bundled compiler, or when using the
3859 @option{-Ac} option. These defines aren't necessary with @option{-Ae}.
3861 It is best to explicitly configure the @samp{hppa64-hp-hpux11*} target
3862 with the @option{--with-ld=@dots{}} option. This overrides the standard
3863 search for ld. The two linkers supported on this target require different
3864 commands. The default linker is determined during configuration. As a
3865 result, it's not possible to switch linkers in the middle of a GCC build.
3866 This has been reported to sometimes occur in unified builds of binutils
3869 A recent linker patch must be installed for the correct operation of
3870 GCC 3.3 and later. @code{PHSS_26559} and @code{PHSS_24304} are the
3871 oldest linker patches that are known to work. They are for HP-UX
3872 11.00 and 11.11, respectively. @code{PHSS_24303}, the companion to
3873 @code{PHSS_24304}, might be usable but it hasn't been tested. These
3874 patches have been superseded. Consult the HP patch database to obtain
3875 the currently recommended linker patch for your system.
3877 The patches are necessary for the support of weak symbols on the
3878 32-bit port, and for the running of initializers and finalizers. Weak
3879 symbols are implemented using SOM secondary definition symbols. Prior
3880 to HP-UX 11, there are bugs in the linker support for secondary symbols.
3881 The patches correct a problem of linker core dumps creating shared
3882 libraries containing secondary symbols, as well as various other
3883 linking issues involving secondary symbols.
3885 GCC 3.3 uses the ELF DT_INIT_ARRAY and DT_FINI_ARRAY capabilities to
3886 run initializers and finalizers on the 64-bit port. The 32-bit port
3887 uses the linker @option{+init} and @option{+fini} options for the same
3888 purpose. The patches correct various problems with the +init/+fini
3889 options, including program core dumps. Binutils 2.14 corrects a
3890 problem on the 64-bit port resulting from HP's non-standard use of
3891 the .init and .fini sections for array initializers and finalizers.
3893 Although the HP and GNU linkers are both supported for the
3894 @samp{hppa64-hp-hpux11*} target, it is strongly recommended that the
3895 HP linker be used for link editing on this target.
3897 At this time, the GNU linker does not support the creation of long
3898 branch stubs. As a result, it can't successfully link binaries
3899 containing branch offsets larger than 8 megabytes. In addition,
3900 there are problems linking shared libraries, linking executables
3901 with @option{-static}, and with dwarf2 unwind and exception support.
3902 It also doesn't provide stubs for internal calls to global functions
3903 in shared libraries, so these calls can't be overloaded.
3905 The HP dynamic loader does not support GNU symbol versioning, so symbol
3906 versioning is not supported. It may be necessary to disable symbol
3907 versioning with @option{--disable-symvers} when using GNU ld.
3909 POSIX threads are the default. The optional DCE thread library is not
3910 supported, so @option{--enable-threads=dce} does not work.
3915 @anchor{x-x-linux-gnu}
3916 @heading *-*-linux-gnu
3917 Versions of libstdc++-v3 starting with 3.2.1 require bug fixes present
3918 in glibc 2.2.5 and later. More information is available in the
3919 libstdc++-v3 documentation.
3924 @anchor{ix86-x-linux}
3925 @heading i?86-*-linux*
3926 As of GCC 3.3, binutils 2.13.1 or later is required for this platform.
3927 See @uref{http://gcc.gnu.org/PR10877,,bug 10877} for more information.
3929 If you receive Signal 11 errors when building on GNU/Linux, then it is
3930 possible you have a hardware problem. Further information on this can be
3931 found on @uref{http://www.bitwizard.nl/sig11/,,www.bitwizard.nl}.
3936 @anchor{ix86-x-solaris210}
3937 @heading i?86-*-solaris2.10
3938 Use this for Solaris 10 or later on x86 and x86-64 systems. Starting
3939 with GCC 4.7, there is also a 64-bit @samp{amd64-*-solaris2.1[0-9]*} or
3940 @samp{x86_64-*-solaris2.1[0-9]*} configuration that corresponds to
3941 @samp{sparcv9-sun-solaris2*}.
3943 It is recommended that you configure GCC to use the GNU assembler. The
3944 versions included in Solaris 10, from GNU binutils 2.15 (in
3945 @file{/usr/sfw/bin/gas}), and Solaris 11, from GNU binutils 2.19 or
3946 newer (also available as @file{/usr/bin/gas} and
3947 @file{/usr/gnu/bin/as}), work fine. Please note that the current
3948 version, from GNU binutils 2.26, only works on Solaris 12 when using the
3949 Solaris linker. On Solaris 10 and 11, you either have to wait for GNU
3950 binutils 2.26.1 or newer, or stay with GNU binutils 2.25.1. Recent
3951 versions of the Solaris assembler in @file{/usr/ccs/bin/as} work almost
3953 @c FIXME: as patch requirements?
3955 For linking, the Solaris linker, is preferred. If you want to use the GNU
3956 linker instead, note that due to a packaging bug the version in Solaris
3957 10, from GNU binutils 2.15 (in @file{/usr/sfw/bin/gld}), cannot be used,
3958 while the version in Solaris 11, from GNU binutils 2.19 or newer (also
3959 in @file{/usr/gnu/bin/ld} and @file{/usr/bin/gld}), works, as does the
3960 latest version, from GNU binutils 2.26.
3962 To use GNU @command{as}, configure with the options
3963 @option{--with-gnu-as --with-as=@//usr/@/sfw/@/bin/@/gas}. It may be necessary
3964 to configure with @option{--without-gnu-ld --with-ld=@//usr/@/ccs/@/bin/@/ld} to
3965 guarantee use of Sun @command{ld}.
3966 @c FIXME: why --without-gnu-ld --with-ld?
3971 @anchor{ia64-x-linux}
3972 @heading ia64-*-linux
3973 IA-64 processor (also known as IPF, or Itanium Processor Family)
3976 If you are using the installed system libunwind library with
3977 @option{--with-system-libunwind}, then you must use libunwind 0.98 or
3980 None of the following versions of GCC has an ABI that is compatible
3981 with any of the other versions in this list, with the exception that
3982 Red Hat 2.96 and Trillian 000171 are compatible with each other:
3983 3.1, 3.0.2, 3.0.1, 3.0, Red Hat 2.96, and Trillian 000717.
3984 This primarily affects C++ programs and programs that create shared libraries.
3985 GCC 3.1 or later is recommended for compiling linux, the kernel.
3986 As of version 3.1 GCC is believed to be fully ABI compliant, and hence no
3987 more major ABI changes are expected.
3992 @anchor{ia64-x-hpux}
3993 @heading ia64-*-hpux*
3994 Building GCC on this target requires the GNU Assembler. The bundled HP
3995 assembler will not work. To prevent GCC from using the wrong assembler,
3996 the option @option{--with-gnu-as} may be necessary.
3998 The GCC libunwind library has not been ported to HPUX@. This means that for
3999 GCC versions 3.2.3 and earlier, @option{--enable-libunwind-exceptions}
4000 is required to build GCC@. For GCC 3.3 and later, this is the default.
4001 For gcc 3.4.3 and later, @option{--enable-libunwind-exceptions} is
4002 removed and the system libunwind library will always be used.
4006 <!-- rs6000-ibm-aix*, powerpc-ibm-aix* -->
4010 Support for AIX version 3 and older was discontinued in GCC 3.4.
4011 Support for AIX version 4.2 and older was discontinued in GCC 4.5.
4013 ``out of memory'' bootstrap failures may indicate a problem with
4014 process resource limits (ulimit). Hard limits are configured in the
4015 @file{/etc/security/limits} system configuration file.
4017 GCC 4.9 and above require a C++ compiler for bootstrap. IBM VAC++ / xlC
4018 cannot bootstrap GCC. xlc can bootstrap an older version of GCC and
4019 G++ can bootstrap recent releases of GCC.
4021 GCC can bootstrap with recent versions of IBM XLC, but bootstrapping
4022 with an earlier release of GCC is recommended. Bootstrapping with XLC
4023 requires a larger data segment, which can be enabled through the
4024 @var{LDR_CNTRL} environment variable, e.g.,
4027 % LDR_CNTRL=MAXDATA=0x50000000
4031 One can start with a pre-compiled version of GCC to build from
4032 sources. One may delete GCC's ``fixed'' header files when starting
4033 with a version of GCC built for an earlier release of AIX.
4035 To speed up the configuration phases of bootstrapping and installing GCC,
4036 one may use GNU Bash instead of AIX @command{/bin/sh}, e.g.,
4039 % CONFIG_SHELL=/opt/freeware/bin/bash
4040 % export CONFIG_SHELL
4043 and then proceed as described in @uref{build.html,,the build
4044 instructions}, where we strongly recommend specifying an absolute path
4045 to invoke @var{srcdir}/configure.
4047 Because GCC on AIX is built as a 32-bit executable by default,
4048 (although it can generate 64-bit programs) the GMP and MPFR libraries
4049 required by gfortran must be 32-bit libraries. Building GMP and MPFR
4050 as static archive libraries works better than shared libraries.
4052 Errors involving @code{alloca} when building GCC generally are due
4053 to an incorrect definition of @code{CC} in the Makefile or mixing files
4054 compiled with the native C compiler and GCC@. During the stage1 phase of
4055 the build, the native AIX compiler @strong{must} be invoked as @command{cc}
4056 (not @command{xlc}). Once @command{configure} has been informed of
4057 @command{xlc}, one needs to use @samp{make distclean} to remove the
4058 configure cache files and ensure that @env{CC} environment variable
4059 does not provide a definition that will confuse @command{configure}.
4060 If this error occurs during stage2 or later, then the problem most likely
4061 is the version of Make (see above).
4063 The native @command{as} and @command{ld} are recommended for
4064 bootstrapping on AIX@. The GNU Assembler, GNU Linker, and GNU
4065 Binutils version 2.20 is the minimum level that supports bootstrap on
4066 AIX 5@. The GNU Assembler has not been updated to support AIX 6@ or
4067 AIX 7. The native AIX tools do interoperate with GCC@.
4069 AIX 7.1 added partial support for DWARF debugging, but full support
4070 requires AIX 7.1 TL03 SP7 that supports additional DWARF sections and
4071 fixes a bug in the assembler. AIX 7.1 TL03 SP5 distributed a version
4072 of libm.a missing important symbols; a fix for IV77796 will be
4075 AIX 5.3 TL10, AIX 6.1 TL05 and AIX 7.1 TL00 introduced an AIX
4076 assembler change that sometimes produces corrupt assembly files
4077 causing AIX linker errors. The bug breaks GCC bootstrap on AIX and
4078 can cause compilation failures with existing GCC installations. An
4079 AIX iFix for AIX 5.3 is available (APAR IZ98385 for AIX 5.3 TL10, APAR
4080 IZ98477 for AIX 5.3 TL11 and IZ98134 for AIX 5.3 TL12). AIX 5.3 TL11 SP8,
4081 AIX 5.3 TL12 SP5, AIX 6.1 TL04 SP11, AIX 6.1 TL05 SP7, AIX 6.1 TL06 SP6,
4082 AIX 6.1 TL07 and AIX 7.1 TL01 should include the fix.
4084 Building @file{libstdc++.a} requires a fix for an AIX Assembler bug
4085 APAR IY26685 (AIX 4.3) or APAR IY25528 (AIX 5.1). It also requires a
4086 fix for another AIX Assembler bug and a co-dependent AIX Archiver fix
4087 referenced as APAR IY53606 (AIX 5.2) or as APAR IY54774 (AIX 5.1)
4089 @anchor{TransferAixShobj}
4090 @samp{libstdc++} in GCC 3.4 increments the major version number of the
4091 shared object and GCC installation places the @file{libstdc++.a}
4092 shared library in a common location which will overwrite the and GCC
4093 3.3 version of the shared library. Applications either need to be
4094 re-linked against the new shared library or the GCC 3.1 and GCC 3.3
4095 versions of the @samp{libstdc++} shared object needs to be available
4096 to the AIX runtime loader. The GCC 3.1 @samp{libstdc++.so.4}, if
4097 present, and GCC 3.3 @samp{libstdc++.so.5} shared objects can be
4098 installed for runtime dynamic loading using the following steps to set
4099 the @samp{F_LOADONLY} flag in the shared object for @emph{each}
4100 multilib @file{libstdc++.a} installed:
4102 Extract the shared objects from the currently installed
4103 @file{libstdc++.a} archive:
4105 % ar -x libstdc++.a libstdc++.so.4 libstdc++.so.5
4108 Enable the @samp{F_LOADONLY} flag so that the shared object will be
4109 available for runtime dynamic loading, but not linking:
4111 % strip -e libstdc++.so.4 libstdc++.so.5
4114 Archive the runtime-only shared object in the GCC 3.4
4115 @file{libstdc++.a} archive:
4117 % ar -q libstdc++.a libstdc++.so.4 libstdc++.so.5
4121 @uref{./configure.html#WithAixSoname,,@option{--with-aix-soname=svr4}}
4122 configure option may drop the need for this procedure for libraries that
4125 Linking executables and shared libraries may produce warnings of
4126 duplicate symbols. The assembly files generated by GCC for AIX always
4127 have included multiple symbol definitions for certain global variable
4128 and function declarations in the original program. The warnings should
4129 not prevent the linker from producing a correct library or runnable
4132 AIX 4.3 utilizes a ``large format'' archive to support both 32-bit and
4133 64-bit object modules. The routines provided in AIX 4.3.0 and AIX 4.3.1
4134 to parse archive libraries did not handle the new format correctly.
4135 These routines are used by GCC and result in error messages during
4136 linking such as ``not a COFF file''. The version of the routines shipped
4137 with AIX 4.3.1 should work for a 32-bit environment. The @option{-g}
4138 option of the archive command may be used to create archives of 32-bit
4139 objects using the original ``small format''. A correct version of the
4140 routines is shipped with AIX 4.3.2 and above.
4142 Some versions of the AIX binder (linker) can fail with a relocation
4143 overflow severe error when the @option{-bbigtoc} option is used to link
4144 GCC-produced object files into an executable that overflows the TOC@. A fix
4145 for APAR IX75823 (OVERFLOW DURING LINK WHEN USING GCC AND -BBIGTOC) is
4146 available from IBM Customer Support and from its
4147 @uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com}
4148 website as PTF U455193.
4150 The AIX 4.3.2.1 linker (bos.rte.bind_cmds Level 4.3.2.1) will dump core
4151 with a segmentation fault when invoked by any version of GCC@. A fix for
4152 APAR IX87327 is available from IBM Customer Support and from its
4153 @uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com}
4154 website as PTF U461879. This fix is incorporated in AIX 4.3.3 and above.
4156 The initial assembler shipped with AIX 4.3.0 generates incorrect object
4157 files. A fix for APAR IX74254 (64BIT DISASSEMBLED OUTPUT FROM COMPILER FAILS
4158 TO ASSEMBLE/BIND) is available from IBM Customer Support and from its
4159 @uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com}
4160 website as PTF U453956. This fix is incorporated in AIX 4.3.1 and above.
4162 AIX provides National Language Support (NLS)@. Compilers and assemblers
4163 use NLS to support locale-specific representations of various data
4164 formats including floating-point numbers (e.g., @samp{.} vs @samp{,} for
4165 separating decimal fractions). There have been problems reported where
4166 GCC does not produce the same floating-point formats that the assembler
4167 expects. If one encounters this problem, set the @env{LANG}
4168 environment variable to @samp{C} or @samp{En_US}.
4170 A default can be specified with the @option{-mcpu=@var{cpu_type}}
4171 switch and using the configure option @option{--with-cpu-@var{cpu_type}}.
4176 @anchor{iq2000-x-elf}
4177 @heading iq2000-*-elf
4178 Vitesse IQ2000 processors. These are used in embedded
4179 applications. There are no standard Unix configurations.
4186 Lattice Mico32 processor.
4187 This configuration is intended for embedded systems.
4192 @anchor{lm32-x-uclinux}
4193 @heading lm32-*-uclinux
4194 Lattice Mico32 processor.
4195 This configuration is intended for embedded systems running uClinux.
4202 Renesas M32C processor.
4203 This configuration is intended for embedded systems.
4210 Renesas M32R processor.
4211 This configuration is intended for embedded systems.
4219 @samp{m68k-*-elf*}, @samp{m68k-*-rtems}, @samp{m68k-*-uclinux} and
4221 build libraries for both M680x0 and ColdFire processors. If you only
4222 need the M680x0 libraries, you can omit the ColdFire ones by passing
4223 @option{--with-arch=m68k} to @command{configure}. Alternatively, you
4224 can omit the M680x0 libraries by passing @option{--with-arch=cf} to
4225 @command{configure}. These targets default to 5206 or 5475 code as
4226 appropriate for the target system when
4227 configured with @option{--with-arch=cf} and 68020 code otherwise.
4229 The @samp{m68k-*-netbsd} and
4230 @samp{m68k-*-openbsd} targets also support the @option{--with-arch}
4231 option. They will generate ColdFire CFV4e code when configured with
4232 @option{--with-arch=cf} and 68020 code otherwise.
4234 You can override the default processors listed above by configuring
4235 with @option{--with-cpu=@var{target}}. This @var{target} can either
4236 be a @option{-mcpu} argument or one of the following values:
4237 @samp{m68000}, @samp{m68010}, @samp{m68020}, @samp{m68030},
4238 @samp{m68040}, @samp{m68060}, @samp{m68020-40} and @samp{m68020-60}.
4240 GCC requires at least binutils version 2.17 on these targets.
4245 @anchor{m68k-x-uclinux}
4246 @heading m68k-*-uclinux
4247 GCC 4.3 changed the uClinux configuration so that it uses the
4248 @samp{m68k-linux-gnu} ABI rather than the @samp{m68k-elf} ABI.
4249 It also added improved support for C++ and flat shared libraries,
4250 both of which were ABI changes.
4255 @anchor{microblaze-x-elf}
4256 @heading microblaze-*-elf
4257 Xilinx MicroBlaze processor.
4258 This configuration is intended for embedded systems.
4265 If on a MIPS system you get an error message saying ``does not have gp
4266 sections for all it's [sic] sectons [sic]'', don't worry about it. This
4267 happens whenever you use GAS with the MIPS linker, but there is not
4268 really anything wrong, and it is okay to use the output file. You can
4269 stop such warnings by installing the GNU linker.
4271 It would be nice to extend GAS to produce the gp tables, but they are
4272 optional, and there should not be a warning about their absence.
4274 The libstdc++ atomic locking routines for MIPS targets requires MIPS II
4275 and later. A patch went in just after the GCC 3.3 release to
4276 make @samp{mips*-*-*} use the generic implementation instead. You can also
4277 configure for @samp{mipsel-elf} as a workaround. The
4278 @samp{mips*-*-linux*} target continues to use the MIPS II routines. More
4279 work on this is expected in future releases.
4281 @c If you make --with-llsc the default for another target, please also
4282 @c update the description of the --with-llsc option.
4284 The built-in @code{__sync_*} functions are available on MIPS II and
4285 later systems and others that support the @samp{ll}, @samp{sc} and
4286 @samp{sync} instructions. This can be overridden by passing
4287 @option{--with-llsc} or @option{--without-llsc} when configuring GCC.
4288 Since the Linux kernel emulates these instructions if they are
4289 missing, the default for @samp{mips*-*-linux*} targets is
4290 @option{--with-llsc}. The @option{--with-llsc} and
4291 @option{--without-llsc} configure options may be overridden at compile
4292 time by passing the @option{-mllsc} or @option{-mno-llsc} options to
4295 MIPS systems check for division by zero (unless
4296 @option{-mno-check-zero-division} is passed to the compiler) by
4297 generating either a conditional trap or a break instruction. Using
4298 trap results in smaller code, but is only supported on MIPS II and
4299 later. Also, some versions of the Linux kernel have a bug that
4300 prevents trap from generating the proper signal (@code{SIGFPE}). To enable
4301 the use of break, use the @option{--with-divide=breaks}
4302 @command{configure} option when configuring GCC@. The default is to
4303 use traps on systems that support them.
4305 The assembler from GNU binutils 2.17 and earlier has a bug in the way
4306 it sorts relocations for REL targets (o32, o64, EABI). This can cause
4307 bad code to be generated for simple C++ programs. Also the linker
4308 from GNU binutils versions prior to 2.17 has a bug which causes the
4309 runtime linker stubs in very large programs, like @file{libgcj.so}, to
4310 be incorrectly generated. GNU Binutils 2.18 and later (and snapshots
4311 made after Nov. 9, 2006) should be free from both of these problems.
4316 @anchor{mips-sgi-irix5}
4317 @heading mips-sgi-irix5
4318 Support for IRIX 5 has been removed in GCC 4.6.
4323 @anchor{mips-sgi-irix6}
4324 @heading mips-sgi-irix6
4325 Support for IRIX 6.5 has been removed in GCC 4.8. Support for IRIX 6
4326 releases before 6.5 has been removed in GCC 4.6, as well as support for
4332 @anchor{moxie-x-elf}
4333 @heading moxie-*-elf
4334 The moxie processor.
4339 @anchor{msp430-x-elf}
4340 @heading msp430-*-elf
4341 TI MSP430 processor.
4342 This configuration is intended for embedded systems.
4347 @anchor{nds32le-x-elf}
4348 @heading nds32le-*-elf
4349 Andes NDS32 target in little endian mode.
4354 @anchor{nds32be-x-elf}
4355 @heading nds32be-*-elf
4356 Andes NDS32 target in big endian mode.
4361 @anchor{nvptx-x-none}
4362 @heading nvptx-*-none
4365 Instead of GNU binutils, you will need to install
4366 @uref{https://github.com/MentorEmbedded/nvptx-tools/,,nvptx-tools}.
4367 Tell GCC where to find it:
4368 @option{--with-build-time-tools=[install-nvptx-tools]/nvptx-none/bin}.
4370 A nvptx port of newlib is available at
4371 @uref{https://github.com/MentorEmbedded/nvptx-newlib/,,nvptx-newlib}.
4372 It can be automatically built together with GCC@. For this, add a
4373 symbolic link to nvptx-newlib's @file{newlib} directory to the
4374 directory containing the GCC sources.
4376 Use the @option{--disable-sjlj-exceptions} and
4377 @option{--enable-newlib-io-long-long} options when configuring.
4382 @anchor{powerpc-x-x}
4383 @heading powerpc-*-*
4384 You can specify a default version for the @option{-mcpu=@var{cpu_type}}
4385 switch by using the configure option @option{--with-cpu-@var{cpu_type}}.
4388 @uref{ftp://ftp.kernel.org/pub/linux/devel/binutils,,binutils 2.15}
4389 or newer for a working GCC@.
4394 @anchor{powerpc-x-darwin}
4395 @heading powerpc-*-darwin*
4396 PowerPC running Darwin (Mac OS X kernel).
4398 Pre-installed versions of Mac OS X may not include any developer tools,
4399 meaning that you will not be able to build GCC from source. Tool
4400 binaries are available at
4401 @uref{http://opensource.apple.com/}.
4403 This version of GCC requires at least cctools-590.36. The
4404 cctools-590.36 package referenced from
4405 @uref{http://gcc.gnu.org/ml/gcc/2006-03/msg00507.html} will not work
4406 on systems older than 10.3.9 (aka darwin7.9.0).
4411 @anchor{powerpc-x-elf}
4412 @heading powerpc-*-elf
4413 PowerPC system in big endian mode, running System V.4.
4418 @anchor{powerpc-x-linux-gnu}
4419 @heading powerpc*-*-linux-gnu*
4420 PowerPC system in big endian mode running Linux.
4425 @anchor{powerpc-x-netbsd}
4426 @heading powerpc-*-netbsd*
4427 PowerPC system in big endian mode running NetBSD@.
4432 @anchor{powerpc-x-eabisim}
4433 @heading powerpc-*-eabisim
4434 Embedded PowerPC system in big endian mode for use in running under the
4440 @anchor{powerpc-x-eabi}
4441 @heading powerpc-*-eabi
4442 Embedded PowerPC system in big endian mode.
4447 @anchor{powerpcle-x-elf}
4448 @heading powerpcle-*-elf
4449 PowerPC system in little endian mode, running System V.4.
4454 @anchor{powerpcle-x-eabisim}
4455 @heading powerpcle-*-eabisim
4456 Embedded PowerPC system in little endian mode for use in running under
4462 @anchor{powerpcle-x-eabi}
4463 @heading powerpcle-*-eabi
4464 Embedded PowerPC system in little endian mode.
4471 The Renesas RL78 processor.
4472 This configuration is intended for embedded systems.
4479 The Renesas RX processor. See
4480 @uref{http://eu.renesas.com/fmwk.jsp?cnt=rx600_series_landing.jsp&fp=/products/mpumcu/rx_family/rx600_series}
4481 for more information about this processor.
4486 @anchor{s390-x-linux}
4487 @heading s390-*-linux*
4488 S/390 system running GNU/Linux for S/390@.
4493 @anchor{s390x-x-linux}
4494 @heading s390x-*-linux*
4495 zSeries system (64-bit) running GNU/Linux for zSeries@.
4500 @anchor{s390x-ibm-tpf}
4501 @heading s390x-ibm-tpf*
4502 zSeries system (64-bit) running TPF@. This platform is
4503 supported as cross-compilation target only.
4508 @c Please use Solaris 2 to refer to all release of Solaris, starting
4509 @c with 2.0 until 2.6, 7, 8, etc. Solaris 1 was a marketing name for
4510 @c SunOS 4 releases which we don't use to avoid confusion. Solaris
4511 @c alone is too unspecific and must be avoided.
4512 @anchor{x-x-solaris2}
4513 @heading *-*-solaris2*
4514 Support for Solaris 9 has been removed in GCC 4.10. Support for Solaris
4515 8 has been removed in GCC 4.8. Support for Solaris 7 has been removed
4518 Sun does not ship a C compiler with Solaris 2 before Solaris 10, though
4519 you can download the Sun Studio compilers for free. In Solaris 10 and
4520 11, GCC 3.4.3 is available as @command{/usr/sfw/bin/gcc}. Solaris 11
4521 also provides GCC 4.5.2, 4.7.3, and 4.8.2 as
4522 @command{/usr/gcc/4.5/bin/gcc} or similar. Alternatively,
4523 you can install a pre-built GCC to bootstrap and install GCC. See the
4524 @uref{binaries.html,,binaries page} for details.
4526 The Solaris 2 @command{/bin/sh} will often fail to configure
4527 @samp{libstdc++-v3}, @samp{boehm-gc} or @samp{libjava}. We therefore
4528 recommend using the following initial sequence of commands
4531 % CONFIG_SHELL=/bin/ksh
4532 % export CONFIG_SHELL
4536 and proceed as described in @uref{configure.html,,the configure instructions}.
4537 In addition we strongly recommend specifying an absolute path to invoke
4538 @command{@var{srcdir}/configure}.
4540 Solaris 10 comes with a number of optional OS packages. Some of these
4541 are needed to use GCC fully, namely @code{SUNWarc},
4542 @code{SUNWbtool}, @code{SUNWesu}, @code{SUNWhea}, @code{SUNWlibm},
4543 @code{SUNWsprot}, and @code{SUNWtoo}. If you did not install all
4544 optional packages when installing Solaris 10, you will need to verify that
4545 the packages that GCC needs are installed.
4546 To check whether an optional package is installed, use
4547 the @command{pkginfo} command. To add an optional package, use the
4548 @command{pkgadd} command. For further details, see the Solaris 10
4551 Starting with Solaris 11, the package management has changed, so you
4552 need to check for @code{system/header}, @code{system/linker}, and
4553 @code{developer/assembler} packages. Checking for and installing
4554 packages is done with the @command{pkg} command now.
4556 Trying to use the linker and other tools in
4557 @file{/usr/ucb} to install GCC has been observed to cause trouble.
4558 For example, the linker may hang indefinitely. The fix is to remove
4559 @file{/usr/ucb} from your @env{PATH}.
4561 The build process works more smoothly with the legacy Sun tools so, if you
4562 have @file{/usr/xpg4/bin} in your @env{PATH}, we recommend that you place
4563 @file{/usr/bin} before @file{/usr/xpg4/bin} for the duration of the build.
4565 We recommend the use of the Solaris assembler or the GNU assembler, in
4566 conjunction with the Solaris linker. The GNU @command{as}
4567 versions included in Solaris 10, from GNU binutils 2.15 (in
4568 @file{/usr/sfw/bin/gas}), and Solaris 11,
4569 from GNU binutils 2.19 or newer (also in @file{/usr/bin/gas} and
4570 @file{/usr/gnu/bin/as}), are known to work.
4571 Current versions of GNU binutils (2.26)
4572 are known to work as well, with the caveat mentioned in
4573 @uref{#ix86-x-solaris210,,i?86-*-solaris2.10} . Note that your mileage may vary
4574 if you use a combination of the GNU tools and the Solaris tools: while the
4575 combination GNU @command{as} + Sun @command{ld} should reasonably work,
4576 the reverse combination Sun @command{as} + GNU @command{ld} may fail to
4577 build or cause memory corruption at runtime in some cases for C++ programs.
4579 GNU @command{ld} usually works as well, although the version included in
4580 Solaris 10 cannot be used due to several bugs. Again, the current
4581 version (2.26) is known to work, but generally lacks platform specific
4582 features, so better stay with Solaris @command{ld}. To use the LTO linker
4583 plugin (@option{-fuse-linker-plugin}) with GNU @command{ld}, GNU
4584 binutils @emph{must} be configured with @option{--enable-largefile}.
4586 To enable symbol versioning in @samp{libstdc++} with Solaris @command{ld},
4587 you need to have any version of GNU @command{c++filt}, which is part of
4588 GNU binutils. @samp{libstdc++} symbol versioning will be disabled if no
4589 appropriate version is found. Solaris @command{c++filt} from the Solaris Studio
4590 compilers does @emph{not} work.
4592 Sun bug 4927647 sometimes causes random spurious testsuite failures
4593 related to missing diagnostic output. This bug doesn't affect GCC
4594 itself, rather it is a kernel bug triggered by the @command{expect}
4595 program which is used only by the GCC testsuite driver. When the bug
4596 causes the @command{expect} program to miss anticipated output, extra
4597 testsuite failures appear.
4604 This section contains general configuration information for all
4605 SPARC-based platforms. In addition to reading this section, please
4606 read all other sections that match your target.
4608 Newer versions of the GNU Multiple Precision Library (GMP), the MPFR
4609 library and the MPC library are known to be miscompiled by earlier
4610 versions of GCC on these platforms. We therefore recommend the use
4611 of the exact versions of these libraries listed as minimal versions
4612 in @uref{prerequisites.html,,the prerequisites}.
4617 @anchor{sparc-sun-solaris2}
4618 @heading sparc-sun-solaris2*
4619 When GCC is configured to use GNU binutils 2.14 or later, the binaries
4620 produced are smaller than the ones produced using Sun's native tools;
4621 this difference is quite significant for binaries containing debugging
4624 Starting with Solaris 7, the operating system is capable of executing
4625 64-bit SPARC V9 binaries. GCC 3.1 and later properly supports
4626 this; the @option{-m64} option enables 64-bit code generation.
4627 However, if all you want is code tuned for the UltraSPARC CPU, you
4628 should try the @option{-mtune=ultrasparc} option instead, which produces
4629 code that, unlike full 64-bit code, can still run on non-UltraSPARC
4632 When configuring the GNU Multiple Precision Library (GMP), the MPFR
4633 library or the MPC library on a Solaris 7 or later system, the canonical
4634 target triplet must be specified as the @command{build} parameter on the
4635 configure line. This target triplet can be obtained by invoking @command{./config.guess} in the toplevel source directory of GCC (and
4636 not that of GMP or MPFR or MPC). For example on a Solaris 9 system:
4639 % ./configure --build=sparc-sun-solaris2.9 --prefix=xxx
4645 @anchor{sparc-sun-solaris210}
4646 @heading sparc-sun-solaris2.10
4647 There is a bug in older versions of the Sun assembler which breaks
4648 thread-local storage (TLS). A typical error message is
4651 ld: fatal: relocation error: R_SPARC_TLS_LE_HIX22: file /var/tmp//ccamPA1v.o:
4652 symbol <unknown>: bad symbol type SECT: symbol type must be TLS
4656 This bug is fixed in Sun patch 118683-03 or later.
4661 @anchor{sparc-x-linux}
4662 @heading sparc-*-linux*
4664 GCC versions 3.0 and higher require binutils 2.11.2 and glibc 2.2.4
4665 or newer on this platform. All earlier binutils and glibc
4666 releases mishandled unaligned relocations on @code{sparc-*-*} targets.
4672 @anchor{sparc64-x-solaris2}
4673 @heading sparc64-*-solaris2*
4674 When configuring the GNU Multiple Precision Library (GMP), the MPFR
4675 library or the MPC library, the canonical target triplet must be specified
4676 as the @command{build} parameter on the configure line. For example
4677 on a Solaris 9 system:
4680 % ./configure --build=sparc64-sun-solaris2.9 --prefix=xxx
4686 @anchor{sparcv9-x-solaris2}
4687 @heading sparcv9-*-solaris2*
4688 This is a synonym for @samp{sparc64-*-solaris2*}.
4695 The C6X family of processors. This port requires binutils-2.22 or newer.
4700 @anchor{tilegx-*-linux}
4701 @heading tilegx-*-linux*
4702 The TILE-Gx processor in little endian mode, running GNU/Linux. This
4703 port requires binutils-2.22 or newer.
4708 @anchor{tilegxbe-*-linux}
4709 @heading tilegxbe-*-linux*
4710 The TILE-Gx processor in big endian mode, running GNU/Linux. This
4711 port requires binutils-2.23 or newer.
4716 @anchor{tilepro-*-linux}
4717 @heading tilepro-*-linux*
4718 The TILEPro processor running GNU/Linux. This port requires
4719 binutils-2.22 or newer.
4724 @anchor{visium-x-elf}
4725 @heading visium-*-elf
4726 CDS VISIUMcore processor.
4727 This configuration is intended for embedded systems.
4732 @anchor{x-x-vxworks}
4733 @heading *-*-vxworks*
4734 Support for VxWorks is in flux. At present GCC supports @emph{only} the
4735 very recent VxWorks 5.5 (aka Tornado 2.2) release, and only on PowerPC@.
4736 We welcome patches for other architectures supported by VxWorks 5.5.
4737 Support for VxWorks AE would also be welcome; we believe this is merely
4738 a matter of writing an appropriate ``configlette'' (see below). We are
4739 not interested in supporting older, a.out or COFF-based, versions of
4742 VxWorks comes with an older version of GCC installed in
4743 @file{@var{$WIND_BASE}/host}; we recommend you do not overwrite it.
4744 Choose an installation @var{prefix} entirely outside @var{$WIND_BASE}.
4745 Before running @command{configure}, create the directories @file{@var{prefix}}
4746 and @file{@var{prefix}/bin}. Link or copy the appropriate assembler,
4747 linker, etc.@: into @file{@var{prefix}/bin}, and set your @var{PATH} to
4748 include that directory while running both @command{configure} and
4751 You must give @command{configure} the
4752 @option{--with-headers=@var{$WIND_BASE}/target/h} switch so that it can
4753 find the VxWorks system headers. Since VxWorks is a cross compilation
4754 target only, you must also specify @option{--target=@var{target}}.
4755 @command{configure} will attempt to create the directory
4756 @file{@var{prefix}/@var{target}/sys-include} and copy files into it;
4757 make sure the user running @command{configure} has sufficient privilege
4760 GCC's exception handling runtime requires a special ``configlette''
4761 module, @file{contrib/gthr_supp_vxw_5x.c}. Follow the instructions in
4762 that file to add the module to your kernel build. (Future versions of
4763 VxWorks will incorporate this module.)
4769 @heading x86_64-*-*, amd64-*-*
4770 GCC supports the x86-64 architecture implemented by the AMD64 processor
4771 (amd64-*-* is an alias for x86_64-*-*) on GNU/Linux, FreeBSD and NetBSD@.
4772 On GNU/Linux the default is a bi-arch compiler which is able to generate
4773 both 64-bit x86-64 and 32-bit x86 code (via the @option{-m32} switch).
4778 @anchor{x86-64-x-solaris210}
4779 @heading x86_64-*-solaris2.1[0-9]*
4780 GCC also supports the x86-64 architecture implemented by the AMD64
4781 processor (@samp{amd64-*-*} is an alias for @samp{x86_64-*-*}) on
4782 Solaris 10 or later. Unlike other systems, without special options a
4783 bi-arch compiler is built which generates 32-bit code by default, but
4784 can generate 64-bit x86-64 code with the @option{-m64} switch. Since
4785 GCC 4.7, there is also a configuration that defaults to 64-bit code, but
4786 can generate 32-bit code with @option{-m32}. To configure and build
4787 this way, you have to provide all support libraries like @file{libgmp}
4788 as 64-bit code, configure with @option{--target=x86_64-pc-solaris2.1x}
4789 and @samp{CC=gcc -m64}.
4794 @anchor{xtensa-x-elf}
4795 @heading xtensa*-*-elf
4796 This target is intended for embedded Xtensa systems using the
4797 @samp{newlib} C library. It uses ELF but does not support shared
4798 objects. Designed-defined instructions specified via the
4799 Tensilica Instruction Extension (TIE) language are only supported
4800 through inline assembly.
4802 The Xtensa configuration information must be specified prior to
4803 building GCC@. The @file{include/xtensa-config.h} header
4804 file contains the configuration information. If you created your
4805 own Xtensa configuration with the Xtensa Processor Generator, the
4806 downloaded files include a customized copy of this header file,
4807 which you can use to replace the default header file.
4812 @anchor{xtensa-x-linux}
4813 @heading xtensa*-*-linux*
4814 This target is for Xtensa systems running GNU/Linux. It supports ELF
4815 shared objects and the GNU C library (glibc). It also generates
4816 position-independent code (PIC) regardless of whether the
4817 @option{-fpic} or @option{-fPIC} options are used. In other
4818 respects, this target is the same as the
4819 @uref{#xtensa*-*-elf,,@samp{xtensa*-*-elf}} target.
4825 @heading Microsoft Windows
4827 @subheading Intel 16-bit versions
4828 The 16-bit versions of Microsoft Windows, such as Windows 3.1, are not
4831 However, the 32-bit port has limited support for Microsoft
4832 Windows 3.11 in the Win32s environment, as a target only. See below.
4834 @subheading Intel 32-bit versions
4835 The 32-bit versions of Windows, including Windows 95, Windows NT, Windows
4836 XP, and Windows Vista, are supported by several different target
4837 platforms. These targets differ in which Windows subsystem they target
4838 and which C libraries are used.
4841 @item Cygwin @uref{#x-x-cygwin,,*-*-cygwin}: Cygwin provides a user-space
4842 Linux API emulation layer in the Win32 subsystem.
4843 @item MinGW @uref{#x-x-mingw32,,*-*-mingw32}: MinGW is a native GCC port for
4844 the Win32 subsystem that provides a subset of POSIX.
4845 @item MKS i386-pc-mks: NuTCracker from MKS. See
4846 @uref{http://www.mkssoftware.com/} for more information.
4849 @subheading Intel 64-bit versions
4850 GCC contains support for x86-64 using the mingw-w64
4851 runtime library, available from @uref{http://mingw-w64.sourceforge.net/}.
4852 This library should be used with the target triple x86_64-pc-mingw32.
4854 Presently Windows for Itanium is not supported.
4856 @subheading Windows CE
4857 Windows CE is supported as a target only on Hitachi
4858 SuperH (sh-wince-pe), and MIPS (mips-wince-pe).
4860 @subheading Other Windows Platforms
4861 GCC no longer supports Windows NT on the Alpha or PowerPC.
4863 GCC no longer supports the Windows POSIX subsystem. However, it does
4864 support the Interix subsystem. See above.
4866 Old target names including *-*-winnt and *-*-windowsnt are no longer used.
4868 PW32 (i386-pc-pw32) support was never completed, and the project seems to
4869 be inactive. See @uref{http://pw32.sourceforge.net/} for more information.
4871 UWIN support has been removed due to a lack of maintenance.
4878 Ports of GCC are included with the
4879 @uref{http://www.cygwin.com/,,Cygwin environment}.
4881 GCC will build under Cygwin without modification; it does not build
4882 with Microsoft's C++ compiler and there are no plans to make it do so.
4884 The Cygwin native compiler can be configured to target any 32-bit x86
4885 cpu architecture desired; the default is i686-pc-cygwin. It should be
4886 used with as up-to-date a version of binutils as possible; use either
4887 the latest official GNU binutils release in the Cygwin distribution,
4888 or version 2.20 or above if building your own.
4893 @anchor{x-x-mingw32}
4894 @heading *-*-mingw32
4895 GCC will build with and support only MinGW runtime 3.12 and later.
4896 Earlier versions of headers are incompatible with the new default semantics
4897 of @code{extern inline} in @code{-std=c99} and @code{-std=gnu99} modes.
4903 @heading Older systems
4904 GCC contains support files for many older (1980s and early
4905 1990s) Unix variants. For the most part, support for these systems
4906 has not been deliberately removed, but it has not been maintained for
4907 several years and may suffer from bitrot.
4909 Starting with GCC 3.1, each release has a list of ``obsoleted'' systems.
4910 Support for these systems is still present in that release, but
4911 @command{configure} will fail unless the @option{--enable-obsolete}
4912 option is given. Unless a maintainer steps forward, support for these
4913 systems will be removed from the next release of GCC@.
4915 Support for old systems as hosts for GCC can cause problems if the
4916 workarounds for compiler, library and operating system bugs affect the
4917 cleanliness or maintainability of the rest of GCC@. In some cases, to
4918 bring GCC up on such a system, if still possible with current GCC, may
4919 require first installing an old version of GCC which did work on that
4920 system, and using it to compile a more recent GCC, to avoid bugs in the
4921 vendor compiler. Old releases of GCC 1 and GCC 2 are available in the
4922 @file{old-releases} directory on the @uref{../mirrors.html,,GCC mirror
4923 sites}. Header bugs may generally be avoided using
4924 @command{fixincludes}, but bugs or deficiencies in libraries and the
4925 operating system may still cause problems.
4927 Support for older systems as targets for cross-compilation is less
4928 problematic than support for them as hosts for GCC; if an enthusiast
4929 wishes to make such a target work again (including resurrecting any of
4930 the targets that never worked with GCC 2, starting from the last
4931 version before they were removed), patches
4932 @uref{../contribute.html,,following the usual requirements} would be
4933 likely to be accepted, since they should not affect the support for more
4936 For some systems, old versions of GNU binutils may also be useful,
4937 and are available from @file{pub/binutils/old-releases} on
4938 @uref{http://sourceware.org/mirrors.html,,sourceware.org mirror sites}.
4940 Some of the information on specific systems above relates to
4941 such older systems, but much of the information
4942 about GCC on such systems (which may no longer be applicable to
4943 current GCC) is to be found in the GCC texinfo manual.
4949 @heading all ELF targets (SVR4, Solaris 2, etc.)
4950 C++ support is significantly better on ELF targets if you use the
4951 @uref{./configure.html#with-gnu-ld,,GNU linker}; duplicate copies of
4952 inlines, vtables and template instantiations will be discarded
4961 @uref{./index.html,,Return to the GCC Installation page}
4965 @c ***Old documentation******************************************************
4967 @include install-old.texi
4973 @uref{./index.html,,Return to the GCC Installation page}
4977 @c ***GFDL********************************************************************
4985 @uref{./index.html,,Return to the GCC Installation page}
4989 @c ***************************************************************************
4990 @c Part 6 The End of the Document
4992 @comment node-name, next, previous, up
4993 @node Concept Index, , GNU Free Documentation License, Top
4997 @unnumbered Concept Index