1\input texinfo.tex @c -*-texinfo-*- 2@c @ifnothtml 3@c %**start of header 4@setfilename gccinstall.info 5@settitle Installing GCC 6@setchapternewpage odd 7@c %**end of header 8@c @end ifnothtml 9 10@include gcc-common.texi 11 12@c Specify title for specific html page 13@ifset indexhtml 14@settitle Installing GCC 15@end ifset 16@ifset specifichtml 17@settitle Host/Target specific installation notes for GCC 18@end ifset 19@ifset prerequisiteshtml 20@settitle Prerequisites for GCC 21@end ifset 22@ifset downloadhtml 23@settitle Downloading GCC 24@end ifset 25@ifset configurehtml 26@settitle Installing GCC: Configuration 27@end ifset 28@ifset buildhtml 29@settitle Installing GCC: Building 30@end ifset 31@ifset testhtml 32@settitle Installing GCC: Testing 33@end ifset 34@ifset finalinstallhtml 35@settitle Installing GCC: Final installation 36@end ifset 37@ifset binarieshtml 38@settitle Installing GCC: Binaries 39@end ifset 40@ifset oldhtml 41@settitle Installing GCC: Old documentation 42@end ifset 43@ifset gfdlhtml 44@settitle Installing GCC: GNU Free Documentation License 45@end ifset 46 47@c Copyright (C) 1988-2015 Free Software Foundation, Inc. 48@c *** Converted to texinfo by Dean Wakerley, dean@wakerley.com 49 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. 52@c 53@c Do not use @footnote{} in this file as it breaks install.texi2html! 54 55@c Include everything if we're not making html 56@ifnothtml 57@set indexhtml 58@set specifichtml 59@set prerequisiteshtml 60@set downloadhtml 61@set configurehtml 62@set buildhtml 63@set testhtml 64@set finalinstallhtml 65@set binarieshtml 66@set oldhtml 67@set gfdlhtml 68@end ifnothtml 69 70@c Part 2 Summary Description and Copyright 71@copying 72Copyright @copyright{} 1988-2015 Free Software Foundation, Inc. 73@sp 1 74Permission is granted to copy, distribute and/or modify this document 75under the terms of the GNU Free Documentation License, Version 1.3 or 76any later version published by the Free Software Foundation; with no 77Invariant Sections, the Front-Cover texts being (a) (see below), and 78with the Back-Cover Texts being (b) (see below). A copy of the 79license is included in the section entitled ``@uref{./gfdl.html,,GNU 80Free Documentation License}''. 81 82(a) The FSF's Front-Cover Text is: 83 84 A GNU Manual 85 86(b) The FSF's Back-Cover Text is: 87 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. 91@end copying 92@ifinfo 93@insertcopying 94@end ifinfo 95@dircategory Software development 96@direntry 97* gccinstall: (gccinstall). Installing the GNU Compiler Collection. 98@end direntry 99 100@c Part 3 Titlepage and Copyright 101@titlepage 102@title Installing GCC 103@versionsubtitle 104 105@c The following two commands start the copyright page. 106@page 107@vskip 0pt plus 1filll 108@insertcopying 109@end titlepage 110 111@c Part 4 Top node, Master Menu, and/or Table of Contents 112@ifinfo 113@node Top, , , (dir) 114@comment node-name, next, Previous, up 115 116@menu 117* Installing GCC:: This document describes the generic installation 118 procedure for GCC as well as detailing some target 119 specific installation instructions. 120 121* Specific:: Host/target specific installation notes for GCC. 122* Binaries:: Where to get pre-compiled binaries. 123 124* Old:: Old installation documentation. 125 126* GNU Free Documentation License:: How you can copy and share this manual. 127* Concept Index:: This index has two entries. 128@end menu 129@end ifinfo 130 131@iftex 132@contents 133@end iftex 134 135@c Part 5 The Body of the Document 136@c ***Installing GCC********************************************************** 137@ifnothtml 138@comment node-name, next, previous, up 139@node Installing GCC, Binaries, , Top 140@end ifnothtml 141@ifset indexhtml 142@ifnothtml 143@chapter Installing GCC 144@end ifnothtml 145 146The latest version of this document is always available at 147@uref{http://gcc.gnu.org/install/,,http://gcc.gnu.org/install/}. 148It refers to the current development sources, instructions for 149specific released versions are included with the sources. 150 151This document describes the generic installation procedure for GCC as well 152as detailing some target specific installation instructions. 153 154GCC includes several components that previously were separate distributions 155with their own installation instructions. This document supersedes all 156package-specific installation instructions. 157 158@emph{Before} starting the build/install procedure please check the 159@ifnothtml 160@ref{Specific, host/target specific installation notes}. 161@end ifnothtml 162@ifhtml 163@uref{specific.html,,host/target specific installation notes}. 164@end ifhtml 165We recommend you browse the entire generic installation instructions before 166you proceed. 167 168Lists of successful builds for released versions of GCC are 169available at @uref{http://gcc.gnu.org/buildstat.html}. 170These lists are updated as new information becomes available. 171 172The installation procedure itself is broken into five steps. 173 174@ifinfo 175@menu 176* Prerequisites:: 177* Downloading the source:: 178* Configuration:: 179* Building:: 180* Testing:: (optional) 181* Final install:: 182@end menu 183@end ifinfo 184@ifhtml 185@enumerate 186@item 187@uref{prerequisites.html,,Prerequisites} 188@item 189@uref{download.html,,Downloading the source} 190@item 191@uref{configure.html,,Configuration} 192@item 193@uref{build.html,,Building} 194@item 195@uref{test.html,,Testing} (optional) 196@item 197@uref{finalinstall.html,,Final install} 198@end enumerate 199@end ifhtml 200 201Please note that GCC does not support @samp{make uninstall} and probably 202won't do so in the near future as this would open a can of worms. Instead, 203we suggest that you install GCC into a directory of its own and simply 204remove that directory when you do not need that specific version of GCC 205any longer, and, if shared libraries are installed there as well, no 206more binaries exist that use them. 207 208@ifhtml 209There are also some @uref{old.html,,old installation instructions}, 210which are mostly obsolete but still contain some information which has 211not yet been merged into the main part of this manual. 212@end ifhtml 213 214@html 215<hr /> 216<p> 217@end html 218@ifhtml 219@uref{./index.html,,Return to the GCC Installation page} 220 221@insertcopying 222@end ifhtml 223@end ifset 224 225@c ***Prerequisites************************************************** 226@ifnothtml 227@comment node-name, next, previous, up 228@node Prerequisites, Downloading the source, , Installing GCC 229@end ifnothtml 230@ifset prerequisiteshtml 231@ifnothtml 232@chapter Prerequisites 233@end ifnothtml 234@cindex Prerequisites 235 236GCC requires that various tools and packages be available for use in the 237build procedure. Modifying GCC sources requires additional tools 238described below. 239 240@heading Tools/packages necessary for building GCC 241@table @asis 242@item ISO C++98 compiler 243Necessary to bootstrap GCC, although versions of GCC prior 244to 4.8 also allow bootstrapping with a ISO C89 compiler and versions 245of GCC prior to 3.4 also allow bootstrapping with a traditional 246(K&R) C compiler. 247 248To build all languages in a cross-compiler or other configuration where 2493-stage bootstrap is not performed, you need to start with an existing 250GCC binary (version 3.4 or later) because source code for language 251frontends other than C might use GCC extensions. 252 253Note that to bootstrap GCC with versions of GCC earlier than 3.4, you 254may need to use @option{--disable-stage1-checking}, though 255bootstrapping the compiler with such earlier compilers is strongly 256discouraged. 257 258@item C standard library and headers 259 260In order to build GCC, the C standard library and headers must be present 261for all target variants for which target libraries will be built (and not 262only the variant of the host C++ compiler). 263 264This affects the popular @samp{x86_64-unknown-linux-gnu} platform (among 265other 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 267build of a native compiler on @samp{x86_64-unknown-linux-gnu}, make sure you 268either have the 32-bit libc developer package properly installed (the exact 269name of the package depends on your distro) or you must build GCC as a 27064-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} 273 274@item GNAT 275 276In order to build the Ada compiler (GNAT) you must already have GNAT 277installed because portions of the Ada frontend are written in Ada (with 278GNAT extensions.) Refer to the Ada installation instructions for more 279specific information. 280 281@item A ``working'' POSIX compatible shell, or GNU bash 282 283Necessary when running @command{configure} because some 284@command{/bin/sh} shells have bugs and may crash when configuring the 285target libraries. In other cases, @command{/bin/sh} or @command{ksh} 286have disastrous corner-case performance problems. This 287can cause target @command{configure} runs to literally take days to 288complete in some cases. 289 290So on some platforms @command{/bin/ksh} is sufficient, on others it 291isn't. See the host/target specific instructions for your platform, or 292use @command{bash} to be sure. Then set @env{CONFIG_SHELL} in your 293environment to your ``good'' shell prior to running 294@command{configure}/@command{make}. 295 296@command{zsh} is not a fully compliant POSIX shell and will not 297work when configuring GCC@. 298 299@item A POSIX or SVR4 awk 300 301Necessary for creating some of the generated source files for GCC@. 302If in doubt, use a recent GNU awk version, as some of the older ones 303are broken. GNU awk version 3.1.5 is known to work. 304 305@item GNU binutils 306 307Necessary in some circumstances, optional in others. See the 308host/target specific instructions for your platform for the exact 309requirements. 310 311@item gzip version 1.2.4 (or later) or 312@itemx bzip2 version 1.0.2 (or later) 313 314Necessary to uncompress GCC @command{tar} files when source code is 315obtained via FTP mirror sites. 316 317@item GNU make version 3.80 (or later) 318 319You must have GNU make installed to build GCC@. 320 321@item GNU tar version 1.14 (or later) 322 323Necessary (only on some platforms) to untar the source code. Many 324systems' @command{tar} programs will also work, only try GNU 325@command{tar} if you have problems. 326 327@item Perl version 5.6.1 (or later) 328 329Necessary when targeting Darwin, building @samp{libstdc++}, 330and not using @option{--disable-symvers}. 331Necessary when targeting Solaris 2 with Sun @command{ld} and not using 332@option{--disable-symvers}. The bundled @command{perl} in Solaris@tie{}8 333and up works. 334 335Necessary when regenerating @file{Makefile} dependencies in libiberty. 336Necessary when regenerating @file{libiberty/functions.texi}. 337Necessary when generating manpages from Texinfo manuals. 338Used by various scripts to generate some files included in SVN (mainly 339Unicode-related and rarely changing) from source tables. 340 341@item @command{jar}, or InfoZIP (@command{zip} and @command{unzip}) 342 343Necessary to build libgcj, the GCJ runtime. 344 345@end table 346 347Several support libraries are necessary to build GCC, some are required, 348others optional. While any sufficiently new version of required tools 349usually work, library requirements are generally stricter. Newer 350versions may work in some cases, but it's safer to use the exact 351versions documented. We appreciate bug reports about problems with 352newer versions, though. If your OS vendor provides packages for the 353support libraries then using those packages may be the simplest way to 354install the libraries. 355 356@table @asis 357@item GNU Multiple Precision Library (GMP) version 4.3.2 (or later) 358 359Necessary to build GCC@. If a GMP source distribution is found in a 360subdirectory of your GCC sources named @file{gmp}, it will be built 361together with GCC. Alternatively, if GMP is already installed but it 362is 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} 364and @option{--with-gmp-include}. 365 366@item MPFR Library version 2.4.2 (or later) 367 368Necessary to build GCC@. It can be downloaded from 369@uref{http://www.mpfr.org/}. If an MPFR source distribution is found 370in a subdirectory of your GCC sources named @file{mpfr}, it will be 371built together with GCC. Alternatively, if MPFR is already installed 372but it is not in your default library search path, the 373@option{--with-mpfr} configure option should be used. See also 374@option{--with-mpfr-lib} and @option{--with-mpfr-include}. 375 376@item MPC Library version 0.8.1 (or later) 377 378Necessary to build GCC@. It can be downloaded from 379@uref{http://www.multiprecision.org/}. If an MPC source distribution 380is found in a subdirectory of your GCC sources named @file{mpc}, it 381will be built together with GCC. Alternatively, if MPC is already 382installed but it is not in your default library search path, the 383@option{--with-mpc} configure option should be used. See also 384@option{--with-mpc-lib} and @option{--with-mpc-include}. 385 386@item ISL Library version 0.14 (or 0.12.2) 387 388Necessary to build GCC with the Graphite loop optimizations. 389It can be downloaded from @uref{ftp://gcc.gnu.org/pub/gcc/infrastructure/} 390as @file{isl-0.12.2.tar.bz2}. If an ISL source distribution is found 391in a subdirectory of your GCC sources named @file{isl}, it will be 392built together with GCC. Alternatively, the @option{--with-isl} configure 393option should be used if ISL is not installed in your default library 394search path. 395 396@end table 397 398@heading Tools/packages necessary for modifying GCC 399@table @asis 400@item autoconf version 2.64 401@itemx GNU m4 version 1.4.6 (or later) 402 403Necessary when modifying @file{configure.ac}, @file{aclocal.m4}, etc.@: 404to regenerate @file{configure} and @file{config.in} files. 405 406@item automake version 1.11.1 407 408Necessary when modifying a @file{Makefile.am} file to regenerate its 409associated @file{Makefile.in}. 410 411Much of GCC does not use automake, so directly edit the @file{Makefile.in} 412file. Specifically this applies to the @file{gcc}, @file{intl}, 413@file{libcpp}, @file{libiberty}, @file{libobjc} directories as well 414as any of their subdirectories. 415 416For directories that use automake, GCC requires the latest release in 417the 1.11 series, which is currently 1.11.1. When regenerating a directory 418to a newer version, please update all the directories using an older 1.11 419to the latest released version. 420 421@item gettext version 0.14.5 (or later) 422 423Needed to regenerate @file{gcc.pot}. 424 425@item gperf version 2.7.2 (or later) 426 427Necessary when modifying @command{gperf} input files, e.g.@: 428@file{gcc/cp/cfns.gperf} to regenerate its associated header file, e.g.@: 429@file{gcc/cp/cfns.h}. 430 431@item DejaGnu 1.4.4 432@itemx Expect 433@itemx Tcl 434 435Necessary to run the GCC testsuite; see the section on testing for 436details. Tcl 8.6 has a known regression in RE pattern handling that 437make parts of the testsuite fail. See 438@uref{http://core.tcl.tk/tcl/tktview/267b7e2334ee2e9de34c4b00d6e72e2f1997085f} 439for more information. This bug has been fixed in 8.6.1. 440 441@item autogen version 5.5.4 (or later) and 442@itemx guile version 1.4.1 (or later) 443 444Necessary to regenerate @file{fixinc/fixincl.x} from 445@file{fixinc/inclhack.def} and @file{fixinc/*.tpl}. 446 447Necessary to run @samp{make check} for @file{fixinc}. 448 449Necessary to regenerate the top level @file{Makefile.in} file from 450@file{Makefile.tpl} and @file{Makefile.def}. 451 452@item Flex version 2.5.4 (or later) 453 454Necessary when modifying @file{*.l} files. 455 456Necessary to build GCC during development because the generated output 457files are not included in the SVN repository. They are included in 458releases. 459 460@item Texinfo version 4.7 (or later) 461 462Necessary for running @command{makeinfo} when modifying @file{*.texi} 463files to test your changes. 464 465Necessary for running @command{make dvi} or @command{make pdf} to 466create printable documentation in DVI or PDF format. Texinfo version 4674.8 or later is required for @command{make pdf}. 468 469Necessary to build GCC documentation during development because the 470generated output files are not included in the SVN repository. They are 471included in releases. 472 473@item @TeX{} (any working version) 474 475Necessary for running @command{texi2dvi} and @command{texi2pdf}, which 476are used when running @command{make dvi} or @command{make pdf} to create 477DVI or PDF files, respectively. 478 479@item Sphinx version 1.0 (or later) 480 481Necessary to regenerate @file{jit/docs/_build/texinfo} from the @file{.rst} 482files in the directories below @file{jit/docs}. 483 484@item SVN (any version) 485@itemx SSH (any version) 486 487Necessary to access the SVN repository. Public releases and weekly 488snapshots of the development sources are also available via FTP@. 489 490@item GNU diffutils version 2.7 (or later) 491 492Useful when submitting patches for the GCC source code. 493 494@item patch version 2.5.4 (or later) 495 496Necessary when applying patches, created with @command{diff}, to one's 497own sources. 498 499@item ecj1 500@itemx gjavah 501 502If you wish to modify @file{.java} files in libjava, you will need to 503configure with @option{--enable-java-maintainer-mode}, and you will need 504to have executables named @command{ecj1} and @command{gjavah} in your path. 505The @command{ecj1} executable should run the Eclipse Java compiler via 506the GCC-specific entry point. You can download a suitable jar from 507@uref{ftp://sourceware.org/pub/java/}, or by running the script 508@command{contrib/download_ecj}. 509 510@item antlr.jar version 2.7.1 (or later) 511@itemx antlr binary 512 513If you wish to build the @command{gjdoc} binary in libjava, you will 514need to have an @file{antlr.jar} library available. The library is 515searched for in system locations but can be specified with 516@option{--with-antlr-jar=} instead. When configuring with 517@option{--enable-java-maintainer-mode}, you will need to have one of 518the executables named @command{cantlr}, @command{runantlr} or 519@command{antlr} in your path. 520 521@end table 522 523@html 524<hr /> 525<p> 526@end html 527@ifhtml 528@uref{./index.html,,Return to the GCC Installation page} 529@end ifhtml 530@end ifset 531 532@c ***Downloading the source************************************************** 533@ifnothtml 534@comment node-name, next, previous, up 535@node Downloading the source, Configuration, Prerequisites, Installing GCC 536@end ifnothtml 537@ifset downloadhtml 538@ifnothtml 539@chapter Downloading GCC 540@end ifnothtml 541@cindex Downloading GCC 542@cindex Downloading the Source 543 544GCC is distributed via @uref{http://gcc.gnu.org/svn.html,,SVN} and FTP 545tarballs compressed with @command{gzip} or 546@command{bzip2}. 547 548Please refer to the @uref{http://gcc.gnu.org/releases.html,,releases web page} 549for information on how to obtain GCC@. 550 551The source distribution includes the C, C++, Objective-C, Fortran, Java, 552and Ada (in the case of GCC 3.1 and later) compilers, as well as 553runtime libraries for C++, Objective-C, Fortran, and Java. 554For previous versions these were downloadable as separate components such 555as the core GCC distribution, which included the C language front end and 556shared components, and language-specific distributions including the 557language front end and the language runtime (where appropriate). 558 559If you also intend to build binutils (either to upgrade an existing 560installation or for use in place of the corresponding tools of your 561OS), unpack the binutils distribution either in the same directory or 562a separate one. In the latter case, add symbolic links to any 563components of the binutils you intend to build alongside the compiler 564(@file{bfd}, @file{binutils}, @file{gas}, @file{gprof}, @file{ld}, 565@file{opcodes}, @dots{}) to the directory containing the GCC sources. 566 567Likewise the GMP, MPFR and MPC libraries can be automatically built 568together with GCC. Unpack the GMP, MPFR and/or MPC source 569distributions in the directory containing the GCC sources and rename 570their directories to @file{gmp}, @file{mpfr} and @file{mpc}, 571respectively (or use symbolic links with the same name). 572 573@html 574<hr /> 575<p> 576@end html 577@ifhtml 578@uref{./index.html,,Return to the GCC Installation page} 579@end ifhtml 580@end ifset 581 582@c ***Configuration*********************************************************** 583@ifnothtml 584@comment node-name, next, previous, up 585@node Configuration, Building, Downloading the source, Installing GCC 586@end ifnothtml 587@ifset configurehtml 588@ifnothtml 589@chapter Installing GCC: Configuration 590@end ifnothtml 591@cindex Configuration 592@cindex Installing GCC: Configuration 593 594Like most GNU software, GCC must be configured before it can be built. 595This document describes the recommended configuration procedure 596for both native and cross targets. 597 598We use @var{srcdir} to refer to the toplevel source directory for 599GCC; we use @var{objdir} to refer to the toplevel build/object directory. 600 601If you obtained the sources via SVN, @var{srcdir} must refer to the top 602@file{gcc} directory, the one where the @file{MAINTAINERS} file can be 603found, and not its @file{gcc} subdirectory, otherwise the build will fail. 604 605If either @var{srcdir} or @var{objdir} is located on an automounted NFS 606file system, the shell's built-in @command{pwd} command will return 607temporary pathnames. Using these can lead to various sorts of build 608problems. To avoid this issue, set the @env{PWDCMD} environment 609variable to an automounter-aware @command{pwd} command, e.g., 610@command{pawd} or @samp{amq -w}, during the configuration and build 611phases. 612 613First, we @strong{highly} recommend that GCC be built into a 614separate directory from the sources which does @strong{not} reside 615within the source tree. This is how we generally build GCC; building 616where @var{srcdir} == @var{objdir} should still work, but doesn't 617get extensive testing; building where @var{objdir} is a subdirectory 618of @var{srcdir} is unsupported. 619 620If you have previously built GCC in the same directory for a 621different target machine, do @samp{make distclean} to delete all files 622that might be invalid. One of the files this deletes is @file{Makefile}; 623if @samp{make distclean} complains that @file{Makefile} does not exist 624or issues a message like ``don't know how to make distclean'' it probably 625means that the directory is already suitably clean. However, with the 626recommended method of building in a separate @var{objdir}, you should 627simply use a different @var{objdir} for each target. 628 629Second, when configuring a native system, either @command{cc} or 630@command{gcc} must be in your path or you must set @env{CC} in 631your environment before running configure. Otherwise the configuration 632scripts may fail. 633 634@ignore 635Note that the bootstrap compiler and the resulting GCC must be link 636compatible, else the bootstrap will fail with linker errors about 637incompatible object file formats. Several multilibed targets are 638affected by this requirement, see 639@ifnothtml 640@ref{Specific, host/target specific installation notes}. 641@end ifnothtml 642@ifhtml 643@uref{specific.html,,host/target specific installation notes}. 644@end ifhtml 645@end ignore 646 647To configure GCC: 648 649@smallexample 650% mkdir @var{objdir} 651% cd @var{objdir} 652% @var{srcdir}/configure [@var{options}] [@var{target}] 653@end smallexample 654 655@heading Distributor options 656 657If you will be distributing binary versions of GCC, with modifications 658to the source code, you should use the options described in this 659section to make clear that your version contains modifications. 660 661@table @code 662@item --with-pkgversion=@var{version} 663Specify a string that identifies your package. You may wish 664to include a build number or build date. This version string will be 665included in the output of @command{gcc --version}. This suffix does 666not replace the default version string, only the @samp{GCC} part. 667 668The default value is @samp{GCC}. 669 670@item --with-bugurl=@var{url} 671Specify the URL that users should visit if they wish to report a bug. 672You are of course welcome to forward bugs reported to you to the FSF, 673if you determine that they are not bugs in your modifications. 674 675The default value refers to the FSF's GCC bug tracker. 676 677@end table 678 679@heading Target specification 680@itemize @bullet 681@item 682GCC has code to correctly determine the correct value for @var{target} 683for nearly all native systems. Therefore, we highly recommend you do 684not provide a configure target when configuring a native compiler. 685 686@item 687@var{target} must be specified as @option{--target=@var{target}} 688when configuring a cross compiler; examples of valid targets would be 689m68k-elf, sh-elf, etc. 690 691@item 692Specifying just @var{target} instead of @option{--target=@var{target}} 693implies that the host defaults to @var{target}. 694@end itemize 695 696 697@heading Options specification 698 699Use @var{options} to override several configure time options for 700GCC@. A list of supported @var{options} follows; @samp{configure 701--help} may list other options, but those not listed below may not 702work and should not normally be used. 703 704Note that each @option{--enable} option has a corresponding 705@option{--disable} option and that each @option{--with} option has a 706corresponding @option{--without} option. 707 708@table @code 709@item --prefix=@var{dirname} 710Specify the toplevel installation 711directory. This is the recommended way to install the tools into a directory 712other than the default. The toplevel installation directory defaults to 713@file{/usr/local}. 714 715We @strong{highly} recommend against @var{dirname} being the same or a 716subdirectory of @var{objdir} or vice versa. If specifying a directory 717beneath a user's home directory tree, some shells will not expand 718@var{dirname} correctly if it contains the @samp{~} metacharacter; use 719@env{$HOME} instead. 720 721The following standard @command{autoconf} options are supported. Normally you 722should not need to use these options. 723@table @code 724@item --exec-prefix=@var{dirname} 725Specify the toplevel installation directory for architecture-dependent 726files. The default is @file{@var{prefix}}. 727 728@item --bindir=@var{dirname} 729Specify the installation directory for the executables called by users 730(such as @command{gcc} and @command{g++}). The default is 731@file{@var{exec-prefix}/bin}. 732 733@item --libdir=@var{dirname} 734Specify the installation directory for object code libraries and 735internal data files of GCC@. The default is @file{@var{exec-prefix}/lib}. 736 737@item --libexecdir=@var{dirname} 738Specify the installation directory for internal executables of GCC@. 739The default is @file{@var{exec-prefix}/libexec}. 740 741@item --with-slibdir=@var{dirname} 742Specify the installation directory for the shared libgcc library. The 743default is @file{@var{libdir}}. 744 745@item --datarootdir=@var{dirname} 746Specify the root of the directory tree for read-only architecture-independent 747data files referenced by GCC@. The default is @file{@var{prefix}/share}. 748 749@item --infodir=@var{dirname} 750Specify the installation directory for documentation in info format. 751The default is @file{@var{datarootdir}/info}. 752 753@item --datadir=@var{dirname} 754Specify the installation directory for some architecture-independent 755data files referenced by GCC@. The default is @file{@var{datarootdir}}. 756 757@item --docdir=@var{dirname} 758Specify the installation directory for documentation files (other 759than Info) for GCC@. The default is @file{@var{datarootdir}/doc}. 760 761@item --htmldir=@var{dirname} 762Specify the installation directory for HTML documentation files. 763The default is @file{@var{docdir}}. 764 765@item --pdfdir=@var{dirname} 766Specify the installation directory for PDF documentation files. 767The default is @file{@var{docdir}}. 768 769@item --mandir=@var{dirname} 770Specify the installation directory for manual pages. The default is 771@file{@var{datarootdir}/man}. (Note that the manual pages are only extracts 772from the full GCC manuals, which are provided in Texinfo format. The manpages 773are derived by an automatic conversion process from parts of the full 774manual.) 775 776@item --with-gxx-include-dir=@var{dirname} 777Specify 778the installation directory for G++ header files. The default depends 779on other configuration options, and differs between cross and native 780configurations. 781 782@item --with-specs=@var{specs} 783Specify additional command line driver SPECS. 784This can be useful if you need to turn on a non-standard feature by 785default without modifying the compiler's source code, for instance 786@option{--with-specs=%@{!fcommon:%@{!fno-common:-fno-common@}@}}. 787@ifnothtml 788@xref{Spec Files,, Specifying subprocesses and the switches to pass to them, 789gcc, Using the GNU Compiler Collection (GCC)}, 790@end ifnothtml 791@ifhtml 792See ``Spec Files'' in the main manual 793@end ifhtml 794 795@end table 796 797@item --program-prefix=@var{prefix} 798GCC supports some transformations of the names of its programs when 799installing them. This option prepends @var{prefix} to the names of 800programs to install in @var{bindir} (see above). For example, specifying 801@option{--program-prefix=foo-} would result in @samp{gcc} 802being installed as @file{/usr/local/bin/foo-gcc}. 803 804@item --program-suffix=@var{suffix} 805Appends @var{suffix} to the names of programs to install in @var{bindir} 806(see above). For example, specifying @option{--program-suffix=-3.1} 807would result in @samp{gcc} being installed as 808@file{/usr/local/bin/gcc-3.1}. 809 810@item --program-transform-name=@var{pattern} 811Applies the @samp{sed} script @var{pattern} to be applied to the names 812of programs to install in @var{bindir} (see above). @var{pattern} has to 813consist of one or more basic @samp{sed} editing commands, separated by 814semicolons. For example, if you want the @samp{gcc} program name to be 815transformed to the installed program @file{/usr/local/bin/myowngcc} and 816the @samp{g++} program name to be transformed to 817@file{/usr/local/bin/gspecial++} without changing other program names, 818you could use the pattern 819@option{--program-transform-name='s/^gcc$/myowngcc/; s/^g++$/gspecial++/'} 820to achieve this effect. 821 822All three options can be combined and used together, resulting in more 823complex conversion patterns. As a basic rule, @var{prefix} (and 824@var{suffix}) are prepended (appended) before further transformations 825can happen with a special transformation script @var{pattern}. 826 827As currently implemented, this option only takes effect for native 828builds; cross compiler binaries' names are not transformed even when a 829transformation is explicitly asked for by one of these options. 830 831For native builds, some of the installed programs are also installed 832with the target alias in front of their name, as in 833@samp{i686-pc-linux-gnu-gcc}. All of the above transformations happen 834before the target alias is prepended to the name---so, specifying 835@option{--program-prefix=foo-} and @option{program-suffix=-3.1}, the 836resulting binary would be installed as 837@file{/usr/local/bin/i686-pc-linux-gnu-foo-gcc-3.1}. 838 839As a last shortcoming, none of the installed Ada programs are 840transformed yet, which will be fixed in some time. 841 842@item --with-local-prefix=@var{dirname} 843Specify the 844installation directory for local include files. The default is 845@file{/usr/local}. Specify this option if you want the compiler to 846search directory @file{@var{dirname}/include} for locally installed 847header files @emph{instead} of @file{/usr/local/include}. 848 849You should specify @option{--with-local-prefix} @strong{only} if your 850site has a different convention (not @file{/usr/local}) for where to put 851site-specific files. 852 853The default value for @option{--with-local-prefix} is @file{/usr/local} 854regardless of the value of @option{--prefix}. Specifying 855@option{--prefix} has no effect on which directory GCC searches for 856local header files. This may seem counterintuitive, but actually it is 857logical. 858 859The purpose of @option{--prefix} is to specify where to @emph{install 860GCC}. The local header files in @file{/usr/local/include}---if you put 861any in that directory---are not part of GCC@. They are part of other 862programs---perhaps many others. (GCC installs its own header files in 863another directory which is based on the @option{--prefix} value.) 864 865Both the local-prefix include directory and the GCC-prefix include 866directory are part of GCC's ``system include'' directories. Although these 867two directories are not fixed, they need to be searched in the proper 868order for the correct processing of the include_next directive. The 869local-prefix include directory is searched before the GCC-prefix 870include directory. Another characteristic of system include directories 871is that pedantic warnings are turned off for headers in these directories. 872 873Some autoconf macros add @option{-I @var{directory}} options to the 874compiler command line, to ensure that directories containing installed 875packages' headers are searched. When @var{directory} is one of GCC's 876system include directories, GCC will ignore the option so that system 877directories continue to be processed in the correct order. This 878may result in a search order different from what was specified but the 879directory will still be searched. 880 881GCC automatically searches for ordinary libraries using 882@env{GCC_EXEC_PREFIX}. Thus, when the same installation prefix is 883used for both GCC and packages, GCC will automatically search for 884both headers and libraries. This provides a configuration that is 885easy to use. GCC behaves in a manner similar to that when it is 886installed as a system compiler in @file{/usr}. 887 888Sites that need to install multiple versions of GCC may not want to 889use the above simple configuration. It is possible to use the 890@option{--program-prefix}, @option{--program-suffix} and 891@option{--program-transform-name} options to install multiple versions 892into a single directory, but it may be simpler to use different prefixes 893and the @option{--with-local-prefix} option to specify the location of the 894site-specific files for each version. It will then be necessary for 895users to specify explicitly the location of local site libraries 896(e.g., with @env{LIBRARY_PATH}). 897 898The same value can be used for both @option{--with-local-prefix} and 899@option{--prefix} provided it is not @file{/usr}. This can be used 900to avoid the default search of @file{/usr/local/include}. 901 902@strong{Do not} specify @file{/usr} as the @option{--with-local-prefix}! 903The directory you use for @option{--with-local-prefix} @strong{must not} 904contain any of the system's standard header files. If it did contain 905them, certain programs would be miscompiled (including GNU Emacs, on 906certain targets), because this would override and nullify the header 907file corrections made by the @command{fixincludes} script. 908 909Indications are that people who use this option use it based on mistaken 910ideas of what it is for. People use it as if it specified where to 911install part of GCC@. Perhaps they make this assumption because 912installing GCC creates the directory. 913 914@item --with-native-system-header-dir=@var{dirname} 915Specifies that @var{dirname} is the directory that contains native system 916header files, rather than @file{/usr/include}. This option is most useful 917if you are creating a compiler that should be isolated from the system 918as much as possible. It is most commonly used with the 919@option{--with-sysroot} option and will cause GCC to search 920@var{dirname} inside the system root specified by that option. 921 922@item --enable-shared[=@var{package}[,@dots{}]] 923Build shared versions of libraries, if shared libraries are supported on 924the target platform. Unlike GCC 2.95.x and earlier, shared libraries 925are enabled by default on all platforms that support shared libraries. 926 927If a list of packages is given as an argument, build shared libraries 928only for the listed packages. For other packages, only static libraries 929will be built. Package names currently recognized in the GCC tree are 930@samp{libgcc} (also known as @samp{gcc}), @samp{libstdc++} (not 931@samp{libstdc++-v3}), @samp{libffi}, @samp{zlib}, @samp{boehm-gc}, 932@samp{ada}, @samp{libada}, @samp{libjava}, @samp{libgo}, and @samp{libobjc}. 933Note @samp{libiberty} does not support shared libraries at all. 934 935Use @option{--disable-shared} to build only static libraries. Note that 936@option{--disable-shared} does not accept a list of package names as 937argument, only @option{--enable-shared} does. 938 939Contrast with @option{--enable-host-shared}, which affects @emph{host} 940code. 941 942@item --enable-host-shared 943Specify that the @emph{host} code should be built into position-independent 944machine code (with -fPIC), allowing it to be used within shared libraries, 945but yielding a slightly slower compiler. 946 947This option is required when building the libgccjit.so library. 948 949Contrast with @option{--enable-shared}, which affects @emph{target} 950libraries. 951 952@item @anchor{with-gnu-as}--with-gnu-as 953Specify that the compiler should assume that the 954assembler it finds is the GNU assembler. However, this does not modify 955the rules to find an assembler and will result in confusion if the 956assembler found is not actually the GNU assembler. (Confusion may also 957result if the compiler finds the GNU assembler but has not been 958configured with @option{--with-gnu-as}.) If you have more than one 959assembler installed on your system, you may want to use this option in 960connection with @option{--with-as=@var{pathname}} or 961@option{--with-build-time-tools=@var{pathname}}. 962 963The following systems are the only ones where it makes a difference 964whether you use the GNU assembler. On any other system, 965@option{--with-gnu-as} has no effect. 966 967@itemize @bullet 968@item @samp{hppa1.0-@var{any}-@var{any}} 969@item @samp{hppa1.1-@var{any}-@var{any}} 970@item @samp{sparc-sun-solaris2.@var{any}} 971@item @samp{sparc64-@var{any}-solaris2.@var{any}} 972@end itemize 973 974@item @anchor{with-as}--with-as=@var{pathname} 975Specify that the compiler should use the assembler pointed to by 976@var{pathname}, rather than the one found by the standard rules to find 977an assembler, which are: 978@itemize @bullet 979@item 980Unless GCC is being built with a cross compiler, check the 981@file{@var{libexec}/gcc/@var{target}/@var{version}} directory. 982@var{libexec} defaults to @file{@var{exec-prefix}/libexec}; 983@var{exec-prefix} defaults to @var{prefix}, which 984defaults to @file{/usr/local} unless overridden by the 985@option{--prefix=@var{pathname}} switch described above. @var{target} 986is the target system triple, such as @samp{sparc-sun-solaris2.7}, and 987@var{version} denotes the GCC version, such as 3.0. 988 989@item 990If the target system is the same that you are building on, check 991operating system specific directories (e.g.@: @file{/usr/ccs/bin} on 992Sun Solaris 2). 993 994@item 995Check in the @env{PATH} for a tool whose name is prefixed by the 996target system triple. 997 998@item 999Check in the @env{PATH} for a tool whose name is not prefixed by the 1000target system triple, if the host and target system triple are 1001the same (in other words, we use a host tool if it can be used for 1002the target as well). 1003@end itemize 1004 1005You may want to use @option{--with-as} if no assembler 1006is installed in the directories listed above, or if you have multiple 1007assemblers installed and want to choose one that is not found by the 1008above rules. 1009 1010@item @anchor{with-gnu-ld}--with-gnu-ld 1011Same as @uref{#with-gnu-as,,@option{--with-gnu-as}} 1012but for the linker. 1013 1014@item --with-ld=@var{pathname} 1015Same as @uref{#with-as,,@option{--with-as}} 1016but for the linker. 1017 1018@item --with-stabs 1019Specify that stabs debugging 1020information should be used instead of whatever format the host normally 1021uses. Normally GCC uses the same debug format as the host system. 1022 1023On MIPS based systems and on Alphas, you must specify whether you want 1024GCC to create the normal ECOFF debugging format, or to use BSD-style 1025stabs passed through the ECOFF symbol table. The normal ECOFF debug 1026format cannot fully handle languages other than C@. BSD stabs format can 1027handle other languages, but it only works with the GNU debugger GDB@. 1028 1029Normally, GCC uses the ECOFF debugging format by default; if you 1030prefer BSD stabs, specify @option{--with-stabs} when you configure GCC@. 1031 1032No matter which default you choose when you configure GCC, the user 1033can use the @option{-gcoff} and @option{-gstabs+} options to specify explicitly 1034the debug format for a particular compilation. 1035 1036@option{--with-stabs} is meaningful on the ISC system on the 386, also, if 1037@option{--with-gas} is used. It selects use of stabs debugging 1038information embedded in COFF output. This kind of debugging information 1039supports C++ well; ordinary COFF debugging information does not. 1040 1041@option{--with-stabs} is also meaningful on 386 systems running SVR4. It 1042selects use of stabs debugging information embedded in ELF output. The 1043C++ compiler currently (2.6.0) does not support the DWARF debugging 1044information normally used on 386 SVR4 platforms; stabs provide a 1045workable alternative. This requires gas and gdb, as the normal SVR4 1046tools can not generate or interpret stabs. 1047 1048@item --with-tls=@var{dialect} 1049Specify the default TLS dialect, for systems were there is a choice. 1050For ARM targets, possible values for @var{dialect} are @code{gnu} or 1051@code{gnu2}, which select between the original GNU dialect and the GNU TLS 1052descriptor-based dialect. 1053 1054@item --enable-multiarch 1055Specify whether to enable or disable multiarch support. The default is 1056to check for glibc start files in a multiarch location, and enable it 1057if the files are found. The auto detection is enabled for native builds, 1058and for cross builds configured with @option{--with-sysroot}, and without 1059@option{--with-native-system-header-dir}. 1060More documentation about multiarch can be found at 1061@uref{http://wiki.debian.org/Multiarch}. 1062 1063@item --enable-vtable-verify 1064Specify whether to enable or disable the vtable verification feature. 1065Enabling this feature causes libstdc++ to be built with its virtual calls 1066in verifiable mode. This means that, when linked with libvtv, every 1067virtual call in libstdc++ will verify the vtable pointer through which the 1068call will be made before actually making the call. If not linked with libvtv, 1069the verifier will call stub functions (in libstdc++ itself) and do nothing. 1070If vtable verification is disabled, then libstdc++ is not built with its 1071virtual calls in verifiable mode at all. However the libvtv library will 1072still be built (see @option{--disable-libvtv} to turn off building libvtv). 1073@option{--disable-vtable-verify} is the default. 1074 1075@item --disable-multilib 1076Specify that multiple target 1077libraries to support different target variants, calling 1078conventions, etc.@: should not be built. The default is to build a 1079predefined set of them. 1080 1081Some targets provide finer-grained control over which multilibs are built 1082(e.g., @option{--disable-softfloat}): 1083@table @code 1084@item arm-*-* 1085fpu, 26bit, underscore, interwork, biendian, nofmult. 1086 1087@item m68*-*-* 1088softfloat, m68881, m68000, m68020. 1089 1090@item mips*-*-* 1091single-float, biendian, softfloat. 1092 1093@item powerpc*-*-*, rs6000*-*-* 1094aix64, pthread, softfloat, powercpu, powerpccpu, powerpcos, biendian, 1095sysv, aix. 1096 1097@end table 1098 1099@item --with-multilib-list=@var{list} 1100@itemx --without-multilib-list 1101Specify what multilibs to build. 1102Currently only implemented for sh*-*-* and x86-64-*-linux*. 1103 1104@table @code 1105@item sh*-*-* 1106@var{list} is a comma separated list of CPU names. These must be of the 1107form @code{sh*} or @code{m*} (in which case they match the compiler option 1108for that processor). The list should not contain any endian options - 1109these are handled by @option{--with-endian}. 1110 1111If @var{list} is empty, then there will be no multilibs for extra 1112processors. The multilib for the secondary endian remains enabled. 1113 1114As a special case, if an entry in the list starts with a @code{!} 1115(exclamation point), then it is added to the list of excluded multilibs. 1116Entries of this sort should be compatible with @samp{MULTILIB_EXCLUDES} 1117(once the leading @code{!} has been stripped). 1118 1119If @option{--with-multilib-list} is not given, then a default set of 1120multilibs is selected based on the value of @option{--target}. This is 1121usually the complete set of libraries, but some targets imply a more 1122specialized subset. 1123 1124Example 1: to configure a compiler for SH4A only, but supporting both 1125endians, with little endian being the default: 1126@smallexample 1127--with-cpu=sh4a --with-endian=little,big --with-multilib-list= 1128@end smallexample 1129 1130Example 2: to configure a compiler for both SH4A and SH4AL-DSP, but with 1131only little endian SH4AL: 1132@smallexample 1133--with-cpu=sh4a --with-endian=little,big \ 1134--with-multilib-list=sh4al,!mb/m4al 1135@end smallexample 1136 1137@item x86-64-*-linux* 1138@var{list} is a comma separated list of @code{m32}, @code{m64} and 1139@code{mx32} to enable 32-bit, 64-bit and x32 run-time libraries, 1140respectively. If @var{list} is empty, then there will be no multilibs 1141and only the default run-time library will be enabled. 1142 1143If @option{--with-multilib-list} is not given, then only 32-bit and 114464-bit run-time libraries will be enabled. 1145@end table 1146 1147@item --with-endian=@var{endians} 1148Specify what endians to use. 1149Currently only implemented for sh*-*-*. 1150 1151@var{endians} may be one of the following: 1152@table @code 1153@item big 1154Use big endian exclusively. 1155@item little 1156Use little endian exclusively. 1157@item big,little 1158Use big endian by default. Provide a multilib for little endian. 1159@item little,big 1160Use little endian by default. Provide a multilib for big endian. 1161@end table 1162 1163@item --enable-threads 1164Specify that the target 1165supports threads. This affects the Objective-C compiler and runtime 1166library, and exception handling for other languages like C++ and Java. 1167On some systems, this is the default. 1168 1169In general, the best (and, in many cases, the only known) threading 1170model available will be configured for use. Beware that on some 1171systems, GCC has not been taught what threading models are generally 1172available for the system. In this case, @option{--enable-threads} is an 1173alias for @option{--enable-threads=single}. 1174 1175@item --disable-threads 1176Specify that threading support should be disabled for the system. 1177This is an alias for @option{--enable-threads=single}. 1178 1179@item --enable-threads=@var{lib} 1180Specify that 1181@var{lib} is the thread support library. This affects the Objective-C 1182compiler and runtime library, and exception handling for other languages 1183like C++ and Java. The possibilities for @var{lib} are: 1184 1185@table @code 1186@item aix 1187AIX thread support. 1188@item dce 1189DCE thread support. 1190@item lynx 1191LynxOS thread support. 1192@item mipssde 1193MIPS SDE thread support. 1194@item no 1195This is an alias for @samp{single}. 1196@item posix 1197Generic POSIX/Unix98 thread support. 1198@item rtems 1199RTEMS thread support. 1200@item single 1201Disable thread support, should work for all platforms. 1202@item tpf 1203TPF thread support. 1204@item vxworks 1205VxWorks thread support. 1206@item win32 1207Microsoft Win32 API thread support. 1208@end table 1209 1210@item --enable-tls 1211Specify that the target supports TLS (Thread Local Storage). Usually 1212configure can correctly determine if TLS is supported. In cases where 1213it guesses incorrectly, TLS can be explicitly enabled or disabled with 1214@option{--enable-tls} or @option{--disable-tls}. This can happen if 1215the assembler supports TLS but the C library does not, or if the 1216assumptions made by the configure test are incorrect. 1217 1218@item --disable-tls 1219Specify that the target does not support TLS. 1220This is an alias for @option{--enable-tls=no}. 1221 1222@item --with-cpu=@var{cpu} 1223@itemx --with-cpu-32=@var{cpu} 1224@itemx --with-cpu-64=@var{cpu} 1225Specify which cpu variant the compiler should generate code for by default. 1226@var{cpu} will be used as the default value of the @option{-mcpu=} switch. 1227This option is only supported on some targets, including ARC, ARM, i386, M68k, 1228PowerPC, and SPARC@. It is mandatory for ARC@. The @option{--with-cpu-32} and 1229@option{--with-cpu-64} options specify separate default CPUs for 123032-bit and 64-bit modes; these options are only supported for i386, 1231x86-64 and PowerPC. 1232 1233@item --with-schedule=@var{cpu} 1234@itemx --with-arch=@var{cpu} 1235@itemx --with-arch-32=@var{cpu} 1236@itemx --with-arch-64=@var{cpu} 1237@itemx --with-tune=@var{cpu} 1238@itemx --with-tune-32=@var{cpu} 1239@itemx --with-tune-64=@var{cpu} 1240@itemx --with-abi=@var{abi} 1241@itemx --with-fpu=@var{type} 1242@itemx --with-float=@var{type} 1243These configure options provide default values for the @option{-mschedule=}, 1244@option{-march=}, @option{-mtune=}, @option{-mabi=}, and @option{-mfpu=} 1245options and for @option{-mhard-float} or @option{-msoft-float}. As with 1246@option{--with-cpu}, which switches will be accepted and acceptable values 1247of the arguments depend on the target. 1248 1249@item --with-mode=@var{mode} 1250Specify if the compiler should default to @option{-marm} or @option{-mthumb}. 1251This option is only supported on ARM targets. 1252 1253@item --with-stack-offset=@var{num} 1254This option sets the default for the -mstack-offset=@var{num} option, 1255and will thus generally also control the setting of this option for 1256libraries. This option is only supported on Epiphany targets. 1257 1258@item --with-fpmath=@var{isa} 1259This options sets @option{-mfpmath=sse} by default and specifies the default 1260ISA for floating-point arithmetics. You can select either @samp{sse} which 1261enables @option{-msse2} or @samp{avx} which enables @option{-mavx} by default. 1262This option is only supported on i386 and x86-64 targets. 1263 1264@item --with-fp-32=@var{mode} 1265On MIPS targets, set the default value for the @option{-mfp} option when using 1266the o32 ABI. The possibilities for @var{mode} are: 1267@table @code 1268@item 32 1269Use the o32 FP32 ABI extension, as with the @option{-mfp32} command-line 1270option. 1271@item xx 1272Use the o32 FPXX ABI extension, as with the @option{-mfpxx} command-line 1273option. 1274@item 64 1275Use the o32 FP64 ABI extension, as with the @option{-mfp64} command-line 1276option. 1277@end table 1278In the absence of this configuration option the default is to use the o32 1279FP32 ABI extension. 1280 1281@item --with-odd-spreg-32 1282On MIPS targets, set the @option{-modd-spreg} option by default when using 1283the o32 ABI. 1284 1285@item --without-odd-spreg-32 1286On MIPS targets, set the @option{-mno-odd-spreg} option by default when using 1287the o32 ABI. This is normally used in conjunction with 1288@option{--with-fp-32=64} in order to target the o32 FP64A ABI extension. 1289 1290@item --with-nan=@var{encoding} 1291On MIPS targets, set the default encoding convention to use for the 1292special not-a-number (NaN) IEEE 754 floating-point data. The 1293possibilities for @var{encoding} are: 1294@table @code 1295@item legacy 1296Use the legacy encoding, as with the @option{-mnan=legacy} command-line 1297option. 1298@item 2008 1299Use the 754-2008 encoding, as with the @option{-mnan=2008} command-line 1300option. 1301@end table 1302To use this configuration option you must have an assembler version 1303installed that supports the @option{-mnan=} command-line option too. 1304In the absence of this configuration option the default convention is 1305the legacy encoding, as when neither of the @option{-mnan=2008} and 1306@option{-mnan=legacy} command-line options has been used. 1307 1308@item --with-divide=@var{type} 1309Specify how the compiler should generate code for checking for 1310division by zero. This option is only supported on the MIPS target. 1311The possibilities for @var{type} are: 1312@table @code 1313@item traps 1314Division by zero checks use conditional traps (this is the default on 1315systems that support conditional traps). 1316@item breaks 1317Division by zero checks use the break instruction. 1318@end table 1319 1320@c If you make --with-llsc the default for additional targets, 1321@c update the --with-llsc description in the MIPS section below. 1322 1323@item --with-llsc 1324On MIPS targets, make @option{-mllsc} the default when no 1325@option{-mno-llsc} option is passed. This is the default for 1326Linux-based targets, as the kernel will emulate them if the ISA does 1327not provide them. 1328 1329@item --without-llsc 1330On MIPS targets, make @option{-mno-llsc} the default when no 1331@option{-mllsc} option is passed. 1332 1333@item --with-synci 1334On MIPS targets, make @option{-msynci} the default when no 1335@option{-mno-synci} option is passed. 1336 1337@item --without-synci 1338On MIPS targets, make @option{-mno-synci} the default when no 1339@option{-msynci} option is passed. This is the default. 1340 1341@item --with-mips-plt 1342On MIPS targets, make use of copy relocations and PLTs. 1343These features are extensions to the traditional 1344SVR4-based MIPS ABIs and require support from GNU binutils 1345and the runtime C library. 1346 1347@item --enable-__cxa_atexit 1348Define if you want to use __cxa_atexit, rather than atexit, to 1349register C++ destructors for local statics and global objects. 1350This is essential for fully standards-compliant handling of 1351destructors, but requires __cxa_atexit in libc. This option is currently 1352only available on systems with GNU libc. When enabled, this will cause 1353@option{-fuse-cxa-atexit} to be passed by default. 1354 1355@item --enable-gnu-indirect-function 1356Define if you want to enable the @code{ifunc} attribute. This option is 1357currently only available on systems with GNU libc on certain targets. 1358 1359@item --enable-target-optspace 1360Specify that target 1361libraries should be optimized for code space instead of code speed. 1362This is the default for the m32r platform. 1363 1364@item --with-cpp-install-dir=@var{dirname} 1365Specify that the user visible @command{cpp} program should be installed 1366in @file{@var{prefix}/@var{dirname}/cpp}, in addition to @var{bindir}. 1367 1368@item --enable-comdat 1369Enable COMDAT group support. This is primarily used to override the 1370automatically detected value. 1371 1372@item --enable-initfini-array 1373Force the use of sections @code{.init_array} and @code{.fini_array} 1374(instead of @code{.init} and @code{.fini}) for constructors and 1375destructors. Option @option{--disable-initfini-array} has the 1376opposite effect. If neither option is specified, the configure script 1377will try to guess whether the @code{.init_array} and 1378@code{.fini_array} sections are supported and, if they are, use them. 1379 1380@item --enable-link-mutex 1381When building GCC, use a mutex to avoid linking the compilers for 1382multiple languages at the same time, to avoid thrashing on build 1383systems with limited free memory. The default is not to use such a mutex. 1384 1385@item --enable-maintainer-mode 1386The build rules that regenerate the Autoconf and Automake output files as 1387well as the GCC master message catalog @file{gcc.pot} are normally 1388disabled. This is because it can only be rebuilt if the complete source 1389tree is present. If you have changed the sources and want to rebuild the 1390catalog, configuring with @option{--enable-maintainer-mode} will enable 1391this. Note that you need a recent version of the @code{gettext} tools 1392to do so. 1393 1394@item --disable-bootstrap 1395For a native build, the default configuration is to perform 1396a 3-stage bootstrap of the compiler when @samp{make} is invoked, 1397testing that GCC can compile itself correctly. If you want to disable 1398this process, you can configure with @option{--disable-bootstrap}. 1399 1400@item --enable-bootstrap 1401In special cases, you may want to perform a 3-stage build 1402even if the target and host triplets are different. 1403This is possible when the host can run code compiled for 1404the target (e.g.@: host is i686-linux, target is i486-linux). 1405Starting from GCC 4.2, to do this you have to configure explicitly 1406with @option{--enable-bootstrap}. 1407 1408@item --enable-generated-files-in-srcdir 1409Neither the .c and .h files that are generated from Bison and flex nor the 1410info manuals and man pages that are built from the .texi files are present 1411in the SVN development tree. When building GCC from that development tree, 1412or from one of our snapshots, those generated files are placed in your 1413build directory, which allows for the source to be in a readonly 1414directory. 1415 1416If you configure with @option{--enable-generated-files-in-srcdir} then those 1417generated files will go into the source directory. This is mainly intended 1418for generating release or prerelease tarballs of the GCC sources, since it 1419is not a requirement that the users of source releases to have flex, Bison, 1420or makeinfo. 1421 1422@item --enable-version-specific-runtime-libs 1423Specify 1424that runtime libraries should be installed in the compiler specific 1425subdirectory (@file{@var{libdir}/gcc}) rather than the usual places. In 1426addition, @samp{libstdc++}'s include files will be installed into 1427@file{@var{libdir}} unless you overruled it by using 1428@option{--with-gxx-include-dir=@var{dirname}}. Using this option is 1429particularly useful if you intend to use several versions of GCC in 1430parallel. This is currently supported by @samp{libgfortran}, 1431@samp{libjava}, @samp{libstdc++}, and @samp{libobjc}. 1432 1433@item @anchor{WithAixSoname}--with-aix-soname=@samp{aix}, @samp{svr4} or @samp{both} 1434Traditional AIX shared library versioning (versioned @code{Shared Object} 1435files as members of unversioned @code{Archive Library} files named 1436@samp{lib.a}) causes numerous headaches for package managers. However, 1437@code{Import Files} as members of @code{Archive Library} files allow for 1438@strong{filename-based versioning} of shared libraries as seen on Linux/SVR4, 1439where this is called the "SONAME". But as they prevent static linking, 1440@code{Import Files} may be used with @code{Runtime Linking} only, where the 1441linker does search for @samp{libNAME.so} before @samp{libNAME.a} library 1442filenames with the @samp{-lNAME} linker flag. 1443 1444@anchor{AixLdCommand}For detailed information please refer to the AIX 1445@uref{http://www-01.ibm.com/support/knowledgecenter/search/%22the%20ld%20command%2C%20also%20called%20the%20linkage%20editor%20or%20binder%22,,ld 1446Command} reference. 1447 1448As long as shared library creation is enabled, upon: 1449@table @code 1450@item --with-aix-soname=aix 1451@item --with-aix-soname=both 1452 A (traditional AIX) @code{Shared Archive Library} file is created: 1453 @itemize @bullet 1454 @item using the @samp{libNAME.a} filename scheme 1455 @item with the @code{Shared Object} file as archive member named 1456 @samp{libNAME.so.V} (except for @samp{libgcc_s}, where the @code{Shared 1457 Object} file is named @samp{shr.o} for backwards compatibility), which 1458 @itemize @minus 1459 @item is used for runtime loading from inside the @samp{libNAME.a} file 1460 @item is used for dynamic loading via 1461 @code{dlopen("libNAME.a(libNAME.so.V)", RTLD_MEMBER)} 1462 @item is used for shared linking 1463 @item is used for static linking, so no separate @code{Static Archive 1464 Library} file is needed 1465 @end itemize 1466 @end itemize 1467@item --with-aix-soname=both 1468@item --with-aix-soname=svr4 1469 A (second) @code{Shared Archive Library} file is created: 1470 @itemize @bullet 1471 @item using the @samp{libNAME.so.V} filename scheme 1472 @item with the @code{Shared Object} file as archive member named 1473 @samp{shr.o}, which 1474 @itemize @minus 1475 @item is created with the @code{-G linker flag} 1476 @item has the @code{F_LOADONLY} flag set 1477 @item is used for runtime loading from inside the @samp{libNAME.so.V} file 1478 @item is used for dynamic loading via @code{dlopen("libNAME.so.V(shr.o)", 1479 RTLD_MEMBER)} 1480 @end itemize 1481 @item with the @code{Import File} as archive member named @samp{shr.imp}, 1482 which 1483 @itemize @minus 1484 @item refers to @samp{libNAME.so.V(shr.o)} as the "SONAME", to be recorded 1485 in the @code{Loader Section} of subsequent binaries 1486 @item indicates whether @samp{libNAME.so.V(shr.o)} is 32 or 64 bit 1487 @item lists all the public symbols exported by @samp{lib.so.V(shr.o)}, 1488 eventually decorated with the @code{@samp{weak} Keyword} 1489 @item is necessary for shared linking against @samp{lib.so.V(shr.o)} 1490 @end itemize 1491 @end itemize 1492 A symbolic link using the @samp{libNAME.so} filename scheme is created: 1493 @itemize @bullet 1494 @item pointing to the @samp{libNAME.so.V} @code{Shared Archive Library} file 1495 @item to permit the @code{ld Command} to find @samp{lib.so.V(shr.imp)} via 1496 the @samp{-lNAME} argument (requires @code{Runtime Linking} to be enabled) 1497 @item to permit dynamic loading of @samp{lib.so.V(shr.o)} without the need 1498 to specify the version number via @code{dlopen("libNAME.so(shr.o)", 1499 RTLD_MEMBER)} 1500 @end itemize 1501@end table 1502 1503As long as static library creation is enabled, upon: 1504@table @code 1505@item --with-aix-soname=svr4 1506 A @code{Static Archive Library} is created: 1507 @itemize @bullet 1508 @item using the @samp{libNAME.a} filename scheme 1509 @item with all the @code{Static Object} files as archive members, which 1510 @itemize @minus 1511 @item are used for static linking 1512 @end itemize 1513 @end itemize 1514@end table 1515 1516While the aix-soname=@samp{svr4} option does not create @code{Shared Object} 1517files as members of unversioned @code{Archive Library} files any more, package 1518managers still are responsible to 1519@uref{./specific.html#TransferAixShobj,,transfer} @code{Shared Object} files 1520found as member of a previously installed unversioned @code{Archive Library} 1521file into the newly installed @code{Archive Library} file with the same 1522filename. 1523 1524@emph{WARNING:} Creating @code{Shared Object} files with @code{Runtime Linking} 1525enabled may bloat the TOC, eventually leading to @code{TOC overflow} errors, 1526requiring the use of either the @option{-Wl,-bbigtoc} linker flag (seen to 1527break with the @code{GDB} debugger) or some of the TOC-related compiler flags, 1528@ifnothtml 1529@xref{RS/6000 and PowerPC Options,, RS/6000 and PowerPC Options, gcc, 1530Using the GNU Compiler Collection (GCC)}. 1531@end ifnothtml 1532@ifhtml 1533see ``RS/6000 and PowerPC Options'' in the main manual. 1534@end ifhtml 1535 1536@option{--with-aix-soname} is currently supported by @samp{libgcc_s} only, so 1537this option is still experimental and not for normal use yet. 1538 1539Default is the traditional behaviour @option{--with-aix-soname=@samp{aix}}. 1540 1541@item --enable-languages=@var{lang1},@var{lang2},@dots{} 1542Specify that only a particular subset of compilers and 1543their runtime libraries should be built. For a list of valid values for 1544@var{langN} you can issue the following command in the 1545@file{gcc} directory of your GCC source tree:@* 1546@smallexample 1547grep language= */config-lang.in 1548@end smallexample 1549Currently, you can use any of the following: 1550@code{all}, @code{ada}, @code{c}, @code{c++}, @code{fortran}, 1551@code{go}, @code{java}, @code{objc}, @code{obj-c++}. 1552Building the Ada compiler has special requirements, see below. 1553If you do not pass this flag, or specify the option @code{all}, then all 1554default languages available in the @file{gcc} sub-tree will be configured. 1555Ada, Go and Objective-C++ are not default languages; the rest are. 1556 1557@item --enable-stage1-languages=@var{lang1},@var{lang2},@dots{} 1558Specify that a particular subset of compilers and their runtime 1559libraries should be built with the system C compiler during stage 1 of 1560the bootstrap process, rather than only in later stages with the 1561bootstrapped C compiler. The list of valid values is the same as for 1562@option{--enable-languages}, and the option @code{all} will select all 1563of the languages enabled by @option{--enable-languages}. This option is 1564primarily useful for GCC development; for instance, when a development 1565version of the compiler cannot bootstrap due to compiler bugs, or when 1566one is debugging front ends other than the C front end. When this 1567option is used, one can then build the target libraries for the 1568specified languages with the stage-1 compiler by using @command{make 1569stage1-bubble all-target}, or run the testsuite on the stage-1 compiler 1570for the specified languages using @command{make stage1-start check-gcc}. 1571 1572@item --disable-libada 1573Specify that the run-time libraries and tools used by GNAT should not 1574be built. This can be useful for debugging, or for compatibility with 1575previous Ada build procedures, when it was required to explicitly 1576do a @samp{make -C gcc gnatlib_and_tools}. 1577 1578@item --disable-libsanitizer 1579Specify that the run-time libraries for the various sanitizers should 1580not be built. 1581 1582@item --disable-libssp 1583Specify that the run-time libraries for stack smashing protection 1584should not be built. 1585 1586@item --disable-libquadmath 1587Specify that the GCC quad-precision math library should not be built. 1588On some systems, the library is required to be linkable when building 1589the Fortran front end, unless @option{--disable-libquadmath-support} 1590is used. 1591 1592@item --disable-libquadmath-support 1593Specify that the Fortran front end and @code{libgfortran} do not add 1594support for @code{libquadmath} on systems supporting it. 1595 1596@item --disable-libgomp 1597Specify that the GNU Offloading and Multi Processing Runtime Library 1598should not be built. 1599 1600@item --disable-libvtv 1601Specify that the run-time libraries used by vtable verification 1602should not be built. 1603 1604@item --with-dwarf2 1605Specify that the compiler should 1606use DWARF 2 debugging information as the default. 1607 1608@item --enable-targets=all 1609@itemx --enable-targets=@var{target_list} 1610Some GCC targets, e.g.@: powerpc64-linux, build bi-arch compilers. 1611These are compilers that are able to generate either 64-bit or 32-bit 1612code. Typically, the corresponding 32-bit target, e.g.@: 1613powerpc-linux for powerpc64-linux, only generates 32-bit code. This 1614option enables the 32-bit target to be a bi-arch compiler, which is 1615useful when you want a bi-arch compiler that defaults to 32-bit, and 1616you are building a bi-arch or multi-arch binutils in a combined tree. 1617On mips-linux, this will build a tri-arch compiler (ABI o32/n32/64), 1618defaulted to o32. 1619Currently, this option only affects sparc-linux, powerpc-linux, x86-linux, 1620mips-linux and s390-linux. 1621 1622@item --enable-secureplt 1623This option enables @option{-msecure-plt} by default for powerpc-linux. 1624@ifnothtml 1625@xref{RS/6000 and PowerPC Options,, RS/6000 and PowerPC Options, gcc, 1626Using the GNU Compiler Collection (GCC)}, 1627@end ifnothtml 1628@ifhtml 1629See ``RS/6000 and PowerPC Options'' in the main manual 1630@end ifhtml 1631 1632@item --enable-cld 1633This option enables @option{-mcld} by default for 32-bit x86 targets. 1634@ifnothtml 1635@xref{i386 and x86-64 Options,, i386 and x86-64 Options, gcc, 1636Using the GNU Compiler Collection (GCC)}, 1637@end ifnothtml 1638@ifhtml 1639See ``i386 and x86-64 Options'' in the main manual 1640@end ifhtml 1641 1642@item --enable-win32-registry 1643@itemx --enable-win32-registry=@var{key} 1644@itemx --disable-win32-registry 1645The @option{--enable-win32-registry} option enables Microsoft Windows-hosted GCC 1646to look up installations paths in the registry using the following key: 1647 1648@smallexample 1649@code{HKEY_LOCAL_MACHINE\SOFTWARE\Free Software Foundation\@var{key}} 1650@end smallexample 1651 1652@var{key} defaults to GCC version number, and can be overridden by the 1653@option{--enable-win32-registry=@var{key}} option. Vendors and distributors 1654who use custom installers are encouraged to provide a different key, 1655perhaps one comprised of vendor name and GCC version number, to 1656avoid conflict with existing installations. This feature is enabled 1657by default, and can be disabled by @option{--disable-win32-registry} 1658option. This option has no effect on the other hosts. 1659 1660@item --nfp 1661Specify that the machine does not have a floating point unit. This 1662option only applies to @samp{m68k-sun-sunos@var{n}}. On any other 1663system, @option{--nfp} has no effect. 1664 1665@item --enable-werror 1666@itemx --disable-werror 1667@itemx --enable-werror=yes 1668@itemx --enable-werror=no 1669When you specify this option, it controls whether certain files in the 1670compiler are built with @option{-Werror} in bootstrap stage2 and later. 1671If you don't specify it, @option{-Werror} is turned on for the main 1672development trunk. However it defaults to off for release branches and 1673final releases. The specific files which get @option{-Werror} are 1674controlled by the Makefiles. 1675 1676@item --enable-checking 1677@itemx --enable-checking=@var{list} 1678When you specify this option, the compiler is built to perform internal 1679consistency checks of the requested complexity. This does not change the 1680generated code, but adds error checking within the compiler. This will 1681slow down the compiler and may only work properly if you are building 1682the compiler with GCC@. This is @samp{yes} by default when building 1683from SVN or snapshots, but @samp{release} for releases. The default 1684for building the stage1 compiler is @samp{yes}. More control 1685over the checks may be had by specifying @var{list}. The categories of 1686checks available are @samp{yes} (most common checks 1687@samp{assert,misc,tree,gc,rtlflag,runtime}), @samp{no} (no checks at 1688all), @samp{all} (all but @samp{valgrind}), @samp{release} (cheapest 1689checks @samp{assert,runtime}) or @samp{none} (same as @samp{no}). 1690Individual checks can be enabled with these flags @samp{assert}, 1691@samp{df}, @samp{fold}, @samp{gc}, @samp{gcac} @samp{misc}, @samp{rtl}, 1692@samp{rtlflag}, @samp{runtime}, @samp{tree}, and @samp{valgrind}. 1693 1694The @samp{valgrind} check requires the external @command{valgrind} 1695simulator, available from @uref{http://valgrind.org/}. The 1696@samp{df}, @samp{rtl}, @samp{gcac} and @samp{valgrind} checks are very expensive. 1697To disable all checking, @samp{--disable-checking} or 1698@samp{--enable-checking=none} must be explicitly requested. Disabling 1699assertions will make the compiler and runtime slightly faster but 1700increase the risk of undetected internal errors causing wrong code to be 1701generated. 1702 1703@item --disable-stage1-checking 1704@itemx --enable-stage1-checking 1705@itemx --enable-stage1-checking=@var{list} 1706If no @option{--enable-checking} option is specified the stage1 1707compiler will be built with @samp{yes} checking enabled, otherwise 1708the stage1 checking flags are the same as specified by 1709@option{--enable-checking}. To build the stage1 compiler with 1710different checking options use @option{--enable-stage1-checking}. 1711The list of checking options is the same as for @option{--enable-checking}. 1712If your system is too slow or too small to bootstrap a released compiler 1713with checking for stage1 enabled, you can use @samp{--disable-stage1-checking} 1714to disable checking for the stage1 compiler. 1715 1716@item --enable-coverage 1717@itemx --enable-coverage=@var{level} 1718With this option, the compiler is built to collect self coverage 1719information, every time it is run. This is for internal development 1720purposes, and only works when the compiler is being built with gcc. The 1721@var{level} argument controls whether the compiler is built optimized or 1722not, values are @samp{opt} and @samp{noopt}. For coverage analysis you 1723want to disable optimization, for performance analysis you want to 1724enable optimization. When coverage is enabled, the default level is 1725without optimization. 1726 1727@item --enable-gather-detailed-mem-stats 1728When this option is specified more detailed information on memory 1729allocation is gathered. This information is printed when using 1730@option{-fmem-report}. 1731 1732@item --enable-nls 1733@itemx --disable-nls 1734The @option{--enable-nls} option enables Native Language Support (NLS), 1735which lets GCC output diagnostics in languages other than American 1736English. Native Language Support is enabled by default if not doing a 1737canadian cross build. The @option{--disable-nls} option disables NLS@. 1738 1739@item --with-included-gettext 1740If NLS is enabled, the @option{--with-included-gettext} option causes the build 1741procedure to prefer its copy of GNU @command{gettext}. 1742 1743@item --with-catgets 1744If NLS is enabled, and if the host lacks @code{gettext} but has the 1745inferior @code{catgets} interface, the GCC build procedure normally 1746ignores @code{catgets} and instead uses GCC's copy of the GNU 1747@code{gettext} library. The @option{--with-catgets} option causes the 1748build procedure to use the host's @code{catgets} in this situation. 1749 1750@item --with-libiconv-prefix=@var{dir} 1751Search for libiconv header files in @file{@var{dir}/include} and 1752libiconv library files in @file{@var{dir}/lib}. 1753 1754@item --enable-obsolete 1755Enable configuration for an obsoleted system. If you attempt to 1756configure GCC for a system (build, host, or target) which has been 1757obsoleted, and you do not specify this flag, configure will halt with an 1758error message. 1759 1760All support for systems which have been obsoleted in one release of GCC 1761is removed entirely in the next major release, unless someone steps 1762forward to maintain the port. 1763 1764@item --enable-decimal-float 1765@itemx --enable-decimal-float=yes 1766@itemx --enable-decimal-float=no 1767@itemx --enable-decimal-float=bid 1768@itemx --enable-decimal-float=dpd 1769@itemx --disable-decimal-float 1770Enable (or disable) support for the C decimal floating point extension 1771that is in the IEEE 754-2008 standard. This is enabled by default only 1772on PowerPC, i386, and x86_64 GNU/Linux systems. Other systems may also 1773support it, but require the user to specifically enable it. You can 1774optionally control which decimal floating point format is used (either 1775@samp{bid} or @samp{dpd}). The @samp{bid} (binary integer decimal) 1776format is default on i386 and x86_64 systems, and the @samp{dpd} 1777(densely packed decimal) format is default on PowerPC systems. 1778 1779@item --enable-fixed-point 1780@itemx --disable-fixed-point 1781Enable (or disable) support for C fixed-point arithmetic. 1782This option is enabled by default for some targets (such as MIPS) which 1783have hardware-support for fixed-point operations. On other targets, you 1784may enable this option manually. 1785 1786@item --with-long-double-128 1787Specify if @code{long double} type should be 128-bit by default on selected 1788GNU/Linux architectures. If using @code{--without-long-double-128}, 1789@code{long double} will be by default 64-bit, the same as @code{double} type. 1790When neither of these configure options are used, the default will be 1791128-bit @code{long double} when built against GNU C Library 2.4 and later, 179264-bit @code{long double} otherwise. 1793 1794@item --with-gmp=@var{pathname} 1795@itemx --with-gmp-include=@var{pathname} 1796@itemx --with-gmp-lib=@var{pathname} 1797@itemx --with-mpfr=@var{pathname} 1798@itemx --with-mpfr-include=@var{pathname} 1799@itemx --with-mpfr-lib=@var{pathname} 1800@itemx --with-mpc=@var{pathname} 1801@itemx --with-mpc-include=@var{pathname} 1802@itemx --with-mpc-lib=@var{pathname} 1803If you want to build GCC but do not have the GMP library, the MPFR 1804library and/or the MPC library installed in a standard location and 1805do not have their sources present in the GCC source tree then you 1806can explicitly specify the directory where they are installed 1807(@samp{--with-gmp=@var{gmpinstalldir}}, 1808@samp{--with-mpfr=@/@var{mpfrinstalldir}}, 1809@samp{--with-mpc=@/@var{mpcinstalldir}}). The 1810@option{--with-gmp=@/@var{gmpinstalldir}} option is shorthand for 1811@option{--with-gmp-lib=@/@var{gmpinstalldir}/lib} and 1812@option{--with-gmp-include=@/@var{gmpinstalldir}/include}. Likewise the 1813@option{--with-mpfr=@/@var{mpfrinstalldir}} option is shorthand for 1814@option{--with-mpfr-lib=@/@var{mpfrinstalldir}/lib} and 1815@option{--with-mpfr-include=@/@var{mpfrinstalldir}/include}, also the 1816@option{--with-mpc=@/@var{mpcinstalldir}} option is shorthand for 1817@option{--with-mpc-lib=@/@var{mpcinstalldir}/lib} and 1818@option{--with-mpc-include=@/@var{mpcinstalldir}/include}. If these 1819shorthand assumptions are not correct, you can use the explicit 1820include and lib options directly. You might also need to ensure the 1821shared libraries can be found by the dynamic linker when building and 1822using GCC, for example by setting the runtime shared library path 1823variable (@env{LD_LIBRARY_PATH} on GNU/Linux and Solaris systems). 1824 1825These flags are applicable to the host platform only. When building 1826a cross compiler, they will not be used to configure target libraries. 1827 1828@item --with-isl=@var{pathname} 1829@itemx --with-isl-include=@var{pathname} 1830@itemx --with-isl-lib=@var{pathname} 1831If you do not have the ISL library installed in a standard location and you 1832want to build GCC, you can explicitly specify the directory where it is 1833installed (@samp{--with-isl=@/@var{islinstalldir}}). The 1834@option{--with-isl=@/@var{islinstalldir}} option is shorthand for 1835@option{--with-isl-lib=@/@var{islinstalldir}/lib} and 1836@option{--with-isl-include=@/@var{islinstalldir}/include}. If this 1837shorthand assumption is not correct, you can use the explicit 1838include and lib options directly. 1839 1840These flags are applicable to the host platform only. When building 1841a cross compiler, they will not be used to configure target libraries. 1842 1843@item --with-host-libstdcxx=@var{linker-args} 1844If you are linking with a static copy of PPL, you can use this option 1845to specify how the linker should find the standard C++ library used 1846internally by PPL. Typical values of @var{linker-args} might be 1847@samp{-lstdc++} or @samp{-Wl,-Bstatic,-lstdc++,-Bdynamic -lm}. If you are 1848linking with a shared copy of PPL, you probably do not need this 1849option; shared library dependencies will cause the linker to search 1850for the standard C++ library automatically. 1851 1852@item --with-stage1-ldflags=@var{flags} 1853This option may be used to set linker flags to be used when linking 1854stage 1 of GCC. These are also used when linking GCC if configured with 1855@option{--disable-bootstrap}. By default no special flags are used. 1856 1857@item --with-stage1-libs=@var{libs} 1858This option may be used to set libraries to be used when linking stage 1 1859of GCC. These are also used when linking GCC if configured with 1860@option{--disable-bootstrap}. The default is the argument to 1861@option{--with-host-libstdcxx}, if specified. 1862 1863@item --with-boot-ldflags=@var{flags} 1864This option may be used to set linker flags to be used when linking 1865stage 2 and later when bootstrapping GCC. If neither --with-boot-libs 1866nor --with-host-libstdcxx is set to a value, then the default is 1867@samp{-static-libstdc++ -static-libgcc}. 1868 1869@item --with-boot-libs=@var{libs} 1870This option may be used to set libraries to be used when linking stage 2 1871and later when bootstrapping GCC. The default is the argument to 1872@option{--with-host-libstdcxx}, if specified. 1873 1874@item --with-debug-prefix-map=@var{map} 1875Convert source directory names using @option{-fdebug-prefix-map} when 1876building runtime libraries. @samp{@var{map}} is a space-separated 1877list of maps of the form @samp{@var{old}=@var{new}}. 1878 1879@item --enable-linker-build-id 1880Tells GCC to pass @option{--build-id} option to the linker for all final 1881links (links performed without the @option{-r} or @option{--relocatable} 1882option), if the linker supports it. If you specify 1883@option{--enable-linker-build-id}, but your linker does not 1884support @option{--build-id} option, a warning is issued and the 1885@option{--enable-linker-build-id} option is ignored. The default is off. 1886 1887@item --with-linker-hash-style=@var{choice} 1888Tells GCC to pass @option{--hash-style=@var{choice}} option to the 1889linker for all final links. @var{choice} can be one of 1890@samp{sysv}, @samp{gnu}, and @samp{both} where @samp{sysv} is the default. 1891 1892@item --enable-gnu-unique-object 1893@itemx --disable-gnu-unique-object 1894Tells GCC to use the gnu_unique_object relocation for C++ template 1895static data members and inline function local statics. Enabled by 1896default for a toolchain with an assembler that accepts it and 1897GLIBC 2.11 or above, otherwise disabled. 1898 1899@item --with-diagnostics-color=@var{choice} 1900Tells GCC to use @var{choice} as the default for @option{-fdiagnostics-color=} 1901option (if not used explicitly on the command line). @var{choice} 1902can be one of @samp{never}, @samp{auto}, @samp{always}, and @samp{auto-if-env} 1903where @samp{auto} is the default. @samp{auto-if-env} means that 1904@option{-fdiagnostics-color=auto} will be the default if @code{GCC_COLORS} 1905is present and non-empty in the environment, and 1906@option{-fdiagnostics-color=never} otherwise. 1907 1908@item --enable-lto 1909@itemx --disable-lto 1910Enable support for link-time optimization (LTO). This is enabled by 1911default, and may be disabled using @option{--disable-lto}. 1912 1913@item --enable-linker-plugin-configure-flags=FLAGS 1914@itemx --enable-linker-plugin-flags=FLAGS 1915By default, linker plugins (such as the LTO plugin) are built for the 1916host system architecture. For the case that the linker has a 1917different (but run-time compatible) architecture, these flags can be 1918specified to build plugins that are compatible to the linker. For 1919example, if you are building GCC for a 64-bit x86_64 1920(@samp{x86_64-unknown-linux-gnu}) host system, but have a 32-bit x86 1921GNU/Linux (@samp{i686-pc-linux-gnu}) linker executable (which is 1922executable on the former system), you can configure GCC as follows for 1923getting compatible linker plugins: 1924 1925@smallexample 1926% @var{srcdir}/configure \ 1927 --host=x86_64-unknown-linux-gnu \ 1928 --enable-linker-plugin-configure-flags=--host=i686-pc-linux-gnu \ 1929 --enable-linker-plugin-flags='CC=gcc\ -m32\ -Wl,-rpath,[...]/i686-pc-linux-gnu/lib' 1930@end smallexample 1931 1932@item --with-plugin-ld=@var{pathname} 1933Enable an alternate linker to be used at link-time optimization (LTO) 1934link time when @option{-fuse-linker-plugin} is enabled. 1935This linker should have plugin support such as gold starting with 1936version 2.20 or GNU ld starting with version 2.21. 1937See @option{-fuse-linker-plugin} for details. 1938 1939@item --enable-canonical-system-headers 1940@itemx --disable-canonical-system-headers 1941Enable system header path canonicalization for @file{libcpp}. This can 1942produce shorter header file paths in diagnostics and dependency output 1943files, but these changed header paths may conflict with some compilation 1944environments. Enabled by default, and may be disabled using 1945@option{--disable-canonical-system-headers}. 1946 1947@item --with-glibc-version=@var{major}.@var{minor} 1948Tell GCC that when the GNU C Library (glibc) is used on the target it 1949will be version @var{major}.@var{minor} or later. Normally this can 1950be detected from the C library's header files, but this option may be 1951needed when bootstrapping a cross toolchain without the header files 1952available for building the initial bootstrap compiler. 1953 1954If GCC is configured with some multilibs that use glibc and some that 1955do not, this option applies only to the multilibs that use glibc. 1956However, such configurations may not work well as not all the relevant 1957configuration in GCC is on a per-multilib basis. 1958 1959@item --enable-as-accelerator-for=@var{target} 1960Build as offload target compiler. Specify offload host triple by @var{target}. 1961 1962@item --enable-offload-targets=@var{target1}[=@var{path1}],@dots{},@var{targetN}[=@var{pathN}] 1963Enable offloading to targets @var{target1}, @dots{}, @var{targetN}. 1964Offload compilers are expected to be already installed. Default search 1965path for them is @file{@var{exec-prefix}}, but it can be changed by 1966specifying paths @var{path1}, @dots{}, @var{pathN}. 1967 1968@smallexample 1969% @var{srcdir}/configure \ 1970 --enable-offload-target=i686-unknown-linux-gnu=/path/to/i686/compiler,x86_64-pc-linux-gnu 1971@end smallexample 1972@end table 1973 1974@subheading Cross-Compiler-Specific Options 1975The following options only apply to building cross compilers. 1976 1977@table @code 1978@item --with-sysroot 1979@itemx --with-sysroot=@var{dir} 1980Tells GCC to consider @var{dir} as the root of a tree that contains 1981(a subset of) the root filesystem of the target operating system. 1982Target system headers, libraries and run-time object files will be 1983searched for in there. More specifically, this acts as if 1984@option{--sysroot=@var{dir}} was added to the default options of the built 1985compiler. The specified directory is not copied into the 1986install tree, unlike the options @option{--with-headers} and 1987@option{--with-libs} that this option obsoletes. The default value, 1988in case @option{--with-sysroot} is not given an argument, is 1989@option{$@{gcc_tooldir@}/sys-root}. If the specified directory is a 1990subdirectory of @option{$@{exec_prefix@}}, then it will be found relative to 1991the GCC binaries if the installation tree is moved. 1992 1993This option affects the system root for the compiler used to build 1994target libraries (which runs on the build system) and the compiler newly 1995installed with @code{make install}; it does not affect the compiler which is 1996used to build GCC itself. 1997 1998If you specify the @option{--with-native-system-header-dir=@var{dirname}} 1999option then the compiler will search that directory within @var{dirname} for 2000native system headers rather than the default @file{/usr/include}. 2001 2002@item --with-build-sysroot 2003@itemx --with-build-sysroot=@var{dir} 2004Tells GCC to consider @var{dir} as the system root (see 2005@option{--with-sysroot}) while building target libraries, instead of 2006the directory specified with @option{--with-sysroot}. This option is 2007only useful when you are already using @option{--with-sysroot}. You 2008can use @option{--with-build-sysroot} when you are configuring with 2009@option{--prefix} set to a directory that is different from the one in 2010which you are installing GCC and your target libraries. 2011 2012This option affects the system root for the compiler used to build 2013target libraries (which runs on the build system); it does not affect 2014the compiler which is used to build GCC itself. 2015 2016If you specify the @option{--with-native-system-header-dir=@var{dirname}} 2017option then the compiler will search that directory within @var{dirname} for 2018native system headers rather than the default @file{/usr/include}. 2019 2020@item --with-headers 2021@itemx --with-headers=@var{dir} 2022Deprecated in favor of @option{--with-sysroot}. 2023Specifies that target headers are available when building a cross compiler. 2024The @var{dir} argument specifies a directory which has the target include 2025files. These include files will be copied into the @file{gcc} install 2026directory. @emph{This option with the @var{dir} argument is required} when 2027building a cross compiler, if @file{@var{prefix}/@var{target}/sys-include} 2028doesn't pre-exist. If @file{@var{prefix}/@var{target}/sys-include} does 2029pre-exist, the @var{dir} argument may be omitted. @command{fixincludes} 2030will be run on these files to make them compatible with GCC@. 2031 2032@item --without-headers 2033Tells GCC not use any target headers from a libc when building a cross 2034compiler. When crossing to GNU/Linux, you need the headers so GCC 2035can build the exception handling for libgcc. 2036 2037@item --with-libs 2038@itemx --with-libs="@var{dir1} @var{dir2} @dots{} @var{dirN}" 2039Deprecated in favor of @option{--with-sysroot}. 2040Specifies a list of directories which contain the target runtime 2041libraries. These libraries will be copied into the @file{gcc} install 2042directory. If the directory list is omitted, this option has no 2043effect. 2044 2045@item --with-newlib 2046Specifies that @samp{newlib} is 2047being used as the target C library. This causes @code{__eprintf} to be 2048omitted from @file{libgcc.a} on the assumption that it will be provided by 2049@samp{newlib}. 2050 2051@item --with-avrlibc 2052Specifies that @samp{AVR-Libc} is 2053being used as the target C library. This causes float support 2054functions like @code{__addsf3} to be omitted from @file{libgcc.a} on 2055the assumption that it will be provided by @file{libm.a}. For more 2056technical details, cf. @uref{http://gcc.gnu.org/PR54461,,PR54461}. 2057This option is only supported for the AVR target. It is not supported for 2058RTEMS configurations, which currently use newlib. The option is 2059supported since version 4.7.2 and is the default in 4.8.0 and newer. 2060 2061@item --with-nds32-lib=@var{library} 2062Specifies that @var{library} setting is used for building @file{libgcc.a}. 2063Currently, the valid @var{library} is @samp{newlib} or @samp{mculib}. 2064This option is only supported for the NDS32 target. 2065 2066@item --with-build-time-tools=@var{dir} 2067Specifies where to find the set of target tools (assembler, linker, etc.) 2068that will be used while building GCC itself. This option can be useful 2069if the directory layouts are different between the system you are building 2070GCC on, and the system where you will deploy it. 2071 2072For example, on an @samp{ia64-hp-hpux} system, you may have the GNU 2073assembler and linker in @file{/usr/bin}, and the native tools in a 2074different path, and build a toolchain that expects to find the 2075native tools in @file{/usr/bin}. 2076 2077When you use this option, you should ensure that @var{dir} includes 2078@command{ar}, @command{as}, @command{ld}, @command{nm}, 2079@command{ranlib} and @command{strip} if necessary, and possibly 2080@command{objdump}. Otherwise, GCC may use an inconsistent set of 2081tools. 2082@end table 2083 2084@subheading Java-Specific Options 2085 2086The following option applies to the build of the Java front end. 2087 2088@table @code 2089@item --disable-libgcj 2090Specify that the run-time libraries 2091used by GCJ should not be built. This is useful in case you intend 2092to use GCJ with some other run-time, or you're going to install it 2093separately, or it just happens not to build on your particular 2094machine. In general, if the Java front end is enabled, the GCJ 2095libraries will be enabled too, unless they're known to not work on 2096the target platform. If GCJ is enabled but @samp{libgcj} isn't built, you 2097may need to port it; in this case, before modifying the top-level 2098@file{configure.in} so that @samp{libgcj} is enabled by default on this platform, 2099you may use @option{--enable-libgcj} to override the default. 2100 2101@end table 2102 2103The following options apply to building @samp{libgcj}. 2104 2105@subsubheading General Options 2106 2107@table @code 2108@item --enable-java-maintainer-mode 2109By default the @samp{libjava} build will not attempt to compile the 2110@file{.java} source files to @file{.class}. Instead, it will use the 2111@file{.class} files from the source tree. If you use this option you 2112must have executables named @command{ecj1} and @command{gjavah} in your path 2113for use by the build. You must use this option if you intend to 2114modify any @file{.java} files in @file{libjava}. 2115 2116@item --with-java-home=@var{dirname} 2117This @samp{libjava} option overrides the default value of the 2118@samp{java.home} system property. It is also used to set 2119@samp{sun.boot.class.path} to @file{@var{dirname}/lib/rt.jar}. By 2120default @samp{java.home} is set to @file{@var{prefix}} and 2121@samp{sun.boot.class.path} to 2122@file{@var{datadir}/java/libgcj-@var{version}.jar}. 2123 2124@item --with-ecj-jar=@var{filename} 2125This option can be used to specify the location of an external jar 2126file containing the Eclipse Java compiler. A specially modified 2127version of this compiler is used by @command{gcj} to parse 2128@file{.java} source files. If this option is given, the 2129@samp{libjava} build will create and install an @file{ecj1} executable 2130which uses this jar file at runtime. 2131 2132If this option is not given, but an @file{ecj.jar} file is found in 2133the topmost source tree at configure time, then the @samp{libgcj} 2134build will create and install @file{ecj1}, and will also install the 2135discovered @file{ecj.jar} into a suitable place in the install tree. 2136 2137If @file{ecj1} is not installed, then the user will have to supply one 2138on his path in order for @command{gcj} to properly parse @file{.java} 2139source files. A suitable jar is available from 2140@uref{ftp://sourceware.org/pub/java/}. 2141 2142@item --disable-getenv-properties 2143Don't set system properties from @env{GCJ_PROPERTIES}. 2144 2145@item --enable-hash-synchronization 2146Use a global hash table for monitor locks. Ordinarily, 2147@samp{libgcj}'s @samp{configure} script automatically makes 2148the correct choice for this option for your platform. Only use 2149this if you know you need the library to be configured differently. 2150 2151@item --enable-interpreter 2152Enable the Java interpreter. The interpreter is automatically 2153enabled by default on all platforms that support it. This option 2154is really only useful if you want to disable the interpreter 2155(using @option{--disable-interpreter}). 2156 2157@item --disable-java-net 2158Disable java.net. This disables the native part of java.net only, 2159using non-functional stubs for native method implementations. 2160 2161@item --disable-jvmpi 2162Disable JVMPI support. 2163 2164@item --disable-libgcj-bc 2165Disable BC ABI compilation of certain parts of libgcj. By default, 2166some portions of libgcj are compiled with @option{-findirect-dispatch} 2167and @option{-fno-indirect-classes}, allowing them to be overridden at 2168run-time. 2169 2170If @option{--disable-libgcj-bc} is specified, libgcj is built without 2171these options. This allows the compile-time linker to resolve 2172dependencies when statically linking to libgcj. However it makes it 2173impossible to override the affected portions of libgcj at run-time. 2174 2175@item --enable-reduced-reflection 2176Build most of libgcj with @option{-freduced-reflection}. This reduces 2177the size of libgcj at the expense of not being able to do accurate 2178reflection on the classes it contains. This option is safe if you 2179know that code using libgcj will never use reflection on the standard 2180runtime classes in libgcj (including using serialization, RMI or CORBA). 2181 2182@item --with-ecos 2183Enable runtime eCos target support. 2184 2185@item --without-libffi 2186Don't use @samp{libffi}. This will disable the interpreter and JNI 2187support as well, as these require @samp{libffi} to work. 2188 2189@item --enable-libgcj-debug 2190Enable runtime debugging code. 2191 2192@item --enable-libgcj-multifile 2193If specified, causes all @file{.java} source files to be 2194compiled into @file{.class} files in one invocation of 2195@samp{gcj}. This can speed up build time, but is more 2196resource-intensive. If this option is unspecified or 2197disabled, @samp{gcj} is invoked once for each @file{.java} 2198file to compile into a @file{.class} file. 2199 2200@item --with-libiconv-prefix=DIR 2201Search for libiconv in @file{DIR/include} and @file{DIR/lib}. 2202 2203@item --enable-sjlj-exceptions 2204Force use of the @code{setjmp}/@code{longjmp}-based scheme for exceptions. 2205@samp{configure} ordinarily picks the correct value based on the platform. 2206Only use this option if you are sure you need a different setting. 2207 2208@item --with-system-zlib 2209Use installed @samp{zlib} rather than that included with GCC@. 2210 2211@item --with-win32-nlsapi=ansi, unicows or unicode 2212Indicates how MinGW @samp{libgcj} translates between UNICODE 2213characters and the Win32 API@. 2214 2215@item --enable-java-home 2216If enabled, this creates a JPackage compatible SDK environment during install. 2217Note that if --enable-java-home is used, --with-arch-directory=ARCH must also 2218be specified. 2219 2220@item --with-arch-directory=ARCH 2221Specifies the name to use for the @file{jre/lib/ARCH} directory in the SDK 2222environment created when --enable-java-home is passed. Typical names for this 2223directory include i386, amd64, ia64, etc. 2224 2225@item --with-os-directory=DIR 2226Specifies the OS directory for the SDK include directory. This is set to auto 2227detect, and is typically 'linux'. 2228 2229@item --with-origin-name=NAME 2230Specifies the JPackage origin name. This defaults to the 'gcj' in 2231java-1.5.0-gcj. 2232 2233@item --with-arch-suffix=SUFFIX 2234Specifies the suffix for the sdk directory. Defaults to the empty string. 2235Examples include '.x86_64' in 'java-1.5.0-gcj-1.5.0.0.x86_64'. 2236 2237@item --with-jvm-root-dir=DIR 2238Specifies where to install the SDK. Default is $(prefix)/lib/jvm. 2239 2240@item --with-jvm-jar-dir=DIR 2241Specifies where to install jars. Default is $(prefix)/lib/jvm-exports. 2242 2243@item --with-python-dir=DIR 2244Specifies where to install the Python modules used for aot-compile. DIR should 2245not include the prefix used in installation. For example, if the Python modules 2246are to be installed in /usr/lib/python2.5/site-packages, then 2247--with-python-dir=/lib/python2.5/site-packages should be passed. If this is 2248not specified, then the Python modules are installed in $(prefix)/share/python. 2249 2250@item --enable-aot-compile-rpm 2251Adds aot-compile-rpm to the list of installed scripts. 2252 2253@item --enable-browser-plugin 2254Build the gcjwebplugin web browser plugin. 2255 2256@item --enable-static-libjava 2257Build static libraries in libjava. The default is to only build shared 2258libraries. 2259 2260@table @code 2261@item ansi 2262Use the single-byte @code{char} and the Win32 A functions natively, 2263translating to and from UNICODE when using these functions. If 2264unspecified, this is the default. 2265 2266@item unicows 2267Use the @code{WCHAR} and Win32 W functions natively. Adds 2268@code{-lunicows} to @file{libgcj.spec} to link with @samp{libunicows}. 2269@file{unicows.dll} needs to be deployed on Microsoft Windows 9X machines 2270running built executables. @file{libunicows.a}, an open-source 2271import library around Microsoft's @code{unicows.dll}, is obtained from 2272@uref{http://libunicows.sourceforge.net/}, which also gives details 2273on getting @file{unicows.dll} from Microsoft. 2274 2275@item unicode 2276Use the @code{WCHAR} and Win32 W functions natively. Does @emph{not} 2277add @code{-lunicows} to @file{libgcj.spec}. The built executables will 2278only run on Microsoft Windows NT and above. 2279@end table 2280@end table 2281 2282@subsubheading AWT-Specific Options 2283 2284@table @code 2285@item --with-x 2286Use the X Window System. 2287 2288@item --enable-java-awt=PEER(S) 2289Specifies the AWT peer library or libraries to build alongside 2290@samp{libgcj}. If this option is unspecified or disabled, AWT 2291will be non-functional. Current valid values are @option{gtk} and 2292@option{xlib}. Multiple libraries should be separated by a 2293comma (i.e.@: @option{--enable-java-awt=gtk,xlib}). 2294 2295@item --enable-gtk-cairo 2296Build the cairo Graphics2D implementation on GTK@. 2297 2298@item --enable-java-gc=TYPE 2299Choose garbage collector. Defaults to @option{boehm} if unspecified. 2300 2301@item --disable-gtktest 2302Do not try to compile and run a test GTK+ program. 2303 2304@item --disable-glibtest 2305Do not try to compile and run a test GLIB program. 2306 2307@item --with-libart-prefix=PFX 2308Prefix where libart is installed (optional). 2309 2310@item --with-libart-exec-prefix=PFX 2311Exec prefix where libart is installed (optional). 2312 2313@item --disable-libarttest 2314Do not try to compile and run a test libart program. 2315 2316@end table 2317 2318@subsubheading Overriding @command{configure} test results 2319 2320Sometimes, it might be necessary to override the result of some 2321@command{configure} test, for example in order to ease porting to a new 2322system or work around a bug in a test. The toplevel @command{configure} 2323script provides three variables for this: 2324 2325@table @code 2326 2327@item build_configargs 2328@cindex @code{build_configargs} 2329The contents of this variable is passed to all build @command{configure} 2330scripts. 2331 2332@item host_configargs 2333@cindex @code{host_configargs} 2334The contents of this variable is passed to all host @command{configure} 2335scripts. 2336 2337@item target_configargs 2338@cindex @code{target_configargs} 2339The contents of this variable is passed to all target @command{configure} 2340scripts. 2341 2342@end table 2343 2344In order to avoid shell and @command{make} quoting issues for complex 2345overrides, you can pass a setting for @env{CONFIG_SITE} and set 2346variables in the site file. 2347 2348@html 2349<hr /> 2350<p> 2351@end html 2352@ifhtml 2353@uref{./index.html,,Return to the GCC Installation page} 2354@end ifhtml 2355@end ifset 2356 2357@c ***Building**************************************************************** 2358@ifnothtml 2359@comment node-name, next, previous, up 2360@node Building, Testing, Configuration, Installing GCC 2361@end ifnothtml 2362@ifset buildhtml 2363@ifnothtml 2364@chapter Building 2365@end ifnothtml 2366@cindex Installing GCC: Building 2367 2368Now that GCC is configured, you are ready to build the compiler and 2369runtime libraries. 2370 2371Some commands executed when making the compiler may fail (return a 2372nonzero status) and be ignored by @command{make}. These failures, which 2373are often due to files that were not found, are expected, and can safely 2374be ignored. 2375 2376It is normal to have compiler warnings when compiling certain files. 2377Unless you are a GCC developer, you can generally ignore these warnings 2378unless they cause compilation to fail. Developers should attempt to fix 2379any warnings encountered, however they can temporarily continue past 2380warnings-as-errors by specifying the configure flag 2381@option{--disable-werror}. 2382 2383On certain old systems, defining certain environment variables such as 2384@env{CC} can interfere with the functioning of @command{make}. 2385 2386If you encounter seemingly strange errors when trying to build the 2387compiler in a directory other than the source directory, it could be 2388because you have previously configured the compiler in the source 2389directory. Make sure you have done all the necessary preparations. 2390 2391If you build GCC on a BSD system using a directory stored in an old System 2392V file system, problems may occur in running @command{fixincludes} if the 2393System V file system doesn't support symbolic links. These problems 2394result in a failure to fix the declaration of @code{size_t} in 2395@file{sys/types.h}. If you find that @code{size_t} is a signed type and 2396that type mismatches occur, this could be the cause. 2397 2398The solution is not to use such a directory for building GCC@. 2399 2400Similarly, when building from SVN or snapshots, or if you modify 2401@file{*.l} files, you need the Flex lexical analyzer generator 2402installed. If you do not modify @file{*.l} files, releases contain 2403the Flex-generated files and you do not need Flex installed to build 2404them. There is still one Flex-based lexical analyzer (part of the 2405build machinery, not of GCC itself) that is used even if you only 2406build the C front end. 2407 2408When building from SVN or snapshots, or if you modify Texinfo 2409documentation, you need version 4.7 or later of Texinfo installed if you 2410want Info documentation to be regenerated. Releases contain Info 2411documentation pre-built for the unmodified documentation in the release. 2412 2413@section Building a native compiler 2414 2415For a native build, the default configuration is to perform 2416a 3-stage bootstrap of the compiler when @samp{make} is invoked. 2417This will build the entire GCC system and ensure that it compiles 2418itself correctly. It can be disabled with the @option{--disable-bootstrap} 2419parameter to @samp{configure}, but bootstrapping is suggested because 2420the compiler will be tested more completely and could also have 2421better performance. 2422 2423The bootstrapping process will complete the following steps: 2424 2425@itemize @bullet 2426@item 2427Build tools necessary to build the compiler. 2428 2429@item 2430Perform a 3-stage bootstrap of the compiler. This includes building 2431three times the target tools for use by the compiler such as binutils 2432(bfd, binutils, gas, gprof, ld, and opcodes) if they have been 2433individually linked or moved into the top level GCC source tree before 2434configuring. 2435 2436@item 2437Perform a comparison test of the stage2 and stage3 compilers. 2438 2439@item 2440Build runtime libraries using the stage3 compiler from the previous step. 2441 2442@end itemize 2443 2444If you are short on disk space you might consider @samp{make 2445bootstrap-lean} instead. The sequence of compilation is the 2446same described above, but object files from the stage1 and 2447stage2 of the 3-stage bootstrap of the compiler are deleted as 2448soon as they are no longer needed. 2449 2450If you wish to use non-default GCC flags when compiling the stage2 2451and stage3 compilers, set @code{BOOT_CFLAGS} on the command line when 2452doing @samp{make}. For example, if you want to save additional space 2453during the bootstrap and in the final installation as well, you can 2454build the compiler binaries without debugging information as in the 2455following example. This will save roughly 40% of disk space both for 2456the bootstrap and the final installation. (Libraries will still contain 2457debugging information.) 2458 2459@smallexample 2460make BOOT_CFLAGS='-O' bootstrap 2461@end smallexample 2462 2463You can place non-default optimization flags into @code{BOOT_CFLAGS}; they 2464are less well tested here than the default of @samp{-g -O2}, but should 2465still work. In a few cases, you may find that you need to specify special 2466flags such as @option{-msoft-float} here to complete the bootstrap; or, 2467if the native compiler miscompiles the stage1 compiler, you may need 2468to work around this, by choosing @code{BOOT_CFLAGS} to avoid the parts 2469of the stage1 compiler that were miscompiled, or by using @samp{make 2470bootstrap4} to increase the number of stages of bootstrap. 2471 2472@code{BOOT_CFLAGS} does not apply to bootstrapped target libraries. 2473Since these are always compiled with the compiler currently being 2474bootstrapped, you can use @code{CFLAGS_FOR_TARGET} to modify their 2475compilation flags, as for non-bootstrapped target libraries. 2476Again, if the native compiler miscompiles the stage1 compiler, you may 2477need to work around this by avoiding non-working parts of the stage1 2478compiler. Use @code{STAGE1_TFLAGS} to this end. 2479 2480If you used the flag @option{--enable-languages=@dots{}} to restrict 2481the compilers to be built, only those you've actually enabled will be 2482built. This will of course only build those runtime libraries, for 2483which the particular compiler has been built. Please note, 2484that re-defining @env{LANGUAGES} when calling @samp{make} 2485@strong{does not} work anymore! 2486 2487If the comparison of stage2 and stage3 fails, this normally indicates 2488that the stage2 compiler has compiled GCC incorrectly, and is therefore 2489a potentially serious bug which you should investigate and report. (On 2490a few systems, meaningful comparison of object files is impossible; they 2491always appear ``different''. If you encounter this problem, you will 2492need to disable comparison in the @file{Makefile}.) 2493 2494If you do not want to bootstrap your compiler, you can configure with 2495@option{--disable-bootstrap}. In particular cases, you may want to 2496bootstrap your compiler even if the target system is not the same as 2497the one you are building on: for example, you could build a 2498@code{powerpc-unknown-linux-gnu} toolchain on a 2499@code{powerpc64-unknown-linux-gnu} host. In this case, pass 2500@option{--enable-bootstrap} to the configure script. 2501 2502@code{BUILD_CONFIG} can be used to bring in additional customization 2503to the build. It can be set to a whitespace-separated list of names. 2504For each such @code{NAME}, top-level @file{config/@code{NAME}.mk} will 2505be included by the top-level @file{Makefile}, bringing in any settings 2506it contains. The default @code{BUILD_CONFIG} can be set using the 2507configure option @option{--with-build-config=@code{NAME}...}. Some 2508examples of supported build configurations are: 2509 2510@table @asis 2511@item @samp{bootstrap-O1} 2512Removes any @option{-O}-started option from @code{BOOT_CFLAGS}, and adds 2513@option{-O1} to it. @samp{BUILD_CONFIG=bootstrap-O1} is equivalent to 2514@samp{BOOT_CFLAGS='-g -O1'}. 2515 2516@item @samp{bootstrap-O3} 2517Analogous to @code{bootstrap-O1}. 2518 2519@item @samp{bootstrap-lto} 2520Enables Link-Time Optimization for host tools during bootstrapping. 2521@samp{BUILD_CONFIG=bootstrap-lto} is equivalent to adding 2522@option{-flto} to @samp{BOOT_CFLAGS}. This option assumes that the host 2523supports the linker plugin (e.g. GNU ld version 2.21 or later or GNU gold 2524version 2.21 or later). 2525 2526@item @samp{bootstrap-lto-noplugin} 2527This option is similar to @code{bootstrap-lto}, but is intended for 2528hosts that do not support the linker plugin. Without the linker plugin 2529static libraries are not compiled with link-time optimizations. Since 2530the GCC middle end and back end are in @file{libbackend.a} this means 2531that only the front end is actually LTO optimized. 2532 2533@item @samp{bootstrap-debug} 2534Verifies that the compiler generates the same executable code, whether 2535or not it is asked to emit debug information. To this end, this 2536option builds stage2 host programs without debug information, and uses 2537@file{contrib/compare-debug} to compare them with the stripped stage3 2538object files. If @code{BOOT_CFLAGS} is overridden so as to not enable 2539debug information, stage2 will have it, and stage3 won't. This option 2540is enabled by default when GCC bootstrapping is enabled, if 2541@code{strip} can turn object files compiled with and without debug 2542info into identical object files. In addition to better test 2543coverage, this option makes default bootstraps faster and leaner. 2544 2545@item @samp{bootstrap-debug-big} 2546Rather than comparing stripped object files, as in 2547@code{bootstrap-debug}, this option saves internal compiler dumps 2548during stage2 and stage3 and compares them as well, which helps catch 2549additional potential problems, but at a great cost in terms of disk 2550space. It can be specified in addition to @samp{bootstrap-debug}. 2551 2552@item @samp{bootstrap-debug-lean} 2553This option saves disk space compared with @code{bootstrap-debug-big}, 2554but at the expense of some recompilation. Instead of saving the dumps 2555of stage2 and stage3 until the final compare, it uses 2556@option{-fcompare-debug} to generate, compare and remove the dumps 2557during stage3, repeating the compilation that already took place in 2558stage2, whose dumps were not saved. 2559 2560@item @samp{bootstrap-debug-lib} 2561This option tests executable code invariance over debug information 2562generation on target libraries, just like @code{bootstrap-debug-lean} 2563tests it on host programs. It builds stage3 libraries with 2564@option{-fcompare-debug}, and it can be used along with any of the 2565@code{bootstrap-debug} options above. 2566 2567There aren't @code{-lean} or @code{-big} counterparts to this option 2568because most libraries are only build in stage3, so bootstrap compares 2569would not get significant coverage. Moreover, the few libraries built 2570in stage2 are used in stage3 host programs, so we wouldn't want to 2571compile stage2 libraries with different options for comparison purposes. 2572 2573@item @samp{bootstrap-debug-ckovw} 2574Arranges for error messages to be issued if the compiler built on any 2575stage is run without the option @option{-fcompare-debug}. This is 2576useful to verify the full @option{-fcompare-debug} testing coverage. It 2577must be used along with @code{bootstrap-debug-lean} and 2578@code{bootstrap-debug-lib}. 2579 2580@item @samp{bootstrap-time} 2581Arranges for the run time of each program started by the GCC driver, 2582built in any stage, to be logged to @file{time.log}, in the top level of 2583the build tree. 2584 2585@end table 2586 2587@section Building a cross compiler 2588 2589When building a cross compiler, it is not generally possible to do a 25903-stage bootstrap of the compiler. This makes for an interesting problem 2591as parts of GCC can only be built with GCC@. 2592 2593To build a cross compiler, we recommend first building and installing a 2594native compiler. You can then use the native GCC compiler to build the 2595cross compiler. The installed native compiler needs to be GCC version 25962.95 or later. 2597 2598If the cross compiler is to be built with support for the Java 2599programming language and the ability to compile .java source files is 2600desired, the installed native compiler used to build the cross 2601compiler needs to be the same GCC version as the cross compiler. In 2602addition the cross compiler needs to be configured with 2603@option{--with-ecj-jar=@dots{}}. 2604 2605Assuming you have already installed a native copy of GCC and configured 2606your cross compiler, issue the command @command{make}, which performs the 2607following steps: 2608 2609@itemize @bullet 2610@item 2611Build host tools necessary to build the compiler. 2612 2613@item 2614Build target tools for use by the compiler such as binutils (bfd, 2615binutils, gas, gprof, ld, and opcodes) 2616if they have been individually linked or moved into the top level GCC source 2617tree before configuring. 2618 2619@item 2620Build the compiler (single stage only). 2621 2622@item 2623Build runtime libraries using the compiler from the previous step. 2624@end itemize 2625 2626Note that if an error occurs in any step the make process will exit. 2627 2628If you are not building GNU binutils in the same source tree as GCC, 2629you will need a cross-assembler and cross-linker installed before 2630configuring GCC@. Put them in the directory 2631@file{@var{prefix}/@var{target}/bin}. Here is a table of the tools 2632you should put in this directory: 2633 2634@table @file 2635@item as 2636This should be the cross-assembler. 2637 2638@item ld 2639This should be the cross-linker. 2640 2641@item ar 2642This should be the cross-archiver: a program which can manipulate 2643archive files (linker libraries) in the target machine's format. 2644 2645@item ranlib 2646This should be a program to construct a symbol table in an archive file. 2647@end table 2648 2649The installation of GCC will find these programs in that directory, 2650and copy or link them to the proper place to for the cross-compiler to 2651find them when run later. 2652 2653The easiest way to provide these files is to build the Binutils package. 2654Configure it with the same @option{--host} and @option{--target} 2655options that you use for configuring GCC, then build and install 2656them. They install their executables automatically into the proper 2657directory. Alas, they do not support all the targets that GCC 2658supports. 2659 2660If you are not building a C library in the same source tree as GCC, 2661you should also provide the target libraries and headers before 2662configuring GCC, specifying the directories with 2663@option{--with-sysroot} or @option{--with-headers} and 2664@option{--with-libs}. Many targets also require ``start files'' such 2665as @file{crt0.o} and 2666@file{crtn.o} which are linked into each executable. There may be several 2667alternatives for @file{crt0.o}, for use with profiling or other 2668compilation options. Check your target's definition of 2669@code{STARTFILE_SPEC} to find out what start files it uses. 2670 2671@section Building in parallel 2672 2673GNU Make 3.80 and above, which is necessary to build GCC, support 2674building in parallel. To activate this, you can use @samp{make -j 2} 2675instead of @samp{make}. You can also specify a bigger number, and 2676in most cases using a value greater than the number of processors in 2677your machine will result in fewer and shorter I/O latency hits, thus 2678improving overall throughput; this is especially true for slow drives 2679and network filesystems. 2680 2681@section Building the Ada compiler 2682 2683In order to build GNAT, the Ada compiler, you need a working GNAT 2684compiler (GCC version 4.0 or later). 2685This includes GNAT tools such as @command{gnatmake} and 2686@command{gnatlink}, since the Ada front end is written in Ada and 2687uses some GNAT-specific extensions. 2688 2689In order to build a cross compiler, it is suggested to install 2690the new compiler as native first, and then use it to build the cross 2691compiler. 2692 2693@command{configure} does not test whether the GNAT installation works 2694and has a sufficiently recent version; if too old a GNAT version is 2695installed, the build will fail unless @option{--enable-languages} is 2696used to disable building the Ada front end. 2697 2698@env{ADA_INCLUDE_PATH} and @env{ADA_OBJECT_PATH} environment variables 2699must not be set when building the Ada compiler, the Ada tools, or the 2700Ada runtime libraries. You can check that your build environment is clean 2701by verifying that @samp{gnatls -v} lists only one explicit path in each 2702section. 2703 2704@section Building with profile feedback 2705 2706It is possible to use profile feedback to optimize the compiler itself. This 2707should result in a faster compiler binary. Experiments done on x86 using gcc 27083.3 showed approximately 7 percent speedup on compiling C programs. To 2709bootstrap the compiler with profile feedback, use @code{make profiledbootstrap}. 2710 2711When @samp{make profiledbootstrap} is run, it will first build a @code{stage1} 2712compiler. This compiler is used to build a @code{stageprofile} compiler 2713instrumented to collect execution counts of instruction and branch 2714probabilities. Then runtime libraries are compiled with profile collected. 2715Finally a @code{stagefeedback} compiler is built using the information collected. 2716 2717Unlike standard bootstrap, several additional restrictions apply. The 2718compiler used to build @code{stage1} needs to support a 64-bit integral type. 2719It is recommended to only use GCC for this. 2720 2721@html 2722<hr /> 2723<p> 2724@end html 2725@ifhtml 2726@uref{./index.html,,Return to the GCC Installation page} 2727@end ifhtml 2728@end ifset 2729 2730@c ***Testing***************************************************************** 2731@ifnothtml 2732@comment node-name, next, previous, up 2733@node Testing, Final install, Building, Installing GCC 2734@end ifnothtml 2735@ifset testhtml 2736@ifnothtml 2737@chapter Installing GCC: Testing 2738@end ifnothtml 2739@cindex Testing 2740@cindex Installing GCC: Testing 2741@cindex Testsuite 2742 2743Before you install GCC, we encourage you to run the testsuites and to 2744compare your results with results from a similar configuration that have 2745been submitted to the 2746@uref{http://gcc.gnu.org/ml/gcc-testresults/,,gcc-testresults mailing list}. 2747Some of these archived results are linked from the build status lists 2748at @uref{http://gcc.gnu.org/buildstat.html}, although not everyone who 2749reports a successful build runs the testsuites and submits the results. 2750This step is optional and may require you to download additional software, 2751but it can give you confidence in your new GCC installation or point out 2752problems before you install and start using your new GCC@. 2753 2754First, you must have @uref{download.html,,downloaded the testsuites}. 2755These are part of the full distribution, but if you downloaded the 2756``core'' compiler plus any front ends, you must download the testsuites 2757separately. 2758 2759Second, you must have the testing tools installed. This includes 2760@uref{http://www.gnu.org/software/dejagnu/,,DejaGnu}, Tcl, and Expect; 2761the DejaGnu site has links to these. 2762 2763If the directories where @command{runtest} and @command{expect} were 2764installed are not in the @env{PATH}, you may need to set the following 2765environment variables appropriately, as in the following example (which 2766assumes that DejaGnu has been installed under @file{/usr/local}): 2767 2768@smallexample 2769TCL_LIBRARY = /usr/local/share/tcl8.0 2770DEJAGNULIBS = /usr/local/share/dejagnu 2771@end smallexample 2772 2773(On systems such as Cygwin, these paths are required to be actual 2774paths, not mounts or links; presumably this is due to some lack of 2775portability in the DejaGnu code.) 2776 2777 2778Finally, you can run the testsuite (which may take a long time): 2779@smallexample 2780cd @var{objdir}; make -k check 2781@end smallexample 2782 2783This will test various components of GCC, such as compiler 2784front ends and runtime libraries. While running the testsuite, DejaGnu 2785might emit some harmless messages resembling 2786@samp{WARNING: Couldn't find the global config file.} or 2787@samp{WARNING: Couldn't find tool init file} that can be ignored. 2788 2789If you are testing a cross-compiler, you may want to run the testsuite 2790on a simulator as described at @uref{http://gcc.gnu.org/simtest-howto.html}. 2791 2792@section How can you run the testsuite on selected tests? 2793 2794In order to run sets of tests selectively, there are targets 2795@samp{make check-gcc} and language specific @samp{make check-c}, 2796@samp{make check-c++}, @samp{make check-fortran}, @samp{make check-java}, 2797@samp{make check-ada}, @samp{make check-objc}, @samp{make check-obj-c++}, 2798@samp{make check-lto} 2799in the @file{gcc} subdirectory of the object directory. You can also 2800just run @samp{make check} in a subdirectory of the object directory. 2801 2802 2803A more selective way to just run all @command{gcc} execute tests in the 2804testsuite is to use 2805 2806@smallexample 2807make check-gcc RUNTESTFLAGS="execute.exp @var{other-options}" 2808@end smallexample 2809 2810Likewise, in order to run only the @command{g++} ``old-deja'' tests in 2811the testsuite with filenames matching @samp{9805*}, you would use 2812 2813@smallexample 2814make check-g++ RUNTESTFLAGS="old-deja.exp=9805* @var{other-options}" 2815@end smallexample 2816 2817The @file{*.exp} files are located in the testsuite directories of the GCC 2818source, the most important ones being @file{compile.exp}, 2819@file{execute.exp}, @file{dg.exp} and @file{old-deja.exp}. 2820To get a list of the possible @file{*.exp} files, pipe the 2821output of @samp{make check} into a file and look at the 2822@samp{Running @dots{} .exp} lines. 2823 2824@section Passing options and running multiple testsuites 2825 2826You can pass multiple options to the testsuite using the 2827@samp{--target_board} option of DejaGNU, either passed as part of 2828@samp{RUNTESTFLAGS}, or directly to @command{runtest} if you prefer to 2829work outside the makefiles. For example, 2830 2831@smallexample 2832make check-g++ RUNTESTFLAGS="--target_board=unix/-O3/-fmerge-constants" 2833@end smallexample 2834 2835will run the standard @command{g++} testsuites (``unix'' is the target name 2836for a standard native testsuite situation), passing 2837@samp{-O3 -fmerge-constants} to the compiler on every test, i.e., 2838slashes separate options. 2839 2840You can run the testsuites multiple times using combinations of options 2841with a syntax similar to the brace expansion of popular shells: 2842 2843@smallexample 2844@dots{}"--target_board=arm-sim\@{-mhard-float,-msoft-float\@}\@{-O1,-O2,-O3,\@}" 2845@end smallexample 2846 2847(Note the empty option caused by the trailing comma in the final group.) 2848The following will run each testsuite eight times using the @samp{arm-sim} 2849target, as if you had specified all possible combinations yourself: 2850 2851@smallexample 2852--target_board='arm-sim/-mhard-float/-O1 \ 2853 arm-sim/-mhard-float/-O2 \ 2854 arm-sim/-mhard-float/-O3 \ 2855 arm-sim/-mhard-float \ 2856 arm-sim/-msoft-float/-O1 \ 2857 arm-sim/-msoft-float/-O2 \ 2858 arm-sim/-msoft-float/-O3 \ 2859 arm-sim/-msoft-float' 2860@end smallexample 2861 2862They can be combined as many times as you wish, in arbitrary ways. This 2863list: 2864 2865@smallexample 2866@dots{}"--target_board=unix/-Wextra\@{-O3,-fno-strength\@}\@{-fomit-frame,\@}" 2867@end smallexample 2868 2869will generate four combinations, all involving @samp{-Wextra}. 2870 2871The disadvantage to this method is that the testsuites are run in serial, 2872which is a waste on multiprocessor systems. For users with GNU Make and 2873a shell which performs brace expansion, you can run the testsuites in 2874parallel by having the shell perform the combinations and @command{make} 2875do the parallel runs. Instead of using @samp{--target_board}, use a 2876special makefile target: 2877 2878@smallexample 2879make -j@var{N} check-@var{testsuite}//@var{test-target}/@var{option1}/@var{option2}/@dots{} 2880@end smallexample 2881 2882For example, 2883 2884@smallexample 2885make -j3 check-gcc//sh-hms-sim/@{-m1,-m2,-m3,-m3e,-m4@}/@{,-nofpu@} 2886@end smallexample 2887 2888will run three concurrent ``make-gcc'' testsuites, eventually testing all 2889ten combinations as described above. Note that this is currently only 2890supported in the @file{gcc} subdirectory. (To see how this works, try 2891typing @command{echo} before the example given here.) 2892 2893 2894@section Additional testing for Java Class Libraries 2895 2896The Java runtime tests can be executed via @samp{make check} 2897in the @file{@var{target}/libjava/testsuite} directory in 2898the build tree. 2899 2900The @uref{http://sourceware.org/mauve/,,Mauve Project} provides 2901a suite of tests for the Java Class Libraries. This suite can be run 2902as part of libgcj testing by placing the Mauve tree within the libjava 2903testsuite at @file{libjava/testsuite/libjava.mauve/mauve}, or by 2904specifying the location of that tree when invoking @samp{make}, as in 2905@samp{make MAUVEDIR=~/mauve check}. 2906 2907@section How to interpret test results 2908 2909The result of running the testsuite are various @file{*.sum} and @file{*.log} 2910files in the testsuite subdirectories. The @file{*.log} files contain a 2911detailed log of the compiler invocations and the corresponding 2912results, the @file{*.sum} files summarize the results. These summaries 2913contain status codes for all tests: 2914 2915@itemize @bullet 2916@item 2917PASS: the test passed as expected 2918@item 2919XPASS: the test unexpectedly passed 2920@item 2921FAIL: the test unexpectedly failed 2922@item 2923XFAIL: the test failed as expected 2924@item 2925UNSUPPORTED: the test is not supported on this platform 2926@item 2927ERROR: the testsuite detected an error 2928@item 2929WARNING: the testsuite detected a possible problem 2930@end itemize 2931 2932It is normal for some tests to report unexpected failures. At the 2933current time the testing harness does not allow fine grained control 2934over whether or not a test is expected to fail. This problem should 2935be fixed in future releases. 2936 2937 2938@section Submitting test results 2939 2940If you want to report the results to the GCC project, use the 2941@file{contrib/test_summary} shell script. Start it in the @var{objdir} with 2942 2943@smallexample 2944@var{srcdir}/contrib/test_summary -p your_commentary.txt \ 2945 -m gcc-testresults@@gcc.gnu.org |sh 2946@end smallexample 2947 2948This script uses the @command{Mail} program to send the results, so 2949make sure it is in your @env{PATH}. The file @file{your_commentary.txt} is 2950prepended to the testsuite summary and should contain any special 2951remarks you have on your results or your build environment. Please 2952do not edit the testsuite result block or the subject line, as these 2953messages may be automatically processed. 2954 2955@html 2956<hr /> 2957<p> 2958@end html 2959@ifhtml 2960@uref{./index.html,,Return to the GCC Installation page} 2961@end ifhtml 2962@end ifset 2963 2964@c ***Final install*********************************************************** 2965@ifnothtml 2966@comment node-name, next, previous, up 2967@node Final install, , Testing, Installing GCC 2968@end ifnothtml 2969@ifset finalinstallhtml 2970@ifnothtml 2971@chapter Installing GCC: Final installation 2972@end ifnothtml 2973 2974Now that GCC has been built (and optionally tested), you can install it with 2975@smallexample 2976cd @var{objdir} && make install 2977@end smallexample 2978 2979We strongly recommend to install into a target directory where there is 2980no previous version of GCC present. Also, the GNAT runtime should not 2981be stripped, as this would break certain features of the debugger that 2982depend on this debugging information (catching Ada exceptions for 2983instance). 2984 2985That step completes the installation of GCC; user level binaries can 2986be found in @file{@var{prefix}/bin} where @var{prefix} is the value 2987you specified with the @option{--prefix} to configure (or 2988@file{/usr/local} by default). (If you specified @option{--bindir}, 2989that directory will be used instead; otherwise, if you specified 2990@option{--exec-prefix}, @file{@var{exec-prefix}/bin} will be used.) 2991Headers for the C++ and Java libraries are installed in 2992@file{@var{prefix}/include}; libraries in @file{@var{libdir}} 2993(normally @file{@var{prefix}/lib}); internal parts of the compiler in 2994@file{@var{libdir}/gcc} and @file{@var{libexecdir}/gcc}; documentation 2995in info format in @file{@var{infodir}} (normally 2996@file{@var{prefix}/info}). 2997 2998When installing cross-compilers, GCC's executables 2999are not only installed into @file{@var{bindir}}, that 3000is, @file{@var{exec-prefix}/bin}, but additionally into 3001@file{@var{exec-prefix}/@var{target-alias}/bin}, if that directory 3002exists. Typically, such @dfn{tooldirs} hold target-specific 3003binutils, including assembler and linker. 3004 3005Installation into a temporary staging area or into a @command{chroot} 3006jail can be achieved with the command 3007 3008@smallexample 3009make DESTDIR=@var{path-to-rootdir} install 3010@end smallexample 3011 3012@noindent 3013where @var{path-to-rootdir} is the absolute path of 3014a directory relative to which all installation paths will be 3015interpreted. Note that the directory specified by @code{DESTDIR} 3016need not exist yet; it will be created if necessary. 3017 3018There is a subtle point with tooldirs and @code{DESTDIR}: 3019If you relocate a cross-compiler installation with 3020e.g.@: @samp{DESTDIR=@var{rootdir}}, then the directory 3021@file{@var{rootdir}/@var{exec-prefix}/@var{target-alias}/bin} will 3022be filled with duplicated GCC executables only if it already exists, 3023it will not be created otherwise. This is regarded as a feature, 3024not as a bug, because it gives slightly more control to the packagers 3025using the @code{DESTDIR} feature. 3026 3027You can install stripped programs and libraries with 3028 3029@smallexample 3030make install-strip 3031@end smallexample 3032 3033If you are bootstrapping a released version of GCC then please 3034quickly review the build status page for your release, available from 3035@uref{http://gcc.gnu.org/buildstat.html}. 3036If your system is not listed for the version of GCC that you built, 3037send a note to 3038@email{gcc@@gcc.gnu.org} indicating 3039that you successfully built and installed GCC@. 3040Include the following information: 3041 3042@itemize @bullet 3043@item 3044Output from running @file{@var{srcdir}/config.guess}. Do not send 3045that file itself, just the one-line output from running it. 3046 3047@item 3048The output of @samp{gcc -v} for your newly installed @command{gcc}. 3049This tells us which version of GCC you built and the options you passed to 3050configure. 3051 3052@item 3053Whether you enabled all languages or a subset of them. If you used a 3054full distribution then this information is part of the configure 3055options in the output of @samp{gcc -v}, but if you downloaded the 3056``core'' compiler plus additional front ends then it isn't apparent 3057which ones you built unless you tell us about it. 3058 3059@item 3060If the build was for GNU/Linux, also include: 3061@itemize @bullet 3062@item 3063The distribution name and version (e.g., Red Hat 7.1 or Debian 2.2.3); 3064this information should be available from @file{/etc/issue}. 3065 3066@item 3067The version of the Linux kernel, available from @samp{uname --version} 3068or @samp{uname -a}. 3069 3070@item 3071The version of glibc you used; for RPM-based systems like Red Hat, 3072Mandrake, and SuSE type @samp{rpm -q glibc} to get the glibc version, 3073and on systems like Debian and Progeny use @samp{dpkg -l libc6}. 3074@end itemize 3075For other systems, you can include similar information if you think it is 3076relevant. 3077 3078@item 3079Any other information that you think would be useful to people building 3080GCC on the same configuration. The new entry in the build status list 3081will include a link to the archived copy of your message. 3082@end itemize 3083 3084We'd also like to know if the 3085@ifnothtml 3086@ref{Specific, host/target specific installation notes} 3087@end ifnothtml 3088@ifhtml 3089@uref{specific.html,,host/target specific installation notes} 3090@end ifhtml 3091didn't include your host/target information or if that information is 3092incomplete or out of date. Send a note to 3093@email{gcc@@gcc.gnu.org} detailing how the information should be changed. 3094 3095If you find a bug, please report it following the 3096@uref{../bugs/,,bug reporting guidelines}. 3097 3098If you want to print the GCC manuals, do @samp{cd @var{objdir}; make 3099dvi}. You will need to have @command{texi2dvi} (version at least 4.7) 3100and @TeX{} installed. This creates a number of @file{.dvi} files in 3101subdirectories of @file{@var{objdir}}; these may be converted for 3102printing with programs such as @command{dvips}. Alternately, by using 3103@samp{make pdf} in place of @samp{make dvi}, you can create documentation 3104in the form of @file{.pdf} files; this requires @command{texi2pdf}, which 3105is included with Texinfo version 4.8 and later. You can also 3106@uref{http://shop.fsf.org/,,buy printed manuals from the 3107Free Software Foundation}, though such manuals may not be for the most 3108recent version of GCC@. 3109 3110If you would like to generate online HTML documentation, do @samp{cd 3111@var{objdir}; make html} and HTML will be generated for the gcc manuals in 3112@file{@var{objdir}/gcc/HTML}. 3113 3114@html 3115<hr /> 3116<p> 3117@end html 3118@ifhtml 3119@uref{./index.html,,Return to the GCC Installation page} 3120@end ifhtml 3121@end ifset 3122 3123@c ***Binaries**************************************************************** 3124@ifnothtml 3125@comment node-name, next, previous, up 3126@node Binaries, Specific, Installing GCC, Top 3127@end ifnothtml 3128@ifset binarieshtml 3129@ifnothtml 3130@chapter Installing GCC: Binaries 3131@end ifnothtml 3132@cindex Binaries 3133@cindex Installing GCC: Binaries 3134 3135We are often asked about pre-compiled versions of GCC@. While we cannot 3136provide these for all platforms, below you'll find links to binaries for 3137various platforms where creating them by yourself is not easy due to various 3138reasons. 3139 3140Please note that we did not create these binaries, nor do we 3141support them. If you have any problems installing them, please 3142contact their makers. 3143 3144@itemize 3145@item 3146AIX: 3147@itemize 3148@item 3149@uref{http://www.bullfreeware.com,,Bull's Freeware and Shareware Archive for AIX}; 3150 3151@item 3152@uref{http://pware.hvcc.edu,,Hudson Valley Community College Open Source Software for IBM System p}; 3153 3154@item 3155@uref{http://www.perzl.org/aix/,,AIX 5L and 6 Open Source Packages}. 3156@end itemize 3157 3158@item 3159DOS---@uref{http://www.delorie.com/djgpp/,,DJGPP}. 3160 3161@item 3162Renesas H8/300[HS]---@uref{http://h8300-hms.sourceforge.net/,,GNU 3163Development Tools for the Renesas H8/300[HS] Series}. 3164 3165@item 3166HP-UX: 3167@itemize 3168@item 3169@uref{http://hpux.connect.org.uk/,,HP-UX Porting Center}; 3170 3171@item 3172@uref{ftp://sunsite.informatik.rwth-aachen.de/pub/packages/gcc_hpux/,,Binaries for HP-UX 11.00 at Aachen University of Technology}. 3173@end itemize 3174 3175@item 3176@uref{http://www.sco.com/skunkware/devtools/index.html#gcc,,SCO 3177OpenServer/Unixware}. 3178 3179@item 3180Solaris 2 (SPARC, Intel): 3181@itemize 3182@item 3183@uref{http://www.opencsw.org/,,OpenCSW} 3184 3185@item 3186@uref{http://jupiterrise.com/tgcware/,,TGCware} 3187@end itemize 3188 3189@item 3190Microsoft Windows: 3191@itemize 3192@item 3193The @uref{http://sourceware.org/cygwin/,,Cygwin} project; 3194@item 3195The @uref{http://www.mingw.org/,,MinGW} project. 3196@end itemize 3197 3198@item 3199@uref{ftp://ftp.thewrittenword.com/packages/by-name/,,The 3200Written Word} offers binaries for 3201AIX 4.3.3, 5.1 and 5.2, 3202GNU/Linux (i386), 3203HP-UX 10.20, 11.00, and 11.11, and 3204Solaris/SPARC 2.5.1, 2.6, 7, 8, 9 and 10. 3205 3206@item 3207@uref{http://www.openpkg.org/,,OpenPKG} offers binaries for quite a 3208number of platforms. 3209 3210@item 3211The @uref{http://gcc.gnu.org/wiki/GFortranBinaries,,GFortran Wiki} has 3212links to GNU Fortran binaries for several platforms. 3213@end itemize 3214 3215@html 3216<hr /> 3217<p> 3218@end html 3219@ifhtml 3220@uref{./index.html,,Return to the GCC Installation page} 3221@end ifhtml 3222@end ifset 3223 3224@c ***Specific**************************************************************** 3225@ifnothtml 3226@comment node-name, next, previous, up 3227@node Specific, Old, Binaries, Top 3228@end ifnothtml 3229@ifset specifichtml 3230@ifnothtml 3231@chapter Host/target specific installation notes for GCC 3232@end ifnothtml 3233@cindex Specific 3234@cindex Specific installation notes 3235@cindex Target specific installation 3236@cindex Host specific installation 3237@cindex Target specific installation notes 3238 3239Please read this document carefully @emph{before} installing the 3240GNU Compiler Collection on your machine. 3241 3242Note that this list of install notes is @emph{not} a list of supported 3243hosts or targets. Not all supported hosts and targets are listed 3244here, only the ones that require host-specific or target-specific 3245information have to. 3246 3247@ifhtml 3248@itemize 3249@item 3250@uref{#aarch64-x-x,,aarch64*-*-*} 3251@item 3252@uref{#alpha-x-x,,alpha*-*-*} 3253@item 3254@uref{#alpha-dec-osf51,,alpha*-dec-osf5.1} 3255@item 3256@uref{#amd64-x-solaris210,,amd64-*-solaris2.10} 3257@item 3258@uref{#arm-x-eabi,,arm-*-eabi} 3259@item 3260@uref{#avr,,avr} 3261@item 3262@uref{#bfin,,Blackfin} 3263@item 3264@uref{#dos,,DOS} 3265@item 3266@uref{#x-x-freebsd,,*-*-freebsd*} 3267@item 3268@uref{#h8300-hms,,h8300-hms} 3269@item 3270@uref{#hppa-hp-hpux,,hppa*-hp-hpux*} 3271@item 3272@uref{#hppa-hp-hpux10,,hppa*-hp-hpux10} 3273@item 3274@uref{#hppa-hp-hpux11,,hppa*-hp-hpux11} 3275@item 3276@uref{#x-x-linux-gnu,,*-*-linux-gnu} 3277@item 3278@uref{#ix86-x-linux,,i?86-*-linux*} 3279@item 3280@uref{#ix86-x-solaris210,,i?86-*-solaris2.10} 3281@item 3282@uref{#ia64-x-linux,,ia64-*-linux} 3283@item 3284@uref{#ia64-x-hpux,,ia64-*-hpux*} 3285@item 3286@uref{#x-ibm-aix,,*-ibm-aix*} 3287@item 3288@uref{#iq2000-x-elf,,iq2000-*-elf} 3289@item 3290@uref{#lm32-x-elf,,lm32-*-elf} 3291@item 3292@uref{#lm32-x-uclinux,,lm32-*-uclinux} 3293@item 3294@uref{#m32c-x-elf,,m32c-*-elf} 3295@item 3296@uref{#m32r-x-elf,,m32r-*-elf} 3297@item 3298@uref{#m68k-x-x,,m68k-*-*} 3299@item 3300@uref{#m68k-uclinux,,m68k-uclinux} 3301@item 3302@uref{#mep-x-elf,,mep-*-elf} 3303@item 3304@uref{#microblaze-x-elf,,microblaze-*-elf} 3305@item 3306@uref{#mips-x-x,,mips-*-*} 3307@item 3308@uref{#mips-sgi-irix5,,mips-sgi-irix5} 3309@item 3310@uref{#mips-sgi-irix6,,mips-sgi-irix6} 3311@item 3312@uref{#nds32le-x-elf,,nds32le-*-elf} 3313@item 3314@uref{#nds32be-x-elf,,nds32be-*-elf} 3315@item 3316@uref{#nvptx-x-none,,nvptx-*-none} 3317@item 3318@uref{#powerpc-x-x,,powerpc*-*-*} 3319@item 3320@uref{#powerpc-x-darwin,,powerpc-*-darwin*} 3321@item 3322@uref{#powerpc-x-elf,,powerpc-*-elf} 3323@item 3324@uref{#powerpc-x-linux-gnu,,powerpc*-*-linux-gnu*} 3325@item 3326@uref{#powerpc-x-netbsd,,powerpc-*-netbsd*} 3327@item 3328@uref{#powerpc-x-eabisim,,powerpc-*-eabisim} 3329@item 3330@uref{#powerpc-x-eabi,,powerpc-*-eabi} 3331@item 3332@uref{#powerpcle-x-elf,,powerpcle-*-elf} 3333@item 3334@uref{#powerpcle-x-eabisim,,powerpcle-*-eabisim} 3335@item 3336@uref{#powerpcle-x-eabi,,powerpcle-*-eabi} 3337@item 3338@uref{#s390-x-linux,,s390-*-linux*} 3339@item 3340@uref{#s390x-x-linux,,s390x-*-linux*} 3341@item 3342@uref{#s390x-ibm-tpf,,s390x-ibm-tpf*} 3343@item 3344@uref{#x-x-solaris2,,*-*-solaris2*} 3345@item 3346@uref{#sparc-x-x,,sparc*-*-*} 3347@item 3348@uref{#sparc-sun-solaris2,,sparc-sun-solaris2*} 3349@item 3350@uref{#sparc-sun-solaris210,,sparc-sun-solaris2.10} 3351@item 3352@uref{#sparc-x-linux,,sparc-*-linux*} 3353@item 3354@uref{#sparc64-x-solaris2,,sparc64-*-solaris2*} 3355@item 3356@uref{#sparcv9-x-solaris2,,sparcv9-*-solaris2*} 3357@item 3358@uref{#c6x-x-x,,c6x-*-*} 3359@item 3360@uref{#tilegx-x-linux,,tilegx-*-linux*} 3361@item 3362@uref{#tilegxbe-x-linux,,tilegxbe-*-linux*} 3363@item 3364@uref{#tilepro-x-linux,,tilepro-*-linux*} 3365@item 3366@uref{#visium-x-elf, visium-*-elf} 3367@item 3368@uref{#x-x-vxworks,,*-*-vxworks*} 3369@item 3370@uref{#x86-64-x-x,,x86_64-*-*, amd64-*-*} 3371@item 3372@uref{#x86-64-x-solaris210,,x86_64-*-solaris2.1[0-9]*} 3373@item 3374@uref{#xtensa-x-elf,,xtensa*-*-elf} 3375@item 3376@uref{#xtensa-x-linux,,xtensa*-*-linux*} 3377@item 3378@uref{#windows,,Microsoft Windows} 3379@item 3380@uref{#x-x-cygwin,,*-*-cygwin} 3381@item 3382@uref{#x-x-interix,,*-*-interix} 3383@item 3384@uref{#x-x-mingw32,,*-*-mingw32} 3385@item 3386@uref{#os2,,OS/2} 3387@item 3388@uref{#older,,Older systems} 3389@end itemize 3390 3391@itemize 3392@item 3393@uref{#elf,,all ELF targets} (SVR4, Solaris 2, etc.) 3394@end itemize 3395@end ifhtml 3396 3397 3398@html 3399<!-- -------- host/target specific issues start here ---------------- --> 3400<hr /> 3401@end html 3402@anchor{aarch64-x-x} 3403@heading aarch64*-*-* 3404Binutils pre 2.24 does not have support for selecting @option{-mabi} and 3405does not support ILP32. If it is used to build GCC 4.9 or later, GCC will 3406not support option @option{-mabi=ilp32}. 3407 3408To enable a workaround for the Cortex-A53 erratum number 835769 by default 3409(for all CPUs regardless of -mcpu option given) at configure time use the 3410@option{--enable-fix-cortex-a53-835769} option. This will enable the fix by 3411default and can be explicitly disabled during compilation by passing the 3412@option{-mno-fix-cortex-a53-835769} option. Conversely, 3413@option{--disable-fix-cortex-a53-835769} will disable the workaround by 3414default. The workaround is disabled by default if neither of 3415@option{--enable-fix-cortex-a53-835769} or 3416@option{--disable-fix-cortex-a53-835769} is given at configure time. 3417 3418To enable a workaround for the Cortex-A53 erratum number 843419 by default 3419(for all CPUs regardless of -mcpu option given) at configure time use the 3420@option{--enable-fix-cortex-a53-843419} option. This workaround is applied at 3421link time. Enabling the workaround will cause GCC to pass the relevant option 3422to the linker. It can be explicitly disabled during compilation by passing the 3423@option{-mno-fix-cortex-a53-843419} option. Conversely, 3424@option{--disable-fix-cortex-a53-843419} will disable the workaround by default. 3425The workaround is disabled by default if neither of 3426@option{--enable-fix-cortex-a53-843419} or 3427@option{--disable-fix-cortex-a53-843419} is given at configure time. 3428 3429@html 3430<hr /> 3431@end html 3432@anchor{alpha-x-x} 3433@heading alpha*-*-* 3434This section contains general configuration information for all 3435alpha-based platforms using ELF (in particular, ignore this section for 3436DEC OSF/1, Digital UNIX and Tru64 UNIX)@. In addition to reading this 3437section, please read all other sections that match your target. 3438 3439We require binutils 2.11.2 or newer. 3440Previous binutils releases had a number of problems with DWARF 2 3441debugging information, not the least of which is incorrect linking of 3442shared libraries. 3443 3444@html 3445<hr /> 3446@end html 3447@anchor{alpha-dec-osf51} 3448@heading alpha*-dec-osf5.1 3449Systems using processors that implement the DEC Alpha architecture and 3450are running the DEC/Compaq/HP Unix (DEC OSF/1, Digital UNIX, or Compaq/HP 3451Tru64 UNIX) operating system, for example the DEC Alpha AXP systems. 3452 3453Support for Tru64 UNIX V5.1 has been removed in GCC 4.8. As of GCC 4.6, 3454support for Tru64 UNIX V4.0 and V5.0 has been removed. As of GCC 3.2, 3455versions before @code{alpha*-dec-osf4} are no longer supported. (These 3456are the versions which identify themselves as DEC OSF/1.) 3457 3458@html 3459<hr /> 3460@end html 3461@anchor{amd64-x-solaris210} 3462@heading amd64-*-solaris2.1[0-9]* 3463This is a synonym for @samp{x86_64-*-solaris2.1[0-9]*}. 3464 3465@html 3466<hr /> 3467@end html 3468@anchor{arc-x-elf32} 3469@heading arc-*-elf32 3470 3471Use @samp{configure --target=arc-elf32 --with-cpu=@var{cpu} --enable-languages="c,c++"} 3472to configure GCC, with @var{cpu} being one of @samp{arc600}, @samp{arc601}, 3473or @samp{arc700}@. 3474 3475@html 3476<hr /> 3477@end html 3478@anchor{arc-linux-uclibc} 3479@heading arc-linux-uclibc 3480 3481Use @samp{configure --target=arc-linux-uclibc --with-cpu=arc700 --enable-languages="c,c++"} to configure GCC@. 3482 3483@html 3484<hr /> 3485@end html 3486@anchor{arm-x-eabi} 3487@heading arm-*-eabi 3488ARM-family processors. Subtargets that use the ELF object format 3489require GNU binutils 2.13 or newer. Such subtargets include: 3490@code{arm-*-netbsdelf}, @code{arm-*-*linux-*} 3491and @code{arm-*-rtemseabi}. 3492 3493@html 3494<hr /> 3495@end html 3496@anchor{avr} 3497@heading avr 3498ATMEL AVR-family micro controllers. These are used in embedded 3499applications. There are no standard Unix configurations. 3500@ifnothtml 3501@xref{AVR Options,, AVR Options, gcc, Using the GNU Compiler 3502Collection (GCC)}, 3503@end ifnothtml 3504@ifhtml 3505See ``AVR Options'' in the main manual 3506@end ifhtml 3507for the list of supported MCU types. 3508 3509Use @samp{configure --target=avr --enable-languages="c"} to configure GCC@. 3510 3511Further installation notes and other useful information about AVR tools 3512can also be obtained from: 3513 3514@itemize @bullet 3515@item 3516@uref{http://www.nongnu.org/avr/,,http://www.nongnu.org/avr/} 3517@item 3518@uref{http://www.amelek.gda.pl/avr/,,http://www.amelek.gda.pl/avr/} 3519@end itemize 3520 3521We @emph{strongly} recommend using binutils 2.13 or newer. 3522 3523The following error: 3524@smallexample 3525Error: register required 3526@end smallexample 3527 3528indicates that you should upgrade to a newer version of the binutils. 3529 3530@html 3531<hr /> 3532@end html 3533@anchor{bfin} 3534@heading Blackfin 3535The Blackfin processor, an Analog Devices DSP. 3536@ifnothtml 3537@xref{Blackfin Options,, Blackfin Options, gcc, Using the GNU Compiler 3538Collection (GCC)}, 3539@end ifnothtml 3540@ifhtml 3541See ``Blackfin Options'' in the main manual 3542@end ifhtml 3543 3544More information, and a version of binutils with support for this processor, 3545is available at @uref{http://blackfin.uclinux.org} 3546 3547@html 3548<hr /> 3549@end html 3550@anchor{cr16} 3551@heading CR16 3552The CR16 CompactRISC architecture is a 16-bit architecture. This 3553architecture is used in embedded applications. 3554 3555@ifnothtml 3556@xref{CR16 Options,, CR16 Options, gcc, Using and Porting the GNU Compiler 3557Collection (GCC)}, 3558@end ifnothtml 3559 3560@ifhtml 3561See ``CR16 Options'' in the main manual for a list of CR16-specific options. 3562@end ifhtml 3563 3564Use @samp{configure --target=cr16-elf --enable-languages=c,c++} to configure 3565GCC@ for building a CR16 elf cross-compiler. 3566 3567Use @samp{configure --target=cr16-uclinux --enable-languages=c,c++} to 3568configure GCC@ for building a CR16 uclinux cross-compiler. 3569 3570@html 3571<hr /> 3572@end html 3573@anchor{cris} 3574@heading CRIS 3575CRIS is the CPU architecture in Axis Communications ETRAX system-on-a-chip 3576series. These are used in embedded applications. 3577 3578@ifnothtml 3579@xref{CRIS Options,, CRIS Options, gcc, Using the GNU Compiler 3580Collection (GCC)}, 3581@end ifnothtml 3582@ifhtml 3583See ``CRIS Options'' in the main manual 3584@end ifhtml 3585for a list of CRIS-specific options. 3586 3587There are a few different CRIS targets: 3588@table @code 3589@item cris-axis-elf 3590Mainly for monolithic embedded systems. Includes a multilib for the 3591@samp{v10} core used in @samp{ETRAX 100 LX}. 3592@item cris-axis-linux-gnu 3593A GNU/Linux port for the CRIS architecture, currently targeting 3594@samp{ETRAX 100 LX} by default. 3595@end table 3596 3597For @code{cris-axis-elf} you need binutils 2.11 3598or newer. For @code{cris-axis-linux-gnu} you need binutils 2.12 or newer. 3599 3600Pre-packaged tools can be obtained from 3601@uref{ftp://ftp.axis.com/@/pub/@/axis/@/tools/@/cris/@/compiler-kit/}. More 3602information about this platform is available at 3603@uref{http://developer.axis.com/}. 3604 3605@html 3606<hr /> 3607@end html 3608@anchor{dos} 3609@heading DOS 3610Please have a look at the @uref{binaries.html,,binaries page}. 3611 3612You cannot install GCC by itself on MSDOS; it will not compile under 3613any MSDOS compiler except itself. You need to get the complete 3614compilation package DJGPP, which includes binaries as well as sources, 3615and includes all the necessary compilation tools and libraries. 3616 3617@html 3618<hr /> 3619@end html 3620@anchor{epiphany-x-elf} 3621@heading epiphany-*-elf 3622Adapteva Epiphany. 3623This configuration is intended for embedded systems. 3624 3625@html 3626<hr /> 3627@end html 3628@anchor{x-x-freebsd} 3629@heading *-*-freebsd* 3630Support for FreeBSD 1 was discontinued in GCC 3.2. Support for 3631FreeBSD 2 (and any mutant a.out variants of FreeBSD 3) was 3632discontinued in GCC 4.0. 3633 3634In order to better utilize FreeBSD base system functionality and match 3635the configuration of the system compiler, GCC 4.5 and above as well as 3636GCC 4.4 past 2010-06-20 leverage SSP support in libc (which is present 3637on FreeBSD 7 or later) and the use of @code{__cxa_atexit} by default 3638(on FreeBSD 6 or later). The use of @code{dl_iterate_phdr} inside 3639@file{libgcc_s.so.1} and boehm-gc (on FreeBSD 7 or later) is enabled 3640by GCC 4.5 and above. 3641 3642We support FreeBSD using the ELF file format with DWARF 2 debugging 3643for all CPU architectures. You may use @option{-gstabs} instead of 3644@option{-g}, if you really want the old debugging format. There are 3645no known issues with mixing object files and libraries with different 3646debugging formats. Otherwise, this release of GCC should now match 3647more of the configuration used in the stock FreeBSD configuration of 3648GCC@. In particular, @option{--enable-threads} is now configured by 3649default. However, as a general user, do not attempt to replace the 3650system compiler with this release. Known to bootstrap and check with 3651good results on FreeBSD 7.2-STABLE@. In the past, known to bootstrap 3652and check with good results on FreeBSD 3.0, 3.4, 4.0, 4.2, 4.3, 4.4, 36534.5, 4.8, 4.9 and 5-CURRENT@. 3654 3655The version of binutils installed in @file{/usr/bin} probably works 3656with this release of GCC@. Bootstrapping against the latest GNU 3657binutils and/or the version found in @file{/usr/ports/devel/binutils} has 3658been known to enable additional features and improve overall testsuite 3659results. However, it is currently known that boehm-gc (which itself 3660is required for java) may not configure properly on FreeBSD prior to 3661the FreeBSD 7.0 release with GNU binutils after 2.16.1. 3662 3663@html 3664<hr /> 3665@end html 3666@anchor{h8300-hms} 3667@heading h8300-hms 3668Renesas H8/300 series of processors. 3669 3670Please have a look at the @uref{binaries.html,,binaries page}. 3671 3672The calling convention and structure layout has changed in release 2.6. 3673All code must be recompiled. The calling convention now passes the 3674first three arguments in function calls in registers. Structures are no 3675longer a multiple of 2 bytes. 3676 3677@html 3678<hr /> 3679@end html 3680@anchor{hppa-hp-hpux} 3681@heading hppa*-hp-hpux* 3682Support for HP-UX version 9 and older was discontinued in GCC 3.4. 3683 3684We require using gas/binutils on all hppa platforms. Version 2.19 or 3685later is recommended. 3686 3687It may be helpful to configure GCC with the 3688@uref{./configure.html#with-gnu-as,,@option{--with-gnu-as}} and 3689@option{--with-as=@dots{}} options to ensure that GCC can find GAS@. 3690 3691The HP assembler should not be used with GCC. It is rarely tested and may 3692not work. It shouldn't be used with any languages other than C due to its 3693many limitations. 3694 3695Specifically, @option{-g} does not work (HP-UX uses a peculiar debugging 3696format which GCC does not know about). It also inserts timestamps 3697into each object file it creates, causing the 3-stage comparison test to 3698fail during a bootstrap. You should be able to continue by saying 3699@samp{make all-host all-target} after getting the failure from @samp{make}. 3700 3701Various GCC features are not supported. For example, it does not support weak 3702symbols or alias definitions. As a result, explicit template instantiations 3703are required when using C++. This makes it difficult if not impossible to 3704build many C++ applications. 3705 3706There are two default scheduling models for instructions. These are 3707PROCESSOR_7100LC and PROCESSOR_8000. They are selected from the pa-risc 3708architecture specified for the target machine when configuring. 3709PROCESSOR_8000 is the default. PROCESSOR_7100LC is selected when 3710the target is a @samp{hppa1*} machine. 3711 3712The PROCESSOR_8000 model is not well suited to older processors. Thus, 3713it is important to completely specify the machine architecture when 3714configuring if you want a model other than PROCESSOR_8000. The macro 3715TARGET_SCHED_DEFAULT can be defined in BOOT_CFLAGS if a different 3716default scheduling model is desired. 3717 3718As of GCC 4.0, GCC uses the UNIX 95 namespace for HP-UX 10.10 3719through 11.00, and the UNIX 98 namespace for HP-UX 11.11 and later. 3720This namespace change might cause problems when bootstrapping with 3721an earlier version of GCC or the HP compiler as essentially the same 3722namespace is required for an entire build. This problem can be avoided 3723in a number of ways. With HP cc, @env{UNIX_STD} can be set to @samp{95} 3724or @samp{98}. Another way is to add an appropriate set of predefines 3725to @env{CC}. The description for the @option{munix=} option contains 3726a list of the predefines used with each standard. 3727 3728More specific information to @samp{hppa*-hp-hpux*} targets follows. 3729 3730@html 3731<hr /> 3732@end html 3733@anchor{hppa-hp-hpux10} 3734@heading hppa*-hp-hpux10 3735For hpux10.20, we @emph{highly} recommend you pick up the latest sed patch 3736@code{PHCO_19798} from HP@. 3737 3738The C++ ABI has changed incompatibly in GCC 4.0. COMDAT subspaces are 3739used for one-only code and data. This resolves many of the previous 3740problems in using C++ on this target. However, the ABI is not compatible 3741with the one implemented under HP-UX 11 using secondary definitions. 3742 3743@html 3744<hr /> 3745@end html 3746@anchor{hppa-hp-hpux11} 3747@heading hppa*-hp-hpux11 3748GCC 3.0 and up support HP-UX 11. GCC 2.95.x is not supported and cannot 3749be used to compile GCC 3.0 and up. 3750 3751The libffi and libjava libraries haven't been ported to 64-bit HP-UX@ 3752and don't build. 3753 3754Refer to @uref{binaries.html,,binaries} for information about obtaining 3755precompiled GCC binaries for HP-UX@. Precompiled binaries must be obtained 3756to build the Ada language as it can't be bootstrapped using C@. Ada is 3757only available for the 32-bit PA-RISC runtime. 3758 3759Starting with GCC 3.4 an ISO C compiler is required to bootstrap. The 3760bundled compiler supports only traditional C; you will need either HP's 3761unbundled compiler, or a binary distribution of GCC@. 3762 3763It is possible to build GCC 3.3 starting with the bundled HP compiler, 3764but the process requires several steps. GCC 3.3 can then be used to 3765build later versions. The fastjar program contains ISO C code and 3766can't be built with the HP bundled compiler. This problem can be 3767avoided by not building the Java language. For example, use the 3768@option{--enable-languages="c,c++,f77,objc"} option in your configure 3769command. 3770 3771There are several possible approaches to building the distribution. 3772Binutils can be built first using the HP tools. Then, the GCC 3773distribution can be built. The second approach is to build GCC 3774first using the HP tools, then build binutils, then rebuild GCC@. 3775There have been problems with various binary distributions, so it 3776is best not to start from a binary distribution. 3777 3778On 64-bit capable systems, there are two distinct targets. Different 3779installation prefixes must be used if both are to be installed on 3780the same system. The @samp{hppa[1-2]*-hp-hpux11*} target generates code 3781for the 32-bit PA-RISC runtime architecture and uses the HP linker. 3782The @samp{hppa64-hp-hpux11*} target generates 64-bit code for the 3783PA-RISC 2.0 architecture. 3784 3785The script config.guess now selects the target type based on the compiler 3786detected during configuration. You must define @env{PATH} or @env{CC} so 3787that configure finds an appropriate compiler for the initial bootstrap. 3788When @env{CC} is used, the definition should contain the options that are 3789needed whenever @env{CC} is used. 3790 3791Specifically, options that determine the runtime architecture must be 3792in @env{CC} to correctly select the target for the build. It is also 3793convenient to place many other compiler options in @env{CC}. For example, 3794@env{CC="cc -Ac +DA2.0W -Wp,-H16376 -D_CLASSIC_TYPES -D_HPUX_SOURCE"} 3795can be used to bootstrap the GCC 3.3 branch with the HP compiler in 379664-bit K&R/bundled mode. The @option{+DA2.0W} option will result in 3797the automatic selection of the @samp{hppa64-hp-hpux11*} target. The 3798macro definition table of cpp needs to be increased for a successful 3799build with the HP compiler. _CLASSIC_TYPES and _HPUX_SOURCE need to 3800be defined when building with the bundled compiler, or when using the 3801@option{-Ac} option. These defines aren't necessary with @option{-Ae}. 3802 3803It is best to explicitly configure the @samp{hppa64-hp-hpux11*} target 3804with the @option{--with-ld=@dots{}} option. This overrides the standard 3805search for ld. The two linkers supported on this target require different 3806commands. The default linker is determined during configuration. As a 3807result, it's not possible to switch linkers in the middle of a GCC build. 3808This has been reported to sometimes occur in unified builds of binutils 3809and GCC@. 3810 3811A recent linker patch must be installed for the correct operation of 3812GCC 3.3 and later. @code{PHSS_26559} and @code{PHSS_24304} are the 3813oldest linker patches that are known to work. They are for HP-UX 381411.00 and 11.11, respectively. @code{PHSS_24303}, the companion to 3815@code{PHSS_24304}, might be usable but it hasn't been tested. These 3816patches have been superseded. Consult the HP patch database to obtain 3817the currently recommended linker patch for your system. 3818 3819The patches are necessary for the support of weak symbols on the 382032-bit port, and for the running of initializers and finalizers. Weak 3821symbols are implemented using SOM secondary definition symbols. Prior 3822to HP-UX 11, there are bugs in the linker support for secondary symbols. 3823The patches correct a problem of linker core dumps creating shared 3824libraries containing secondary symbols, as well as various other 3825linking issues involving secondary symbols. 3826 3827GCC 3.3 uses the ELF DT_INIT_ARRAY and DT_FINI_ARRAY capabilities to 3828run initializers and finalizers on the 64-bit port. The 32-bit port 3829uses the linker @option{+init} and @option{+fini} options for the same 3830purpose. The patches correct various problems with the +init/+fini 3831options, including program core dumps. Binutils 2.14 corrects a 3832problem on the 64-bit port resulting from HP's non-standard use of 3833the .init and .fini sections for array initializers and finalizers. 3834 3835Although the HP and GNU linkers are both supported for the 3836@samp{hppa64-hp-hpux11*} target, it is strongly recommended that the 3837HP linker be used for link editing on this target. 3838 3839At this time, the GNU linker does not support the creation of long 3840branch stubs. As a result, it can't successfully link binaries 3841containing branch offsets larger than 8 megabytes. In addition, 3842there are problems linking shared libraries, linking executables 3843with @option{-static}, and with dwarf2 unwind and exception support. 3844It also doesn't provide stubs for internal calls to global functions 3845in shared libraries, so these calls can't be overloaded. 3846 3847The HP dynamic loader does not support GNU symbol versioning, so symbol 3848versioning is not supported. It may be necessary to disable symbol 3849versioning with @option{--disable-symvers} when using GNU ld. 3850 3851POSIX threads are the default. The optional DCE thread library is not 3852supported, so @option{--enable-threads=dce} does not work. 3853 3854@html 3855<hr /> 3856@end html 3857@anchor{x-x-linux-gnu} 3858@heading *-*-linux-gnu 3859Versions of libstdc++-v3 starting with 3.2.1 require bug fixes present 3860in glibc 2.2.5 and later. More information is available in the 3861libstdc++-v3 documentation. 3862 3863@html 3864<hr /> 3865@end html 3866@anchor{ix86-x-linux} 3867@heading i?86-*-linux* 3868As of GCC 3.3, binutils 2.13.1 or later is required for this platform. 3869See @uref{http://gcc.gnu.org/PR10877,,bug 10877} for more information. 3870 3871If you receive Signal 11 errors when building on GNU/Linux, then it is 3872possible you have a hardware problem. Further information on this can be 3873found on @uref{http://www.bitwizard.nl/sig11/,,www.bitwizard.nl}. 3874 3875@html 3876<hr /> 3877@end html 3878@anchor{ix86-x-solaris210} 3879@heading i?86-*-solaris2.10 3880Use this for Solaris 10 or later on x86 and x86-64 systems. Starting 3881with GCC 4.7, there is also a 64-bit @samp{amd64-*-solaris2.1[0-9]*} or 3882@samp{x86_64-*-solaris2.1[0-9]*} configuration that corresponds to 3883@samp{sparcv9-sun-solaris2*}. 3884 3885It is recommended that you configure GCC to use the GNU assembler, in 3886@file{/usr/sfw/bin/gas}. The versions included in Solaris 10, from GNU 3887binutils 2.15, and Solaris 11, from GNU binutils 2.19, work fine, 3888although the current version, from GNU binutils 38892.22, is known to work, too. Recent versions of the Sun assembler in 3890@file{/usr/ccs/bin/as} work almost as well, though. 3891@c FIXME: as patch requirements? 3892 3893For linking, the Sun linker, is preferred. If you want to use the GNU 3894linker instead, which is available in @file{/usr/sfw/bin/gld}, note that 3895due to a packaging bug the version in Solaris 10, from GNU binutils 38962.15, cannot be used, while the version in Solaris 11, from GNU binutils 38972.19, works, as does the latest version, from GNU binutils 2.22. 3898 3899To use GNU @command{as}, configure with the options 3900@option{--with-gnu-as --with-as=@//usr/@/sfw/@/bin/@/gas}. It may be necessary 3901to configure with @option{--without-gnu-ld --with-ld=@//usr/@/ccs/@/bin/@/ld} to 3902guarantee use of Sun @command{ld}. 3903@c FIXME: why --without-gnu-ld --with-ld? 3904 3905@html 3906<hr /> 3907@end html 3908@anchor{ia64-x-linux} 3909@heading ia64-*-linux 3910IA-64 processor (also known as IPF, or Itanium Processor Family) 3911running GNU/Linux. 3912 3913If you are using the installed system libunwind library with 3914@option{--with-system-libunwind}, then you must use libunwind 0.98 or 3915later. 3916 3917None of the following versions of GCC has an ABI that is compatible 3918with any of the other versions in this list, with the exception that 3919Red Hat 2.96 and Trillian 000171 are compatible with each other: 39203.1, 3.0.2, 3.0.1, 3.0, Red Hat 2.96, and Trillian 000717. 3921This primarily affects C++ programs and programs that create shared libraries. 3922GCC 3.1 or later is recommended for compiling linux, the kernel. 3923As of version 3.1 GCC is believed to be fully ABI compliant, and hence no 3924more major ABI changes are expected. 3925 3926@html 3927<hr /> 3928@end html 3929@anchor{ia64-x-hpux} 3930@heading ia64-*-hpux* 3931Building GCC on this target requires the GNU Assembler. The bundled HP 3932assembler will not work. To prevent GCC from using the wrong assembler, 3933the option @option{--with-gnu-as} may be necessary. 3934 3935The GCC libunwind library has not been ported to HPUX@. This means that for 3936GCC versions 3.2.3 and earlier, @option{--enable-libunwind-exceptions} 3937is required to build GCC@. For GCC 3.3 and later, this is the default. 3938For gcc 3.4.3 and later, @option{--enable-libunwind-exceptions} is 3939removed and the system libunwind library will always be used. 3940 3941@html 3942<hr /> 3943<!-- rs6000-ibm-aix*, powerpc-ibm-aix* --> 3944@end html 3945@anchor{x-ibm-aix} 3946@heading *-ibm-aix* 3947Support for AIX version 3 and older was discontinued in GCC 3.4. 3948Support for AIX version 4.2 and older was discontinued in GCC 4.5. 3949 3950``out of memory'' bootstrap failures may indicate a problem with 3951process resource limits (ulimit). Hard limits are configured in the 3952@file{/etc/security/limits} system configuration file. 3953 3954GCC can bootstrap with recent versions of IBM XLC, but bootstrapping 3955with an earlier release of GCC is recommended. Bootstrapping with XLC 3956requires a larger data segment, which can be enabled through the 3957@var{LDR_CNTRL} environment variable, e.g., 3958 3959@smallexample 3960% LDR_CNTRL=MAXDATA=0x50000000 3961% export LDR_CNTRL 3962@end smallexample 3963 3964One can start with a pre-compiled version of GCC to build from 3965sources. One may delete GCC's ``fixed'' header files when starting 3966with a version of GCC built for an earlier release of AIX. 3967 3968To speed up the configuration phases of bootstrapping and installing GCC, 3969one may use GNU Bash instead of AIX @command{/bin/sh}, e.g., 3970 3971@smallexample 3972% CONFIG_SHELL=/opt/freeware/bin/bash 3973% export CONFIG_SHELL 3974@end smallexample 3975 3976and then proceed as described in @uref{build.html,,the build 3977instructions}, where we strongly recommend specifying an absolute path 3978to invoke @var{srcdir}/configure. 3979 3980Because GCC on AIX is built as a 32-bit executable by default, 3981(although it can generate 64-bit programs) the GMP and MPFR libraries 3982required by gfortran must be 32-bit libraries. Building GMP and MPFR 3983as static archive libraries works better than shared libraries. 3984 3985Errors involving @code{alloca} when building GCC generally are due 3986to an incorrect definition of @code{CC} in the Makefile or mixing files 3987compiled with the native C compiler and GCC@. During the stage1 phase of 3988the build, the native AIX compiler @strong{must} be invoked as @command{cc} 3989(not @command{xlc}). Once @command{configure} has been informed of 3990@command{xlc}, one needs to use @samp{make distclean} to remove the 3991configure cache files and ensure that @env{CC} environment variable 3992does not provide a definition that will confuse @command{configure}. 3993If this error occurs during stage2 or later, then the problem most likely 3994is the version of Make (see above). 3995 3996The native @command{as} and @command{ld} are recommended for 3997bootstrapping on AIX@. The GNU Assembler, GNU Linker, and GNU 3998Binutils version 2.20 is the minimum level that supports bootstrap on 3999AIX 5@. The GNU Assembler has not been updated to support AIX 6@ or 4000AIX 7. The native AIX tools do interoperate with GCC@. 4001 4002AIX 5.3 TL10, AIX 6.1 TL05 and AIX 7.1 TL00 introduced an AIX 4003assembler change that sometimes produces corrupt assembly files 4004causing AIX linker errors. The bug breaks GCC bootstrap on AIX and 4005can cause compilation failures with existing GCC installations. An 4006AIX iFix for AIX 5.3 is available (APAR IZ98385 for AIX 5.3 TL10, APAR 4007IZ98477 for AIX 5.3 TL11 and IZ98134 for AIX 5.3 TL12). AIX 5.3 TL11 SP8, 4008AIX 5.3 TL12 SP5, AIX 6.1 TL04 SP11, AIX 6.1 TL05 SP7, AIX 6.1 TL06 SP6, 4009AIX 6.1 TL07 and AIX 7.1 TL01 should include the fix. 4010 4011Building @file{libstdc++.a} requires a fix for an AIX Assembler bug 4012APAR IY26685 (AIX 4.3) or APAR IY25528 (AIX 5.1). It also requires a 4013fix for another AIX Assembler bug and a co-dependent AIX Archiver fix 4014referenced as APAR IY53606 (AIX 5.2) or as APAR IY54774 (AIX 5.1) 4015 4016@anchor{TransferAixShobj} 4017@samp{libstdc++} in GCC 3.4 increments the major version number of the 4018shared object and GCC installation places the @file{libstdc++.a} 4019shared library in a common location which will overwrite the and GCC 40203.3 version of the shared library. Applications either need to be 4021re-linked against the new shared library or the GCC 3.1 and GCC 3.3 4022versions of the @samp{libstdc++} shared object needs to be available 4023to the AIX runtime loader. The GCC 3.1 @samp{libstdc++.so.4}, if 4024present, and GCC 3.3 @samp{libstdc++.so.5} shared objects can be 4025installed for runtime dynamic loading using the following steps to set 4026the @samp{F_LOADONLY} flag in the shared object for @emph{each} 4027multilib @file{libstdc++.a} installed: 4028 4029Extract the shared objects from the currently installed 4030@file{libstdc++.a} archive: 4031@smallexample 4032% ar -x libstdc++.a libstdc++.so.4 libstdc++.so.5 4033@end smallexample 4034 4035Enable the @samp{F_LOADONLY} flag so that the shared object will be 4036available for runtime dynamic loading, but not linking: 4037@smallexample 4038% strip -e libstdc++.so.4 libstdc++.so.5 4039@end smallexample 4040 4041Archive the runtime-only shared object in the GCC 3.4 4042@file{libstdc++.a} archive: 4043@smallexample 4044% ar -q libstdc++.a libstdc++.so.4 libstdc++.so.5 4045@end smallexample 4046 4047Eventually, the 4048@uref{./configure.html#WithAixSoname,,@option{--with-aix-soname=svr4}} 4049configure option may drop the need for this procedure for libraries that 4050support it. 4051 4052Linking executables and shared libraries may produce warnings of 4053duplicate symbols. The assembly files generated by GCC for AIX always 4054have included multiple symbol definitions for certain global variable 4055and function declarations in the original program. The warnings should 4056not prevent the linker from producing a correct library or runnable 4057executable. 4058 4059AIX 4.3 utilizes a ``large format'' archive to support both 32-bit and 406064-bit object modules. The routines provided in AIX 4.3.0 and AIX 4.3.1 4061to parse archive libraries did not handle the new format correctly. 4062These routines are used by GCC and result in error messages during 4063linking such as ``not a COFF file''. The version of the routines shipped 4064with AIX 4.3.1 should work for a 32-bit environment. The @option{-g} 4065option of the archive command may be used to create archives of 32-bit 4066objects using the original ``small format''. A correct version of the 4067routines is shipped with AIX 4.3.2 and above. 4068 4069Some versions of the AIX binder (linker) can fail with a relocation 4070overflow severe error when the @option{-bbigtoc} option is used to link 4071GCC-produced object files into an executable that overflows the TOC@. A fix 4072for APAR IX75823 (OVERFLOW DURING LINK WHEN USING GCC AND -BBIGTOC) is 4073available from IBM Customer Support and from its 4074@uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com} 4075website as PTF U455193. 4076 4077The AIX 4.3.2.1 linker (bos.rte.bind_cmds Level 4.3.2.1) will dump core 4078with a segmentation fault when invoked by any version of GCC@. A fix for 4079APAR IX87327 is available from IBM Customer Support and from its 4080@uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com} 4081website as PTF U461879. This fix is incorporated in AIX 4.3.3 and above. 4082 4083The initial assembler shipped with AIX 4.3.0 generates incorrect object 4084files. A fix for APAR IX74254 (64BIT DISASSEMBLED OUTPUT FROM COMPILER FAILS 4085TO ASSEMBLE/BIND) is available from IBM Customer Support and from its 4086@uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com} 4087website as PTF U453956. This fix is incorporated in AIX 4.3.1 and above. 4088 4089AIX provides National Language Support (NLS)@. Compilers and assemblers 4090use NLS to support locale-specific representations of various data 4091formats including floating-point numbers (e.g., @samp{.} vs @samp{,} for 4092separating decimal fractions). There have been problems reported where 4093GCC does not produce the same floating-point formats that the assembler 4094expects. If one encounters this problem, set the @env{LANG} 4095environment variable to @samp{C} or @samp{En_US}. 4096 4097A default can be specified with the @option{-mcpu=@var{cpu_type}} 4098switch and using the configure option @option{--with-cpu-@var{cpu_type}}. 4099 4100@html 4101<hr /> 4102@end html 4103@anchor{iq2000-x-elf} 4104@heading iq2000-*-elf 4105Vitesse IQ2000 processors. These are used in embedded 4106applications. There are no standard Unix configurations. 4107 4108@html 4109<hr /> 4110@end html 4111@anchor{lm32-x-elf} 4112@heading lm32-*-elf 4113Lattice Mico32 processor. 4114This configuration is intended for embedded systems. 4115 4116@html 4117<hr /> 4118@end html 4119@anchor{lm32-x-uclinux} 4120@heading lm32-*-uclinux 4121Lattice Mico32 processor. 4122This configuration is intended for embedded systems running uClinux. 4123 4124@html 4125<hr /> 4126@end html 4127@anchor{m32c-x-elf} 4128@heading m32c-*-elf 4129Renesas M32C processor. 4130This configuration is intended for embedded systems. 4131 4132@html 4133<hr /> 4134@end html 4135@anchor{m32r-x-elf} 4136@heading m32r-*-elf 4137Renesas M32R processor. 4138This configuration is intended for embedded systems. 4139 4140@html 4141<hr /> 4142@end html 4143@anchor{m68k-x-x} 4144@heading m68k-*-* 4145By default, 4146@samp{m68k-*-elf*}, @samp{m68k-*-rtems}, @samp{m68k-*-uclinux} and 4147@samp{m68k-*-linux} 4148build libraries for both M680x0 and ColdFire processors. If you only 4149need the M680x0 libraries, you can omit the ColdFire ones by passing 4150@option{--with-arch=m68k} to @command{configure}. Alternatively, you 4151can omit the M680x0 libraries by passing @option{--with-arch=cf} to 4152@command{configure}. These targets default to 5206 or 5475 code as 4153appropriate for the target system when 4154configured with @option{--with-arch=cf} and 68020 code otherwise. 4155 4156The @samp{m68k-*-netbsd} and 4157@samp{m68k-*-openbsd} targets also support the @option{--with-arch} 4158option. They will generate ColdFire CFV4e code when configured with 4159@option{--with-arch=cf} and 68020 code otherwise. 4160 4161You can override the default processors listed above by configuring 4162with @option{--with-cpu=@var{target}}. This @var{target} can either 4163be a @option{-mcpu} argument or one of the following values: 4164@samp{m68000}, @samp{m68010}, @samp{m68020}, @samp{m68030}, 4165@samp{m68040}, @samp{m68060}, @samp{m68020-40} and @samp{m68020-60}. 4166 4167GCC requires at least binutils version 2.17 on these targets. 4168 4169@html 4170<hr /> 4171@end html 4172@anchor{m68k-x-uclinux} 4173@heading m68k-*-uclinux 4174GCC 4.3 changed the uClinux configuration so that it uses the 4175@samp{m68k-linux-gnu} ABI rather than the @samp{m68k-elf} ABI. 4176It also added improved support for C++ and flat shared libraries, 4177both of which were ABI changes. 4178 4179@html 4180<hr /> 4181@end html 4182@anchor{mep-x-elf} 4183@heading mep-*-elf 4184Toshiba Media embedded Processor. 4185This configuration is intended for embedded systems. 4186 4187@html 4188<hr /> 4189@end html 4190@anchor{microblaze-x-elf} 4191@heading microblaze-*-elf 4192Xilinx MicroBlaze processor. 4193This configuration is intended for embedded systems. 4194 4195@html 4196<hr /> 4197@end html 4198@anchor{mips-x-x} 4199@heading mips-*-* 4200If on a MIPS system you get an error message saying ``does not have gp 4201sections for all it's [sic] sectons [sic]'', don't worry about it. This 4202happens whenever you use GAS with the MIPS linker, but there is not 4203really anything wrong, and it is okay to use the output file. You can 4204stop such warnings by installing the GNU linker. 4205 4206It would be nice to extend GAS to produce the gp tables, but they are 4207optional, and there should not be a warning about their absence. 4208 4209The libstdc++ atomic locking routines for MIPS targets requires MIPS II 4210and later. A patch went in just after the GCC 3.3 release to 4211make @samp{mips*-*-*} use the generic implementation instead. You can also 4212configure for @samp{mipsel-elf} as a workaround. The 4213@samp{mips*-*-linux*} target continues to use the MIPS II routines. More 4214work on this is expected in future releases. 4215 4216@c If you make --with-llsc the default for another target, please also 4217@c update the description of the --with-llsc option. 4218 4219The built-in @code{__sync_*} functions are available on MIPS II and 4220later systems and others that support the @samp{ll}, @samp{sc} and 4221@samp{sync} instructions. This can be overridden by passing 4222@option{--with-llsc} or @option{--without-llsc} when configuring GCC. 4223Since the Linux kernel emulates these instructions if they are 4224missing, the default for @samp{mips*-*-linux*} targets is 4225@option{--with-llsc}. The @option{--with-llsc} and 4226@option{--without-llsc} configure options may be overridden at compile 4227time by passing the @option{-mllsc} or @option{-mno-llsc} options to 4228the compiler. 4229 4230MIPS systems check for division by zero (unless 4231@option{-mno-check-zero-division} is passed to the compiler) by 4232generating either a conditional trap or a break instruction. Using 4233trap results in smaller code, but is only supported on MIPS II and 4234later. Also, some versions of the Linux kernel have a bug that 4235prevents trap from generating the proper signal (@code{SIGFPE}). To enable 4236the use of break, use the @option{--with-divide=breaks} 4237@command{configure} option when configuring GCC@. The default is to 4238use traps on systems that support them. 4239 4240The assembler from GNU binutils 2.17 and earlier has a bug in the way 4241it sorts relocations for REL targets (o32, o64, EABI). This can cause 4242bad code to be generated for simple C++ programs. Also the linker 4243from GNU binutils versions prior to 2.17 has a bug which causes the 4244runtime linker stubs in very large programs, like @file{libgcj.so}, to 4245be incorrectly generated. GNU Binutils 2.18 and later (and snapshots 4246made after Nov. 9, 2006) should be free from both of these problems. 4247 4248@html 4249<hr /> 4250@end html 4251@anchor{mips-sgi-irix5} 4252@heading mips-sgi-irix5 4253Support for IRIX 5 has been removed in GCC 4.6. 4254 4255@html 4256<hr /> 4257@end html 4258@anchor{mips-sgi-irix6} 4259@heading mips-sgi-irix6 4260Support for IRIX 6.5 has been removed in GCC 4.8. Support for IRIX 6 4261releases before 6.5 has been removed in GCC 4.6, as well as support for 4262the O32 ABI. 4263 4264@html 4265<hr /> 4266@end html 4267@anchor{moxie-x-elf} 4268@heading moxie-*-elf 4269The moxie processor. 4270 4271@html 4272<hr /> 4273@end html 4274@anchor{msp430-x-elf} 4275@heading msp430-*-elf 4276TI MSP430 processor. 4277This configuration is intended for embedded systems. 4278 4279@html 4280<hr /> 4281@end html 4282@anchor{nds32le-x-elf} 4283@heading nds32le-*-elf 4284Andes NDS32 target in little endian mode. 4285 4286@html 4287<hr /> 4288@end html 4289@anchor{nds32be-x-elf} 4290@heading nds32be-*-elf 4291Andes NDS32 target in big endian mode. 4292 4293@html 4294<hr /> 4295@end html 4296@anchor{nvptx-x-none} 4297@heading nvptx-*-none 4298Nvidia PTX target. 4299 4300Instead of GNU binutils, you will need to install 4301@uref{https://github.com/MentorEmbedded/nvptx-tools/,,nvptx-tools}. 4302Tell GCC where to find it: 4303@option{--with-build-time-tools=[install-nvptx-tools]/nvptx-none/bin}. 4304 4305A nvptx port of newlib is available at 4306@uref{https://github.com/MentorEmbedded/nvptx-newlib/,,nvptx-newlib}. 4307It can be automatically built together with GCC@. For this, add a 4308symbolic link to nvptx-newlib's @file{newlib} directory to the 4309directory containing the GCC sources. 4310 4311Use the @option{--disable-sjlj-exceptions} and 4312@option{--enable-newlib-io-long-long} options when configuring. 4313 4314@html 4315<hr /> 4316@end html 4317@anchor{powerpc-x-x} 4318@heading powerpc-*-* 4319You can specify a default version for the @option{-mcpu=@var{cpu_type}} 4320switch by using the configure option @option{--with-cpu-@var{cpu_type}}. 4321 4322You will need 4323@uref{ftp://ftp.kernel.org/pub/linux/devel/binutils,,binutils 2.15} 4324or newer for a working GCC@. 4325 4326@html 4327<hr /> 4328@end html 4329@anchor{powerpc-x-darwin} 4330@heading powerpc-*-darwin* 4331PowerPC running Darwin (Mac OS X kernel). 4332 4333Pre-installed versions of Mac OS X may not include any developer tools, 4334meaning that you will not be able to build GCC from source. Tool 4335binaries are available at 4336@uref{http://opensource.apple.com/}. 4337 4338This version of GCC requires at least cctools-590.36. The 4339cctools-590.36 package referenced from 4340@uref{http://gcc.gnu.org/ml/gcc/2006-03/msg00507.html} will not work 4341on systems older than 10.3.9 (aka darwin7.9.0). 4342 4343@html 4344<hr /> 4345@end html 4346@anchor{powerpc-x-elf} 4347@heading powerpc-*-elf 4348PowerPC system in big endian mode, running System V.4. 4349 4350@html 4351<hr /> 4352@end html 4353@anchor{powerpc-x-linux-gnu} 4354@heading powerpc*-*-linux-gnu* 4355PowerPC system in big endian mode running Linux. 4356 4357@html 4358<hr /> 4359@end html 4360@anchor{powerpc-x-netbsd} 4361@heading powerpc-*-netbsd* 4362PowerPC system in big endian mode running NetBSD@. 4363 4364@html 4365<hr /> 4366@end html 4367@anchor{powerpc-x-eabisim} 4368@heading powerpc-*-eabisim 4369Embedded PowerPC system in big endian mode for use in running under the 4370PSIM simulator. 4371 4372@html 4373<hr /> 4374@end html 4375@anchor{powerpc-x-eabi} 4376@heading powerpc-*-eabi 4377Embedded PowerPC system in big endian mode. 4378 4379@html 4380<hr /> 4381@end html 4382@anchor{powerpcle-x-elf} 4383@heading powerpcle-*-elf 4384PowerPC system in little endian mode, running System V.4. 4385 4386@html 4387<hr /> 4388@end html 4389@anchor{powerpcle-x-eabisim} 4390@heading powerpcle-*-eabisim 4391Embedded PowerPC system in little endian mode for use in running under 4392the PSIM simulator. 4393 4394@html 4395<hr /> 4396@end html 4397@anchor{powerpcle-x-eabi} 4398@heading powerpcle-*-eabi 4399Embedded PowerPC system in little endian mode. 4400 4401@html 4402<hr /> 4403@end html 4404@anchor{rl78-x-elf} 4405@heading rl78-*-elf 4406The Renesas RL78 processor. 4407This configuration is intended for embedded systems. 4408 4409@html 4410<hr /> 4411@end html 4412@anchor{rx-x-elf} 4413@heading rx-*-elf 4414The Renesas RX processor. See 4415@uref{http://eu.renesas.com/fmwk.jsp?cnt=rx600_series_landing.jsp&fp=/products/mpumcu/rx_family/rx600_series} 4416for more information about this processor. 4417 4418@html 4419<hr /> 4420@end html 4421@anchor{s390-x-linux} 4422@heading s390-*-linux* 4423S/390 system running GNU/Linux for S/390@. 4424 4425@html 4426<hr /> 4427@end html 4428@anchor{s390x-x-linux} 4429@heading s390x-*-linux* 4430zSeries system (64-bit) running GNU/Linux for zSeries@. 4431 4432@html 4433<hr /> 4434@end html 4435@anchor{s390x-ibm-tpf} 4436@heading s390x-ibm-tpf* 4437zSeries system (64-bit) running TPF@. This platform is 4438supported as cross-compilation target only. 4439 4440@html 4441<hr /> 4442@end html 4443@c Please use Solaris 2 to refer to all release of Solaris, starting 4444@c with 2.0 until 2.6, 7, 8, etc. Solaris 1 was a marketing name for 4445@c SunOS 4 releases which we don't use to avoid confusion. Solaris 4446@c alone is too unspecific and must be avoided. 4447@anchor{x-x-solaris2} 4448@heading *-*-solaris2* 4449Support for Solaris 9 has been removed in GCC 4.10. Support for Solaris 44508 has been removed in GCC 4.8. Support for Solaris 7 has been removed 4451in GCC 4.6. 4452 4453Sun does not ship a C compiler with Solaris 2 before Solaris 10, though 4454you can download the Sun Studio compilers for free. In Solaris 10 and 445511, GCC 3.4.3 is available as @command{/usr/sfw/bin/gcc}. Solaris 11 4456also provides GCC 4.5.2 as @command{/usr/gcc/4.5/bin/gcc}. Alternatively, 4457you can install a pre-built GCC to bootstrap and install GCC. See the 4458@uref{binaries.html,,binaries page} for details. 4459 4460The Solaris 2 @command{/bin/sh} will often fail to configure 4461@samp{libstdc++-v3}, @samp{boehm-gc} or @samp{libjava}. We therefore 4462recommend using the following initial sequence of commands 4463 4464@smallexample 4465% CONFIG_SHELL=/bin/ksh 4466% export CONFIG_SHELL 4467@end smallexample 4468 4469@noindent 4470and proceed as described in @uref{configure.html,,the configure instructions}. 4471In addition we strongly recommend specifying an absolute path to invoke 4472@command{@var{srcdir}/configure}. 4473 4474Solaris 2 comes with a number of optional OS packages. Some of these 4475are needed to use GCC fully, namely @code{SUNWarc}, 4476@code{SUNWbtool}, @code{SUNWesu}, @code{SUNWhea}, @code{SUNWlibm}, 4477@code{SUNWsprot}, and @code{SUNWtoo}. If you did not install all 4478optional packages when installing Solaris 2, you will need to verify that 4479the packages that GCC needs are installed. 4480 4481To check whether an optional package is installed, use 4482the @command{pkginfo} command. To add an optional package, use the 4483@command{pkgadd} command. For further details, see the Solaris 2 4484documentation. 4485 4486Trying to use the linker and other tools in 4487@file{/usr/ucb} to install GCC has been observed to cause trouble. 4488For example, the linker may hang indefinitely. The fix is to remove 4489@file{/usr/ucb} from your @env{PATH}. 4490 4491The build process works more smoothly with the legacy Sun tools so, if you 4492have @file{/usr/xpg4/bin} in your @env{PATH}, we recommend that you place 4493@file{/usr/bin} before @file{/usr/xpg4/bin} for the duration of the build. 4494 4495We recommend the use of the Sun assembler or the GNU assembler, in 4496conjunction with the Sun linker. The GNU @command{as} 4497versions included in Solaris 10, from GNU binutils 2.15, and Solaris 11, 4498from GNU binutils 2.19, are known to work. They can be found in 4499@file{/usr/sfw/bin/gas}. Current versions of GNU binutils (2.22) 4500are known to work as well. Note that your mileage may vary 4501if you use a combination of the GNU tools and the Sun tools: while the 4502combination GNU @command{as} + Sun @command{ld} should reasonably work, 4503the reverse combination Sun @command{as} + GNU @command{ld} may fail to 4504build or cause memory corruption at runtime in some cases for C++ programs. 4505@c FIXME: still? 4506GNU @command{ld} usually works as well, although the version included in 4507Solaris 10 cannot be used due to several bugs. Again, the current 4508version (2.22) is known to work, but generally lacks platform specific 4509features, so better stay with Sun @command{ld}. To use the LTO linker 4510plugin (@option{-fuse-linker-plugin}) with GNU @command{ld}, GNU 4511binutils @emph{must} be configured with @option{--enable-largefile}. 4512 4513To enable symbol versioning in @samp{libstdc++} with Sun @command{ld}, 4514you need to have any version of GNU @command{c++filt}, which is part of 4515GNU binutils. @samp{libstdc++} symbol versioning will be disabled if no 4516appropriate version is found. Sun @command{c++filt} from the Sun Studio 4517compilers does @emph{not} work. 4518 4519Sun bug 4296832 turns up when compiling X11 headers with GCC 2.95 or 4520newer: @command{g++} will complain that types are missing. These headers 4521assume that omitting the type means @code{int}; this assumption worked for 4522C90 but is wrong for C++, and is now wrong for C99 also. 4523 4524Sun bug 4927647 sometimes causes random spurious testsuite failures 4525related to missing diagnostic output. This bug doesn't affect GCC 4526itself, rather it is a kernel bug triggered by the @command{expect} 4527program which is used only by the GCC testsuite driver. When the bug 4528causes the @command{expect} program to miss anticipated output, extra 4529testsuite failures appear. 4530 4531@html 4532<hr /> 4533@end html 4534@anchor{sparc-x-x} 4535@heading sparc*-*-* 4536This section contains general configuration information for all 4537SPARC-based platforms. In addition to reading this section, please 4538read all other sections that match your target. 4539 4540Newer versions of the GNU Multiple Precision Library (GMP), the MPFR 4541library and the MPC library are known to be miscompiled by earlier 4542versions of GCC on these platforms. We therefore recommend the use 4543of the exact versions of these libraries listed as minimal versions 4544in @uref{prerequisites.html,,the prerequisites}. 4545 4546@html 4547<hr /> 4548@end html 4549@anchor{sparc-sun-solaris2} 4550@heading sparc-sun-solaris2* 4551When GCC is configured to use GNU binutils 2.14 or later, the binaries 4552produced are smaller than the ones produced using Sun's native tools; 4553this difference is quite significant for binaries containing debugging 4554information. 4555 4556Starting with Solaris 7, the operating system is capable of executing 455764-bit SPARC V9 binaries. GCC 3.1 and later properly supports 4558this; the @option{-m64} option enables 64-bit code generation. 4559However, if all you want is code tuned for the UltraSPARC CPU, you 4560should try the @option{-mtune=ultrasparc} option instead, which produces 4561code that, unlike full 64-bit code, can still run on non-UltraSPARC 4562machines. 4563 4564When configuring on a Solaris 7 or later system that is running a kernel 4565that supports only 32-bit binaries, one must configure with 4566@option{--disable-multilib}, since we will not be able to build the 456764-bit target libraries. 4568 4569GCC 3.3 and GCC 3.4 trigger code generation bugs in earlier versions of 4570the GNU compiler (especially GCC 3.0.x versions), which lead to the 4571miscompilation of the stage1 compiler and the subsequent failure of the 4572bootstrap process. A workaround is to use GCC 3.2.3 as an intermediary 4573stage, i.e.@: to bootstrap that compiler with the base compiler and then 4574use it to bootstrap the final compiler. 4575 4576GCC 3.4 triggers a code generation bug in versions 5.4 (Sun ONE Studio 7) 4577and 5.5 (Sun ONE Studio 8) of the Sun compiler, which causes a bootstrap 4578failure in form of a miscompilation of the stage1 compiler by the Sun 4579compiler. This is Sun bug 4974440. This is fixed with patch 112760-07. 4580 4581GCC 3.4 changed the default debugging format from Stabs to DWARF-2 for 458232-bit code on Solaris 7 and later. If you use the Sun assembler, this 4583change apparently runs afoul of Sun bug 4910101 (which is referenced as 4584an x86-only problem by Sun, probably because they do not use DWARF-2). 4585A symptom of the problem is that you cannot compile C++ programs like 4586@command{groff} 1.19.1 without getting messages similar to the following: 4587 4588@smallexample 4589ld: warning: relocation error: R_SPARC_UA32: @dots{} 4590 external symbolic relocation against non-allocatable section 4591 .debug_info cannot be processed at runtime: relocation ignored. 4592@end smallexample 4593 4594@noindent 4595To work around this problem, compile with @option{-gstabs+} instead of 4596plain @option{-g}. 4597 4598When configuring the GNU Multiple Precision Library (GMP), the MPFR 4599library or the MPC library on a Solaris 7 or later system, the canonical 4600target triplet must be specified as the @command{build} parameter on the 4601configure line. This target triplet can be obtained by invoking @command{./config.guess} in the toplevel source directory of GCC (and 4602not that of GMP or MPFR or MPC). For example on a Solaris 9 system: 4603 4604@smallexample 4605% ./configure --build=sparc-sun-solaris2.9 --prefix=xxx 4606@end smallexample 4607 4608@html 4609<hr /> 4610@end html 4611@anchor{sparc-sun-solaris210} 4612@heading sparc-sun-solaris2.10 4613There is a bug in older versions of the Sun assembler which breaks 4614thread-local storage (TLS). A typical error message is 4615 4616@smallexample 4617ld: fatal: relocation error: R_SPARC_TLS_LE_HIX22: file /var/tmp//ccamPA1v.o: 4618 symbol <unknown>: bad symbol type SECT: symbol type must be TLS 4619@end smallexample 4620 4621@noindent 4622This bug is fixed in Sun patch 118683-03 or later. 4623 4624@html 4625<hr /> 4626@end html 4627@anchor{sparc-x-linux} 4628@heading sparc-*-linux* 4629 4630GCC versions 3.0 and higher require binutils 2.11.2 and glibc 2.2.4 4631or newer on this platform. All earlier binutils and glibc 4632releases mishandled unaligned relocations on @code{sparc-*-*} targets. 4633 4634 4635@html 4636<hr /> 4637@end html 4638@anchor{sparc64-x-solaris2} 4639@heading sparc64-*-solaris2* 4640When configuring the GNU Multiple Precision Library (GMP), the MPFR 4641library or the MPC library, the canonical target triplet must be specified 4642as the @command{build} parameter on the configure line. For example 4643on a Solaris 9 system: 4644 4645@smallexample 4646% ./configure --build=sparc64-sun-solaris2.9 --prefix=xxx 4647@end smallexample 4648 4649The following compiler flags must be specified in the configure 4650step in order to bootstrap this target with the Sun compiler: 4651 4652@smallexample 4653% CC="cc -xarch=v9 -xildoff" @var{srcdir}/configure [@var{options}] [@var{target}] 4654@end smallexample 4655 4656@noindent 4657@option{-xarch=v9} specifies the SPARC-V9 architecture to the Sun toolchain 4658and @option{-xildoff} turns off the incremental linker. 4659 4660@html 4661<hr /> 4662@end html 4663@anchor{sparcv9-x-solaris2} 4664@heading sparcv9-*-solaris2* 4665This is a synonym for @samp{sparc64-*-solaris2*}. 4666 4667@html 4668<hr /> 4669@end html 4670@anchor{c6x-x-x} 4671@heading c6x-*-* 4672The C6X family of processors. This port requires binutils-2.22 or newer. 4673 4674@html 4675<hr /> 4676@end html 4677@anchor{tilegx-*-linux} 4678@heading tilegx-*-linux* 4679The TILE-Gx processor in little endian mode, running GNU/Linux. This 4680port requires binutils-2.22 or newer. 4681 4682@html 4683<hr /> 4684@end html 4685@anchor{tilegxbe-*-linux} 4686@heading tilegxbe-*-linux* 4687The TILE-Gx processor in big endian mode, running GNU/Linux. This 4688port requires binutils-2.23 or newer. 4689 4690@html 4691<hr /> 4692@end html 4693@anchor{tilepro-*-linux} 4694@heading tilepro-*-linux* 4695The TILEPro processor running GNU/Linux. This port requires 4696binutils-2.22 or newer. 4697 4698@html 4699<hr /> 4700@end html 4701@anchor{visium-x-elf} 4702@heading visium-*-elf 4703CDS VISIUMcore processor. 4704This configuration is intended for embedded systems. 4705 4706@html 4707<hr /> 4708@end html 4709@anchor{x-x-vxworks} 4710@heading *-*-vxworks* 4711Support for VxWorks is in flux. At present GCC supports @emph{only} the 4712very recent VxWorks 5.5 (aka Tornado 2.2) release, and only on PowerPC@. 4713We welcome patches for other architectures supported by VxWorks 5.5. 4714Support for VxWorks AE would also be welcome; we believe this is merely 4715a matter of writing an appropriate ``configlette'' (see below). We are 4716not interested in supporting older, a.out or COFF-based, versions of 4717VxWorks in GCC 3. 4718 4719VxWorks comes with an older version of GCC installed in 4720@file{@var{$WIND_BASE}/host}; we recommend you do not overwrite it. 4721Choose an installation @var{prefix} entirely outside @var{$WIND_BASE}. 4722Before running @command{configure}, create the directories @file{@var{prefix}} 4723and @file{@var{prefix}/bin}. Link or copy the appropriate assembler, 4724linker, etc.@: into @file{@var{prefix}/bin}, and set your @var{PATH} to 4725include that directory while running both @command{configure} and 4726@command{make}. 4727 4728You must give @command{configure} the 4729@option{--with-headers=@var{$WIND_BASE}/target/h} switch so that it can 4730find the VxWorks system headers. Since VxWorks is a cross compilation 4731target only, you must also specify @option{--target=@var{target}}. 4732@command{configure} will attempt to create the directory 4733@file{@var{prefix}/@var{target}/sys-include} and copy files into it; 4734make sure the user running @command{configure} has sufficient privilege 4735to do so. 4736 4737GCC's exception handling runtime requires a special ``configlette'' 4738module, @file{contrib/gthr_supp_vxw_5x.c}. Follow the instructions in 4739that file to add the module to your kernel build. (Future versions of 4740VxWorks will incorporate this module.) 4741 4742@html 4743<hr /> 4744@end html 4745@anchor{x86-64-x-x} 4746@heading x86_64-*-*, amd64-*-* 4747GCC supports the x86-64 architecture implemented by the AMD64 processor 4748(amd64-*-* is an alias for x86_64-*-*) on GNU/Linux, FreeBSD and NetBSD@. 4749On GNU/Linux the default is a bi-arch compiler which is able to generate 4750both 64-bit x86-64 and 32-bit x86 code (via the @option{-m32} switch). 4751 4752@html 4753<hr /> 4754@end html 4755@anchor{x86-64-x-solaris210} 4756@heading x86_64-*-solaris2.1[0-9]* 4757GCC also supports the x86-64 architecture implemented by the AMD64 4758processor (@samp{amd64-*-*} is an alias for @samp{x86_64-*-*}) on 4759Solaris 10 or later. Unlike other systems, without special options a 4760bi-arch compiler is built which generates 32-bit code by default, but 4761can generate 64-bit x86-64 code with the @option{-m64} switch. Since 4762GCC 4.7, there is also configuration that defaults to 64-bit code, but 4763can generate 32-bit code with @option{-m32}. To configure and build 4764this way, you have to provide all support libraries like @file{libgmp} 4765as 64-bit code, configure with @option{--target=x86_64-pc-solaris2.1x} 4766and @samp{CC=gcc -m64}. 4767 4768@html 4769<hr /> 4770@end html 4771@anchor{xtensa-x-elf} 4772@heading xtensa*-*-elf 4773This target is intended for embedded Xtensa systems using the 4774@samp{newlib} C library. It uses ELF but does not support shared 4775objects. Designed-defined instructions specified via the 4776Tensilica Instruction Extension (TIE) language are only supported 4777through inline assembly. 4778 4779The Xtensa configuration information must be specified prior to 4780building GCC@. The @file{include/xtensa-config.h} header 4781file contains the configuration information. If you created your 4782own Xtensa configuration with the Xtensa Processor Generator, the 4783downloaded files include a customized copy of this header file, 4784which you can use to replace the default header file. 4785 4786@html 4787<hr /> 4788@end html 4789@anchor{xtensa-x-linux} 4790@heading xtensa*-*-linux* 4791This target is for Xtensa systems running GNU/Linux. It supports ELF 4792shared objects and the GNU C library (glibc). It also generates 4793position-independent code (PIC) regardless of whether the 4794@option{-fpic} or @option{-fPIC} options are used. In other 4795respects, this target is the same as the 4796@uref{#xtensa*-*-elf,,@samp{xtensa*-*-elf}} target. 4797 4798@html 4799<hr /> 4800@end html 4801@anchor{windows} 4802@heading Microsoft Windows 4803 4804@subheading Intel 16-bit versions 4805The 16-bit versions of Microsoft Windows, such as Windows 3.1, are not 4806supported. 4807 4808However, the 32-bit port has limited support for Microsoft 4809Windows 3.11 in the Win32s environment, as a target only. See below. 4810 4811@subheading Intel 32-bit versions 4812The 32-bit versions of Windows, including Windows 95, Windows NT, Windows 4813XP, and Windows Vista, are supported by several different target 4814platforms. These targets differ in which Windows subsystem they target 4815and which C libraries are used. 4816 4817@itemize 4818@item Cygwin @uref{#x-x-cygwin,,*-*-cygwin}: Cygwin provides a user-space 4819Linux API emulation layer in the Win32 subsystem. 4820@item Interix @uref{#x-x-interix,,*-*-interix}: The Interix subsystem 4821provides native support for POSIX. 4822@item MinGW @uref{#x-x-mingw32,,*-*-mingw32}: MinGW is a native GCC port for 4823the Win32 subsystem that provides a subset of POSIX. 4824@item MKS i386-pc-mks: NuTCracker from MKS. See 4825@uref{http://www.mkssoftware.com/} for more information. 4826@end itemize 4827 4828@subheading Intel 64-bit versions 4829GCC contains support for x86-64 using the mingw-w64 4830runtime library, available from @uref{http://mingw-w64.sourceforge.net/}. 4831This library should be used with the target triple x86_64-pc-mingw32. 4832 4833Presently Windows for Itanium is not supported. 4834 4835@subheading Windows CE 4836Windows CE is supported as a target only on Hitachi 4837SuperH (sh-wince-pe), and MIPS (mips-wince-pe). 4838 4839@subheading Other Windows Platforms 4840GCC no longer supports Windows NT on the Alpha or PowerPC. 4841 4842GCC no longer supports the Windows POSIX subsystem. However, it does 4843support the Interix subsystem. See above. 4844 4845Old target names including *-*-winnt and *-*-windowsnt are no longer used. 4846 4847PW32 (i386-pc-pw32) support was never completed, and the project seems to 4848be inactive. See @uref{http://pw32.sourceforge.net/} for more information. 4849 4850UWIN support has been removed due to a lack of maintenance. 4851 4852@html 4853<hr /> 4854@end html 4855@anchor{x-x-cygwin} 4856@heading *-*-cygwin 4857Ports of GCC are included with the 4858@uref{http://www.cygwin.com/,,Cygwin environment}. 4859 4860GCC will build under Cygwin without modification; it does not build 4861with Microsoft's C++ compiler and there are no plans to make it do so. 4862 4863The Cygwin native compiler can be configured to target any 32-bit x86 4864cpu architecture desired; the default is i686-pc-cygwin. It should be 4865used with as up-to-date a version of binutils as possible; use either 4866the latest official GNU binutils release in the Cygwin distribution, 4867or version 2.20 or above if building your own. 4868 4869@html 4870<hr /> 4871@end html 4872@anchor{x-x-interix} 4873@heading *-*-interix 4874The Interix target is used by OpenNT, Interix, Services For UNIX (SFU), 4875and Subsystem for UNIX-based Applications (SUA). Applications compiled 4876with this target run in the Interix subsystem, which is separate from 4877the Win32 subsystem. This target was last known to work in GCC 3.3. 4878 4879@html 4880<hr /> 4881@end html 4882@anchor{x-x-mingw32} 4883@heading *-*-mingw32 4884GCC will build with and support only MinGW runtime 3.12 and later. 4885Earlier versions of headers are incompatible with the new default semantics 4886of @code{extern inline} in @code{-std=c99} and @code{-std=gnu99} modes. 4887 4888@html 4889<hr /> 4890@end html 4891@anchor{older} 4892@heading Older systems 4893GCC contains support files for many older (1980s and early 48941990s) Unix variants. For the most part, support for these systems 4895has not been deliberately removed, but it has not been maintained for 4896several years and may suffer from bitrot. 4897 4898Starting with GCC 3.1, each release has a list of ``obsoleted'' systems. 4899Support for these systems is still present in that release, but 4900@command{configure} will fail unless the @option{--enable-obsolete} 4901option is given. Unless a maintainer steps forward, support for these 4902systems will be removed from the next release of GCC@. 4903 4904Support for old systems as hosts for GCC can cause problems if the 4905workarounds for compiler, library and operating system bugs affect the 4906cleanliness or maintainability of the rest of GCC@. In some cases, to 4907bring GCC up on such a system, if still possible with current GCC, may 4908require first installing an old version of GCC which did work on that 4909system, and using it to compile a more recent GCC, to avoid bugs in the 4910vendor compiler. Old releases of GCC 1 and GCC 2 are available in the 4911@file{old-releases} directory on the @uref{../mirrors.html,,GCC mirror 4912sites}. Header bugs may generally be avoided using 4913@command{fixincludes}, but bugs or deficiencies in libraries and the 4914operating system may still cause problems. 4915 4916Support for older systems as targets for cross-compilation is less 4917problematic than support for them as hosts for GCC; if an enthusiast 4918wishes to make such a target work again (including resurrecting any of 4919the targets that never worked with GCC 2, starting from the last 4920version before they were removed), patches 4921@uref{../contribute.html,,following the usual requirements} would be 4922likely to be accepted, since they should not affect the support for more 4923modern targets. 4924 4925For some systems, old versions of GNU binutils may also be useful, 4926and are available from @file{pub/binutils/old-releases} on 4927@uref{http://sourceware.org/mirrors.html,,sourceware.org mirror sites}. 4928 4929Some of the information on specific systems above relates to 4930such older systems, but much of the information 4931about GCC on such systems (which may no longer be applicable to 4932current GCC) is to be found in the GCC texinfo manual. 4933 4934@html 4935<hr /> 4936@end html 4937@anchor{elf} 4938@heading all ELF targets (SVR4, Solaris 2, etc.) 4939C++ support is significantly better on ELF targets if you use the 4940@uref{./configure.html#with-gnu-ld,,GNU linker}; duplicate copies of 4941inlines, vtables and template instantiations will be discarded 4942automatically. 4943 4944 4945@html 4946<hr /> 4947<p> 4948@end html 4949@ifhtml 4950@uref{./index.html,,Return to the GCC Installation page} 4951@end ifhtml 4952@end ifset 4953 4954@c ***Old documentation****************************************************** 4955@ifset oldhtml 4956@include install-old.texi 4957@html 4958<hr /> 4959<p> 4960@end html 4961@ifhtml 4962@uref{./index.html,,Return to the GCC Installation page} 4963@end ifhtml 4964@end ifset 4965 4966@c ***GFDL******************************************************************** 4967@ifset gfdlhtml 4968@include fdl.texi 4969@html 4970<hr /> 4971<p> 4972@end html 4973@ifhtml 4974@uref{./index.html,,Return to the GCC Installation page} 4975@end ifhtml 4976@end ifset 4977 4978@c *************************************************************************** 4979@c Part 6 The End of the Document 4980@ifinfo 4981@comment node-name, next, previous, up 4982@node Concept Index, , GNU Free Documentation License, Top 4983@end ifinfo 4984 4985@ifinfo 4986@unnumbered Concept Index 4987 4988@printindex cp 4989 4990@contents 4991@end ifinfo 4992@bye 4993