Home | History | Annotate | Line # | Download | only in INSTALL
configure.html revision 1.19
      1  1.10    mrg <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
      2  1.10    mrg <html>
      3  1.19    mrg <!-- Copyright (C) 1988-2024 Free Software Foundation, Inc.
      4   1.1    mrg 
      5  1.10    mrg Permission is granted to copy, distribute and/or modify this document
      6   1.3  skrll under the terms of the GNU Free Documentation License, Version 1.3 or
      7   1.1    mrg any later version published by the Free Software Foundation; with no
      8   1.1    mrg Invariant Sections, the Front-Cover texts being (a) (see below), and
      9   1.1    mrg with the Back-Cover Texts being (b) (see below).  A copy of the
     10  1.10    mrg license is included in the section entitled "GNU
     11  1.10    mrg Free Documentation License".
     12   1.1    mrg 
     13   1.1    mrg (a) The FSF's Front-Cover Text is:
     14   1.1    mrg 
     15  1.10    mrg A GNU Manual
     16   1.1    mrg 
     17   1.1    mrg (b) The FSF's Back-Cover Text is:
     18   1.1    mrg 
     19  1.10    mrg You have freedom to copy and modify this GNU Manual, like GNU
     20   1.1    mrg      software.  Copies published by the Free Software Foundation raise
     21  1.10    mrg      funds for GNU development. -->
     22  1.18    mrg <!-- Created by GNU Texinfo 6.5, http://www.gnu.org/software/texinfo/ -->
     23  1.18    mrg <head>
     24  1.18    mrg <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
     25  1.13    mrg <title>Installing GCC: Configuration</title>
     26  1.10    mrg 
     27  1.13    mrg <meta name="description" content="Installing GCC: Configuration">
     28  1.13    mrg <meta name="keywords" content="Installing GCC: Configuration">
     29  1.10    mrg <meta name="resource-type" content="document">
     30  1.10    mrg <meta name="distribution" content="global">
     31  1.10    mrg <meta name="Generator" content="makeinfo">
     32  1.10    mrg <style type="text/css">
     33  1.10    mrg <!--
     34  1.10    mrg a.summary-letter {text-decoration: none}
     35  1.10    mrg blockquote.indentedblock {margin-right: 0em}
     36  1.18    mrg blockquote.smallindentedblock {margin-right: 0em; font-size: smaller}
     37  1.18    mrg blockquote.smallquotation {font-size: smaller}
     38  1.10    mrg div.display {margin-left: 3.2em}
     39  1.10    mrg div.example {margin-left: 3.2em}
     40  1.18    mrg div.lisp {margin-left: 3.2em}
     41  1.18    mrg div.smalldisplay {margin-left: 3.2em}
     42  1.18    mrg div.smallexample {margin-left: 3.2em}
     43  1.18    mrg div.smalllisp {margin-left: 3.2em}
     44  1.10    mrg kbd {font-style: oblique}
     45  1.10    mrg pre.display {font-family: inherit}
     46  1.10    mrg pre.format {font-family: inherit}
     47  1.10    mrg pre.menu-comment {font-family: serif}
     48  1.10    mrg pre.menu-preformatted {font-family: serif}
     49  1.18    mrg pre.smalldisplay {font-family: inherit; font-size: smaller}
     50  1.18    mrg pre.smallexample {font-size: smaller}
     51  1.18    mrg pre.smallformat {font-family: inherit; font-size: smaller}
     52  1.18    mrg pre.smalllisp {font-size: smaller}
     53  1.10    mrg span.nolinebreak {white-space: nowrap}
     54  1.10    mrg span.roman {font-family: initial; font-weight: normal}
     55  1.10    mrg span.sansserif {font-family: sans-serif; font-weight: normal}
     56  1.10    mrg ul.no-bullet {list-style: none}
     57  1.10    mrg -->
     58  1.10    mrg </style>
     59  1.10    mrg 
     60  1.10    mrg 
     61   1.1    mrg </head>
     62  1.10    mrg 
     63  1.10    mrg <body lang="en">
     64  1.13    mrg <h1 class="settitle" align="center">Installing GCC: Configuration</h1>
     65  1.10    mrg 
     66  1.10    mrg 
     67  1.10    mrg 
     68  1.10    mrg 
     69  1.10    mrg 
     70  1.10    mrg 
     71  1.10    mrg 
     72  1.10    mrg 
     73  1.10    mrg 
     74  1.10    mrg 
     75  1.10    mrg 
     76  1.10    mrg 
     77  1.10    mrg 
     78  1.10    mrg 
     79  1.10    mrg 
     80  1.10    mrg 
     81  1.10    mrg 
     82  1.10    mrg 
     83  1.10    mrg 
     84  1.19    mrg 
     85  1.19    mrg 
     86  1.19    mrg 
     87  1.18    mrg <a name="index-Configuration"></a>
     88  1.18    mrg <a name="index-Installing-GCC_003a-Configuration"></a>
     89  1.10    mrg 
     90  1.10    mrg <p>Like most GNU software, GCC must be configured before it can be built.
     91   1.9    mrg This document describes the recommended configuration procedure
     92   1.9    mrg for both native and cross targets.
     93  1.10    mrg </p>
     94  1.10    mrg <p>We use <var>srcdir</var> to refer to the toplevel source directory for
     95   1.9    mrg GCC; we use <var>objdir</var> to refer to the toplevel build/object directory.
     96  1.10    mrg </p>
     97  1.11    mrg <p>If you obtained the sources by cloning the repository, <var>srcdir</var>
     98  1.11    mrg must refer to the top <samp>gcc</samp> directory, the one where the
     99  1.11    mrg <samp>MAINTAINERS</samp> file can be found, and not its <samp>gcc</samp>
    100  1.11    mrg subdirectory, otherwise the build will fail.
    101  1.10    mrg </p>
    102  1.10    mrg <p>If either <var>srcdir</var> or <var>objdir</var> is located on an automounted NFS
    103  1.10    mrg file system, the shell&rsquo;s built-in <code>pwd</code> command will return
    104   1.1    mrg temporary pathnames.  Using these can lead to various sorts of build
    105  1.10    mrg problems.  To avoid this issue, set the <code>PWDCMD</code> environment
    106  1.10    mrg variable to an automounter-aware <code>pwd</code> command, e.g.,
    107  1.10    mrg <code>pawd</code> or &lsquo;<samp>amq -w</samp>&rsquo;, during the configuration and build
    108   1.1    mrg phases.
    109  1.10    mrg </p>
    110  1.10    mrg <p>First, we <strong>highly</strong> recommend that GCC be built into a
    111   1.1    mrg separate directory from the sources which does <strong>not</strong> reside
    112   1.1    mrg within the source tree.  This is how we generally build GCC; building
    113  1.19    mrg where <var>objdir</var> is a subdirectory of <var>srcdir</var> should work as well;
    114  1.19    mrg building where <var>objdir</var> == <var>srcdir</var> is unsupported.
    115  1.10    mrg </p>
    116  1.10    mrg <p>If you have previously built GCC in the same directory for a
    117  1.10    mrg different target machine, do &lsquo;<samp>make distclean</samp>&rsquo; to delete all files
    118  1.10    mrg that might be invalid.  One of the files this deletes is <samp>Makefile</samp>;
    119  1.10    mrg if &lsquo;<samp>make distclean</samp>&rsquo; complains that <samp>Makefile</samp> does not exist
    120  1.10    mrg or issues a message like &ldquo;don&rsquo;t know how to make distclean&rdquo; it probably
    121   1.1    mrg means that the directory is already suitably clean.  However, with the
    122   1.1    mrg recommended method of building in a separate <var>objdir</var>, you should
    123   1.1    mrg simply use a different <var>objdir</var> for each target.
    124  1.10    mrg </p>
    125  1.10    mrg <p>Second, when configuring a native system, either <code>cc</code> or
    126  1.10    mrg <code>gcc</code> must be in your path or you must set <code>CC</code> in
    127   1.1    mrg your environment before running configure.  Otherwise the configuration
    128   1.1    mrg scripts may fail.
    129  1.10    mrg </p>
    130   1.1    mrg 
    131  1.10    mrg <p>To configure GCC:
    132  1.10    mrg </p>
    133  1.18    mrg <div class="smallexample">
    134  1.18    mrg <pre class="smallexample">% mkdir <var>objdir</var>
    135  1.10    mrg % cd <var>objdir</var>
    136  1.10    mrg % <var>srcdir</var>/configure [<var>options</var>] [<var>target</var>]
    137  1.10    mrg </pre></div>
    138   1.1    mrg 
    139  1.18    mrg <a name="Distributor-options"></a>
    140  1.18    mrg <h3 class="heading">Distributor options</h3>
    141   1.1    mrg 
    142   1.1    mrg <p>If you will be distributing binary versions of GCC, with modifications
    143   1.1    mrg to the source code, you should use the options described in this
    144   1.1    mrg section to make clear that your version contains modifications.
    145  1.10    mrg </p>
    146  1.10    mrg <dl compact="compact">
    147  1.18    mrg <dt><code>--with-pkgversion=<var>version</var></code></dt>
    148  1.10    mrg <dd><p>Specify a string that identifies your package.  You may wish
    149   1.1    mrg to include a build number or build date.  This version string will be
    150  1.10    mrg included in the output of <code>gcc --version</code>.  This suffix does
    151  1.10    mrg not replace the default version string, only the &lsquo;<samp>GCC</samp>&rsquo; part.
    152  1.10    mrg </p>
    153  1.10    mrg <p>The default value is &lsquo;<samp>GCC</samp>&rsquo;.
    154  1.10    mrg </p>
    155  1.10    mrg </dd>
    156  1.18    mrg <dt><code>--with-bugurl=<var>url</var></code></dt>
    157  1.10    mrg <dd><p>Specify the URL that users should visit if they wish to report a bug.
    158   1.1    mrg You are of course welcome to forward bugs reported to you to the FSF,
    159   1.1    mrg if you determine that they are not bugs in your modifications.
    160  1.10    mrg </p>
    161  1.10    mrg <p>The default value refers to the FSF&rsquo;s GCC bug tracker.
    162  1.10    mrg </p>
    163  1.10    mrg </dd>
    164  1.18    mrg <dt><code>--with-documentation-root-url=<var>url</var></code></dt>
    165  1.13    mrg <dd><p>Specify the URL root that contains GCC option documentation.  The <var>url</var>
    166  1.13    mrg should end with a <code>/</code> character.
    167  1.13    mrg </p>
    168  1.19    mrg <p>The default value is <a href="https://gcc.gnu.org/onlinedocs/">https://gcc.gnu.org/onlinedocs/</a>
    169  1.19    mrg on the GCC main development trunk.  On release branches, the default
    170  1.19    mrg is <code>https://gcc.gnu.org/onlinedocs/gcc-<var>major</var>.<var>minor</var>.0/</code>.
    171  1.13    mrg </p>
    172  1.13    mrg </dd>
    173  1.18    mrg <dt><code>--with-changes-root-url=<var>url</var></code></dt>
    174  1.13    mrg <dd><p>Specify the URL root that contains information about changes in GCC
    175  1.13    mrg releases like <code>gcc-<var>version</var>/changes.html</code>.
    176  1.13    mrg The <var>url</var> should end with a <code>/</code> character.
    177  1.13    mrg </p>
    178  1.13    mrg <p>The default value is <a href="https://gcc.gnu.org/">https://gcc.gnu.org/</a>.
    179  1.13    mrg </p>
    180  1.13    mrg </dd>
    181  1.10    mrg </dl>
    182   1.1    mrg 
    183  1.18    mrg <a name="Host_002c-Build-and-Target-specification"></a>
    184  1.18    mrg <h3 class="heading">Host, Build and Target specification</h3>
    185   1.1    mrg 
    186  1.16    mrg <p>Specify the host, build and target machine configurations.  You do this
    187  1.16    mrg when you run the <samp>configure</samp> script.
    188  1.16    mrg </p>
    189  1.16    mrg <p>The <em>build</em> machine is the system which you are using, the
    190  1.16    mrg <em>host</em> machine is the system where you want to run the resulting
    191  1.16    mrg compiler (normally the build machine), and the <em>target</em> machine is
    192  1.16    mrg the system for which you want the compiler to generate code.
    193  1.16    mrg </p>
    194  1.16    mrg <p>If you are building a compiler to produce code for the machine it runs
    195  1.16    mrg on (a native compiler), you normally do not need to specify any operands
    196  1.16    mrg to <samp>configure</samp>; it will try to guess the type of machine you are on
    197  1.16    mrg and use that as the build, host and target machines.  So you don&rsquo;t need
    198  1.16    mrg to specify a configuration when building a native compiler unless
    199  1.16    mrg <samp>configure</samp> cannot figure out what your configuration is or guesses
    200  1.16    mrg wrong.
    201  1.16    mrg </p>
    202  1.16    mrg <p>In those cases, specify the build machine&rsquo;s <em>configuration name</em>
    203  1.16    mrg with the <samp>--host</samp> option; the host and target will default to be
    204  1.16    mrg the same as the host machine.
    205  1.16    mrg </p>
    206  1.16    mrg <p>Here is an example:
    207  1.16    mrg </p>
    208  1.18    mrg <div class="smallexample">
    209  1.18    mrg <pre class="smallexample">./configure --host=x86_64-pc-linux-gnu
    210  1.16    mrg </pre></div>
    211  1.10    mrg 
    212  1.16    mrg <p>A configuration name may be canonical or it may be more or less
    213  1.16    mrg abbreviated (<samp>config.sub</samp> script produces canonical versions).
    214  1.16    mrg </p>
    215  1.16    mrg <p>A canonical configuration name has three parts, separated by dashes.
    216  1.16    mrg It looks like this: &lsquo;<samp><var>cpu</var>-<var>company</var>-<var>system</var></samp>&rsquo;.
    217  1.16    mrg </p>
    218  1.16    mrg <p>Here are the possible CPU types:
    219  1.16    mrg </p>
    220  1.16    mrg <blockquote>
    221  1.16    mrg <p>aarch64, aarch64_be, alpha, alpha64, amdgcn, arc, arceb, arm, armeb, avr, bfin,
    222  1.19    mrg bpf, cris, csky, epiphany, fido, fr30, frv, ft32, h8300, hppa, hppa2.0,
    223  1.16    mrg hppa64, i486, i686, ia64, iq2000, lm32, loongarch64, m32c, m32r, m32rle, m68k,
    224  1.16    mrg mcore, microblaze, microblazeel, mips, mips64, mips64el, mips64octeon,
    225  1.16    mrg mips64orion, mips64vr, mipsel, mipsisa32, mipsisa32r2, mipsisa64, mipsisa64r2,
    226  1.16    mrg mipsisa64r2el, mipsisa64sb1, mipsisa64sr71k, mipstx39, mmix, mn10300, moxie,
    227  1.16    mrg msp430, nds32be, nds32le, nios2, nvptx, or1k, pdp11, powerpc, powerpc64,
    228  1.16    mrg powerpc64le, powerpcle, pru, riscv32, riscv32be, riscv64, riscv64be, rl78, rx,
    229  1.19    mrg s390, s390x, sh, shle, sparc, sparc64, tic6x, v850,
    230  1.16    mrg v850e, v850e1, vax, visium, x86_64, xstormy16, xtensa
    231  1.16    mrg </p></blockquote>
    232  1.16    mrg 
    233  1.16    mrg <p>Here is a list of system types:
    234  1.16    mrg </p>
    235  1.16    mrg <blockquote>
    236  1.16    mrg <p>aix<var>version</var>, amdhsa, aout, cygwin, darwin<var>version</var>,
    237  1.16    mrg eabi, eabialtivec, eabisim, eabisimaltivec, elf, elf32,
    238  1.16    mrg elfbare, elfoabi, freebsd<var>version</var>, gnu, hpux, hpux<var>version</var>,
    239  1.16    mrg kfreebsd-gnu, kopensolaris-gnu, linux-androideabi, linux-gnu,
    240  1.16    mrg linux-gnu_altivec, linux-musl, linux-uclibc, lynxos, mingw32, mingw32crt,
    241  1.16    mrg mmixware, msdosdjgpp, netbsd, netbsdelf<var>version</var>, nto-qnx, openbsd,
    242  1.16    mrg rtems, solaris<var>version</var>, symbianelf, tpf, uclinux, uclinux_eabi, vms,
    243  1.16    mrg vxworks, vxworksae, vxworksmils
    244  1.16    mrg </p></blockquote>
    245   1.1    mrg 
    246  1.18    mrg <a name="Options-specification"></a>
    247  1.18    mrg <h3 class="heading">Options specification</h3>
    248   1.1    mrg 
    249   1.1    mrg <p>Use <var>options</var> to override several configure time options for
    250  1.10    mrg GCC.  A list of supported <var>options</var> follows; &lsquo;<samp>configure
    251  1.10    mrg --help</samp>&rsquo; may list other options, but those not listed below may not
    252   1.1    mrg work and should not normally be used.
    253  1.10    mrg </p>
    254  1.10    mrg <p>Note that each <samp>--enable</samp> option has a corresponding
    255  1.10    mrg <samp>--disable</samp> option and that each <samp>--with</samp> option has a
    256  1.10    mrg corresponding <samp>--without</samp> option.
    257  1.10    mrg </p>
    258  1.10    mrg <dl compact="compact">
    259  1.18    mrg <dt><code>--prefix=<var>dirname</var></code></dt>
    260  1.10    mrg <dd><p>Specify the toplevel installation
    261   1.1    mrg directory.  This is the recommended way to install the tools into a directory
    262   1.1    mrg other than the default.  The toplevel installation directory defaults to
    263  1.10    mrg <samp>/usr/local</samp>.
    264  1.10    mrg </p>
    265  1.10    mrg <p>We <strong>highly</strong> recommend against <var>dirname</var> being the same or a
    266   1.1    mrg subdirectory of <var>objdir</var> or vice versa.  If specifying a directory
    267  1.10    mrg beneath a user&rsquo;s home directory tree, some shells will not expand
    268  1.10    mrg <var>dirname</var> correctly if it contains the &lsquo;<samp>~</samp>&rsquo; metacharacter; use
    269  1.10    mrg <code>$HOME</code> instead.
    270  1.10    mrg </p>
    271  1.10    mrg <p>The following standard <code>autoconf</code> options are supported.  Normally you
    272   1.1    mrg should not need to use these options.
    273  1.10    mrg </p><dl compact="compact">
    274  1.18    mrg <dt><code>--exec-prefix=<var>dirname</var></code></dt>
    275  1.10    mrg <dd><p>Specify the toplevel installation directory for architecture-dependent
    276   1.1    mrg files.  The default is <samp><var>prefix</var></samp>.
    277  1.10    mrg </p>
    278  1.10    mrg </dd>
    279  1.18    mrg <dt><code>--bindir=<var>dirname</var></code></dt>
    280  1.10    mrg <dd><p>Specify the installation directory for the executables called by users
    281  1.10    mrg (such as <code>gcc</code> and <code>g++</code>).  The default is
    282  1.10    mrg <samp><var>exec-prefix</var>/bin</samp>.
    283  1.10    mrg </p>
    284  1.10    mrg </dd>
    285  1.18    mrg <dt><code>--libdir=<var>dirname</var></code></dt>
    286  1.10    mrg <dd><p>Specify the installation directory for object code libraries and
    287  1.10    mrg internal data files of GCC.  The default is <samp><var>exec-prefix</var>/lib</samp>.
    288  1.10    mrg </p>
    289  1.10    mrg </dd>
    290  1.18    mrg <dt><code>--libexecdir=<var>dirname</var></code></dt>
    291  1.10    mrg <dd><p>Specify the installation directory for internal executables of GCC.
    292  1.10    mrg The default is <samp><var>exec-prefix</var>/libexec</samp>.
    293  1.10    mrg </p>
    294  1.10    mrg </dd>
    295  1.18    mrg <dt><code>--with-slibdir=<var>dirname</var></code></dt>
    296  1.10    mrg <dd><p>Specify the installation directory for the shared libgcc library.  The
    297   1.1    mrg default is <samp><var>libdir</var></samp>.
    298  1.10    mrg </p>
    299  1.10    mrg </dd>
    300  1.18    mrg <dt><code>--datarootdir=<var>dirname</var></code></dt>
    301  1.10    mrg <dd><p>Specify the root of the directory tree for read-only architecture-independent
    302  1.10    mrg data files referenced by GCC.  The default is <samp><var>prefix</var>/share</samp>.
    303  1.10    mrg </p>
    304  1.10    mrg </dd>
    305  1.18    mrg <dt><code>--infodir=<var>dirname</var></code></dt>
    306  1.10    mrg <dd><p>Specify the installation directory for documentation in info format.
    307  1.10    mrg The default is <samp><var>datarootdir</var>/info</samp>.
    308  1.10    mrg </p>
    309  1.10    mrg </dd>
    310  1.18    mrg <dt><code>--datadir=<var>dirname</var></code></dt>
    311  1.10    mrg <dd><p>Specify the installation directory for some architecture-independent
    312   1.1    mrg data files referenced by GCC.  The default is <samp><var>datarootdir</var></samp>.
    313  1.10    mrg </p>
    314  1.10    mrg </dd>
    315  1.18    mrg <dt><code>--docdir=<var>dirname</var></code></dt>
    316  1.10    mrg <dd><p>Specify the installation directory for documentation files (other
    317  1.10    mrg than Info) for GCC.  The default is <samp><var>datarootdir</var>/doc</samp>.
    318  1.10    mrg </p>
    319  1.10    mrg </dd>
    320  1.18    mrg <dt><code>--htmldir=<var>dirname</var></code></dt>
    321  1.10    mrg <dd><p>Specify the installation directory for HTML documentation files.
    322   1.1    mrg The default is <samp><var>docdir</var></samp>.
    323  1.10    mrg </p>
    324  1.10    mrg </dd>
    325  1.18    mrg <dt><code>--pdfdir=<var>dirname</var></code></dt>
    326  1.10    mrg <dd><p>Specify the installation directory for PDF documentation files.
    327   1.1    mrg The default is <samp><var>docdir</var></samp>.
    328  1.10    mrg </p>
    329  1.10    mrg </dd>
    330  1.18    mrg <dt><code>--mandir=<var>dirname</var></code></dt>
    331  1.10    mrg <dd><p>Specify the installation directory for manual pages.  The default is
    332  1.10    mrg <samp><var>datarootdir</var>/man</samp>.  (Note that the manual pages are only extracts
    333   1.1    mrg from the full GCC manuals, which are provided in Texinfo format.  The manpages
    334   1.1    mrg are derived by an automatic conversion process from parts of the full
    335   1.1    mrg manual.)
    336  1.10    mrg </p>
    337  1.10    mrg </dd>
    338  1.18    mrg <dt><code>--with-gxx-include-dir=<var>dirname</var></code></dt>
    339  1.10    mrg <dd><p>Specify
    340   1.1    mrg the installation directory for G++ header files.  The default depends
    341   1.1    mrg on other configuration options, and differs between cross and native
    342   1.1    mrg configurations.
    343  1.10    mrg </p>
    344  1.10    mrg </dd>
    345  1.18    mrg <dt><code>--with-specs=<var>specs</var></code></dt>
    346  1.10    mrg <dd><p>Specify additional command line driver SPECS.
    347   1.3  skrll This can be useful if you need to turn on a non-standard feature by
    348  1.10    mrg default without modifying the compiler&rsquo;s source code, for instance
    349  1.10    mrg <samp>--with-specs=%{!fcommon:%{!fno-common:-fno-common}}</samp>.
    350   1.3  skrll See &ldquo;Spec Files&rdquo; in the main manual
    351  1.10    mrg </p>
    352  1.10    mrg </dd>
    353  1.10    mrg </dl>
    354   1.3  skrll 
    355  1.10    mrg </dd>
    356  1.18    mrg <dt><code>--program-prefix=<var>prefix</var></code></dt>
    357  1.10    mrg <dd><p>GCC supports some transformations of the names of its programs when
    358   1.1    mrg installing them.  This option prepends <var>prefix</var> to the names of
    359   1.1    mrg programs to install in <var>bindir</var> (see above).  For example, specifying
    360  1.10    mrg <samp>--program-prefix=foo-</samp> would result in &lsquo;<samp>gcc</samp>&rsquo;
    361  1.10    mrg being installed as <samp>/usr/local/bin/foo-gcc</samp>.
    362  1.10    mrg </p>
    363  1.10    mrg </dd>
    364  1.18    mrg <dt><code>--program-suffix=<var>suffix</var></code></dt>
    365  1.10    mrg <dd><p>Appends <var>suffix</var> to the names of programs to install in <var>bindir</var>
    366  1.10    mrg (see above).  For example, specifying <samp>--program-suffix=-3.1</samp>
    367  1.10    mrg would result in &lsquo;<samp>gcc</samp>&rsquo; being installed as
    368  1.10    mrg <samp>/usr/local/bin/gcc-3.1</samp>.
    369  1.10    mrg </p>
    370  1.10    mrg </dd>
    371  1.18    mrg <dt><code>--program-transform-name=<var>pattern</var></code></dt>
    372  1.10    mrg <dd><p>Applies the &lsquo;<samp>sed</samp>&rsquo; script <var>pattern</var> to be applied to the names
    373   1.1    mrg of programs to install in <var>bindir</var> (see above).  <var>pattern</var> has to
    374  1.10    mrg consist of one or more basic &lsquo;<samp>sed</samp>&rsquo; editing commands, separated by
    375  1.10    mrg semicolons.  For example, if you want the &lsquo;<samp>gcc</samp>&rsquo; program name to be
    376  1.10    mrg transformed to the installed program <samp>/usr/local/bin/myowngcc</samp> and
    377  1.10    mrg the &lsquo;<samp>g++</samp>&rsquo; program name to be transformed to
    378  1.10    mrg <samp>/usr/local/bin/gspecial++</samp> without changing other program names,
    379   1.1    mrg you could use the pattern
    380  1.10    mrg <samp>--program-transform-name='s/^gcc$/myowngcc/; s/^g++$/gspecial++/'</samp>
    381   1.1    mrg to achieve this effect.
    382  1.10    mrg </p>
    383  1.10    mrg <p>All three options can be combined and used together, resulting in more
    384   1.1    mrg complex conversion patterns.  As a basic rule, <var>prefix</var> (and
    385   1.1    mrg <var>suffix</var>) are prepended (appended) before further transformations
    386   1.1    mrg can happen with a special transformation script <var>pattern</var>.
    387  1.10    mrg </p>
    388  1.10    mrg <p>As currently implemented, this option only takes effect for native
    389  1.10    mrg builds; cross compiler binaries&rsquo; names are not transformed even when a
    390   1.1    mrg transformation is explicitly asked for by one of these options.
    391  1.10    mrg </p>
    392  1.10    mrg <p>For native builds, some of the installed programs are also installed
    393   1.1    mrg with the target alias in front of their name, as in
    394  1.10    mrg &lsquo;<samp>i686-pc-linux-gnu-gcc</samp>&rsquo;.  All of the above transformations happen
    395   1.1    mrg before the target alias is prepended to the name&mdash;so, specifying
    396  1.10    mrg <samp>--program-prefix=foo-</samp> and <samp>program-suffix=-3.1</samp>, the
    397   1.1    mrg resulting binary would be installed as
    398  1.10    mrg <samp>/usr/local/bin/i686-pc-linux-gnu-foo-gcc-3.1</samp>.
    399  1.10    mrg </p>
    400  1.10    mrg <p>As a last shortcoming, none of the installed Ada programs are
    401   1.1    mrg transformed yet, which will be fixed in some time.
    402  1.10    mrg </p>
    403  1.10    mrg </dd>
    404  1.18    mrg <dt><code>--with-local-prefix=<var>dirname</var></code></dt>
    405  1.10    mrg <dd><p>Specify the
    406   1.1    mrg installation directory for local include files.  The default is
    407  1.10    mrg <samp>/usr/local</samp>.  Specify this option if you want the compiler to
    408  1.10    mrg search directory <samp><var>dirname</var>/include</samp> for locally installed
    409  1.10    mrg header files <em>instead</em> of <samp>/usr/local/include</samp>.
    410  1.10    mrg </p>
    411  1.10    mrg <p>You should specify <samp>--with-local-prefix</samp> <strong>only</strong> if your
    412  1.10    mrg site has a different convention (not <samp>/usr/local</samp>) for where to put
    413   1.1    mrg site-specific files.
    414  1.10    mrg </p>
    415  1.10    mrg <p>The default value for <samp>--with-local-prefix</samp> is <samp>/usr/local</samp>
    416  1.10    mrg regardless of the value of <samp>--prefix</samp>.  Specifying
    417  1.10    mrg <samp>--prefix</samp> has no effect on which directory GCC searches for
    418   1.1    mrg local header files.  This may seem counterintuitive, but actually it is
    419   1.1    mrg logical.
    420  1.10    mrg </p>
    421  1.10    mrg <p>The purpose of <samp>--prefix</samp> is to specify where to <em>install
    422  1.10    mrg GCC</em>.  The local header files in <samp>/usr/local/include</samp>&mdash;if you put
    423   1.1    mrg any in that directory&mdash;are not part of GCC.  They are part of other
    424   1.1    mrg programs&mdash;perhaps many others.  (GCC installs its own header files in
    425  1.10    mrg another directory which is based on the <samp>--prefix</samp> value.)
    426  1.10    mrg </p>
    427  1.10    mrg <p>Both the local-prefix include directory and the GCC-prefix include
    428  1.10    mrg directory are part of GCC&rsquo;s &ldquo;system include&rdquo; directories.  Although these
    429   1.1    mrg two directories are not fixed, they need to be searched in the proper
    430   1.1    mrg order for the correct processing of the include_next directive.  The
    431   1.1    mrg local-prefix include directory is searched before the GCC-prefix
    432   1.1    mrg include directory.  Another characteristic of system include directories
    433   1.1    mrg is that pedantic warnings are turned off for headers in these directories.
    434  1.10    mrg </p>
    435  1.10    mrg <p>Some autoconf macros add <samp>-I <var>directory</var></samp> options to the
    436   1.1    mrg compiler command line, to ensure that directories containing installed
    437  1.10    mrg packages&rsquo; headers are searched.  When <var>directory</var> is one of GCC&rsquo;s
    438   1.1    mrg system include directories, GCC will ignore the option so that system
    439   1.1    mrg directories continue to be processed in the correct order.  This
    440   1.1    mrg may result in a search order different from what was specified but the
    441   1.1    mrg directory will still be searched.
    442  1.10    mrg </p>
    443  1.10    mrg <p>GCC automatically searches for ordinary libraries using
    444  1.10    mrg <code>GCC_EXEC_PREFIX</code>.  Thus, when the same installation prefix is
    445   1.1    mrg used for both GCC and packages, GCC will automatically search for
    446   1.1    mrg both headers and libraries.  This provides a configuration that is
    447   1.1    mrg easy to use.  GCC behaves in a manner similar to that when it is
    448  1.10    mrg installed as a system compiler in <samp>/usr</samp>.
    449  1.10    mrg </p>
    450  1.10    mrg <p>Sites that need to install multiple versions of GCC may not want to
    451   1.1    mrg use the above simple configuration.  It is possible to use the
    452  1.10    mrg <samp>--program-prefix</samp>, <samp>--program-suffix</samp> and
    453  1.10    mrg <samp>--program-transform-name</samp> options to install multiple versions
    454   1.1    mrg into a single directory, but it may be simpler to use different prefixes
    455  1.10    mrg and the <samp>--with-local-prefix</samp> option to specify the location of the
    456   1.1    mrg site-specific files for each version.  It will then be necessary for
    457   1.1    mrg users to specify explicitly the location of local site libraries
    458  1.10    mrg (e.g., with <code>LIBRARY_PATH</code>).
    459  1.10    mrg </p>
    460  1.10    mrg <p>The same value can be used for both <samp>--with-local-prefix</samp> and
    461  1.10    mrg <samp>--prefix</samp> provided it is not <samp>/usr</samp>.  This can be used
    462  1.10    mrg to avoid the default search of <samp>/usr/local/include</samp>.
    463  1.10    mrg </p>
    464  1.10    mrg <p><strong>Do not</strong> specify <samp>/usr</samp> as the <samp>--with-local-prefix</samp>!
    465  1.10    mrg The directory you use for <samp>--with-local-prefix</samp> <strong>must not</strong>
    466  1.10    mrg contain any of the system&rsquo;s standard header files.  If it did contain
    467   1.1    mrg them, certain programs would be miscompiled (including GNU Emacs, on
    468   1.1    mrg certain targets), because this would override and nullify the header
    469  1.10    mrg file corrections made by the <code>fixincludes</code> script.
    470  1.10    mrg </p>
    471  1.10    mrg <p>Indications are that people who use this option use it based on mistaken
    472   1.1    mrg ideas of what it is for.  People use it as if it specified where to
    473   1.1    mrg install part of GCC.  Perhaps they make this assumption because
    474   1.1    mrg installing GCC creates the directory.
    475  1.10    mrg </p>
    476  1.10    mrg </dd>
    477  1.18    mrg <dt><code>--with-gcc-major-version-only</code></dt>
    478  1.10    mrg <dd><p>Specifies that GCC should use only the major number rather than
    479   1.9    mrg <var>major</var>.<var>minor</var>.<var>patchlevel</var> in filesystem paths.
    480  1.10    mrg </p>
    481  1.10    mrg </dd>
    482  1.18    mrg <dt><code>--with-native-system-header-dir=<var>dirname</var></code></dt>
    483  1.10    mrg <dd><p>Specifies that <var>dirname</var> is the directory that contains native system
    484  1.10    mrg header files, rather than <samp>/usr/include</samp>.  This option is most useful
    485   1.3  skrll if you are creating a compiler that should be isolated from the system
    486   1.3  skrll as much as possible.  It is most commonly used with the
    487  1.10    mrg <samp>--with-sysroot</samp> option and will cause GCC to search
    488   1.3  skrll <var>dirname</var> inside the system root specified by that option.
    489  1.10    mrg </p>
    490  1.10    mrg </dd>
    491  1.18    mrg <dt><code>--enable-shared[=<var>package</var>[,&hellip;]]</code></dt>
    492  1.10    mrg <dd><p>Build shared versions of libraries, if shared libraries are supported on
    493   1.1    mrg the target platform.  Unlike GCC 2.95.x and earlier, shared libraries
    494   1.1    mrg are enabled by default on all platforms that support shared libraries.
    495  1.10    mrg </p>
    496  1.10    mrg <p>If a list of packages is given as an argument, build shared libraries
    497   1.1    mrg only for the listed packages.  For other packages, only static libraries
    498   1.1    mrg will be built.  Package names currently recognized in the GCC tree are
    499  1.10    mrg &lsquo;<samp>libgcc</samp>&rsquo; (also known as &lsquo;<samp>gcc</samp>&rsquo;), &lsquo;<samp>libstdc++</samp>&rsquo; (not
    500  1.10    mrg &lsquo;<samp>libstdc++-v3</samp>&rsquo;), &lsquo;<samp>libffi</samp>&rsquo;, &lsquo;<samp>zlib</samp>&rsquo;, &lsquo;<samp>boehm-gc</samp>&rsquo;,
    501  1.12    mrg &lsquo;<samp>ada</samp>&rsquo;, &lsquo;<samp>libada</samp>&rsquo;, &lsquo;<samp>libgo</samp>&rsquo;, &lsquo;<samp>libobjc</samp>&rsquo;, and &lsquo;<samp>libphobos</samp>&rsquo;.
    502  1.10    mrg Note &lsquo;<samp>libiberty</samp>&rsquo; does not support shared libraries at all.
    503  1.10    mrg </p>
    504  1.10    mrg <p>Use <samp>--disable-shared</samp> to build only static libraries.  Note that
    505  1.10    mrg <samp>--disable-shared</samp> does not accept a list of package names as
    506  1.10    mrg argument, only <samp>--enable-shared</samp> does.
    507  1.10    mrg </p>
    508  1.10    mrg <p>Contrast with <samp>--enable-host-shared</samp>, which affects <em>host</em>
    509   1.5    mrg code.
    510  1.10    mrg </p>
    511  1.10    mrg </dd>
    512  1.18    mrg <dt><code>--enable-host-shared</code></dt>
    513  1.10    mrg <dd><p>Specify that the <em>host</em> code should be built into position-independent
    514  1.19    mrg machine code (with <samp>-fPIC</samp>), allowing it to be used within shared
    515  1.19    mrg libraries, but yielding a slightly slower compiler.
    516  1.10    mrg </p>
    517  1.10    mrg <p>This option is required when building the libgccjit.so library.
    518  1.10    mrg </p>
    519  1.10    mrg <p>Contrast with <samp>--enable-shared</samp>, which affects <em>target</em>
    520   1.5    mrg libraries.
    521  1.10    mrg </p>
    522  1.10    mrg </dd>
    523  1.19    mrg <dt><code>--enable-host-pie</code></dt>
    524  1.19    mrg <dd><p>Specify that the <em>host</em> executables should be built into
    525  1.19    mrg position-independent executables (with <samp>-fPIE</samp> and <samp>-pie</samp>),
    526  1.19    mrg yielding a slightly slower compiler (but faster than
    527  1.19    mrg <samp>--enable-host-shared</samp>).  Position-independent executables are loaded
    528  1.19    mrg at random addresses each time they are executed, therefore provide additional
    529  1.19    mrg protection against Return Oriented Programming (ROP) attacks.
    530  1.19    mrg </p>
    531  1.19    mrg <p><samp>--enable-host-pie</samp> may be used with <samp>--enable-host-shared</samp>,
    532  1.19    mrg in which case <samp>-fPIC</samp> is used when compiling, and <samp>-pie</samp> when
    533  1.19    mrg linking.
    534  1.19    mrg </p>
    535  1.19    mrg </dd>
    536  1.19    mrg <dt><code>--enable-host-bind-now</code></dt>
    537  1.19    mrg <dd><p>Specify that the <em>host</em> executables should be linked with the option
    538  1.19    mrg <samp>-Wl,-z,now</samp>, which means that the dynamic linker will resolve all
    539  1.19    mrg symbols when the executables are started, and that in turn allows RELRO to
    540  1.19    mrg mark the GOT read-only, resulting in better security.
    541  1.19    mrg </p>
    542  1.19    mrg </dd>
    543  1.18    mrg <dt><code><a name="with-gnu-as"></a>--with-gnu-as</code></dt>
    544  1.10    mrg <dd><p>Specify that the compiler should assume that the
    545   1.1    mrg assembler it finds is the GNU assembler.  However, this does not modify
    546   1.1    mrg the rules to find an assembler and will result in confusion if the
    547   1.1    mrg assembler found is not actually the GNU assembler.  (Confusion may also
    548   1.1    mrg result if the compiler finds the GNU assembler but has not been
    549  1.10    mrg configured with <samp>--with-gnu-as</samp>.)  If you have more than one
    550   1.1    mrg assembler installed on your system, you may want to use this option in
    551  1.10    mrg connection with <samp>--with-as=<var>pathname</var></samp> or
    552  1.10    mrg <samp>--with-build-time-tools=<var>pathname</var></samp>.
    553  1.10    mrg </p>
    554  1.10    mrg <p>The following systems are the only ones where it makes a difference
    555   1.1    mrg whether you use the GNU assembler.  On any other system,
    556  1.10    mrg <samp>--with-gnu-as</samp> has no effect.
    557  1.10    mrg </p>
    558  1.10    mrg <ul>
    559  1.10    mrg <li> &lsquo;<samp>hppa1.0-<var>any</var>-<var>any</var></samp>&rsquo;
    560  1.10    mrg </li><li> &lsquo;<samp>hppa1.1-<var>any</var>-<var>any</var></samp>&rsquo;
    561  1.19    mrg </li><li> &lsquo;<samp>*-*-solaris2.11</samp>&rsquo;
    562  1.10    mrg </li></ul>
    563  1.10    mrg 
    564  1.10    mrg </dd>
    565  1.18    mrg <dt><code><a name="with-as"></a>--with-as=<var>pathname</var></code></dt>
    566  1.10    mrg <dd><p>Specify that the compiler should use the assembler pointed to by
    567   1.1    mrg <var>pathname</var>, rather than the one found by the standard rules to find
    568   1.1    mrg an assembler, which are:
    569  1.10    mrg </p><ul>
    570  1.10    mrg <li> Unless GCC is being built with a cross compiler, check the
    571  1.10    mrg <samp><var>libexec</var>/gcc/<var>target</var>/<var>version</var></samp> directory.
    572  1.10    mrg <var>libexec</var> defaults to <samp><var>exec-prefix</var>/libexec</samp>;
    573   1.1    mrg <var>exec-prefix</var> defaults to <var>prefix</var>, which
    574  1.10    mrg defaults to <samp>/usr/local</samp> unless overridden by the
    575  1.10    mrg <samp>--prefix=<var>pathname</var></samp> switch described above.  <var>target</var>
    576  1.19    mrg is the target system triple, such as &lsquo;<samp>sparc-sun-solaris2.11</samp>&rsquo;, and
    577   1.1    mrg <var>version</var> denotes the GCC version, such as 3.0.
    578   1.1    mrg 
    579  1.10    mrg </li><li> If the target system is the same that you are building on, check
    580  1.19    mrg operating system specific directories.
    581   1.1    mrg 
    582  1.10    mrg </li><li> Check in the <code>PATH</code> for a tool whose name is prefixed by the
    583   1.1    mrg target system triple.
    584   1.1    mrg 
    585  1.10    mrg </li><li> Check in the <code>PATH</code> for a tool whose name is not prefixed by the
    586   1.1    mrg target system triple, if the host and target system triple are
    587   1.1    mrg the same (in other words, we use a host tool if it can be used for
    588  1.10    mrg the target as well).
    589  1.10    mrg </li></ul>
    590   1.1    mrg 
    591  1.10    mrg <p>You may want to use <samp>--with-as</samp> if no assembler
    592   1.1    mrg is installed in the directories listed above, or if you have multiple
    593   1.1    mrg assemblers installed and want to choose one that is not found by the
    594   1.1    mrg above rules.
    595  1.10    mrg </p>
    596  1.10    mrg </dd>
    597  1.18    mrg <dt><code><a name="with-gnu-ld"></a>--with-gnu-ld</code></dt>
    598  1.10    mrg <dd><p>Same as <a href="#with-gnu-as"><samp>--with-gnu-as</samp></a>
    599   1.1    mrg but for the linker.
    600  1.10    mrg </p>
    601  1.10    mrg </dd>
    602  1.18    mrg <dt><code>--with-ld=<var>pathname</var></code></dt>
    603  1.10    mrg <dd><p>Same as <a href="#with-as"><samp>--with-as</samp></a>
    604   1.1    mrg but for the linker.
    605  1.10    mrg </p>
    606  1.10    mrg </dd>
    607  1.18    mrg <dt><code>--with-dsymutil=<var>pathname</var></code></dt>
    608  1.14    mrg <dd><p>Same as <a href="#with-as"><samp>--with-as</samp></a>
    609  1.14    mrg but for the debug linker (only used on Darwin platforms so far).
    610  1.14    mrg </p>
    611  1.14    mrg </dd>
    612  1.18    mrg <dt><code>--with-tls=<var>dialect</var></code></dt>
    613  1.10    mrg <dd><p>Specify the default TLS dialect, for systems were there is a choice.
    614   1.3  skrll For ARM targets, possible values for <var>dialect</var> are <code>gnu</code> or
    615   1.3  skrll <code>gnu2</code>, which select between the original GNU dialect and the GNU TLS
    616   1.3  skrll descriptor-based dialect.
    617  1.19    mrg For RISC-V targets, possible values for <var>dialect</var> are <code>trad</code> or
    618  1.19    mrg <code>desc</code>, which select between the traditional GNU dialect and the GNU TLS
    619  1.19    mrg descriptor-based dialect.
    620  1.10    mrg </p>
    621  1.10    mrg </dd>
    622  1.18    mrg <dt><code>--enable-multiarch</code></dt>
    623  1.10    mrg <dd><p>Specify whether to enable or disable multiarch support.  The default is
    624   1.3  skrll to check for glibc start files in a multiarch location, and enable it
    625   1.3  skrll if the files are found.  The auto detection is enabled for native builds,
    626  1.10    mrg and for cross builds configured with <samp>--with-sysroot</samp>, and without
    627  1.10    mrg <samp>--with-native-system-header-dir</samp>.
    628   1.3  skrll More documentation about multiarch can be found at
    629   1.7    mrg <a href="https://wiki.debian.org/Multiarch">https://wiki.debian.org/Multiarch</a>.
    630  1.10    mrg </p>
    631  1.10    mrg </dd>
    632  1.18    mrg <dt><code>--enable-sjlj-exceptions</code></dt>
    633  1.10    mrg <dd><p>Force use of the <code>setjmp</code>/<code>longjmp</code>-based scheme for exceptions.
    634  1.10    mrg &lsquo;<samp>configure</samp>&rsquo; ordinarily picks the correct value based on the platform.
    635   1.6    mrg Only use this option if you are sure you need a different setting.
    636  1.10    mrg </p>
    637  1.10    mrg </dd>
    638  1.18    mrg <dt><code>--enable-vtable-verify</code></dt>
    639  1.10    mrg <dd><p>Specify whether to enable or disable the vtable verification feature.
    640   1.5    mrg Enabling this feature causes libstdc++ to be built with its virtual calls
    641   1.5    mrg in verifiable mode.  This means that, when linked with libvtv, every
    642   1.5    mrg virtual call in libstdc++ will verify the vtable pointer through which the
    643   1.5    mrg call will be made before actually making the call.  If not linked with libvtv,
    644  1.10    mrg the verifier will call stub functions (in libstdc++ itself) and do nothing.
    645   1.5    mrg If vtable verification is disabled, then libstdc++ is not built with its
    646   1.5    mrg virtual calls in verifiable mode at all.  However the libvtv library will
    647  1.10    mrg still be built (see <samp>--disable-libvtv</samp> to turn off building libvtv).
    648  1.10    mrg <samp>--disable-vtable-verify</samp> is the default.
    649  1.10    mrg </p>
    650  1.10    mrg </dd>
    651  1.18    mrg <dt><code>--disable-gcov</code></dt>
    652  1.12    mrg <dd><p>Specify that the run-time library used for coverage analysis
    653  1.12    mrg and associated host tools should not be built.
    654  1.12    mrg </p>
    655  1.12    mrg </dd>
    656  1.18    mrg <dt><code>--disable-multilib</code></dt>
    657  1.10    mrg <dd><p>Specify that multiple target
    658   1.1    mrg libraries to support different target variants, calling
    659   1.1    mrg conventions, etc. should not be built.  The default is to build a
    660   1.1    mrg predefined set of them.
    661  1.10    mrg </p>
    662  1.10    mrg <p>Some targets provide finer-grained control over which multilibs are built
    663  1.10    mrg (e.g., <samp>--disable-softfloat</samp>):
    664  1.10    mrg </p><dl compact="compact">
    665  1.18    mrg <dt><code>arm-*-*</code></dt>
    666  1.10    mrg <dd><p>fpu, 26bit, underscore, interwork, biendian, nofmult.
    667  1.10    mrg </p>
    668  1.10    mrg </dd>
    669  1.18    mrg <dt><code>m68*-*-*</code></dt>
    670  1.10    mrg <dd><p>softfloat, m68881, m68000, m68020.
    671  1.10    mrg </p>
    672  1.10    mrg </dd>
    673  1.18    mrg <dt><code>mips*-*-*</code></dt>
    674  1.10    mrg <dd><p>single-float, biendian, softfloat.
    675  1.10    mrg </p>
    676  1.10    mrg </dd>
    677  1.18    mrg <dt><code>msp430-*-*</code></dt>
    678  1.13    mrg <dd><p>no-exceptions
    679  1.13    mrg </p>
    680  1.13    mrg </dd>
    681  1.18    mrg <dt><code>powerpc*-*-*, rs6000*-*-*</code></dt>
    682  1.10    mrg <dd><p>aix64, pthread, softfloat, powercpu, powerpccpu, powerpcos, biendian,
    683   1.1    mrg sysv, aix.
    684  1.10    mrg </p>
    685  1.10    mrg </dd>
    686  1.10    mrg </dl>
    687   1.1    mrg 
    688  1.10    mrg </dd>
    689  1.18    mrg <dt><code>--with-multilib-list=<var>list</var></code></dt>
    690  1.18    mrg <dt><code>--without-multilib-list</code></dt>
    691  1.10    mrg <dd><p>Specify what multilibs to build.  <var>list</var> is a comma separated list of
    692   1.9    mrg values, possibly consisting of a single value.  Currently only implemented
    693  1.19    mrg for aarch64*-*-*, amdgcn*-*-*, arm*-*-*, loongarch*-*-*, riscv*-*-*, sh*-*-*
    694  1.19    mrg and x86-64-*-linux*.  The accepted values and meaning for each target is given
    695  1.16    mrg below.
    696  1.10    mrg </p>
    697  1.10    mrg <dl compact="compact">
    698  1.18    mrg <dt><code>aarch64*-*-*</code></dt>
    699  1.10    mrg <dd><p><var>list</var> is a comma separated list of <code>ilp32</code>, and <code>lp64</code>
    700  1.10    mrg to enable ILP32 and LP64 run-time libraries, respectively.  If
    701  1.10    mrg <var>list</var> is empty, then there will be no multilibs and only the
    702  1.10    mrg default run-time library will be built.  If <var>list</var> is
    703  1.10    mrg <code>default</code> or &ndash;with-multilib-list= is not specified, then the
    704  1.10    mrg default set of libraries is selected based on the value of
    705  1.10    mrg <samp>--target</samp>.
    706  1.10    mrg </p>
    707  1.10    mrg </dd>
    708  1.19    mrg <dt><code>amdgcn*-*-*</code></dt>
    709  1.19    mrg <dd><p><var>list</var> is a comma separated list of ISA names (allowed values: <code>fiji</code>,
    710  1.19    mrg <code>gfx900</code>, <code>gfx906</code>, <code>gfx908</code>, <code>gfx90a</code>, <code>gfx90c</code>, 
    711  1.19    mrg <code>gfx1030</code>, <code>gfx1036</code>, <code>gfx1100</code>, <code>gfx1103</code>).
    712  1.19    mrg It ought not include the name of the default
    713  1.19    mrg ISA, specified via <samp>--with-arch</samp>.  If <var>list</var> is empty, then there
    714  1.19    mrg will be no multilibs and only the default run-time library will be built.  If
    715  1.19    mrg <var>list</var> is <code>default</code> or <samp>--with-multilib-list=</samp> is not
    716  1.19    mrg specified, then the default set of libraries is selected.
    717  1.19    mrg </p>
    718  1.19    mrg </dd>
    719  1.18    mrg <dt><code>arm*-*-*</code></dt>
    720  1.12    mrg <dd><p><var>list</var> is a comma separated list of <code>aprofile</code> and
    721  1.12    mrg <code>rmprofile</code> to build multilibs for A or R and M architecture
    722  1.12    mrg profiles respectively.  Note that, due to some limitation of the current
    723  1.12    mrg multilib framework, using the combined <code>aprofile,rmprofile</code>
    724  1.12    mrg multilibs selects in some cases a less optimal multilib than when using
    725  1.12    mrg the multilib profile for the architecture targetted.  The special value
    726  1.12    mrg <code>default</code> is also accepted and is equivalent to omitting the
    727  1.12    mrg option, i.e., only the default run-time library will be enabled.
    728  1.12    mrg </p>
    729  1.12    mrg <p><var>list</var> may instead contain <code>@name</code>, to use the multilib
    730  1.12    mrg configuration Makefile fragment <samp>name</samp> in <samp>gcc/config/arm</samp> in
    731  1.12    mrg the source tree (it is part of the corresponding sources, after all).
    732  1.12    mrg It is recommended, but not required, that files used for this purpose to
    733  1.12    mrg be named starting with <samp>t-ml-</samp>, to make their intended purpose
    734  1.12    mrg self-evident, in line with GCC conventions.  Such files enable custom,
    735  1.12    mrg user-chosen multilib lists to be configured.  Whether multiple such
    736  1.12    mrg files can be used together depends on the contents of the supplied
    737  1.12    mrg files.  See <samp>gcc/config/arm/t-multilib</samp> and its supplementary
    738  1.12    mrg <samp>gcc/config/arm/t-*profile</samp> files for an example of what such
    739  1.12    mrg Makefile fragments might look like for this version of GCC.  The macros
    740  1.12    mrg expected to be defined in these fragments are not stable across GCC
    741  1.12    mrg releases, so make sure they define the <code>MULTILIB</code>-related macros
    742  1.12    mrg expected by the version of GCC you are building.
    743  1.12    mrg See &ldquo;Target Makefile Fragments&rdquo; in the internals manual.
    744  1.10    mrg </p>
    745  1.10    mrg <p>The table below gives the combination of ISAs, architectures, FPUs and
    746  1.12    mrg floating-point ABIs for which multilibs are built for each predefined
    747  1.12    mrg profile.  The union of these options is considered when specifying both
    748  1.12    mrg <code>aprofile</code> and <code>rmprofile</code>.
    749  1.10    mrg </p>
    750  1.10    mrg <table>
    751  1.10    mrg <tr><td width="15%">Option</td><td width="28%">aprofile</td><td width="30%">rmprofile</td></tr>
    752  1.10    mrg <tr><td width="15%">ISAs</td><td width="28%"><code>-marm</code> and <code>-mthumb</code></td><td width="30%"><code>-mthumb</code></td></tr>
    753  1.10    mrg <tr><td width="15%">Architectures<br><br><br><br><br><br></td><td width="28%">default architecture<br>
    754   1.9    mrg <code>-march=armv7-a</code><br>
    755   1.9    mrg <code>-march=armv7ve</code><br>
    756  1.10    mrg <code>-march=armv8-a</code><br><br><br></td><td width="30%">default architecture<br>
    757   1.9    mrg <code>-march=armv6s-m</code><br>
    758   1.9    mrg <code>-march=armv7-m</code><br>
    759   1.9    mrg <code>-march=armv7e-m</code><br>
    760   1.9    mrg <code>-march=armv8-m.base</code><br>
    761   1.9    mrg <code>-march=armv8-m.main</code><br>
    762  1.10    mrg <code>-march=armv7</code></td></tr>
    763  1.10    mrg <tr><td width="15%">FPUs<br><br><br><br><br></td><td width="28%">none<br>
    764   1.9    mrg <code>-mfpu=vfpv3-d16</code><br>
    765   1.9    mrg <code>-mfpu=neon</code><br>
    766   1.9    mrg <code>-mfpu=vfpv4-d16</code><br>
    767   1.9    mrg <code>-mfpu=neon-vfpv4</code><br>
    768  1.10    mrg <code>-mfpu=neon-fp-armv8</code></td><td width="30%">none<br>
    769   1.9    mrg <code>-mfpu=vfpv3-d16</code><br>
    770   1.9    mrg <code>-mfpu=fpv4-sp-d16</code><br>
    771   1.9    mrg <code>-mfpu=fpv5-sp-d16</code><br>
    772  1.10    mrg <code>-mfpu=fpv5-d16</code><br></td></tr>
    773  1.10    mrg <tr><td width="15%">floating-point ABIs<br><br></td><td width="28%"><code>-mfloat-abi=soft</code><br>
    774   1.9    mrg <code>-mfloat-abi=softfp</code><br>
    775  1.10    mrg <code>-mfloat-abi=hard</code></td><td width="30%"><code>-mfloat-abi=soft</code><br>
    776   1.9    mrg <code>-mfloat-abi=softfp</code><br>
    777  1.10    mrg <code>-mfloat-abi=hard</code></td></tr>
    778  1.10    mrg </table>
    779   1.9    mrg 
    780  1.10    mrg </dd>
    781  1.18    mrg <dt><code>loongarch*-*-*</code></dt>
    782  1.19    mrg <dd><p><var>list</var> is a comma-separated list, with each of the element starting with
    783  1.19    mrg the following ABI identifiers: <code>lp64d[/base]</code> <code>lp64f[/base]</code>
    784  1.19    mrg <code>lp64d[/base]</code> (the <code>/base</code> suffix may be omitted)
    785  1.19    mrg to enable their respective run-time libraries.
    786  1.19    mrg </p>
    787  1.19    mrg <p>A suffix <code>[/<var>arch</var>][/<var>option</var>/&hellip;]</code> may follow immediately
    788  1.19    mrg after the ABI identifier to customize the compiler options for building the
    789  1.19    mrg given set of libraries.  <var>arch</var> denotes the architecture name recognized
    790  1.19    mrg by the <samp>-march=<var>arch</var></samp> compiler option, which acts as a basic target
    791  1.19    mrg ISA configuration that can be adjusted using the subsequent <var>option</var>
    792  1.19    mrg suffixes, where each <var>option</var> is a compiler option without a leading dash
    793  1.19    mrg (&rsquo;-&rsquo;).
    794  1.19    mrg </p>
    795  1.19    mrg <p>If no such suffix is present for a given multilib variant, the
    796  1.19    mrg configured value of <samp>--with-multilib-default</samp> is appended as a default
    797  1.19    mrg suffix.  If <samp>--with-multilib-default</samp> is not given, the default build
    798  1.19    mrg option <samp>-march=abi-default</samp> is applied when building the variants
    799  1.19    mrg without a suffix.
    800  1.19    mrg </p>
    801  1.19    mrg <p>As a special case, <code>fixed</code> may be used in the position of <var>arch</var>,
    802  1.19    mrg which means using the architecture configured with
    803  1.19    mrg <samp>--with-arch=<var>arch</var></samp>, or its default value (e.g. <code>loongarch64</code>
    804  1.19    mrg for <code>loongarch64-*</code> targets).
    805  1.19    mrg </p>
    806  1.19    mrg <p>If <var>list</var> is empty or <code>default</code>, or if <samp>--with-multilib-list</samp>
    807  1.19    mrg is not specified, then only the default variant of the libraries are built,
    808  1.19    mrg where the default ABI is implied by the configured target triplet.
    809  1.16    mrg </p>
    810  1.16    mrg </dd>
    811  1.18    mrg <dt><code>riscv*-*-*</code></dt>
    812  1.12    mrg <dd><p><var>list</var> is a single ABI name.  The target architecture must be either
    813  1.12    mrg <code>rv32gc</code> or <code>rv64gc</code>.  This will build a single multilib for the
    814  1.12    mrg specified architecture and ABI pair.  If <code>--with-multilib-list</code> is not
    815  1.12    mrg given, then a default set of multilibs is selected based on the value of
    816  1.12    mrg <samp>--target</samp>.  This is usually a large set of multilibs.
    817  1.12    mrg </p>
    818  1.12    mrg </dd>
    819  1.18    mrg <dt><code>sh*-*-*</code></dt>
    820  1.10    mrg <dd><p><var>list</var> is a comma separated list of CPU names.  These must be of the
    821   1.1    mrg form <code>sh*</code> or <code>m*</code> (in which case they match the compiler option
    822   1.1    mrg for that processor).  The list should not contain any endian options -
    823  1.10    mrg these are handled by <samp>--with-endian</samp>.
    824  1.10    mrg </p>
    825  1.10    mrg <p>If <var>list</var> is empty, then there will be no multilibs for extra
    826   1.1    mrg processors.  The multilib for the secondary endian remains enabled.
    827  1.10    mrg </p>
    828  1.10    mrg <p>As a special case, if an entry in the list starts with a <code>!</code>
    829  1.10    mrg (exclamation point), then it is added to the list of excluded multilibs.
    830  1.10    mrg Entries of this sort should be compatible with &lsquo;<samp>MULTILIB_EXCLUDES</samp>&rsquo;
    831   1.1    mrg (once the leading <code>!</code> has been stripped).
    832  1.10    mrg </p>
    833  1.10    mrg <p>If <samp>--with-multilib-list</samp> is not given, then a default set of
    834  1.10    mrg multilibs is selected based on the value of <samp>--target</samp>.  This is
    835   1.1    mrg usually the complete set of libraries, but some targets imply a more
    836   1.1    mrg specialized subset.
    837  1.10    mrg </p>
    838  1.10    mrg <p>Example 1: to configure a compiler for SH4A only, but supporting both
    839  1.10    mrg endians, with little endian being the default:
    840  1.18    mrg </p><div class="smallexample">
    841  1.18    mrg <pre class="smallexample">--with-cpu=sh4a --with-endian=little,big --with-multilib-list=
    842  1.10    mrg </pre></div>
    843   1.9    mrg 
    844  1.10    mrg <p>Example 2: to configure a compiler for both SH4A and SH4AL-DSP, but with
    845   1.1    mrg only little endian SH4AL:
    846  1.18    mrg </p><div class="smallexample">
    847  1.18    mrg <pre class="smallexample">--with-cpu=sh4a --with-endian=little,big \
    848  1.10    mrg --with-multilib-list=sh4al,!mb/m4al
    849  1.10    mrg </pre></div>
    850  1.10    mrg 
    851  1.10    mrg </dd>
    852  1.18    mrg <dt><code>x86-64-*-linux*</code></dt>
    853  1.10    mrg <dd><p><var>list</var> is a comma separated list of <code>m32</code>, <code>m64</code> and
    854   1.3  skrll <code>mx32</code> to enable 32-bit, 64-bit and x32 run-time libraries,
    855   1.3  skrll respectively.  If <var>list</var> is empty, then there will be no multilibs
    856   1.3  skrll and only the default run-time library will be enabled.
    857  1.10    mrg </p>
    858  1.10    mrg <p>If <samp>--with-multilib-list</samp> is not given, then only 32-bit and
    859  1.10    mrg 64-bit run-time libraries will be enabled.
    860  1.10    mrg </p></dd>
    861   1.3  skrll </dl>
    862   1.3  skrll 
    863  1.10    mrg </dd>
    864  1.19    mrg <dt><code>--with-multilib-default</code></dt>
    865  1.19    mrg <dd><p>On LoongArch targets, set the default build options for enabled multilibs
    866  1.19    mrg without build options appended to their corresponding
    867  1.19    mrg <samp>--with-multilib-list</samp> items.  The format of this value is
    868  1.19    mrg <code>[/<var>arch</var>][/<var>option</var>/&hellip;]</code>, where <var>arch</var> is an
    869  1.19    mrg architecture name recognized by <samp>-march=<var>arch</var></samp> compiler option,
    870  1.19    mrg and subsequent <var>option</var> suffixes are compiler options minus a leading
    871  1.19    mrg dash (&rsquo;-&rsquo;).
    872  1.19    mrg </p>
    873  1.19    mrg <p>Multiple <var>option</var>s may appear consecutively while <var>arch</var> may only
    874  1.19    mrg appear in the beginning or be omitted (which means <samp>-march=abi-default</samp>
    875  1.19    mrg is applied when building the libraries).
    876  1.19    mrg </p>
    877  1.19    mrg </dd>
    878  1.19    mrg <dt><code>--with-strict-align-lib</code></dt>
    879  1.19    mrg <dd><p>On LoongArch targets, build all enabled multilibs with <samp>-mstrict-align</samp>
    880  1.19    mrg (Not enabled by default).
    881  1.19    mrg </p>
    882  1.19    mrg </dd>
    883  1.18    mrg <dt><code>--with-multilib-generator=<var>config</var></code></dt>
    884  1.16    mrg <dd><p>Specify what multilibs to build.  <var>config</var> is a semicolon separated list of
    885  1.16    mrg values, possibly consisting of a single value.  Currently only implemented
    886  1.16    mrg for riscv*-*-elf*.  The accepted values and meanings are given below.
    887  1.16    mrg </p>
    888  1.16    mrg 
    889  1.16    mrg <p>Every config is constructed with four components: architecture string, ABI,
    890  1.16    mrg reuse rule with architecture string and reuse rule with sub-extension.
    891  1.16    mrg </p>
    892  1.16    mrg <p>Example 1: Add multi-lib suppport for rv32i with ilp32.
    893  1.18    mrg </p><div class="smallexample">
    894  1.18    mrg <pre class="smallexample">rv32i-ilp32--
    895  1.16    mrg </pre></div>
    896  1.16    mrg 
    897  1.16    mrg <p>Example 2: Add multi-lib suppport for rv32i with ilp32 and rv32imafd with ilp32.
    898  1.18    mrg </p><div class="smallexample">
    899  1.18    mrg <pre class="smallexample">rv32i-ilp32--;rv32imafd-ilp32--
    900  1.16    mrg </pre></div>
    901  1.16    mrg 
    902  1.16    mrg <p>Example 3: Add multi-lib suppport for rv32i with ilp32; rv32im with ilp32 and
    903  1.16    mrg rv32ic with ilp32 will reuse this multi-lib set.
    904  1.18    mrg </p><div class="smallexample">
    905  1.18    mrg <pre class="smallexample">rv32i-ilp32-rv32im-c
    906  1.16    mrg </pre></div>
    907  1.16    mrg 
    908  1.16    mrg <p>Example 4: Add multi-lib suppport for rv64ima with lp64; rv64imaf with lp64,
    909  1.16    mrg rv64imac with lp64 and rv64imafc with lp64 will reuse this multi-lib set.
    910  1.18    mrg </p><div class="smallexample">
    911  1.18    mrg <pre class="smallexample">rv64ima-lp64--f,c,fc
    912  1.16    mrg </pre></div>
    913  1.16    mrg 
    914  1.16    mrg <p><samp>--with-multilib-generator</samp> have an optional configuration argument
    915  1.16    mrg <samp>--cmodel=val</samp> for code model, this option will expand with other
    916  1.16    mrg config options, <var>val</var> is a comma separated list of possible code model,
    917  1.16    mrg currently we support medlow and medany.
    918  1.16    mrg </p>
    919  1.16    mrg <p>Example 5: Add multi-lib suppport for rv64ima with lp64; rv64ima with lp64 and
    920  1.16    mrg medlow code model
    921  1.18    mrg </p><div class="smallexample">
    922  1.18    mrg <pre class="smallexample">rv64ima-lp64--;--cmodel=medlow
    923  1.16    mrg </pre></div>
    924  1.16    mrg 
    925  1.16    mrg <p>Example 6: Add multi-lib suppport for rv64ima with lp64; rv64ima with lp64 and
    926  1.16    mrg medlow code model; rv64ima with lp64 and medany code model
    927  1.18    mrg </p><div class="smallexample">
    928  1.18    mrg <pre class="smallexample">rv64ima-lp64--;--cmodel=medlow,medany
    929  1.16    mrg </pre></div>
    930  1.16    mrg 
    931  1.16    mrg </dd>
    932  1.18    mrg <dt><code>--with-endian=<var>endians</var></code></dt>
    933  1.10    mrg <dd><p>Specify what endians to use.
    934   1.1    mrg Currently only implemented for sh*-*-*.
    935  1.10    mrg </p>
    936  1.10    mrg <p><var>endians</var> may be one of the following:
    937  1.10    mrg </p><dl compact="compact">
    938  1.18    mrg <dt><code>big</code></dt>
    939  1.10    mrg <dd><p>Use big endian exclusively.
    940  1.10    mrg </p></dd>
    941  1.18    mrg <dt><code>little</code></dt>
    942  1.10    mrg <dd><p>Use little endian exclusively.
    943  1.10    mrg </p></dd>
    944  1.18    mrg <dt><code>big,little</code></dt>
    945  1.10    mrg <dd><p>Use big endian by default.  Provide a multilib for little endian.
    946  1.10    mrg </p></dd>
    947  1.18    mrg <dt><code>little,big</code></dt>
    948  1.10    mrg <dd><p>Use little endian by default.  Provide a multilib for big endian.
    949  1.10    mrg </p></dd>
    950   1.1    mrg </dl>
    951   1.1    mrg 
    952  1.10    mrg </dd>
    953  1.18    mrg <dt><code>--enable-threads</code></dt>
    954  1.10    mrg <dd><p>Specify that the target
    955   1.1    mrg supports threads.  This affects the Objective-C compiler and runtime
    956  1.10    mrg library, and exception handling for other languages like C++.
    957   1.1    mrg On some systems, this is the default.
    958  1.10    mrg </p>
    959  1.10    mrg <p>In general, the best (and, in many cases, the only known) threading
    960   1.1    mrg model available will be configured for use.  Beware that on some
    961   1.1    mrg systems, GCC has not been taught what threading models are generally
    962  1.10    mrg available for the system.  In this case, <samp>--enable-threads</samp> is an
    963  1.10    mrg alias for <samp>--enable-threads=single</samp>.
    964  1.10    mrg </p>
    965  1.10    mrg </dd>
    966  1.18    mrg <dt><code>--disable-threads</code></dt>
    967  1.10    mrg <dd><p>Specify that threading support should be disabled for the system.
    968  1.10    mrg This is an alias for <samp>--enable-threads=single</samp>.
    969  1.10    mrg </p>
    970  1.10    mrg </dd>
    971  1.18    mrg <dt><code>--enable-threads=<var>lib</var></code></dt>
    972  1.10    mrg <dd><p>Specify that
    973   1.1    mrg <var>lib</var> is the thread support library.  This affects the Objective-C
    974   1.1    mrg compiler and runtime library, and exception handling for other languages
    975   1.9    mrg like C++.  The possibilities for <var>lib</var> are:
    976  1.10    mrg </p>
    977  1.10    mrg <dl compact="compact">
    978  1.18    mrg <dt><code>aix</code></dt>
    979  1.10    mrg <dd><p>AIX thread support.
    980  1.10    mrg </p></dd>
    981  1.18    mrg <dt><code>dce</code></dt>
    982  1.10    mrg <dd><p>DCE thread support.
    983  1.10    mrg </p></dd>
    984  1.18    mrg <dt><code>lynx</code></dt>
    985  1.10    mrg <dd><p>LynxOS thread support.
    986  1.10    mrg </p></dd>
    987  1.18    mrg <dt><code>mipssde</code></dt>
    988  1.10    mrg <dd><p>MIPS SDE thread support.
    989  1.10    mrg </p></dd>
    990  1.18    mrg <dt><code>no</code></dt>
    991  1.10    mrg <dd><p>This is an alias for &lsquo;<samp>single</samp>&rsquo;.
    992  1.10    mrg </p></dd>
    993  1.18    mrg <dt><code>posix</code></dt>
    994  1.10    mrg <dd><p>Generic POSIX/Unix98 thread support.
    995  1.10    mrg </p></dd>
    996  1.18    mrg <dt><code>rtems</code></dt>
    997  1.10    mrg <dd><p>RTEMS thread support.
    998  1.10    mrg </p></dd>
    999  1.18    mrg <dt><code>single</code></dt>
   1000  1.10    mrg <dd><p>Disable thread support, should work for all platforms.
   1001  1.10    mrg </p></dd>
   1002  1.18    mrg <dt><code>tpf</code></dt>
   1003  1.10    mrg <dd><p>TPF thread support.
   1004  1.10    mrg </p></dd>
   1005  1.18    mrg <dt><code>vxworks</code></dt>
   1006  1.10    mrg <dd><p>VxWorks thread support.
   1007  1.10    mrg </p></dd>
   1008  1.18    mrg <dt><code>win32</code></dt>
   1009  1.10    mrg <dd><p>Microsoft Win32 API thread support.
   1010  1.10    mrg </p></dd>
   1011   1.1    mrg </dl>
   1012   1.1    mrg 
   1013  1.10    mrg </dd>
   1014  1.18    mrg <dt><code>--enable-tls</code></dt>
   1015  1.10    mrg <dd><p>Specify that the target supports TLS (Thread Local Storage).  Usually
   1016   1.1    mrg configure can correctly determine if TLS is supported.  In cases where
   1017   1.1    mrg it guesses incorrectly, TLS can be explicitly enabled or disabled with
   1018  1.10    mrg <samp>--enable-tls</samp> or <samp>--disable-tls</samp>.  This can happen if
   1019   1.1    mrg the assembler supports TLS but the C library does not, or if the
   1020   1.1    mrg assumptions made by the configure test are incorrect.
   1021  1.10    mrg </p>
   1022  1.10    mrg </dd>
   1023  1.18    mrg <dt><code>--disable-tls</code></dt>
   1024  1.10    mrg <dd><p>Specify that the target does not support TLS.
   1025  1.10    mrg This is an alias for <samp>--enable-tls=no</samp>.
   1026  1.10    mrg </p>
   1027  1.10    mrg </dd>
   1028  1.18    mrg <dt><code>--disable-tm-clone-registry</code></dt>
   1029  1.13    mrg <dd><p>Disable TM clone registry in libgcc. It is enabled in libgcc by default.
   1030  1.13    mrg This option helps to reduce code size for embedded targets which do
   1031  1.13    mrg not use transactional memory.
   1032  1.13    mrg </p>
   1033  1.13    mrg </dd>
   1034  1.18    mrg <dt><code>--with-cpu=<var>cpu</var></code></dt>
   1035  1.18    mrg <dt><code>--with-cpu-32=<var>cpu</var></code></dt>
   1036  1.18    mrg <dt><code>--with-cpu-64=<var>cpu</var></code></dt>
   1037  1.10    mrg <dd><p>Specify which cpu variant the compiler should generate code for by default.
   1038  1.10    mrg <var>cpu</var> will be used as the default value of the <samp>-mcpu=</samp> switch.
   1039   1.5    mrg This option is only supported on some targets, including ARC, ARM, i386, M68k,
   1040  1.10    mrg PowerPC, and SPARC.  It is mandatory for ARC.  The <samp>--with-cpu-32</samp> and
   1041  1.10    mrg <samp>--with-cpu-64</samp> options specify separate default CPUs for
   1042  1.13    mrg 32-bit and 64-bit modes; these options are only supported for aarch64, i386,
   1043   1.7    mrg x86-64, PowerPC, and SPARC.
   1044  1.10    mrg </p>
   1045  1.10    mrg </dd>
   1046  1.18    mrg <dt><code>--with-schedule=<var>cpu</var></code></dt>
   1047  1.18    mrg <dt><code>--with-arch=<var>cpu</var></code></dt>
   1048  1.18    mrg <dt><code>--with-arch-32=<var>cpu</var></code></dt>
   1049  1.18    mrg <dt><code>--with-arch-64=<var>cpu</var></code></dt>
   1050  1.18    mrg <dt><code>--with-tune=<var>cpu</var></code></dt>
   1051  1.18    mrg <dt><code>--with-tune-32=<var>cpu</var></code></dt>
   1052  1.18    mrg <dt><code>--with-tune-64=<var>cpu</var></code></dt>
   1053  1.18    mrg <dt><code>--with-abi=<var>abi</var></code></dt>
   1054  1.18    mrg <dt><code>--with-fpu=<var>type</var></code></dt>
   1055  1.18    mrg <dt><code>--with-float=<var>type</var></code></dt>
   1056  1.19    mrg <dt><code>--with-simd=<var>type</var></code></dt>
   1057  1.10    mrg <dd><p>These configure options provide default values for the <samp>-mschedule=</samp>,
   1058  1.10    mrg <samp>-march=</samp>, <samp>-mtune=</samp>, <samp>-mabi=</samp>, and <samp>-mfpu=</samp>
   1059  1.10    mrg options and for <samp>-mhard-float</samp> or <samp>-msoft-float</samp>.  As with
   1060  1.10    mrg <samp>--with-cpu</samp>, which switches will be accepted and acceptable values
   1061   1.1    mrg of the arguments depend on the target.
   1062  1.10    mrg </p>
   1063  1.10    mrg </dd>
   1064  1.18    mrg <dt><code>--with-mode=<var>mode</var></code></dt>
   1065  1.10    mrg <dd><p>Specify if the compiler should default to <samp>-marm</samp> or <samp>-mthumb</samp>.
   1066   1.1    mrg This option is only supported on ARM targets.
   1067  1.10    mrg </p>
   1068  1.10    mrg </dd>
   1069  1.18    mrg <dt><code>--with-stack-offset=<var>num</var></code></dt>
   1070  1.10    mrg <dd><p>This option sets the default for the -mstack-offset=<var>num</var> option,
   1071   1.3  skrll and will thus generally also control the setting of this option for
   1072   1.3  skrll libraries.  This option is only supported on Epiphany targets.
   1073  1.10    mrg </p>
   1074  1.10    mrg </dd>
   1075  1.18    mrg <dt><code>--with-fpmath=<var>isa</var></code></dt>
   1076  1.10    mrg <dd><p>This options sets <samp>-mfpmath=sse</samp> by default and specifies the default
   1077  1.10    mrg ISA for floating-point arithmetics.  You can select either &lsquo;<samp>sse</samp>&rsquo; which
   1078  1.10    mrg enables <samp>-msse2</samp> or &lsquo;<samp>avx</samp>&rsquo; which enables <samp>-mavx</samp> by default.
   1079   1.3  skrll This option is only supported on i386 and x86-64 targets.
   1080  1.10    mrg </p>
   1081  1.10    mrg </dd>
   1082  1.18    mrg <dt><code>--with-fp-32=<var>mode</var></code></dt>
   1083  1.10    mrg <dd><p>On MIPS targets, set the default value for the <samp>-mfp</samp> option when using
   1084   1.5    mrg the o32 ABI.  The possibilities for <var>mode</var> are:
   1085  1.10    mrg </p><dl compact="compact">
   1086  1.18    mrg <dt><code>32</code></dt>
   1087  1.10    mrg <dd><p>Use the o32 FP32 ABI extension, as with the <samp>-mfp32</samp> command-line
   1088  1.10    mrg option.
   1089  1.10    mrg </p></dd>
   1090  1.18    mrg <dt><code>xx</code></dt>
   1091  1.10    mrg <dd><p>Use the o32 FPXX ABI extension, as with the <samp>-mfpxx</samp> command-line
   1092  1.10    mrg option.
   1093  1.10    mrg </p></dd>
   1094  1.18    mrg <dt><code>64</code></dt>
   1095  1.10    mrg <dd><p>Use the o32 FP64 ABI extension, as with the <samp>-mfp64</samp> command-line
   1096  1.10    mrg option.
   1097  1.10    mrg </p></dd>
   1098   1.5    mrg </dl>
   1099  1.10    mrg <p>In the absence of this configuration option the default is to use the o32
   1100   1.5    mrg FP32 ABI extension.
   1101  1.10    mrg </p>
   1102  1.10    mrg </dd>
   1103  1.18    mrg <dt><code>--with-odd-spreg-32</code></dt>
   1104  1.10    mrg <dd><p>On MIPS targets, set the <samp>-modd-spreg</samp> option by default when using
   1105   1.5    mrg the o32 ABI.
   1106  1.10    mrg </p>
   1107  1.10    mrg </dd>
   1108  1.18    mrg <dt><code>--without-odd-spreg-32</code></dt>
   1109  1.10    mrg <dd><p>On MIPS targets, set the <samp>-mno-odd-spreg</samp> option by default when using
   1110   1.5    mrg the o32 ABI.  This is normally used in conjunction with
   1111  1.10    mrg <samp>--with-fp-32=64</samp> in order to target the o32 FP64A ABI extension.
   1112  1.10    mrg </p>
   1113  1.10    mrg </dd>
   1114  1.18    mrg <dt><code>--with-nan=<var>encoding</var></code></dt>
   1115  1.10    mrg <dd><p>On MIPS targets, set the default encoding convention to use for the
   1116   1.5    mrg special not-a-number (NaN) IEEE 754 floating-point data.  The
   1117   1.5    mrg possibilities for <var>encoding</var> are:
   1118  1.10    mrg </p><dl compact="compact">
   1119  1.18    mrg <dt><code>legacy</code></dt>
   1120  1.10    mrg <dd><p>Use the legacy encoding, as with the <samp>-mnan=legacy</samp> command-line
   1121  1.10    mrg option.
   1122  1.10    mrg </p></dd>
   1123  1.18    mrg <dt><code>2008</code></dt>
   1124  1.10    mrg <dd><p>Use the 754-2008 encoding, as with the <samp>-mnan=2008</samp> command-line
   1125  1.10    mrg option.
   1126  1.10    mrg </p></dd>
   1127   1.5    mrg </dl>
   1128  1.10    mrg <p>To use this configuration option you must have an assembler version
   1129  1.10    mrg installed that supports the <samp>-mnan=</samp> command-line option too.
   1130   1.5    mrg In the absence of this configuration option the default convention is
   1131  1.10    mrg the legacy encoding, as when neither of the <samp>-mnan=2008</samp> and
   1132  1.10    mrg <samp>-mnan=legacy</samp> command-line options has been used.
   1133  1.10    mrg </p>
   1134  1.10    mrg </dd>
   1135  1.18    mrg <dt><code>--with-divide=<var>type</var></code></dt>
   1136  1.10    mrg <dd><p>Specify how the compiler should generate code for checking for
   1137  1.10    mrg division by zero.  This option is only supported on the MIPS target.
   1138   1.1    mrg The possibilities for <var>type</var> are:
   1139  1.10    mrg </p><dl compact="compact">
   1140  1.18    mrg <dt><code>traps</code></dt>
   1141  1.10    mrg <dd><p>Division by zero checks use conditional traps (this is the default on
   1142  1.10    mrg systems that support conditional traps).
   1143  1.10    mrg </p></dd>
   1144  1.18    mrg <dt><code>breaks</code></dt>
   1145  1.10    mrg <dd><p>Division by zero checks use the break instruction.
   1146  1.10    mrg </p></dd>
   1147   1.1    mrg </dl>
   1148   1.1    mrg 
   1149  1.19    mrg </dd>
   1150  1.19    mrg <dt><code>--with-compact-branches=<var>policy</var></code></dt>
   1151  1.19    mrg <dd><p>Specify how the compiler should generate branch instructions.
   1152  1.19    mrg This option is only supported on the MIPS target.
   1153  1.19    mrg The possibilities for <var>type</var> are:
   1154  1.19    mrg </p><dl compact="compact">
   1155  1.19    mrg <dt><code>optimal</code></dt>
   1156  1.19    mrg <dd><p>Cause a delay slot branch to be used if one is available in the
   1157  1.19    mrg current ISA and the delay slot is successfully filled. If the delay slot
   1158  1.19    mrg is not filled, a compact branch will be chosen if one is available.
   1159  1.19    mrg </p></dd>
   1160  1.19    mrg <dt><code>never</code></dt>
   1161  1.19    mrg <dd><p>Ensures that compact branch instructions will never be generated.
   1162  1.19    mrg </p></dd>
   1163  1.19    mrg <dt><code>always</code></dt>
   1164  1.19    mrg <dd><p>Ensures that a compact branch instruction will be generated if available.
   1165  1.19    mrg If a compact branch instruction is not available,
   1166  1.19    mrg a delay slot form of the branch will be used instead.
   1167  1.19    mrg This option is supported from MIPS Release 6 onwards.
   1168  1.19    mrg For pre-R6/microMIPS/MIPS16, this option is just same as never/optimal.
   1169  1.19    mrg </p></dd>
   1170  1.19    mrg </dl>
   1171  1.19    mrg 
   1172  1.10    mrg 
   1173  1.10    mrg </dd>
   1174  1.18    mrg <dt><code>--with-llsc</code></dt>
   1175  1.10    mrg <dd><p>On MIPS targets, make <samp>-mllsc</samp> the default when no
   1176  1.10    mrg <samp>-mno-llsc</samp> option is passed.  This is the default for
   1177   1.1    mrg Linux-based targets, as the kernel will emulate them if the ISA does
   1178   1.1    mrg not provide them.
   1179  1.10    mrg </p>
   1180  1.10    mrg </dd>
   1181  1.18    mrg <dt><code>--without-llsc</code></dt>
   1182  1.10    mrg <dd><p>On MIPS targets, make <samp>-mno-llsc</samp> the default when no
   1183  1.10    mrg <samp>-mllsc</samp> option is passed.
   1184  1.10    mrg </p>
   1185  1.10    mrg </dd>
   1186  1.18    mrg <dt><code>--with-synci</code></dt>
   1187  1.10    mrg <dd><p>On MIPS targets, make <samp>-msynci</samp> the default when no
   1188  1.10    mrg <samp>-mno-synci</samp> option is passed.
   1189  1.10    mrg </p>
   1190  1.10    mrg </dd>
   1191  1.18    mrg <dt><code>--without-synci</code></dt>
   1192  1.10    mrg <dd><p>On MIPS targets, make <samp>-mno-synci</samp> the default when no
   1193  1.10    mrg <samp>-msynci</samp> option is passed.  This is the default.
   1194  1.10    mrg </p>
   1195  1.10    mrg </dd>
   1196  1.18    mrg <dt><code>--with-lxc1-sxc1</code></dt>
   1197  1.10    mrg <dd><p>On MIPS targets, make <samp>-mlxc1-sxc1</samp> the default when no
   1198  1.10    mrg <samp>-mno-lxc1-sxc1</samp> option is passed.  This is the default.
   1199  1.10    mrg </p>
   1200  1.10    mrg </dd>
   1201  1.18    mrg <dt><code>--without-lxc1-sxc1</code></dt>
   1202  1.10    mrg <dd><p>On MIPS targets, make <samp>-mno-lxc1-sxc1</samp> the default when no
   1203  1.10    mrg <samp>-mlxc1-sxc1</samp> option is passed.  The indexed load/store
   1204   1.9    mrg instructions are not directly a problem but can lead to unexpected
   1205   1.9    mrg behaviour when deployed in an application intended for a 32-bit address
   1206   1.9    mrg space but run on a 64-bit processor.  The issue is seen because all
   1207   1.9    mrg known MIPS 64-bit Linux kernels execute o32 and n32 applications
   1208   1.9    mrg with 64-bit addressing enabled which affects the overflow behaviour
   1209   1.9    mrg of the indexed addressing mode.  GCC will assume that ordinary
   1210   1.9    mrg 32-bit arithmetic overflow behaviour is the same whether performed
   1211   1.9    mrg as an <code>addu</code> instruction or as part of the address calculation
   1212   1.9    mrg in <code>lwxc1</code> type instructions.  This assumption holds true in a
   1213   1.9    mrg pure 32-bit environment and can hold true in a 64-bit environment if
   1214   1.9    mrg the address space is accurately set to be 32-bit for o32 and n32.
   1215  1.10    mrg </p>
   1216  1.10    mrg </dd>
   1217  1.18    mrg <dt><code>--with-madd4</code></dt>
   1218  1.10    mrg <dd><p>On MIPS targets, make <samp>-mmadd4</samp> the default when no
   1219  1.10    mrg <samp>-mno-madd4</samp> option is passed.  This is the default.
   1220  1.10    mrg </p>
   1221  1.10    mrg </dd>
   1222  1.18    mrg <dt><code>--without-madd4</code></dt>
   1223  1.10    mrg <dd><p>On MIPS targets, make <samp>-mno-madd4</samp> the default when no
   1224  1.10    mrg <samp>-mmadd4</samp> option is passed.  The <code>madd4</code> instruction
   1225   1.9    mrg family can be problematic when targeting a combination of cores that
   1226   1.9    mrg implement these instructions differently.  There are two known cores
   1227   1.9    mrg that implement these as fused operations instead of unfused (where
   1228   1.9    mrg unfused is normally expected).  Disabling these instructions is the
   1229   1.9    mrg only way to ensure compatible code is generated; this will incur
   1230   1.9    mrg a performance penalty.
   1231  1.10    mrg </p>
   1232  1.10    mrg </dd>
   1233  1.19    mrg <dt><code>--with-msa</code></dt>
   1234  1.19    mrg <dd><p>On MIPS targets, make <samp>-mmsa</samp> the default when no
   1235  1.19    mrg <samp>-mno-msa</samp> option is passed.
   1236  1.19    mrg </p>
   1237  1.19    mrg </dd>
   1238  1.19    mrg <dt><code>--without-msa</code></dt>
   1239  1.19    mrg <dd><p>On MIPS targets, make <samp>-mno-msa</samp> the default when no
   1240  1.19    mrg <samp>-mmsa</samp> option is passed. This is the default.
   1241  1.19    mrg </p>
   1242  1.19    mrg </dd>
   1243  1.18    mrg <dt><code>--with-mips-plt</code></dt>
   1244  1.10    mrg <dd><p>On MIPS targets, make use of copy relocations and PLTs.
   1245   1.1    mrg These features are extensions to the traditional
   1246   1.1    mrg SVR4-based MIPS ABIs and require support from GNU binutils
   1247   1.1    mrg and the runtime C library.
   1248  1.10    mrg </p>
   1249  1.10    mrg </dd>
   1250  1.18    mrg <dt><code>--with-stack-clash-protection-guard-size=<var>size</var></code></dt>
   1251  1.12    mrg <dd><p>On certain targets this option sets the default stack clash protection guard
   1252  1.12    mrg size as a power of two in bytes.  On AArch64 <var>size</var> is required to be either
   1253  1.12    mrg 12 (4KB) or 16 (64KB).
   1254  1.12    mrg </p>
   1255  1.12    mrg </dd>
   1256  1.18    mrg <dt><code>--with-isa-spec=<var>ISA-spec-string</var></code></dt>
   1257  1.16    mrg <dd><p>On RISC-V targets specify the default version of the RISC-V Unprivileged
   1258  1.16    mrg (formerly User-Level) ISA specification to produce code conforming to.
   1259  1.16    mrg The possibilities for <var>ISA-spec-string</var> are:
   1260  1.16    mrg </p><dl compact="compact">
   1261  1.18    mrg <dt><code>2.2</code></dt>
   1262  1.16    mrg <dd><p>Produce code conforming to version 2.2.
   1263  1.16    mrg </p></dd>
   1264  1.18    mrg <dt><code>20190608</code></dt>
   1265  1.16    mrg <dd><p>Produce code conforming to version 20190608.
   1266  1.16    mrg </p></dd>
   1267  1.18    mrg <dt><code>20191213</code></dt>
   1268  1.16    mrg <dd><p>Produce code conforming to version 20191213.
   1269  1.16    mrg </p></dd>
   1270  1.16    mrg </dl>
   1271  1.16    mrg <p>In the absence of this configuration option the default version is 20191213.
   1272  1.16    mrg </p>
   1273  1.16    mrg </dd>
   1274  1.18    mrg <dt><code>--enable-__cxa_atexit</code></dt>
   1275  1.10    mrg <dd><p>Define if you want to use __cxa_atexit, rather than atexit, to
   1276  1.10    mrg register C++ destructors for local statics and global objects.
   1277   1.1    mrg This is essential for fully standards-compliant handling of
   1278   1.1    mrg destructors, but requires __cxa_atexit in libc.  This option is currently
   1279   1.1    mrg only available on systems with GNU libc.  When enabled, this will cause
   1280  1.10    mrg <samp>-fuse-cxa-atexit</samp> to be passed by default.
   1281  1.10    mrg </p>
   1282  1.10    mrg </dd>
   1283  1.18    mrg <dt><code>--enable-gnu-indirect-function</code></dt>
   1284  1.10    mrg <dd><p>Define if you want to enable the <code>ifunc</code> attribute.  This option is
   1285   1.3  skrll currently only available on systems with GNU libc on certain targets.
   1286  1.10    mrg </p>
   1287  1.10    mrg </dd>
   1288  1.18    mrg <dt><code>--enable-target-optspace</code></dt>
   1289  1.10    mrg <dd><p>Specify that target
   1290  1.10    mrg libraries should be optimized for code space instead of code speed.
   1291   1.1    mrg This is the default for the m32r platform.
   1292  1.10    mrg </p>
   1293  1.10    mrg </dd>
   1294  1.18    mrg <dt><code>--with-cpp-install-dir=<var>dirname</var></code></dt>
   1295  1.10    mrg <dd><p>Specify that the user visible <code>cpp</code> program should be installed
   1296  1.10    mrg in <samp><var>prefix</var>/<var>dirname</var>/cpp</samp>, in addition to <var>bindir</var>.
   1297  1.10    mrg </p>
   1298  1.10    mrg </dd>
   1299  1.18    mrg <dt><code>--enable-comdat</code></dt>
   1300  1.10    mrg <dd><p>Enable COMDAT group support.  This is primarily used to override the
   1301   1.1    mrg automatically detected value.
   1302  1.10    mrg </p>
   1303  1.10    mrg </dd>
   1304  1.18    mrg <dt><code>--enable-initfini-array</code></dt>
   1305  1.10    mrg <dd><p>Force the use of sections <code>.init_array</code> and <code>.fini_array</code>
   1306   1.1    mrg (instead of <code>.init</code> and <code>.fini</code>) for constructors and
   1307  1.10    mrg destructors.  Option <samp>--disable-initfini-array</samp> has the
   1308   1.1    mrg opposite effect.  If neither option is specified, the configure script
   1309   1.1    mrg will try to guess whether the <code>.init_array</code> and
   1310   1.1    mrg <code>.fini_array</code> sections are supported and, if they are, use them.
   1311  1.10    mrg </p>
   1312  1.10    mrg </dd>
   1313  1.18    mrg <dt><code>--enable-link-mutex</code></dt>
   1314  1.10    mrg <dd><p>When building GCC, use a mutex to avoid linking the compilers for
   1315   1.5    mrg multiple languages at the same time, to avoid thrashing on build
   1316   1.5    mrg systems with limited free memory.  The default is not to use such a mutex.
   1317  1.10    mrg </p>
   1318  1.10    mrg </dd>
   1319  1.18    mrg <dt><code>--enable-link-serialization</code></dt>
   1320  1.16    mrg <dd><p>When building GCC, use make dependencies to serialize linking the compilers for
   1321  1.16    mrg multiple languages, to avoid thrashing on build
   1322  1.16    mrg systems with limited free memory.  The default is not to add such
   1323  1.16    mrg dependencies and thus with parallel make potentially link different
   1324  1.16    mrg compilers concurrently.  If the argument is a positive integer, allow
   1325  1.16    mrg that number of concurrent link processes for the large binaries.
   1326  1.16    mrg </p>
   1327  1.16    mrg </dd>
   1328  1.18    mrg <dt><code>--enable-maintainer-mode</code></dt>
   1329  1.10    mrg <dd><p>The build rules that regenerate the Autoconf and Automake output files as
   1330  1.10    mrg well as the GCC master message catalog <samp>gcc.pot</samp> are normally
   1331   1.1    mrg disabled.  This is because it can only be rebuilt if the complete source
   1332   1.1    mrg tree is present.  If you have changed the sources and want to rebuild the
   1333  1.10    mrg catalog, configuring with <samp>--enable-maintainer-mode</samp> will enable
   1334   1.1    mrg this.  Note that you need a recent version of the <code>gettext</code> tools
   1335   1.1    mrg to do so.
   1336  1.10    mrg </p>
   1337  1.10    mrg </dd>
   1338  1.18    mrg <dt><code>--disable-bootstrap</code></dt>
   1339  1.10    mrg <dd><p>For a native build, the default configuration is to perform
   1340  1.10    mrg a 3-stage bootstrap of the compiler when &lsquo;<samp>make</samp>&rsquo; is invoked,
   1341   1.1    mrg testing that GCC can compile itself correctly.  If you want to disable
   1342  1.10    mrg this process, you can configure with <samp>--disable-bootstrap</samp>.
   1343  1.10    mrg </p>
   1344  1.10    mrg </dd>
   1345  1.18    mrg <dt><code>--enable-bootstrap</code></dt>
   1346  1.10    mrg <dd><p>In special cases, you may want to perform a 3-stage build
   1347  1.10    mrg even if the target and host triplets are different.
   1348   1.1    mrg This is possible when the host can run code compiled for
   1349  1.10    mrg the target (e.g. host is i686-linux, target is i486-linux).
   1350   1.1    mrg Starting from GCC 4.2, to do this you have to configure explicitly
   1351  1.10    mrg with <samp>--enable-bootstrap</samp>.
   1352  1.10    mrg </p>
   1353  1.10    mrg </dd>
   1354  1.18    mrg <dt><code>--enable-generated-files-in-srcdir</code></dt>
   1355  1.10    mrg <dd><p>Neither the .c and .h files that are generated from Bison and flex nor the
   1356   1.1    mrg info manuals and man pages that are built from the .texi files are present
   1357  1.11    mrg in the repository development tree.  When building GCC from that development tree,
   1358   1.1    mrg or from one of our snapshots, those generated files are placed in your
   1359   1.1    mrg build directory, which allows for the source to be in a readonly
   1360   1.1    mrg directory.
   1361  1.10    mrg </p>
   1362  1.10    mrg <p>If you configure with <samp>--enable-generated-files-in-srcdir</samp> then those
   1363   1.1    mrg generated files will go into the source directory.  This is mainly intended
   1364   1.1    mrg for generating release or prerelease tarballs of the GCC sources, since it
   1365   1.1    mrg is not a requirement that the users of source releases to have flex, Bison,
   1366   1.1    mrg or makeinfo.
   1367  1.10    mrg </p>
   1368  1.10    mrg </dd>
   1369  1.18    mrg <dt><code>--enable-version-specific-runtime-libs</code></dt>
   1370  1.10    mrg <dd><p>Specify
   1371   1.1    mrg that runtime libraries should be installed in the compiler specific
   1372  1.10    mrg subdirectory (<samp><var>libdir</var>/gcc</samp>) rather than the usual places.  In
   1373  1.10    mrg addition, &lsquo;<samp>libstdc++</samp>&rsquo;&rsquo;s include files will be installed into
   1374   1.1    mrg <samp><var>libdir</var></samp> unless you overruled it by using
   1375  1.10    mrg <samp>--with-gxx-include-dir=<var>dirname</var></samp>.  Using this option is
   1376   1.1    mrg particularly useful if you intend to use several versions of GCC in
   1377  1.13    mrg parallel.  The default is &lsquo;<samp>yes</samp>&rsquo; for &lsquo;<samp>libada</samp>&rsquo;, and &lsquo;<samp>no</samp>&rsquo; for
   1378  1.13    mrg the remaining libraries.
   1379  1.10    mrg </p>
   1380  1.10    mrg </dd>
   1381  1.19    mrg <dt><code>--with-darwin-extra-rpath</code></dt>
   1382  1.19    mrg <dd><p>This is provided to allow distributions to add a single additional
   1383  1.19    mrg runpath on Darwin / macOS systems. This allows for cases where the
   1384  1.19    mrg installed GCC library directories are then symlinked to a common
   1385  1.19    mrg directory outside of the GCC installation.
   1386  1.19    mrg </p>
   1387  1.19    mrg </dd>
   1388  1.18    mrg <dt><code><a name="WithAixSoname"></a>--with-aix-soname=&lsquo;<samp>aix</samp>&rsquo;, &lsquo;<samp>svr4</samp>&rsquo; or &lsquo;<samp>both</samp>&rsquo;</code></dt>
   1389  1.10    mrg <dd><p>Traditional AIX shared library versioning (versioned <code>Shared Object</code>
   1390   1.5    mrg files as members of unversioned <code>Archive Library</code> files named
   1391  1.10    mrg &lsquo;<samp>lib.a</samp>&rsquo;) causes numerous headaches for package managers. However,
   1392   1.5    mrg <code>Import Files</code> as members of <code>Archive Library</code> files allow for
   1393   1.5    mrg <strong>filename-based versioning</strong> of shared libraries as seen on Linux/SVR4,
   1394  1.10    mrg where this is called the &quot;SONAME&quot;. But as they prevent static linking,
   1395   1.5    mrg <code>Import Files</code> may be used with <code>Runtime Linking</code> only, where the
   1396  1.10    mrg linker does search for &lsquo;<samp>libNAME.so</samp>&rsquo; before &lsquo;<samp>libNAME.a</samp>&rsquo; library
   1397  1.10    mrg filenames with the &lsquo;<samp>-lNAME</samp>&rsquo; linker flag.
   1398  1.10    mrg </p>
   1399  1.18    mrg <a name="AixLdCommand"></a><p>For detailed information please refer to the AIX
   1400  1.10    mrg <a href="https://www.ibm.com/support/knowledgecenter/search/%22the%20ld%20command%2C%20also%20called%20the%20linkage%20editor%20or%20binder%22">ld
   1401  1.10    mrg Command</a> reference.
   1402  1.10    mrg </p>
   1403  1.10    mrg <p>As long as shared library creation is enabled, upon:
   1404  1.10    mrg </p><dl compact="compact">
   1405  1.18    mrg <dt><code>--with-aix-soname=aix</code></dt>
   1406  1.18    mrg <dt><code>--with-aix-soname=both</code></dt>
   1407  1.10    mrg <dd><p>A (traditional AIX) <code>Shared Archive Library</code> file is created:
   1408  1.10    mrg  </p><ul>
   1409  1.10    mrg <li> using the &lsquo;<samp>libNAME.a</samp>&rsquo; filename scheme
   1410  1.10    mrg   </li><li> with the <code>Shared Object</code> file as archive member named
   1411  1.10    mrg   &lsquo;<samp>libNAME.so.V</samp>&rsquo; (except for &lsquo;<samp>libgcc_s</samp>&rsquo;, where the <code>Shared
   1412  1.10    mrg   Object</code> file is named &lsquo;<samp>shr.o</samp>&rsquo; for backwards compatibility), which
   1413  1.10    mrg   <ul class="no-bullet">
   1414  1.10    mrg <li>- is used for runtime loading from inside the &lsquo;<samp>libNAME.a</samp>&rsquo; file
   1415  1.10    mrg    </li><li>- is used for dynamic loading via
   1416  1.10    mrg    <code>dlopen(&quot;libNAME.a(libNAME.so.V)&quot;, RTLD_MEMBER)</code>
   1417  1.10    mrg    </li><li>- is used for shared linking
   1418  1.10    mrg    </li><li>- is used for static linking, so no separate <code>Static Archive
   1419   1.5    mrg    Library</code> file is needed
   1420  1.10    mrg   </li></ul>
   1421  1.10    mrg </li></ul>
   1422  1.10    mrg </dd>
   1423  1.18    mrg <dt><code>--with-aix-soname=both</code></dt>
   1424  1.18    mrg <dt><code>--with-aix-soname=svr4</code></dt>
   1425  1.10    mrg <dd><p>A (second) <code>Shared Archive Library</code> file is created:
   1426  1.10    mrg  </p><ul>
   1427  1.10    mrg <li> using the &lsquo;<samp>libNAME.so.V</samp>&rsquo; filename scheme
   1428  1.10    mrg  </li><li> with the <code>Shared Object</code> file as archive member named
   1429  1.10    mrg  &lsquo;<samp>shr.o</samp>&rsquo;, which
   1430  1.10    mrg   <ul class="no-bullet">
   1431  1.10    mrg <li>- is created with the <code>-G linker flag</code>
   1432  1.10    mrg    </li><li>- has the <code>F_LOADONLY</code> flag set
   1433  1.10    mrg    </li><li>- is used for runtime loading from inside the &lsquo;<samp>libNAME.so.V</samp>&rsquo; file
   1434  1.10    mrg    </li><li>- is used for dynamic loading via <code>dlopen(&quot;libNAME.so.V(shr.o)&quot;,
   1435   1.5    mrg    RTLD_MEMBER)</code>
   1436  1.10    mrg   </li></ul>
   1437  1.10    mrg </li><li> with the <code>Import File</code> as archive member named &lsquo;<samp>shr.imp</samp>&rsquo;,
   1438   1.5    mrg  which
   1439  1.10    mrg   <ul class="no-bullet">
   1440  1.10    mrg <li>- refers to &lsquo;<samp>libNAME.so.V(shr.o)</samp>&rsquo; as the &quot;SONAME&quot;, to be recorded
   1441   1.5    mrg    in the <code>Loader Section</code> of subsequent binaries
   1442  1.10    mrg    </li><li>- indicates whether &lsquo;<samp>libNAME.so.V(shr.o)</samp>&rsquo; is 32 or 64 bit
   1443  1.10    mrg    </li><li>- lists all the public symbols exported by &lsquo;<samp>lib.so.V(shr.o)</samp>&rsquo;,
   1444  1.10    mrg    eventually decorated with the <code>&lsquo;<samp>weak</samp>&rsquo; Keyword</code>
   1445  1.10    mrg    </li><li>- is necessary for shared linking against &lsquo;<samp>lib.so.V(shr.o)</samp>&rsquo;
   1446  1.10    mrg    </li></ul>
   1447  1.10    mrg </li></ul>
   1448  1.10    mrg <p>A symbolic link using the &lsquo;<samp>libNAME.so</samp>&rsquo; filename scheme is created:
   1449  1.10    mrg   </p><ul>
   1450  1.10    mrg <li> pointing to the &lsquo;<samp>libNAME.so.V</samp>&rsquo; <code>Shared Archive Library</code> file
   1451  1.10    mrg   </li><li> to permit the <code>ld Command</code> to find &lsquo;<samp>lib.so.V(shr.imp)</samp>&rsquo; via
   1452  1.10    mrg   the &lsquo;<samp>-lNAME</samp>&rsquo; argument (requires <code>Runtime Linking</code> to be enabled)
   1453  1.10    mrg   </li><li> to permit dynamic loading of &lsquo;<samp>lib.so.V(shr.o)</samp>&rsquo; without the need
   1454  1.10    mrg   to specify the version number via <code>dlopen(&quot;libNAME.so(shr.o)&quot;,
   1455   1.5    mrg   RTLD_MEMBER)</code>
   1456  1.10    mrg   </li></ul>
   1457  1.10    mrg </dd>
   1458  1.10    mrg </dl>
   1459   1.5    mrg 
   1460  1.10    mrg <p>As long as static library creation is enabled, upon:
   1461  1.10    mrg </p><dl compact="compact">
   1462  1.18    mrg <dt><code>--with-aix-soname=svr4</code></dt>
   1463  1.10    mrg <dd><p>A <code>Static Archive Library</code> is created:
   1464  1.10    mrg  </p><ul>
   1465  1.10    mrg <li> using the &lsquo;<samp>libNAME.a</samp>&rsquo; filename scheme
   1466  1.10    mrg  </li><li> with all the <code>Static Object</code> files as archive members, which
   1467  1.10    mrg   <ul class="no-bullet">
   1468  1.10    mrg <li>- are used for static linking
   1469  1.10    mrg   </li></ul>
   1470  1.10    mrg </li></ul>
   1471  1.10    mrg </dd>
   1472  1.10    mrg </dl>
   1473   1.5    mrg 
   1474  1.10    mrg <p>While the aix-soname=&lsquo;<samp>svr4</samp>&rsquo; option does not create <code>Shared Object</code>
   1475   1.5    mrg files as members of unversioned <code>Archive Library</code> files any more, package
   1476   1.5    mrg managers still are responsible to
   1477   1.5    mrg <a href="./specific.html#TransferAixShobj">transfer</a> <code>Shared Object</code> files
   1478   1.5    mrg found as member of a previously installed unversioned <code>Archive Library</code>
   1479   1.5    mrg file into the newly installed <code>Archive Library</code> file with the same
   1480   1.5    mrg filename.
   1481  1.10    mrg </p>
   1482  1.10    mrg <p><em>WARNING:</em> Creating <code>Shared Object</code> files with <code>Runtime Linking</code>
   1483   1.5    mrg enabled may bloat the TOC, eventually leading to <code>TOC overflow</code> errors,
   1484  1.10    mrg requiring the use of either the <samp>-Wl,-bbigtoc</samp> linker flag (seen to
   1485   1.5    mrg break with the <code>GDB</code> debugger) or some of the TOC-related compiler flags,
   1486   1.5    mrg see &ldquo;RS/6000 and PowerPC Options&rdquo; in the main manual.
   1487  1.10    mrg </p>
   1488  1.10    mrg <p><samp>--with-aix-soname</samp> is currently supported by &lsquo;<samp>libgcc_s</samp>&rsquo; only, so
   1489   1.5    mrg this option is still experimental and not for normal use yet.
   1490  1.10    mrg </p>
   1491  1.10    mrg <p>Default is the traditional behavior <samp>--with-aix-soname=&lsquo;<samp>aix</samp>&rsquo;</samp>.
   1492  1.10    mrg </p>
   1493  1.10    mrg </dd>
   1494  1.18    mrg <dt><code>--enable-languages=<var>lang1</var>,<var>lang2</var>,&hellip;</code></dt>
   1495  1.10    mrg <dd><p>Specify that only a particular subset of compilers and
   1496   1.1    mrg their runtime libraries should be built.  For a list of valid values for
   1497   1.1    mrg <var>langN</var> you can issue the following command in the
   1498  1.10    mrg <samp>gcc</samp> directory of your GCC source tree:<br>
   1499  1.18    mrg </p><div class="smallexample">
   1500  1.18    mrg <pre class="smallexample">grep ^language= */config-lang.in
   1501  1.10    mrg </pre></div>
   1502  1.10    mrg <p>Currently, you can use any of the following:
   1503  1.12    mrg <code>all</code>, <code>default</code>, <code>ada</code>, <code>c</code>, <code>c++</code>, <code>d</code>,
   1504  1.19    mrg <code>fortran</code>, <code>go</code>, <code>jit</code>, <code>lto</code>, <code>m2</code>,
   1505  1.19    mrg <code>objc</code>, <code>obj-c++</code>.
   1506  1.10    mrg Building the Ada compiler has special requirements, see below.
   1507  1.10    mrg If you do not pass this flag, or specify the option <code>default</code>, then the
   1508  1.10    mrg default languages available in the <samp>gcc</samp> sub-tree will be configured.
   1509  1.19    mrg Ada, D, Go, Jit, Objective-C++ and Modula-2 are not default languages.
   1510  1.19    mrg LTO is not a
   1511  1.10    mrg default language, but is built by default because <samp>--enable-lto</samp> is
   1512  1.10    mrg enabled by default.  The other languages are default languages.  If
   1513  1.10    mrg <code>all</code> is specified, then all available languages are built.  An
   1514  1.10    mrg exception is <code>jit</code> language, which requires
   1515  1.10    mrg <samp>--enable-host-shared</samp> to be included with <code>all</code>.
   1516  1.10    mrg </p>
   1517  1.10    mrg </dd>
   1518  1.18    mrg <dt><code>--enable-stage1-languages=<var>lang1</var>,<var>lang2</var>,&hellip;</code></dt>
   1519  1.10    mrg <dd><p>Specify that a particular subset of compilers and their runtime
   1520   1.1    mrg libraries should be built with the system C compiler during stage 1 of
   1521   1.1    mrg the bootstrap process, rather than only in later stages with the
   1522   1.1    mrg bootstrapped C compiler.  The list of valid values is the same as for
   1523  1.10    mrg <samp>--enable-languages</samp>, and the option <code>all</code> will select all
   1524  1.10    mrg of the languages enabled by <samp>--enable-languages</samp>.  This option is
   1525   1.1    mrg primarily useful for GCC development; for instance, when a development
   1526   1.1    mrg version of the compiler cannot bootstrap due to compiler bugs, or when
   1527   1.1    mrg one is debugging front ends other than the C front end.  When this
   1528   1.1    mrg option is used, one can then build the target libraries for the
   1529  1.10    mrg specified languages with the stage-1 compiler by using <code>make
   1530  1.10    mrg stage1-bubble all-target</code>, or run the testsuite on the stage-1 compiler
   1531  1.10    mrg for the specified languages using <code>make stage1-start check-gcc</code>.
   1532  1.10    mrg </p>
   1533  1.10    mrg </dd>
   1534  1.18    mrg <dt><code>--disable-libada</code></dt>
   1535  1.10    mrg <dd><p>Specify that the run-time libraries and tools used by GNAT should not
   1536   1.1    mrg be built.  This can be useful for debugging, or for compatibility with
   1537   1.1    mrg previous Ada build procedures, when it was required to explicitly
   1538  1.10    mrg do a &lsquo;<samp>make -C gcc gnatlib_and_tools</samp>&rsquo;.
   1539  1.10    mrg </p>
   1540  1.10    mrg </dd>
   1541  1.19    mrg <dt><code>--disable-libgm2</code></dt>
   1542  1.19    mrg <dd><p>Specify that the run-time libraries and tools used by Modula-2 should not
   1543  1.19    mrg be built.  This can be useful for debugging.
   1544  1.19    mrg </p>
   1545  1.19    mrg </dd>
   1546  1.18    mrg <dt><code>--disable-libsanitizer</code></dt>
   1547  1.10    mrg <dd><p>Specify that the run-time libraries for the various sanitizers should
   1548   1.5    mrg not be built.
   1549  1.10    mrg </p>
   1550  1.10    mrg </dd>
   1551  1.18    mrg <dt><code>--disable-libssp</code></dt>
   1552  1.10    mrg <dd><p>Specify that the run-time libraries for stack smashing protection
   1553  1.10    mrg should not be built or linked against.  On many targets library support
   1554  1.10    mrg is provided by the C library instead.
   1555  1.10    mrg </p>
   1556  1.10    mrg </dd>
   1557  1.18    mrg <dt><code>--disable-libquadmath</code></dt>
   1558  1.10    mrg <dd><p>Specify that the GCC quad-precision math library should not be built.
   1559   1.3  skrll On some systems, the library is required to be linkable when building
   1560  1.10    mrg the Fortran front end, unless <samp>--disable-libquadmath-support</samp>
   1561   1.3  skrll is used.
   1562  1.10    mrg </p>
   1563  1.10    mrg </dd>
   1564  1.18    mrg <dt><code>--disable-libquadmath-support</code></dt>
   1565  1.10    mrg <dd><p>Specify that the Fortran front end and <code>libgfortran</code> do not add
   1566   1.3  skrll support for <code>libquadmath</code> on systems supporting it.
   1567  1.10    mrg </p>
   1568  1.10    mrg </dd>
   1569  1.18    mrg <dt><code>--disable-libgomp</code></dt>
   1570  1.10    mrg <dd><p>Specify that the GNU Offloading and Multi Processing Runtime Library
   1571   1.5    mrg should not be built.
   1572  1.10    mrg </p>
   1573  1.10    mrg </dd>
   1574  1.18    mrg <dt><code>--disable-libvtv</code></dt>
   1575  1.10    mrg <dd><p>Specify that the run-time libraries used by vtable verification
   1576   1.5    mrg should not be built.
   1577  1.10    mrg </p>
   1578  1.10    mrg </dd>
   1579  1.18    mrg <dt><code>--with-dwarf2</code></dt>
   1580  1.10    mrg <dd><p>Specify that the compiler should
   1581  1.19    mrg use DWARF debugging information as the default; the exact
   1582  1.19    mrg DWARF version that is the default is target-specific.
   1583  1.10    mrg </p>
   1584  1.10    mrg </dd>
   1585  1.18    mrg <dt><code>--with-advance-toolchain=<var>at</var></code></dt>
   1586  1.10    mrg <dd><p>On 64-bit PowerPC Linux systems, configure the compiler to use the
   1587   1.7    mrg header files, library files, and the dynamic linker from the Advance
   1588   1.7    mrg Toolchain release <var>at</var> instead of the default versions that are
   1589   1.7    mrg provided by the Linux distribution.  In general, this option is
   1590   1.7    mrg intended for the developers of GCC, and it is not intended for general
   1591   1.7    mrg use.
   1592  1.10    mrg </p>
   1593  1.10    mrg </dd>
   1594  1.18    mrg <dt><code>--enable-targets=all</code></dt>
   1595  1.18    mrg <dt><code>--enable-targets=<var>target_list</var></code></dt>
   1596  1.10    mrg <dd><p>Some GCC targets, e.g. powerpc64-linux, build bi-arch compilers.
   1597   1.1    mrg These are compilers that are able to generate either 64-bit or 32-bit
   1598  1.10    mrg code.  Typically, the corresponding 32-bit target, e.g.
   1599   1.1    mrg powerpc-linux for powerpc64-linux, only generates 32-bit code.  This
   1600   1.1    mrg option enables the 32-bit target to be a bi-arch compiler, which is
   1601   1.1    mrg useful when you want a bi-arch compiler that defaults to 32-bit, and
   1602  1.10    mrg you are building a bi-arch or multi-arch binutils in a combined tree.
   1603   1.1    mrg On mips-linux, this will build a tri-arch compiler (ABI o32/n32/64),
   1604  1.10    mrg defaulted to o32.
   1605   1.3  skrll Currently, this option only affects sparc-linux, powerpc-linux, x86-linux,
   1606   1.3  skrll mips-linux and s390-linux.
   1607  1.10    mrg </p>
   1608  1.10    mrg </dd>
   1609  1.18    mrg <dt><code>--enable-default-pie</code></dt>
   1610  1.10    mrg <dd><p>Turn on <samp>-fPIE</samp> and <samp>-pie</samp> by default.
   1611  1.10    mrg </p>
   1612  1.10    mrg </dd>
   1613  1.18    mrg <dt><code>--enable-secureplt</code></dt>
   1614  1.10    mrg <dd><p>This option enables <samp>-msecure-plt</samp> by default for powerpc-linux.
   1615   1.1    mrg See &ldquo;RS/6000 and PowerPC Options&rdquo; in the main manual
   1616  1.10    mrg </p>
   1617  1.10    mrg </dd>
   1618  1.18    mrg <dt><code>--enable-default-ssp</code></dt>
   1619  1.10    mrg <dd><p>Turn on <samp>-fstack-protector-strong</samp> by default.
   1620  1.10    mrg </p>
   1621  1.10    mrg </dd>
   1622  1.18    mrg <dt><code>--enable-cld</code></dt>
   1623  1.10    mrg <dd><p>This option enables <samp>-mcld</samp> by default for 32-bit x86 targets.
   1624   1.1    mrg See &ldquo;i386 and x86-64 Options&rdquo; in the main manual
   1625  1.10    mrg </p>
   1626  1.10    mrg </dd>
   1627  1.18    mrg <dt><code>--enable-large-address-aware</code></dt>
   1628  1.12    mrg <dd><p>The <samp>--enable-large-address-aware</samp> option arranges for MinGW
   1629  1.12    mrg executables to be linked using the <samp>--large-address-aware</samp>
   1630  1.12    mrg option, that enables the use of more than 2GB of memory.  If GCC is
   1631  1.12    mrg configured with this option, its effects can be reversed by passing the
   1632  1.12    mrg <samp>-Wl,--disable-large-address-aware</samp> option to the so-configured
   1633  1.12    mrg compiler driver.
   1634  1.12    mrg </p>
   1635  1.12    mrg </dd>
   1636  1.18    mrg <dt><code>--enable-win32-registry</code></dt>
   1637  1.18    mrg <dt><code>--enable-win32-registry=<var>key</var></code></dt>
   1638  1.18    mrg <dt><code>--disable-win32-registry</code></dt>
   1639  1.10    mrg <dd><p>The <samp>--enable-win32-registry</samp> option enables Microsoft Windows-hosted GCC
   1640   1.1    mrg to look up installations paths in the registry using the following key:
   1641  1.10    mrg </p>
   1642  1.18    mrg <div class="smallexample">
   1643  1.18    mrg <pre class="smallexample"><code>HKEY_LOCAL_MACHINE\SOFTWARE\Free Software Foundation\<var>key</var></code>
   1644  1.10    mrg </pre></div>
   1645   1.1    mrg 
   1646  1.10    mrg <p><var>key</var> defaults to GCC version number, and can be overridden by the
   1647  1.10    mrg <samp>--enable-win32-registry=<var>key</var></samp> option.  Vendors and distributors
   1648   1.1    mrg who use custom installers are encouraged to provide a different key,
   1649   1.1    mrg perhaps one comprised of vendor name and GCC version number, to
   1650   1.1    mrg avoid conflict with existing installations.  This feature is enabled
   1651  1.10    mrg by default, and can be disabled by <samp>--disable-win32-registry</samp>
   1652   1.1    mrg option.  This option has no effect on the other hosts.
   1653  1.10    mrg </p>
   1654  1.10    mrg </dd>
   1655  1.18    mrg <dt><code>--nfp</code></dt>
   1656  1.10    mrg <dd><p>Specify that the machine does not have a floating point unit.  This
   1657  1.10    mrg option only applies to &lsquo;<samp>m68k-sun-sunos<var>n</var></samp>&rsquo;.  On any other
   1658  1.10    mrg system, <samp>--nfp</samp> has no effect.
   1659  1.10    mrg </p>
   1660  1.10    mrg </dd>
   1661  1.18    mrg <dt><code>--enable-werror</code></dt>
   1662  1.18    mrg <dt><code>--disable-werror</code></dt>
   1663  1.18    mrg <dt><code>--enable-werror=yes</code></dt>
   1664  1.18    mrg <dt><code>--enable-werror=no</code></dt>
   1665  1.10    mrg <dd><p>When you specify this option, it controls whether certain files in the
   1666  1.10    mrg compiler are built with <samp>-Werror</samp> in bootstrap stage2 and later.
   1667  1.10    mrg If you don&rsquo;t specify it, <samp>-Werror</samp> is turned on for the main
   1668   1.1    mrg development trunk.  However it defaults to off for release branches and
   1669  1.10    mrg final releases.  The specific files which get <samp>-Werror</samp> are
   1670   1.1    mrg controlled by the Makefiles.
   1671  1.10    mrg </p>
   1672  1.10    mrg </dd>
   1673  1.18    mrg <dt><code>--enable-checking</code></dt>
   1674  1.18    mrg <dt><code>--disable-checking</code></dt>
   1675  1.18    mrg <dt><code>--enable-checking=<var>list</var></code></dt>
   1676  1.11    mrg <dd><p>This option controls performing internal consistency checks in the compiler.
   1677  1.11    mrg It does not change the generated code, but adds error checking of the
   1678  1.11    mrg requested complexity.  This slows down the compiler and may only work
   1679  1.11    mrg properly if you are building the compiler with GCC.
   1680  1.11    mrg </p>
   1681  1.11    mrg <p>When the option is not specified, the active set of checks depends on context.
   1682  1.11    mrg Namely, bootstrap stage 1 defaults to &lsquo;<samp>--enable-checking=yes</samp>&rsquo;, builds
   1683  1.11    mrg from release branches or release archives default to
   1684  1.11    mrg &lsquo;<samp>--enable-checking=release</samp>&rsquo;, and otherwise
   1685  1.11    mrg &lsquo;<samp>--enable-checking=yes,extra</samp>&rsquo; is used.  When the option is
   1686  1.11    mrg specified without a <var>list</var>, the result is the same as
   1687  1.11    mrg &lsquo;<samp>--enable-checking=yes</samp>&rsquo;.  Likewise, &lsquo;<samp>--disable-checking</samp>&rsquo; is
   1688  1.11    mrg equivalent to &lsquo;<samp>--enable-checking=no</samp>&rsquo;.
   1689  1.11    mrg </p>
   1690  1.11    mrg <p>The categories of checks available in <var>list</var> are &lsquo;<samp>yes</samp>&rsquo; (most common
   1691  1.11    mrg checks &lsquo;<samp>assert,misc,gc,gimple,rtlflag,runtime,tree,types</samp>&rsquo;), &lsquo;<samp>no</samp>&rsquo;
   1692  1.11    mrg (no checks at all), &lsquo;<samp>all</samp>&rsquo; (all but &lsquo;<samp>valgrind</samp>&rsquo;), &lsquo;<samp>release</samp>&rsquo;
   1693  1.11    mrg (cheapest checks &lsquo;<samp>assert,runtime</samp>&rsquo;) or &lsquo;<samp>none</samp>&rsquo; (same as &lsquo;<samp>no</samp>&rsquo;).
   1694  1.11    mrg &lsquo;<samp>release</samp>&rsquo; checks are always on and to disable them
   1695  1.11    mrg &lsquo;<samp>--disable-checking</samp>&rsquo; or &lsquo;<samp>--enable-checking=no[,&lt;other checks&gt;]</samp>&rsquo;
   1696  1.11    mrg must be explicitly requested.  Disabling assertions makes the compiler and
   1697  1.11    mrg runtime slightly faster but increases the risk of undetected internal errors
   1698  1.11    mrg causing wrong code to be generated.
   1699  1.11    mrg </p>
   1700  1.11    mrg <p>Individual checks can be enabled with these flags: &lsquo;<samp>assert</samp>&rsquo;, &lsquo;<samp>df</samp>&rsquo;,
   1701  1.11    mrg &lsquo;<samp>extra</samp>&rsquo;, &lsquo;<samp>fold</samp>&rsquo;, &lsquo;<samp>gc</samp>&rsquo;, &lsquo;<samp>gcac</samp>&rsquo;, &lsquo;<samp>gimple</samp>&rsquo;,
   1702  1.11    mrg &lsquo;<samp>misc</samp>&rsquo;, &lsquo;<samp>rtl</samp>&rsquo;, &lsquo;<samp>rtlflag</samp>&rsquo;, &lsquo;<samp>runtime</samp>&rsquo;, &lsquo;<samp>tree</samp>&rsquo;,
   1703  1.11    mrg &lsquo;<samp>types</samp>&rsquo; and &lsquo;<samp>valgrind</samp>&rsquo;.  &lsquo;<samp>extra</samp>&rsquo; extends &lsquo;<samp>misc</samp>&rsquo;
   1704  1.11    mrg checking with extra checks that might affect code generation and should
   1705  1.11    mrg therefore not differ between stage1 and later stages in bootstrap.
   1706  1.11    mrg </p>
   1707  1.11    mrg <p>The &lsquo;<samp>valgrind</samp>&rsquo; check requires the external <code>valgrind</code> simulator,
   1708  1.16    mrg available from <a href="https://valgrind.org">https://valgrind.org</a>.  The &lsquo;<samp>rtl</samp>&rsquo; checks are
   1709  1.11    mrg expensive and the &lsquo;<samp>df</samp>&rsquo;, &lsquo;<samp>gcac</samp>&rsquo; and &lsquo;<samp>valgrind</samp>&rsquo; checks are very
   1710  1.11    mrg expensive.
   1711  1.10    mrg </p>
   1712  1.10    mrg </dd>
   1713  1.18    mrg <dt><code>--disable-stage1-checking</code></dt>
   1714  1.18    mrg <dt><code>--enable-stage1-checking</code></dt>
   1715  1.18    mrg <dt><code>--enable-stage1-checking=<var>list</var></code></dt>
   1716  1.11    mrg <dd><p>This option affects only bootstrap build.  If no <samp>--enable-checking</samp>
   1717  1.11    mrg option is specified the stage1 compiler is built with &lsquo;<samp>yes</samp>&rsquo; checking
   1718  1.11    mrg enabled, otherwise the stage1 checking flags are the same as specified by
   1719  1.10    mrg <samp>--enable-checking</samp>.  To build the stage1 compiler with
   1720  1.10    mrg different checking options use <samp>--enable-stage1-checking</samp>.
   1721  1.10    mrg The list of checking options is the same as for <samp>--enable-checking</samp>.
   1722   1.1    mrg If your system is too slow or too small to bootstrap a released compiler
   1723  1.10    mrg with checking for stage1 enabled, you can use &lsquo;<samp>--disable-stage1-checking</samp>&rsquo;
   1724   1.1    mrg to disable checking for the stage1 compiler.
   1725  1.10    mrg </p>
   1726  1.10    mrg </dd>
   1727  1.18    mrg <dt><code>--enable-coverage</code></dt>
   1728  1.18    mrg <dt><code>--enable-coverage=<var>level</var></code></dt>
   1729  1.10    mrg <dd><p>With this option, the compiler is built to collect self coverage
   1730   1.1    mrg information, every time it is run.  This is for internal development
   1731   1.1    mrg purposes, and only works when the compiler is being built with gcc.  The
   1732   1.1    mrg <var>level</var> argument controls whether the compiler is built optimized or
   1733  1.10    mrg not, values are &lsquo;<samp>opt</samp>&rsquo; and &lsquo;<samp>noopt</samp>&rsquo;.  For coverage analysis you
   1734   1.1    mrg want to disable optimization, for performance analysis you want to
   1735   1.1    mrg enable optimization.  When coverage is enabled, the default level is
   1736   1.1    mrg without optimization.
   1737  1.10    mrg </p>
   1738  1.10    mrg </dd>
   1739  1.18    mrg <dt><code>--enable-gather-detailed-mem-stats</code></dt>
   1740  1.10    mrg <dd><p>When this option is specified more detailed information on memory
   1741   1.1    mrg allocation is gathered.  This information is printed when using
   1742  1.10    mrg <samp>-fmem-report</samp>.
   1743  1.10    mrg </p>
   1744  1.10    mrg </dd>
   1745  1.18    mrg <dt><code>--enable-valgrind-annotations</code></dt>
   1746  1.10    mrg <dd><p>Mark selected memory related operations in the compiler when run under
   1747   1.7    mrg valgrind to suppress false positives.
   1748  1.10    mrg </p>
   1749  1.10    mrg </dd>
   1750  1.18    mrg <dt><code>--enable-nls</code></dt>
   1751  1.18    mrg <dt><code>--disable-nls</code></dt>
   1752  1.10    mrg <dd><p>The <samp>--enable-nls</samp> option enables Native Language Support (NLS),
   1753   1.1    mrg which lets GCC output diagnostics in languages other than American
   1754   1.1    mrg English.  Native Language Support is enabled by default if not doing a
   1755  1.10    mrg canadian cross build.  The <samp>--disable-nls</samp> option disables NLS.
   1756  1.10    mrg </p>
   1757  1.19    mrg <p>Note that this functionality requires either libintl (provided by GNU
   1758  1.19    mrg gettext) or C standard library that contains support for gettext (such
   1759  1.19    mrg as the GNU C Library).
   1760  1.19    mrg See <a href="#with-included-gettext">&ndash;with-included-gettext</a> for more
   1761  1.19    mrg information on the conditions required to get gettext support.
   1762  1.19    mrg </p>
   1763  1.19    mrg </dd>
   1764  1.19    mrg <dt><code>--with-libintl-prefix=<var>dir</var></code></dt>
   1765  1.19    mrg <dt><code>--without-libintl-prefix</code></dt>
   1766  1.19    mrg <dd><p>Searches for libintl in <samp><var>dir</var>/include</samp> and
   1767  1.19    mrg <samp><var>dir</var>/lib</samp>, or disables manual searching for it, letting the
   1768  1.19    mrg linker handle it.
   1769  1.19    mrg </p>
   1770  1.10    mrg </dd>
   1771  1.19    mrg <dt><code>--with-libintl-type=<var>type</var></code></dt>
   1772  1.19    mrg <dd><p>Specifies the type of library to search for when looking for libintl.
   1773  1.19    mrg <var>type</var> can be one of <code>auto</code>, <code>static</code> or <code>shared</code>.
   1774  1.19    mrg </p>
   1775  1.19    mrg <a name="with-included-gettext"></a></dd>
   1776  1.18    mrg <dt><code>--with-included-gettext</code></dt>
   1777  1.19    mrg <dd><p>Only available if <samp>gettext</samp> is present in the source tree.
   1778  1.10    mrg </p>
   1779  1.19    mrg <p>Forces the gettext tree to be configured to build and use a new static
   1780  1.19    mrg libintl, overriding the system libintl.  Results in GCC being built
   1781  1.19    mrg against the newly built libintl rather than the system libintl.
   1782  1.19    mrg </p>
   1783  1.19    mrg <p>The build system makes a somewhat complicated choice when picking where
   1784  1.19    mrg to get gettext routines from.  The following table is a summary of the
   1785  1.19    mrg possible options:
   1786  1.19    mrg </p>
   1787  1.19    mrg <table>
   1788  1.19    mrg <thead><tr><th width="12%">GNU gettext present in sources</th><th width="12%">libintl installed on the system</th><th width="12%"><code>gettext</code> present in libc</th><th width="20%"><code>--with-included-gettext</code></th><th width="44%">Effects on localization</th></tr></thead>
   1789  1.19    mrg <tr><td width="12%">No</td><td width="12%">No</td><td width="12%">No</td><td width="20%">(ignored)</td><td width="44%">No localization</td></tr>
   1790  1.19    mrg <tr><td width="12%">No</td><td width="12%">No</td><td width="12%">Yes</td><td width="20%">(ignored)</td><td width="44%">Localized, libc gettext</td></tr>
   1791  1.19    mrg <tr><td width="12%">No</td><td width="12%">Yes</td><td width="12%">No</td><td width="20%">(ignored)</td><td width="44%">Localized, libintl</td></tr>
   1792  1.19    mrg <tr><td width="12%">No</td><td width="12%">Yes</td><td width="12%">Yes</td><td width="20%">(ignored)</td><td width="44%">Localized, libintl</td></tr>
   1793  1.19    mrg <tr><td width="12%">Yes</td><td width="12%">No</td><td width="12%">No</td><td width="20%">No</td><td width="44%">Localized, new, static libintl</td></tr>
   1794  1.19    mrg <tr><td width="12%">Yes</td><td width="12%">No</td><td width="12%">No</td><td width="20%">Yes</td><td width="44%">Localized, new, static libintl</td></tr>
   1795  1.19    mrg <tr><td width="12%">Yes</td><td width="12%">No</td><td width="12%">Yes</td><td width="20%">No</td><td width="44%">Localized, libc gettext</td></tr>
   1796  1.19    mrg <tr><td width="12%">Yes</td><td width="12%">No</td><td width="12%">Yes</td><td width="20%">Yes</td><td width="44%">Localized, new, static libintl</td></tr>
   1797  1.19    mrg <tr><td width="12%">Yes</td><td width="12%">Yes</td><td width="12%">No</td><td width="20%">No</td><td width="44%">Localized, libintl</td></tr>
   1798  1.19    mrg <tr><td width="12%">Yes</td><td width="12%">Yes</td><td width="12%">No</td><td width="20%">Yes</td><td width="44%">Localized, new, static libintl</td></tr>
   1799  1.19    mrg <tr><td width="12%">Yes</td><td width="12%">Yes</td><td width="12%">Yes</td><td width="20%">No</td><td width="44%">Localized, libintl</td></tr>
   1800  1.19    mrg <tr><td width="12%">Yes</td><td width="12%">Yes</td><td width="12%">Yes</td><td width="20%">Yes</td><td width="44%">Localized, new, static libintl</td></tr>
   1801  1.19    mrg </table>
   1802  1.19    mrg 
   1803  1.10    mrg </dd>
   1804  1.18    mrg <dt><code>--with-catgets</code></dt>
   1805  1.10    mrg <dd><p>If NLS is enabled, and if the host lacks <code>gettext</code> but has the
   1806   1.1    mrg inferior <code>catgets</code> interface, the GCC build procedure normally
   1807  1.10    mrg ignores <code>catgets</code> and instead uses GCC&rsquo;s copy of the GNU
   1808  1.10    mrg <code>gettext</code> library.  The <samp>--with-catgets</samp> option causes the
   1809  1.10    mrg build procedure to use the host&rsquo;s <code>catgets</code> in this situation.
   1810  1.10    mrg </p>
   1811  1.10    mrg </dd>
   1812  1.18    mrg <dt><code>--with-libiconv-prefix=<var>dir</var></code></dt>
   1813  1.10    mrg <dd><p>Search for libiconv header files in <samp><var>dir</var>/include</samp> and
   1814  1.10    mrg libiconv library files in <samp><var>dir</var>/lib</samp>.
   1815  1.10    mrg </p>
   1816  1.10    mrg </dd>
   1817  1.18    mrg <dt><code>--enable-obsolete</code></dt>
   1818  1.10    mrg <dd><p>Enable configuration for an obsoleted system.  If you attempt to
   1819   1.1    mrg configure GCC for a system (build, host, or target) which has been
   1820   1.1    mrg obsoleted, and you do not specify this flag, configure will halt with an
   1821   1.1    mrg error message.
   1822  1.10    mrg </p>
   1823  1.10    mrg <p>All support for systems which have been obsoleted in one release of GCC
   1824   1.1    mrg is removed entirely in the next major release, unless someone steps
   1825   1.1    mrg forward to maintain the port.
   1826  1.10    mrg </p>
   1827  1.10    mrg </dd>
   1828  1.18    mrg <dt><code>--enable-decimal-float</code></dt>
   1829  1.18    mrg <dt><code>--enable-decimal-float=yes</code></dt>
   1830  1.18    mrg <dt><code>--enable-decimal-float=no</code></dt>
   1831  1.18    mrg <dt><code>--enable-decimal-float=bid</code></dt>
   1832  1.18    mrg <dt><code>--enable-decimal-float=dpd</code></dt>
   1833  1.18    mrg <dt><code>--disable-decimal-float</code></dt>
   1834  1.10    mrg <dd><p>Enable (or disable) support for the C decimal floating point extension
   1835  1.19    mrg that is in the IEEE 754-2008 standard.  This is enabled by default
   1836  1.19    mrg only on AArch64, PowerPC, i386, and x86_64 GNU/Linux systems.  Other
   1837  1.19    mrg systems may also support it, but require the user to specifically
   1838  1.19    mrg enable it.  You can optionally control which decimal floating point
   1839  1.19    mrg format is used (either &lsquo;<samp>bid</samp>&rsquo; or &lsquo;<samp>dpd</samp>&rsquo;).  The &lsquo;<samp>bid</samp>&rsquo;
   1840  1.19    mrg (binary integer decimal) format is default on AArch64, i386 and x86_64
   1841  1.19    mrg systems, and the &lsquo;<samp>dpd</samp>&rsquo; (densely packed decimal) format is default
   1842  1.19    mrg on PowerPC systems.
   1843  1.10    mrg </p>
   1844  1.10    mrg </dd>
   1845  1.18    mrg <dt><code>--enable-fixed-point</code></dt>
   1846  1.18    mrg <dt><code>--disable-fixed-point</code></dt>
   1847  1.10    mrg <dd><p>Enable (or disable) support for C fixed-point arithmetic.
   1848   1.1    mrg This option is enabled by default for some targets (such as MIPS) which
   1849   1.1    mrg have hardware-support for fixed-point operations.  On other targets, you
   1850   1.1    mrg may enable this option manually.
   1851  1.10    mrg </p>
   1852  1.10    mrg </dd>
   1853  1.18    mrg <dt><code>--with-long-double-128</code></dt>
   1854  1.10    mrg <dd><p>Specify if <code>long double</code> type should be 128-bit by default on selected
   1855   1.1    mrg GNU/Linux architectures.  If using <code>--without-long-double-128</code>,
   1856  1.10    mrg <code>long double</code> will be by default 64-bit, the same as <code>double</code> type.
   1857   1.1    mrg When neither of these configure options are used, the default will be
   1858   1.1    mrg 128-bit <code>long double</code> when built against GNU C Library 2.4 and later,
   1859   1.1    mrg 64-bit <code>long double</code> otherwise.
   1860  1.10    mrg </p>
   1861  1.10    mrg </dd>
   1862  1.18    mrg <dt><code>--with-long-double-format=ibm</code></dt>
   1863  1.18    mrg <dt><code>--with-long-double-format=ieee</code></dt>
   1864  1.10    mrg <dd><p>Specify whether <code>long double</code> uses the IBM extended double format
   1865  1.10    mrg or the IEEE 128-bit floating point format on PowerPC Linux systems.
   1866  1.10    mrg This configuration switch will only work on little endian PowerPC
   1867  1.10    mrg Linux systems and on big endian 64-bit systems where the default cpu
   1868  1.10    mrg is at least power7 (i.e. <samp>--with-cpu=power7</samp>,
   1869  1.10    mrg <samp>--with-cpu=power8</samp>, or <samp>--with-cpu=power9</samp> is used).
   1870  1.10    mrg </p>
   1871  1.10    mrg <p>If you use the <samp>--with-long-double-64</samp> configuration option,
   1872  1.10    mrg the <samp>--with-long-double-format=ibm</samp> and
   1873  1.10    mrg <samp>--with-long-double-format=ieee</samp> options are ignored.
   1874  1.10    mrg </p>
   1875  1.10    mrg <p>The default <code>long double</code> format is to use IBM extended double.
   1876  1.10    mrg Until all of the libraries are converted to use IEEE 128-bit floating
   1877  1.10    mrg point, it is not recommended to use
   1878  1.10    mrg <samp>--with-long-double-format=ieee</samp>.
   1879  1.10    mrg </p>
   1880  1.10    mrg </dd>
   1881  1.18    mrg <dt><code>--enable-fdpic</code></dt>
   1882  1.10    mrg <dd><p>On SH Linux systems, generate ELF FDPIC code.
   1883  1.10    mrg </p>
   1884  1.10    mrg </dd>
   1885  1.18    mrg <dt><code>--with-gmp=<var>pathname</var></code></dt>
   1886  1.18    mrg <dt><code>--with-gmp-include=<var>pathname</var></code></dt>
   1887  1.18    mrg <dt><code>--with-gmp-lib=<var>pathname</var></code></dt>
   1888  1.18    mrg <dt><code>--with-mpfr=<var>pathname</var></code></dt>
   1889  1.18    mrg <dt><code>--with-mpfr-include=<var>pathname</var></code></dt>
   1890  1.18    mrg <dt><code>--with-mpfr-lib=<var>pathname</var></code></dt>
   1891  1.18    mrg <dt><code>--with-mpc=<var>pathname</var></code></dt>
   1892  1.18    mrg <dt><code>--with-mpc-include=<var>pathname</var></code></dt>
   1893  1.18    mrg <dt><code>--with-mpc-lib=<var>pathname</var></code></dt>
   1894  1.10    mrg <dd><p>If you want to build GCC but do not have the GMP library, the MPFR
   1895   1.1    mrg library and/or the MPC library installed in a standard location and
   1896   1.3  skrll do not have their sources present in the GCC source tree then you
   1897   1.3  skrll can explicitly specify the directory where they are installed
   1898  1.10    mrg (&lsquo;<samp>--with-gmp=<var>gmpinstalldir</var></samp>&rsquo;,
   1899  1.10    mrg &lsquo;<samp>--with-mpfr=<var>mpfrinstalldir</var></samp>&rsquo;,
   1900  1.10    mrg &lsquo;<samp>--with-mpc=<var>mpcinstalldir</var></samp>&rsquo;).  The
   1901  1.10    mrg <samp>--with-gmp=<var>gmpinstalldir</var></samp> option is shorthand for
   1902  1.10    mrg <samp>--with-gmp-lib=<var>gmpinstalldir</var>/lib</samp> and
   1903  1.10    mrg <samp>--with-gmp-include=<var>gmpinstalldir</var>/include</samp>.  Likewise the
   1904  1.10    mrg <samp>--with-mpfr=<var>mpfrinstalldir</var></samp> option is shorthand for
   1905  1.10    mrg <samp>--with-mpfr-lib=<var>mpfrinstalldir</var>/lib</samp> and
   1906  1.10    mrg <samp>--with-mpfr-include=<var>mpfrinstalldir</var>/include</samp>, also the
   1907  1.10    mrg <samp>--with-mpc=<var>mpcinstalldir</var></samp> option is shorthand for
   1908  1.10    mrg <samp>--with-mpc-lib=<var>mpcinstalldir</var>/lib</samp> and
   1909  1.10    mrg <samp>--with-mpc-include=<var>mpcinstalldir</var>/include</samp>.  If these
   1910   1.1    mrg shorthand assumptions are not correct, you can use the explicit
   1911   1.3  skrll include and lib options directly.  You might also need to ensure the
   1912   1.3  skrll shared libraries can be found by the dynamic linker when building and
   1913   1.3  skrll using GCC, for example by setting the runtime shared library path
   1914  1.10    mrg variable (<code>LD_LIBRARY_PATH</code> on GNU/Linux and Solaris systems).
   1915  1.10    mrg </p>
   1916  1.10    mrg <p>These flags are applicable to the host platform only.  When building
   1917   1.3  skrll a cross compiler, they will not be used to configure target libraries.
   1918  1.10    mrg </p>
   1919  1.10    mrg </dd>
   1920  1.18    mrg <dt><code>--with-isl=<var>pathname</var></code></dt>
   1921  1.18    mrg <dt><code>--with-isl-include=<var>pathname</var></code></dt>
   1922  1.18    mrg <dt><code>--with-isl-lib=<var>pathname</var></code></dt>
   1923  1.10    mrg <dd><p>If you do not have the isl library installed in a standard location and you
   1924   1.5    mrg want to build GCC, you can explicitly specify the directory where it is
   1925  1.10    mrg installed (&lsquo;<samp>--with-isl=<var>islinstalldir</var></samp>&rsquo;). The
   1926  1.10    mrg <samp>--with-isl=<var>islinstalldir</var></samp> option is shorthand for
   1927  1.10    mrg <samp>--with-isl-lib=<var>islinstalldir</var>/lib</samp> and
   1928  1.10    mrg <samp>--with-isl-include=<var>islinstalldir</var>/include</samp>. If this
   1929   1.5    mrg shorthand assumption is not correct, you can use the explicit
   1930   1.1    mrg include and lib options directly.
   1931  1.10    mrg </p>
   1932  1.10    mrg <p>These flags are applicable to the host platform only.  When building
   1933   1.3  skrll a cross compiler, they will not be used to configure target libraries.
   1934  1.10    mrg </p>
   1935  1.10    mrg </dd>
   1936  1.18    mrg <dt><code>--with-stage1-ldflags=<var>flags</var></code></dt>
   1937  1.10    mrg <dd><p>This option may be used to set linker flags to be used when linking
   1938   1.1    mrg stage 1 of GCC.  These are also used when linking GCC if configured with
   1939  1.10    mrg <samp>--disable-bootstrap</samp>.  If <samp>--with-stage1-libs</samp> is not set to a
   1940  1.10    mrg value, then the default is &lsquo;<samp>-static-libstdc++ -static-libgcc</samp>&rsquo;, if
   1941   1.7    mrg supported.
   1942  1.10    mrg </p>
   1943  1.10    mrg </dd>
   1944  1.18    mrg <dt><code>--with-stage1-libs=<var>libs</var></code></dt>
   1945  1.10    mrg <dd><p>This option may be used to set libraries to be used when linking stage 1
   1946   1.1    mrg of GCC.  These are also used when linking GCC if configured with
   1947  1.10    mrg <samp>--disable-bootstrap</samp>.
   1948  1.10    mrg </p>
   1949  1.10    mrg </dd>
   1950  1.18    mrg <dt><code>--with-boot-ldflags=<var>flags</var></code></dt>
   1951  1.10    mrg <dd><p>This option may be used to set linker flags to be used when linking
   1952   1.7    mrg stage 2 and later when bootstrapping GCC.  If &ndash;with-boot-libs
   1953   1.7    mrg is not is set to a value, then the default is
   1954  1.10    mrg &lsquo;<samp>-static-libstdc++ -static-libgcc</samp>&rsquo;.
   1955  1.10    mrg </p>
   1956  1.10    mrg </dd>
   1957  1.18    mrg <dt><code>--with-boot-libs=<var>libs</var></code></dt>
   1958  1.10    mrg <dd><p>This option may be used to set libraries to be used when linking stage 2
   1959   1.7    mrg and later when bootstrapping GCC.
   1960  1.10    mrg </p>
   1961  1.10    mrg </dd>
   1962  1.18    mrg <dt><code>--with-debug-prefix-map=<var>map</var></code></dt>
   1963  1.10    mrg <dd><p>Convert source directory names using <samp>-fdebug-prefix-map</samp> when
   1964   1.1    mrg building runtime libraries.  &lsquo;<samp><var>map</var></samp>&rsquo; is a space-separated
   1965  1.10    mrg list of maps of the form &lsquo;<samp><var>old</var>=<var>new</var></samp>&rsquo;.
   1966  1.10    mrg </p>
   1967  1.10    mrg </dd>
   1968  1.18    mrg <dt><code>--enable-linker-build-id</code></dt>
   1969  1.10    mrg <dd><p>Tells GCC to pass <samp>--build-id</samp> option to the linker for all final
   1970  1.10    mrg links (links performed without the <samp>-r</samp> or <samp>--relocatable</samp>
   1971   1.1    mrg option), if the linker supports it.  If you specify
   1972  1.10    mrg <samp>--enable-linker-build-id</samp>, but your linker does not
   1973  1.10    mrg support <samp>--build-id</samp> option, a warning is issued and the
   1974  1.10    mrg <samp>--enable-linker-build-id</samp> option is ignored.  The default is off.
   1975  1.10    mrg </p>
   1976  1.10    mrg </dd>
   1977  1.18    mrg <dt><code>--with-linker-hash-style=<var>choice</var></code></dt>
   1978  1.10    mrg <dd><p>Tells GCC to pass <samp>--hash-style=<var>choice</var></samp> option to the
   1979   1.3  skrll linker for all final links. <var>choice</var> can be one of
   1980  1.10    mrg &lsquo;<samp>sysv</samp>&rsquo;, &lsquo;<samp>gnu</samp>&rsquo;, and &lsquo;<samp>both</samp>&rsquo; where &lsquo;<samp>sysv</samp>&rsquo; is the default.
   1981  1.10    mrg </p>
   1982  1.10    mrg </dd>
   1983  1.18    mrg <dt><code>--enable-gnu-unique-object</code></dt>
   1984  1.18    mrg <dt><code>--disable-gnu-unique-object</code></dt>
   1985  1.10    mrg <dd><p>Tells GCC to use the gnu_unique_object relocation for C++ template
   1986   1.1    mrg static data members and inline function local statics.  Enabled by
   1987   1.5    mrg default for a toolchain with an assembler that accepts it and
   1988   1.1    mrg GLIBC 2.11 or above, otherwise disabled.
   1989  1.10    mrg </p>
   1990  1.10    mrg </dd>
   1991  1.18    mrg <dt><code>--with-diagnostics-color=<var>choice</var></code></dt>
   1992  1.10    mrg <dd><p>Tells GCC to use <var>choice</var> as the default for <samp>-fdiagnostics-color=</samp>
   1993   1.5    mrg option (if not used explicitly on the command line).  <var>choice</var>
   1994  1.10    mrg can be one of &lsquo;<samp>never</samp>&rsquo;, &lsquo;<samp>auto</samp>&rsquo;, &lsquo;<samp>always</samp>&rsquo;, and &lsquo;<samp>auto-if-env</samp>&rsquo;
   1995  1.13    mrg where &lsquo;<samp>auto</samp>&rsquo; is the default.  &lsquo;<samp>auto-if-env</samp>&rsquo; makes
   1996  1.13    mrg <samp>-fdiagnostics-color=auto</samp> the default if <code>GCC_COLORS</code>
   1997  1.13    mrg is present and non-empty in the environment of the compiler, and
   1998  1.10    mrg <samp>-fdiagnostics-color=never</samp> otherwise.
   1999  1.10    mrg </p>
   2000  1.10    mrg </dd>
   2001  1.18    mrg <dt><code>--with-diagnostics-urls=<var>choice</var></code></dt>
   2002  1.13    mrg <dd><p>Tells GCC to use <var>choice</var> as the default for <samp>-fdiagnostics-urls=</samp>
   2003  1.13    mrg option (if not used explicitly on the command line).  <var>choice</var>
   2004  1.13    mrg can be one of &lsquo;<samp>never</samp>&rsquo;, &lsquo;<samp>auto</samp>&rsquo;, &lsquo;<samp>always</samp>&rsquo;, and &lsquo;<samp>auto-if-env</samp>&rsquo;
   2005  1.13    mrg where &lsquo;<samp>auto</samp>&rsquo; is the default.  &lsquo;<samp>auto-if-env</samp>&rsquo; makes
   2006  1.13    mrg <samp>-fdiagnostics-urls=auto</samp> the default if <code>GCC_URLS</code>
   2007  1.13    mrg or <code>TERM_URLS</code> is present and non-empty in the environment of the
   2008  1.13    mrg compiler, and <samp>-fdiagnostics-urls=never</samp> otherwise.
   2009  1.13    mrg </p>
   2010  1.13    mrg </dd>
   2011  1.18    mrg <dt><code>--enable-lto</code></dt>
   2012  1.18    mrg <dt><code>--disable-lto</code></dt>
   2013  1.10    mrg <dd><p>Enable support for link-time optimization (LTO).  This is enabled by
   2014  1.10    mrg default, and may be disabled using <samp>--disable-lto</samp>.
   2015  1.10    mrg </p>
   2016  1.10    mrg </dd>
   2017  1.18    mrg <dt><code>--enable-linker-plugin-configure-flags=FLAGS</code></dt>
   2018  1.18    mrg <dt><code>--enable-linker-plugin-flags=FLAGS</code></dt>
   2019  1.10    mrg <dd><p>By default, linker plugins (such as the LTO plugin) are built for the
   2020   1.5    mrg host system architecture.  For the case that the linker has a
   2021   1.5    mrg different (but run-time compatible) architecture, these flags can be
   2022   1.5    mrg specified to build plugins that are compatible to the linker.  For
   2023   1.5    mrg example, if you are building GCC for a 64-bit x86_64
   2024  1.12    mrg (&lsquo;<samp>x86_64-pc-linux-gnu</samp>&rsquo;) host system, but have a 32-bit x86
   2025  1.10    mrg GNU/Linux (&lsquo;<samp>i686-pc-linux-gnu</samp>&rsquo;) linker executable (which is
   2026   1.5    mrg executable on the former system), you can configure GCC as follows for
   2027   1.5    mrg getting compatible linker plugins:
   2028  1.10    mrg </p>
   2029  1.18    mrg <div class="smallexample">
   2030  1.18    mrg <pre class="smallexample">% <var>srcdir</var>/configure \
   2031  1.12    mrg     --host=x86_64-pc-linux-gnu \
   2032  1.10    mrg     --enable-linker-plugin-configure-flags=--host=i686-pc-linux-gnu \
   2033  1.10    mrg     --enable-linker-plugin-flags='CC=gcc\ -m32\ -Wl,-rpath,[...]/i686-pc-linux-gnu/lib'
   2034  1.10    mrg </pre></div>
   2035  1.10    mrg 
   2036  1.10    mrg </dd>
   2037  1.18    mrg <dt><code>--with-plugin-ld=<var>pathname</var></code></dt>
   2038  1.10    mrg <dd><p>Enable an alternate linker to be used at link-time optimization (LTO)
   2039  1.10    mrg link time when <samp>-fuse-linker-plugin</samp> is enabled.
   2040   1.3  skrll This linker should have plugin support such as gold starting with
   2041  1.10    mrg version 2.20 or GNU ld starting with version 2.21.
   2042  1.10    mrg See <samp>-fuse-linker-plugin</samp> for details.
   2043  1.10    mrg </p>
   2044  1.10    mrg </dd>
   2045  1.18    mrg <dt><code>--enable-canonical-system-headers</code></dt>
   2046  1.18    mrg <dt><code>--disable-canonical-system-headers</code></dt>
   2047  1.10    mrg <dd><p>Enable system header path canonicalization for <samp>libcpp</samp>.  This can
   2048   1.3  skrll produce shorter header file paths in diagnostics and dependency output
   2049   1.3  skrll files, but these changed header paths may conflict with some compilation
   2050   1.3  skrll environments.  Enabled by default, and may be disabled using
   2051  1.10    mrg <samp>--disable-canonical-system-headers</samp>.
   2052  1.10    mrg </p>
   2053  1.10    mrg </dd>
   2054  1.18    mrg <dt><code>--with-glibc-version=<var>major</var>.<var>minor</var></code></dt>
   2055  1.10    mrg <dd><p>Tell GCC that when the GNU C Library (glibc) is used on the target it
   2056   1.5    mrg will be version <var>major</var>.<var>minor</var> or later.  Normally this can
   2057  1.10    mrg be detected from the C library&rsquo;s header files, but this option may be
   2058   1.5    mrg needed when bootstrapping a cross toolchain without the header files
   2059   1.5    mrg available for building the initial bootstrap compiler.
   2060  1.10    mrg </p>
   2061  1.10    mrg <p>If GCC is configured with some multilibs that use glibc and some that
   2062  1.10    mrg do not, this option applies only to the multilibs that use glibc.
   2063   1.5    mrg However, such configurations may not work well as not all the relevant
   2064   1.5    mrg configuration in GCC is on a per-multilib basis.
   2065  1.10    mrg </p>
   2066  1.10    mrg </dd>
   2067  1.18    mrg <dt><code>--enable-as-accelerator-for=<var>target</var></code></dt>
   2068  1.10    mrg <dd><p>Build as offload target compiler. Specify offload host triple by <var>target</var>.
   2069  1.10    mrg </p>
   2070  1.10    mrg </dd>
   2071  1.18    mrg <dt><code>--enable-offload-targets=<var>target1</var>[=<var>path1</var>],&hellip;,<var>targetN</var>[=<var>pathN</var>]</code></dt>
   2072  1.10    mrg <dd><p>Enable offloading to targets <var>target1</var>, &hellip;, <var>targetN</var>.
   2073   1.5    mrg Offload compilers are expected to be already installed.  Default search
   2074   1.5    mrg path for them is <samp><var>exec-prefix</var></samp>, but it can be changed by
   2075  1.10    mrg specifying paths <var>path1</var>, &hellip;, <var>pathN</var>.
   2076  1.10    mrg </p>
   2077  1.18    mrg <div class="smallexample">
   2078  1.18    mrg <pre class="smallexample">% <var>srcdir</var>/configure \
   2079  1.19    mrg     --enable-offload-targets=amdgcn-amdhsa,nvptx-none
   2080  1.10    mrg </pre></div>
   2081   1.7    mrg 
   2082  1.16    mrg </dd>
   2083  1.18    mrg <dt><code>--enable-offload-defaulted</code></dt>
   2084  1.16    mrg <dd>
   2085  1.16    mrg <p>Tell GCC that configured but not installed offload compilers and libgomp
   2086  1.16    mrg plugins are silently ignored.  Useful for distribution compilers where
   2087  1.16    mrg those are in separate optional packages and where the presence or absence
   2088  1.16    mrg of those optional packages should determine the actual supported offloading
   2089  1.16    mrg target set rather than the GCC configure-time selection.
   2090  1.10    mrg </p>
   2091  1.10    mrg </dd>
   2092  1.18    mrg <dt><code>--enable-cet</code></dt>
   2093  1.18    mrg <dt><code>--disable-cet</code></dt>
   2094  1.10    mrg <dd><p>Enable building target run-time libraries with control-flow
   2095  1.10    mrg instrumentation, see <samp>-fcf-protection</samp> option.  When
   2096  1.10    mrg <code>--enable-cet</code> is specified target libraries are configured
   2097  1.10    mrg to add <samp>-fcf-protection</samp> and, if needed, other target
   2098  1.10    mrg specific options to a set of building options.
   2099  1.10    mrg </p>
   2100  1.16    mrg <p><code>--enable-cet=auto</code> is default.  CET is enabled on Linux/x86 if
   2101  1.16    mrg target binutils supports <code>Intel CET</code> instructions and disabled
   2102  1.16    mrg otherwise.  In this case, the target libraries are configured to get
   2103  1.16    mrg additional <samp>-fcf-protection</samp> option.
   2104  1.12    mrg </p>
   2105  1.12    mrg </dd>
   2106  1.18    mrg <dt><code>--with-riscv-attribute=&lsquo;<samp>yes</samp>&rsquo;, &lsquo;<samp>no</samp>&rsquo; or &lsquo;<samp>default</samp>&rsquo;</code></dt>
   2107  1.12    mrg <dd><p>Generate RISC-V attribute by default, in order to record extra build
   2108  1.12    mrg information in object.
   2109  1.12    mrg </p>
   2110  1.12    mrg <p>The option is disabled by default. It is enabled on RISC-V/ELF (bare-metal)
   2111  1.12    mrg target if target binutils supported.
   2112  1.16    mrg </p>
   2113  1.16    mrg </dd>
   2114  1.18    mrg <dt><code>--enable-s390-excess-float-precision</code></dt>
   2115  1.18    mrg <dt><code>--disable-s390-excess-float-precision</code></dt>
   2116  1.16    mrg <dd><p>On s390(x) targets, enable treatment of float expressions with double precision
   2117  1.16    mrg when in standards-compliant mode (e.g., when <code>--std=c99</code> or
   2118  1.16    mrg <code>-fexcess-precision=standard</code> are given).
   2119  1.16    mrg </p>
   2120  1.16    mrg <p>For a native build and cross compiles that have target headers, the option&rsquo;s
   2121  1.16    mrg default is derived from glibc&rsquo;s behavior. When glibc clamps float_t to double,
   2122  1.16    mrg GCC follows and enables the option. For other cross compiles, the default is
   2123  1.16    mrg disabled.
   2124  1.16    mrg </p>
   2125  1.16    mrg </dd>
   2126  1.18    mrg <dt><code>--with-zstd=<var>pathname</var></code></dt>
   2127  1.18    mrg <dt><code>--with-zstd-include=<var>pathname</var></code></dt>
   2128  1.18    mrg <dt><code>--with-zstd-lib=<var>pathname</var></code></dt>
   2129  1.16    mrg <dd><p>If you do not have the <code>zstd</code> library installed in a standard
   2130  1.16    mrg location and you want to build GCC, you can explicitly specify the
   2131  1.16    mrg directory where it is installed (&lsquo;<samp>--with-zstd=<var>zstdinstalldir</var></samp>&rsquo;).
   2132  1.16    mrg The <samp>--with-zstd=<var>zstdinstalldir</var></samp> option is shorthand for
   2133  1.16    mrg <samp>--with-zstd-lib=<var>zstdinstalldir</var>/lib</samp> and
   2134  1.16    mrg <samp>--with-zstd-include=<var>zstdinstalldir</var>/include</samp>. If this
   2135  1.16    mrg shorthand assumption is not correct, you can use the explicit
   2136  1.16    mrg include and lib options directly.
   2137  1.16    mrg </p>
   2138  1.16    mrg <p>These flags are applicable to the host platform only.  When building
   2139  1.16    mrg a cross compiler, they will not be used to configure target libraries.
   2140  1.10    mrg </p></dd>
   2141   1.6    mrg </dl>
   2142   1.1    mrg 
   2143  1.18    mrg <a name="Cross-Compiler-Specific-Options"></a>
   2144  1.18    mrg <h4 class="subheading">Cross-Compiler-Specific Options</h4>
   2145   1.1    mrg <p>The following options only apply to building cross compilers.
   2146  1.10    mrg </p>
   2147  1.10    mrg <dl compact="compact">
   2148  1.18    mrg <dt><code>--with-toolexeclibdir=<var>dir</var></code></dt>
   2149  1.13    mrg <dd><p>Specify the installation directory for libraries built with a cross compiler.
   2150  1.13    mrg The default is <samp>${gcc_tooldir}/lib</samp>.
   2151  1.13    mrg </p>
   2152  1.13    mrg </dd>
   2153  1.18    mrg <dt><code>--with-sysroot</code></dt>
   2154  1.18    mrg <dt><code>--with-sysroot=<var>dir</var></code></dt>
   2155  1.10    mrg <dd><p>Tells GCC to consider <var>dir</var> as the root of a tree that contains
   2156  1.10    mrg (a subset of) the root filesystem of the target operating system.
   2157   1.1    mrg Target system headers, libraries and run-time object files will be
   2158   1.3  skrll searched for in there.  More specifically, this acts as if
   2159  1.10    mrg <samp>--sysroot=<var>dir</var></samp> was added to the default options of the built
   2160   1.1    mrg compiler.  The specified directory is not copied into the
   2161  1.10    mrg install tree, unlike the options <samp>--with-headers</samp> and
   2162  1.10    mrg <samp>--with-libs</samp> that this option obsoletes.  The default value,
   2163  1.10    mrg in case <samp>--with-sysroot</samp> is not given an argument, is
   2164  1.10    mrg <samp>${gcc_tooldir}/sys-root</samp>.  If the specified directory is a
   2165  1.10    mrg subdirectory of <samp>${exec_prefix}</samp>, then it will be found relative to
   2166   1.1    mrg the GCC binaries if the installation tree is moved.
   2167  1.10    mrg </p>
   2168  1.10    mrg <p>This option affects the system root for the compiler used to build
   2169   1.1    mrg target libraries (which runs on the build system) and the compiler newly
   2170   1.1    mrg installed with <code>make install</code>; it does not affect the compiler which is
   2171   1.1    mrg used to build GCC itself.
   2172  1.10    mrg </p>
   2173  1.10    mrg <p>If you specify the <samp>--with-native-system-header-dir=<var>dirname</var></samp>
   2174   1.3  skrll option then the compiler will search that directory within <var>dirname</var> for
   2175  1.10    mrg native system headers rather than the default <samp>/usr/include</samp>.
   2176  1.10    mrg </p>
   2177  1.10    mrg </dd>
   2178  1.18    mrg <dt><code>--with-build-sysroot</code></dt>
   2179  1.18    mrg <dt><code>--with-build-sysroot=<var>dir</var></code></dt>
   2180  1.10    mrg <dd><p>Tells GCC to consider <var>dir</var> as the system root (see
   2181  1.10    mrg <samp>--with-sysroot</samp>) while building target libraries, instead of
   2182  1.10    mrg the directory specified with <samp>--with-sysroot</samp>.  This option is
   2183  1.10    mrg only useful when you are already using <samp>--with-sysroot</samp>.  You
   2184  1.10    mrg can use <samp>--with-build-sysroot</samp> when you are configuring with
   2185  1.10    mrg <samp>--prefix</samp> set to a directory that is different from the one in
   2186   1.1    mrg which you are installing GCC and your target libraries.
   2187  1.10    mrg </p>
   2188  1.10    mrg <p>This option affects the system root for the compiler used to build
   2189   1.1    mrg target libraries (which runs on the build system); it does not affect
   2190   1.1    mrg the compiler which is used to build GCC itself.
   2191  1.10    mrg </p>
   2192  1.10    mrg <p>If you specify the <samp>--with-native-system-header-dir=<var>dirname</var></samp>
   2193   1.3  skrll option then the compiler will search that directory within <var>dirname</var> for
   2194  1.10    mrg native system headers rather than the default <samp>/usr/include</samp>.
   2195  1.10    mrg </p>
   2196  1.10    mrg </dd>
   2197  1.18    mrg <dt><code>--with-headers</code></dt>
   2198  1.18    mrg <dt><code>--with-headers=<var>dir</var></code></dt>
   2199  1.10    mrg <dd><p>Deprecated in favor of <samp>--with-sysroot</samp>.
   2200  1.10    mrg Specifies that target headers are available when building a cross compiler.
   2201   1.1    mrg The <var>dir</var> argument specifies a directory which has the target include
   2202  1.10    mrg files.  These include files will be copied into the <samp>gcc</samp> install
   2203  1.10    mrg directory.  <em>This option with the <var>dir</var> argument is required</em> when
   2204  1.10    mrg building a cross compiler, if <samp><var>prefix</var>/<var>target</var>/sys-include</samp>
   2205  1.10    mrg doesn&rsquo;t pre-exist.  If <samp><var>prefix</var>/<var>target</var>/sys-include</samp> does
   2206  1.10    mrg pre-exist, the <var>dir</var> argument may be omitted.  <code>fixincludes</code>
   2207   1.1    mrg will be run on these files to make them compatible with GCC.
   2208  1.10    mrg </p>
   2209  1.10    mrg </dd>
   2210  1.18    mrg <dt><code>--without-headers</code></dt>
   2211  1.10    mrg <dd><p>Tells GCC not use any target headers from a libc when building a cross
   2212   1.1    mrg compiler.  When crossing to GNU/Linux, you need the headers so GCC
   2213   1.1    mrg can build the exception handling for libgcc.
   2214  1.10    mrg </p>
   2215  1.10    mrg </dd>
   2216  1.18    mrg <dt><code>--with-libs</code></dt>
   2217  1.18    mrg <dt><code>--with-libs=&quot;<var>dir1</var> <var>dir2</var> &hellip; <var>dirN</var>&quot;</code></dt>
   2218  1.10    mrg <dd><p>Deprecated in favor of <samp>--with-sysroot</samp>.
   2219   1.1    mrg Specifies a list of directories which contain the target runtime
   2220  1.10    mrg libraries.  These libraries will be copied into the <samp>gcc</samp> install
   2221   1.1    mrg directory.  If the directory list is omitted, this option has no
   2222   1.1    mrg effect.
   2223  1.10    mrg </p>
   2224  1.10    mrg </dd>
   2225  1.18    mrg <dt><code>--with-newlib</code></dt>
   2226  1.10    mrg <dd><p>Specifies that &lsquo;<samp>newlib</samp>&rsquo; is
   2227   1.1    mrg being used as the target C library.  This causes <code>__eprintf</code> to be
   2228  1.10    mrg omitted from <samp>libgcc.a</samp> on the assumption that it will be provided by
   2229  1.10    mrg &lsquo;<samp>newlib</samp>&rsquo;.
   2230  1.10    mrg </p>
   2231  1.13    mrg <a name="avr"></a>
   2232  1.10    mrg </dd>
   2233  1.18    mrg <dt><code>--with-avrlibc</code></dt>
   2234  1.13    mrg <dd><p>Only supported for the AVR target. Specifies that &lsquo;<samp>AVR-Libc</samp>&rsquo; is
   2235  1.13    mrg being used as the target C&nbsp; library.  This causes float support
   2236  1.10    mrg functions like <code>__addsf3</code> to be omitted from <samp>libgcc.a</samp> on
   2237  1.10    mrg the assumption that it will be provided by <samp>libm.a</samp>.  For more
   2238  1.16    mrg technical details, cf. <a href="https://gcc.gnu.org/PR54461">PR54461</a>.
   2239  1.13    mrg It is not supported for
   2240   1.3  skrll RTEMS configurations, which currently use newlib.  The option is
   2241   1.3  skrll supported since version 4.7.2 and is the default in 4.8.0 and newer.
   2242  1.10    mrg </p>
   2243  1.10    mrg </dd>
   2244  1.18    mrg <dt><code>--with-double={32|64|32,64|64,32}</code></dt>
   2245  1.18    mrg <dt><code>--with-long-double={32|64|32,64|64,32|double}</code></dt>
   2246  1.13    mrg <dd><p>Only supported for the AVR target since version&nbsp;10.
   2247  1.13    mrg Specify the default layout available for the C/C++ &lsquo;<samp>double</samp>&rsquo;
   2248  1.13    mrg and &lsquo;<samp>long double</samp>&rsquo; type, respectively. The following rules apply:
   2249  1.13    mrg </p><ul>
   2250  1.13    mrg <li> The first value after the &lsquo;<samp>=</samp>&rsquo; specifies the default layout (in bits)
   2251  1.13    mrg of the type and also the default for the <samp>-mdouble=</samp> resp.
   2252  1.13    mrg <samp>-mlong-double=</samp> compiler option.
   2253  1.13    mrg </li><li> If more than one value is specified, respective multilib variants are
   2254  1.13    mrg available, and  <samp>-mdouble=</samp> resp. <samp>-mlong-double=</samp> acts
   2255  1.13    mrg as a multilib option.
   2256  1.13    mrg </li><li> If <samp>--with-long-double=double</samp> is specified, &lsquo;<samp>double</samp>&rsquo; and
   2257  1.13    mrg &lsquo;<samp>long double</samp>&rsquo; will have the same layout.
   2258  1.13    mrg </li><li> The defaults are <samp>--with-long-double=64,32</samp> and
   2259  1.13    mrg <samp>--with-double=32,64</samp>.  The default &lsquo;<samp>double</samp>&rsquo; layout imposed by
   2260  1.13    mrg the latter is compatible with older versions of the compiler that implement
   2261  1.13    mrg &lsquo;<samp>double</samp>&rsquo; as a 32-bit type, which does not comply to the language standard.
   2262  1.13    mrg </li></ul>
   2263  1.13    mrg <p>Not all combinations of <samp>--with-double=</samp> and
   2264  1.13    mrg <samp>--with-long-double=</samp> are valid.  For example, the combination
   2265  1.13    mrg <samp>--with-double=32,64</samp> <samp>--with-long-double=32</samp> will be
   2266  1.13    mrg rejected because the first option specifies the availability of
   2267  1.13    mrg multilibs for &lsquo;<samp>double</samp>&rsquo;, whereas the second option implies
   2268  1.13    mrg that &lsquo;<samp>long double</samp>&rsquo; &mdash; and hence also &lsquo;<samp>double</samp>&rsquo; &mdash; is always
   2269  1.13    mrg 32&nbsp;bits wide.
   2270  1.13    mrg </p>
   2271  1.13    mrg </dd>
   2272  1.18    mrg <dt><code>--with-double-comparison={tristate|bool|libf7}</code></dt>
   2273  1.13    mrg <dd><p>Only supported for the AVR target since version&nbsp;10.
   2274  1.13    mrg Specify what result format is returned by library functions that
   2275  1.13    mrg compare 64-bit floating point values (<code>DFmode</code>).
   2276  1.13    mrg The GCC default is &lsquo;<samp>tristate</samp>&rsquo;.  If the floating point
   2277  1.13    mrg implementation returns a boolean instead, set it to &lsquo;<samp>bool</samp>&rsquo;.
   2278  1.13    mrg </p>
   2279  1.13    mrg </dd>
   2280  1.18    mrg <dt><code>--with-libf7={libgcc|math|math-symbols|no}</code></dt>
   2281  1.13    mrg <dd><p>Only supported for the AVR target since version&nbsp;10.
   2282  1.13    mrg Specify to which degree code from LibF7 is included in libgcc.
   2283  1.13    mrg LibF7 is an ad-hoc, AVR-specific, 64-bit floating point emulation
   2284  1.13    mrg written in C and (inline) assembly. &lsquo;<samp>libgcc</samp>&rsquo; adds support
   2285  1.13    mrg for functions that one would usually expect in libgcc like double addition,
   2286  1.13    mrg double comparisons and double conversions. &lsquo;<samp>math</samp>&rsquo; also adds routines
   2287  1.13    mrg that one would expect in <samp>libm.a</samp>, but with <code>__</code> (two underscores)
   2288  1.13    mrg prepended to the symbol names as specified by <samp>math.h</samp>.
   2289  1.13    mrg &lsquo;<samp>math-symbols</samp>&rsquo; also defines weak aliases for the functions
   2290  1.13    mrg declared in <samp>math.h</samp>.  However, <code>--with-libf7</code> won&rsquo;t
   2291  1.13    mrg install no <samp>math.h</samp> header file whatsoever, this file must come
   2292  1.13    mrg from elsewhere.  This option sets <samp>--with-double-comparison</samp>
   2293  1.13    mrg to &lsquo;<samp>bool</samp>&rsquo;.
   2294  1.13    mrg </p>
   2295  1.13    mrg </dd>
   2296  1.18    mrg <dt><code>--with-nds32-lib=<var>library</var></code></dt>
   2297  1.10    mrg <dd><p>Specifies that <var>library</var> setting is used for building <samp>libgcc.a</samp>.
   2298  1.10    mrg Currently, the valid <var>library</var> is &lsquo;<samp>newlib</samp>&rsquo; or &lsquo;<samp>mculib</samp>&rsquo;.
   2299   1.5    mrg This option is only supported for the NDS32 target.
   2300  1.10    mrg </p>
   2301  1.10    mrg </dd>
   2302  1.18    mrg <dt><code>--with-build-time-tools=<var>dir</var></code></dt>
   2303  1.10    mrg <dd><p>Specifies where to find the set of target tools (assembler, linker, etc.)
   2304   1.1    mrg that will be used while building GCC itself.  This option can be useful
   2305   1.1    mrg if the directory layouts are different between the system you are building
   2306   1.1    mrg GCC on, and the system where you will deploy it.
   2307  1.10    mrg </p>
   2308  1.10    mrg <p>For example, on an &lsquo;<samp>ia64-hp-hpux</samp>&rsquo; system, you may have the GNU
   2309  1.10    mrg assembler and linker in <samp>/usr/bin</samp>, and the native tools in a
   2310   1.6    mrg different path, and build a toolchain that expects to find the
   2311  1.10    mrg native tools in <samp>/usr/bin</samp>.
   2312  1.10    mrg </p>
   2313  1.10    mrg <p>When you use this option, you should ensure that <var>dir</var> includes
   2314  1.10    mrg <code>ar</code>, <code>as</code>, <code>ld</code>, <code>nm</code>,
   2315  1.10    mrg <code>ranlib</code> and <code>strip</code> if necessary, and possibly
   2316  1.10    mrg <code>objdump</code>.  Otherwise, GCC may use an inconsistent set of
   2317  1.10    mrg tools.
   2318  1.10    mrg </p></dd>
   2319   1.6    mrg </dl>
   2320   1.1    mrg 
   2321  1.18    mrg <a name="Overriding-configure-test-results"></a>
   2322  1.18    mrg <h4 class="subsubheading">Overriding <code>configure</code> test results</h4>
   2323   1.1    mrg 
   2324   1.6    mrg <p>Sometimes, it might be necessary to override the result of some
   2325  1.10    mrg <code>configure</code> test, for example in order to ease porting to a new
   2326  1.10    mrg system or work around a bug in a test.  The toplevel <code>configure</code>
   2327   1.6    mrg script provides three variables for this:
   2328  1.10    mrg </p>
   2329  1.10    mrg <dl compact="compact">
   2330  1.19    mrg <dd>
   2331  1.19    mrg <a name="index-build_005fconfigargs"></a>
   2332  1.19    mrg </dd>
   2333  1.18    mrg <dt><code>build_configargs</code></dt>
   2334  1.19    mrg <dd><p>The contents of this variable is passed to all build <code>configure</code>
   2335   1.6    mrg scripts.
   2336  1.10    mrg </p>
   2337  1.19    mrg <a name="index-host_005fconfigargs"></a>
   2338  1.10    mrg </dd>
   2339  1.18    mrg <dt><code>host_configargs</code></dt>
   2340  1.19    mrg <dd><p>The contents of this variable is passed to all host <code>configure</code>
   2341   1.6    mrg scripts.
   2342  1.10    mrg </p>
   2343  1.19    mrg <a name="index-target_005fconfigargs"></a>
   2344  1.10    mrg </dd>
   2345  1.18    mrg <dt><code>target_configargs</code></dt>
   2346  1.19    mrg <dd><p>The contents of this variable is passed to all target <code>configure</code>
   2347   1.6    mrg scripts.
   2348  1.10    mrg </p>
   2349  1.10    mrg </dd>
   2350  1.10    mrg </dl>
   2351   1.1    mrg 
   2352  1.10    mrg <p>In order to avoid shell and <code>make</code> quoting issues for complex
   2353  1.10    mrg overrides, you can pass a setting for <code>CONFIG_SITE</code> and set
   2354   1.6    mrg variables in the site file.
   2355  1.10    mrg </p>
   2356  1.18    mrg <a name="Objective-C-Specific-Options"></a>
   2357  1.18    mrg <h4 class="subheading">Objective-C-Specific Options</h4>
   2358   1.7    mrg 
   2359   1.9    mrg <p>The following options apply to the build of the Objective-C runtime library.
   2360  1.10    mrg </p>
   2361  1.10    mrg <dl compact="compact">
   2362  1.18    mrg <dt><code>--enable-objc-gc</code></dt>
   2363  1.10    mrg <dd><p>Specify that an additional variant of the GNU Objective-C runtime library
   2364   1.9    mrg is built, using an external build of the Boehm-Demers-Weiser garbage
   2365  1.13    mrg collector (<a href="https://www.hboehm.info/gc/">https://www.hboehm.info/gc/</a>).  This library needs to be
   2366   1.9    mrg available for each multilib variant, unless configured with
   2367  1.10    mrg <samp>--enable-objc-gc=&lsquo;<samp>auto</samp>&rsquo;</samp> in which case the build of the
   2368   1.9    mrg additional runtime library is skipped when not available and the build
   2369   1.9    mrg continues.
   2370  1.10    mrg </p>
   2371  1.10    mrg </dd>
   2372  1.18    mrg <dt><code>--with-target-bdw-gc=<var>list</var></code></dt>
   2373  1.18    mrg <dt><code>--with-target-bdw-gc-include=<var>list</var></code></dt>
   2374  1.18    mrg <dt><code>--with-target-bdw-gc-lib=<var>list</var></code></dt>
   2375  1.10    mrg <dd><p>Specify search directories for the garbage collector header files and
   2376   1.9    mrg libraries. <var>list</var> is a comma separated list of key value pairs of the
   2377  1.10    mrg form &lsquo;<samp><var>multilibdir</var>=<var>path</var></samp>&rsquo;, where the default multilib key
   2378  1.10    mrg is named as &lsquo;<samp>.</samp>&rsquo; (dot), or is omitted (e.g.
   2379  1.10    mrg &lsquo;<samp>--with-target-bdw-gc=/opt/bdw-gc,32=/opt-bdw-gc32</samp>&rsquo;).
   2380  1.10    mrg </p>
   2381  1.10    mrg <p>The options <samp>--with-target-bdw-gc-include</samp> and
   2382  1.10    mrg <samp>--with-target-bdw-gc-lib</samp> must always be specified together
   2383   1.9    mrg for each multilib variant and they take precedence over
   2384  1.10    mrg <samp>--with-target-bdw-gc</samp>.  If <samp>--with-target-bdw-gc-include</samp>
   2385   1.9    mrg is missing values for a multilib, then the value for the default
   2386  1.10    mrg multilib is used (e.g. &lsquo;<samp>--with-target-bdw-gc-include=/opt/bdw-gc/include</samp>&rsquo;
   2387  1.10    mrg &lsquo;<samp>--with-target-bdw-gc-lib=/opt/bdw-gc/lib64,32=/opt-bdw-gc/lib32</samp>&rsquo;).
   2388   1.9    mrg If none of these options are specified, the library is assumed in
   2389  1.10    mrg default locations.
   2390  1.10    mrg </p></dd>
   2391   1.6    mrg </dl>
   2392   1.1    mrg 
   2393  1.18    mrg <a name="D-Specific-Options"></a>
   2394  1.18    mrg <h4 class="subheading">D-Specific Options</h4>
   2395  1.12    mrg 
   2396  1.12    mrg <p>The following options apply to the build of the D runtime library.
   2397  1.12    mrg </p>
   2398  1.12    mrg <dl compact="compact">
   2399  1.18    mrg <dt><code>--enable-libphobos-checking</code></dt>
   2400  1.18    mrg <dt><code>--disable-libphobos-checking</code></dt>
   2401  1.18    mrg <dt><code>--enable-libphobos-checking=<var>list</var></code></dt>
   2402  1.13    mrg <dd><p>This option controls whether run-time checks and contracts are compiled into
   2403  1.13    mrg the D runtime library.  When the option is not specified, the library is built
   2404  1.13    mrg with &lsquo;<samp>release</samp>&rsquo; checking.  When the option is specified without a
   2405  1.13    mrg <var>list</var>, the result is the same as &lsquo;<samp>--enable-libphobos-checking=yes</samp>&rsquo;.
   2406  1.13    mrg Likewise, &lsquo;<samp>--disable-libphobos-checking</samp>&rsquo; is equivalent to
   2407  1.13    mrg &lsquo;<samp>--enable-libphobos-checking=no</samp>&rsquo;.
   2408  1.13    mrg </p>
   2409  1.13    mrg <p>The categories of checks available in <var>list</var> are &lsquo;<samp>yes</samp>&rsquo; (compiles
   2410  1.13    mrg libphobos with <samp>-fno-release</samp>), &lsquo;<samp>no</samp>&rsquo; (compiles libphobos with
   2411  1.13    mrg <samp>-frelease</samp>), &lsquo;<samp>all</samp>&rsquo; (same as &lsquo;<samp>yes</samp>&rsquo;), &lsquo;<samp>none</samp>&rsquo; or
   2412  1.13    mrg &lsquo;<samp>release</samp>&rsquo; (same as &lsquo;<samp>no</samp>&rsquo;).
   2413  1.13    mrg </p>
   2414  1.13    mrg <p>Individual checks available in <var>list</var> are &lsquo;<samp>assert</samp>&rsquo; (compiles libphobos
   2415  1.13    mrg with an extra option <samp>-fassert</samp>).
   2416  1.13    mrg </p>
   2417  1.13    mrg </dd>
   2418  1.18    mrg <dt><code>--with-libphobos-druntime-only</code></dt>
   2419  1.18    mrg <dt><code>--with-libphobos-druntime-only=<var>choice</var></code></dt>
   2420  1.13    mrg <dd><p>Specify whether to build only the core D runtime library (druntime), or both
   2421  1.13    mrg the core and standard library (phobos) into libphobos.  This is useful for
   2422  1.13    mrg targets that have full support in druntime, but no or incomplete support
   2423  1.13    mrg in phobos.  <var>choice</var> can be one of &lsquo;<samp>auto</samp>&rsquo;, &lsquo;<samp>yes</samp>&rsquo;, and &lsquo;<samp>no</samp>&rsquo;
   2424  1.13    mrg where &lsquo;<samp>auto</samp>&rsquo; is the default.
   2425  1.13    mrg </p>
   2426  1.13    mrg <p>When the option is not specified, the default choice &lsquo;<samp>auto</samp>&rsquo; means that it
   2427  1.13    mrg is inferred whether the target has support for the phobos standard library.
   2428  1.13    mrg When the option is specified without a <var>choice</var>,  the result is the same as
   2429  1.13    mrg &lsquo;<samp>--with-libphobos-druntime-only=yes</samp>&rsquo;.
   2430  1.13    mrg </p>
   2431  1.13    mrg </dd>
   2432  1.18    mrg <dt><code>--with-target-system-zlib</code></dt>
   2433  1.12    mrg <dd><p>Use installed &lsquo;<samp>zlib</samp>&rsquo; rather than that included with GCC.  This needs
   2434  1.12    mrg to be available for each multilib variant, unless configured with
   2435  1.12    mrg <samp>--with-target-system-zlib=&lsquo;<samp>auto</samp>&rsquo;</samp> in which case the GCC&nbsp;included
   2436  1.12    mrg &lsquo;<samp>zlib</samp>&rsquo; is only used when the system installed library is not available.
   2437  1.12    mrg </p></dd>
   2438  1.12    mrg </dl>
   2439  1.12    mrg 
   2440  1.10    mrg <hr />
   2441  1.10    mrg <p>
   2442   1.8    mrg <p><a href="./index.html">Return to the GCC Installation page</a>
   2443  1.10    mrg </p>
   2444  1.10    mrg 
   2445  1.10    mrg 
   2446  1.10    mrg 
   2447  1.10    mrg 
   2448  1.10    mrg 
   2449  1.10    mrg 
   2450  1.10    mrg 
   2451  1.10    mrg 
   2452  1.10    mrg 
   2453   1.1    mrg 
   2454  1.10    mrg </body>
   2455  1.10    mrg </html>
   2456