configure.html revision 1.1.1.11
1 <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd"> 2 <html> 3 <!-- Copyright (C) 1988-2018 Free Software Foundation, Inc. 4 5 Permission is granted to copy, distribute and/or modify this document 6 under the terms of the GNU Free Documentation License, Version 1.3 or 7 any later version published by the Free Software Foundation; with no 8 Invariant Sections, the Front-Cover texts being (a) (see below), and 9 with the Back-Cover Texts being (b) (see below). A copy of the 10 license is included in the section entitled "GNU 11 Free Documentation License". 12 13 (a) The FSF's Front-Cover Text is: 14 15 A GNU Manual 16 17 (b) The FSF's Back-Cover Text is: 18 19 You have freedom to copy and modify this GNU Manual, like GNU 20 software. Copies published by the Free Software Foundation raise 21 funds for GNU development. --> 22 <!-- Created by GNU Texinfo 6.5, http://www.gnu.org/software/texinfo/ --> 23 <head> 24 <meta http-equiv="Content-Type" content="text/html; charset=utf-8"> 25 <title>Installing GCC</title> 26 27 <meta name="description" content="Installing GCC"> 28 <meta name="keywords" content="Installing GCC"> 29 <meta name="resource-type" content="document"> 30 <meta name="distribution" content="global"> 31 <meta name="Generator" content="makeinfo"> 32 <style type="text/css"> 33 <!-- 34 a.summary-letter {text-decoration: none} 35 blockquote.indentedblock {margin-right: 0em} 36 blockquote.smallindentedblock {margin-right: 0em; font-size: smaller} 37 blockquote.smallquotation {font-size: smaller} 38 div.display {margin-left: 3.2em} 39 div.example {margin-left: 3.2em} 40 div.lisp {margin-left: 3.2em} 41 div.smalldisplay {margin-left: 3.2em} 42 div.smallexample {margin-left: 3.2em} 43 div.smalllisp {margin-left: 3.2em} 44 kbd {font-style: oblique} 45 pre.display {font-family: inherit} 46 pre.format {font-family: inherit} 47 pre.menu-comment {font-family: serif} 48 pre.menu-preformatted {font-family: serif} 49 pre.smalldisplay {font-family: inherit; font-size: smaller} 50 pre.smallexample {font-size: smaller} 51 pre.smallformat {font-family: inherit; font-size: smaller} 52 pre.smalllisp {font-size: smaller} 53 span.nolinebreak {white-space: nowrap} 54 span.roman {font-family: initial; font-weight: normal} 55 span.sansserif {font-family: sans-serif; font-weight: normal} 56 ul.no-bullet {list-style: none} 57 --> 58 </style> 59 60 61 </head> 62 63 <body lang="en"> 64 <h1 class="settitle" align="center">Installing GCC</h1> 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 <a name="index-Configuration"></a> 85 <a name="index-Installing-GCC_003a-Configuration"></a> 86 87 <p>Like most GNU software, GCC must be configured before it can be built. 88 This document describes the recommended configuration procedure 89 for both native and cross targets. 90 </p> 91 <p>We use <var>srcdir</var> to refer to the toplevel source directory for 92 GCC; we use <var>objdir</var> to refer to the toplevel build/object directory. 93 </p> 94 <p>If you obtained the sources via SVN, <var>srcdir</var> must refer to the top 95 <samp>gcc</samp> directory, the one where the <samp>MAINTAINERS</samp> file can be 96 found, and not its <samp>gcc</samp> subdirectory, otherwise the build will fail. 97 </p> 98 <p>If either <var>srcdir</var> or <var>objdir</var> is located on an automounted NFS 99 file system, the shell’s built-in <code>pwd</code> command will return 100 temporary pathnames. Using these can lead to various sorts of build 101 problems. To avoid this issue, set the <code>PWDCMD</code> environment 102 variable to an automounter-aware <code>pwd</code> command, e.g., 103 <code>pawd</code> or ‘<samp>amq -w</samp>’, during the configuration and build 104 phases. 105 </p> 106 <p>First, we <strong>highly</strong> recommend that GCC be built into a 107 separate directory from the sources which does <strong>not</strong> reside 108 within the source tree. This is how we generally build GCC; building 109 where <var>srcdir</var> == <var>objdir</var> should still work, but doesn’t 110 get extensive testing; building where <var>objdir</var> is a subdirectory 111 of <var>srcdir</var> is unsupported. 112 </p> 113 <p>If you have previously built GCC in the same directory for a 114 different target machine, do ‘<samp>make distclean</samp>’ to delete all files 115 that might be invalid. One of the files this deletes is <samp>Makefile</samp>; 116 if ‘<samp>make distclean</samp>’ complains that <samp>Makefile</samp> does not exist 117 or issues a message like “don’t know how to make distclean” it probably 118 means that the directory is already suitably clean. However, with the 119 recommended method of building in a separate <var>objdir</var>, you should 120 simply use a different <var>objdir</var> for each target. 121 </p> 122 <p>Second, when configuring a native system, either <code>cc</code> or 123 <code>gcc</code> must be in your path or you must set <code>CC</code> in 124 your environment before running configure. Otherwise the configuration 125 scripts may fail. 126 </p> 127 128 <p>To configure GCC: 129 </p> 130 <div class="smallexample"> 131 <pre class="smallexample">% mkdir <var>objdir</var> 132 % cd <var>objdir</var> 133 % <var>srcdir</var>/configure [<var>options</var>] [<var>target</var>] 134 </pre></div> 135 136 <a name="Distributor-options"></a> 137 <h3 class="heading">Distributor options</h3> 138 139 <p>If you will be distributing binary versions of GCC, with modifications 140 to the source code, you should use the options described in this 141 section to make clear that your version contains modifications. 142 </p> 143 <dl compact="compact"> 144 <dt><code>--with-pkgversion=<var>version</var></code></dt> 145 <dd><p>Specify a string that identifies your package. You may wish 146 to include a build number or build date. This version string will be 147 included in the output of <code>gcc --version</code>. This suffix does 148 not replace the default version string, only the ‘<samp>GCC</samp>’ part. 149 </p> 150 <p>The default value is ‘<samp>GCC</samp>’. 151 </p> 152 </dd> 153 <dt><code>--with-bugurl=<var>url</var></code></dt> 154 <dd><p>Specify the URL that users should visit if they wish to report a bug. 155 You are of course welcome to forward bugs reported to you to the FSF, 156 if you determine that they are not bugs in your modifications. 157 </p> 158 <p>The default value refers to the FSF’s GCC bug tracker. 159 </p> 160 </dd> 161 </dl> 162 163 <a name="Target-specification"></a> 164 <h3 class="heading">Target specification</h3> 165 <ul> 166 <li> GCC has code to correctly determine the correct value for <var>target</var> 167 for nearly all native systems. Therefore, we highly recommend you do 168 not provide a configure target when configuring a native compiler. 169 170 </li><li> <var>target</var> must be specified as <samp>--target=<var>target</var></samp> 171 when configuring a cross compiler; examples of valid targets would be 172 m68k-elf, sh-elf, etc. 173 174 </li><li> Specifying just <var>target</var> instead of <samp>--target=<var>target</var></samp> 175 implies that the host defaults to <var>target</var>. 176 </li></ul> 177 178 179 <a name="Options-specification"></a> 180 <h3 class="heading">Options specification</h3> 181 182 <p>Use <var>options</var> to override several configure time options for 183 GCC. A list of supported <var>options</var> follows; ‘<samp>configure 184 --help</samp>’ may list other options, but those not listed below may not 185 work and should not normally be used. 186 </p> 187 <p>Note that each <samp>--enable</samp> option has a corresponding 188 <samp>--disable</samp> option and that each <samp>--with</samp> option has a 189 corresponding <samp>--without</samp> option. 190 </p> 191 <dl compact="compact"> 192 <dt><code>--prefix=<var>dirname</var></code></dt> 193 <dd><p>Specify the toplevel installation 194 directory. This is the recommended way to install the tools into a directory 195 other than the default. The toplevel installation directory defaults to 196 <samp>/usr/local</samp>. 197 </p> 198 <p>We <strong>highly</strong> recommend against <var>dirname</var> being the same or a 199 subdirectory of <var>objdir</var> or vice versa. If specifying a directory 200 beneath a user’s home directory tree, some shells will not expand 201 <var>dirname</var> correctly if it contains the ‘<samp>~</samp>’ metacharacter; use 202 <code>$HOME</code> instead. 203 </p> 204 <p>The following standard <code>autoconf</code> options are supported. Normally you 205 should not need to use these options. 206 </p><dl compact="compact"> 207 <dt><code>--exec-prefix=<var>dirname</var></code></dt> 208 <dd><p>Specify the toplevel installation directory for architecture-dependent 209 files. The default is <samp><var>prefix</var></samp>. 210 </p> 211 </dd> 212 <dt><code>--bindir=<var>dirname</var></code></dt> 213 <dd><p>Specify the installation directory for the executables called by users 214 (such as <code>gcc</code> and <code>g++</code>). The default is 215 <samp><var>exec-prefix</var>/bin</samp>. 216 </p> 217 </dd> 218 <dt><code>--libdir=<var>dirname</var></code></dt> 219 <dd><p>Specify the installation directory for object code libraries and 220 internal data files of GCC. The default is <samp><var>exec-prefix</var>/lib</samp>. 221 </p> 222 </dd> 223 <dt><code>--libexecdir=<var>dirname</var></code></dt> 224 <dd><p>Specify the installation directory for internal executables of GCC. 225 The default is <samp><var>exec-prefix</var>/libexec</samp>. 226 </p> 227 </dd> 228 <dt><code>--with-slibdir=<var>dirname</var></code></dt> 229 <dd><p>Specify the installation directory for the shared libgcc library. The 230 default is <samp><var>libdir</var></samp>. 231 </p> 232 </dd> 233 <dt><code>--datarootdir=<var>dirname</var></code></dt> 234 <dd><p>Specify the root of the directory tree for read-only architecture-independent 235 data files referenced by GCC. The default is <samp><var>prefix</var>/share</samp>. 236 </p> 237 </dd> 238 <dt><code>--infodir=<var>dirname</var></code></dt> 239 <dd><p>Specify the installation directory for documentation in info format. 240 The default is <samp><var>datarootdir</var>/info</samp>. 241 </p> 242 </dd> 243 <dt><code>--datadir=<var>dirname</var></code></dt> 244 <dd><p>Specify the installation directory for some architecture-independent 245 data files referenced by GCC. The default is <samp><var>datarootdir</var></samp>. 246 </p> 247 </dd> 248 <dt><code>--docdir=<var>dirname</var></code></dt> 249 <dd><p>Specify the installation directory for documentation files (other 250 than Info) for GCC. The default is <samp><var>datarootdir</var>/doc</samp>. 251 </p> 252 </dd> 253 <dt><code>--htmldir=<var>dirname</var></code></dt> 254 <dd><p>Specify the installation directory for HTML documentation files. 255 The default is <samp><var>docdir</var></samp>. 256 </p> 257 </dd> 258 <dt><code>--pdfdir=<var>dirname</var></code></dt> 259 <dd><p>Specify the installation directory for PDF documentation files. 260 The default is <samp><var>docdir</var></samp>. 261 </p> 262 </dd> 263 <dt><code>--mandir=<var>dirname</var></code></dt> 264 <dd><p>Specify the installation directory for manual pages. The default is 265 <samp><var>datarootdir</var>/man</samp>. (Note that the manual pages are only extracts 266 from the full GCC manuals, which are provided in Texinfo format. The manpages 267 are derived by an automatic conversion process from parts of the full 268 manual.) 269 </p> 270 </dd> 271 <dt><code>--with-gxx-include-dir=<var>dirname</var></code></dt> 272 <dd><p>Specify 273 the installation directory for G++ header files. The default depends 274 on other configuration options, and differs between cross and native 275 configurations. 276 </p> 277 </dd> 278 <dt><code>--with-specs=<var>specs</var></code></dt> 279 <dd><p>Specify additional command line driver SPECS. 280 This can be useful if you need to turn on a non-standard feature by 281 default without modifying the compiler’s source code, for instance 282 <samp>--with-specs=%{!fcommon:%{!fno-common:-fno-common}}</samp>. 283 See “Spec Files” in the main manual 284 </p> 285 </dd> 286 </dl> 287 288 </dd> 289 <dt><code>--program-prefix=<var>prefix</var></code></dt> 290 <dd><p>GCC supports some transformations of the names of its programs when 291 installing them. This option prepends <var>prefix</var> to the names of 292 programs to install in <var>bindir</var> (see above). For example, specifying 293 <samp>--program-prefix=foo-</samp> would result in ‘<samp>gcc</samp>’ 294 being installed as <samp>/usr/local/bin/foo-gcc</samp>. 295 </p> 296 </dd> 297 <dt><code>--program-suffix=<var>suffix</var></code></dt> 298 <dd><p>Appends <var>suffix</var> to the names of programs to install in <var>bindir</var> 299 (see above). For example, specifying <samp>--program-suffix=-3.1</samp> 300 would result in ‘<samp>gcc</samp>’ being installed as 301 <samp>/usr/local/bin/gcc-3.1</samp>. 302 </p> 303 </dd> 304 <dt><code>--program-transform-name=<var>pattern</var></code></dt> 305 <dd><p>Applies the ‘<samp>sed</samp>’ script <var>pattern</var> to be applied to the names 306 of programs to install in <var>bindir</var> (see above). <var>pattern</var> has to 307 consist of one or more basic ‘<samp>sed</samp>’ editing commands, separated by 308 semicolons. For example, if you want the ‘<samp>gcc</samp>’ program name to be 309 transformed to the installed program <samp>/usr/local/bin/myowngcc</samp> and 310 the ‘<samp>g++</samp>’ program name to be transformed to 311 <samp>/usr/local/bin/gspecial++</samp> without changing other program names, 312 you could use the pattern 313 <samp>--program-transform-name='s/^gcc$/myowngcc/; s/^g++$/gspecial++/'</samp> 314 to achieve this effect. 315 </p> 316 <p>All three options can be combined and used together, resulting in more 317 complex conversion patterns. As a basic rule, <var>prefix</var> (and 318 <var>suffix</var>) are prepended (appended) before further transformations 319 can happen with a special transformation script <var>pattern</var>. 320 </p> 321 <p>As currently implemented, this option only takes effect for native 322 builds; cross compiler binaries’ names are not transformed even when a 323 transformation is explicitly asked for by one of these options. 324 </p> 325 <p>For native builds, some of the installed programs are also installed 326 with the target alias in front of their name, as in 327 ‘<samp>i686-pc-linux-gnu-gcc</samp>’. All of the above transformations happen 328 before the target alias is prepended to the name—so, specifying 329 <samp>--program-prefix=foo-</samp> and <samp>program-suffix=-3.1</samp>, the 330 resulting binary would be installed as 331 <samp>/usr/local/bin/i686-pc-linux-gnu-foo-gcc-3.1</samp>. 332 </p> 333 <p>As a last shortcoming, none of the installed Ada programs are 334 transformed yet, which will be fixed in some time. 335 </p> 336 </dd> 337 <dt><code>--with-local-prefix=<var>dirname</var></code></dt> 338 <dd><p>Specify the 339 installation directory for local include files. The default is 340 <samp>/usr/local</samp>. Specify this option if you want the compiler to 341 search directory <samp><var>dirname</var>/include</samp> for locally installed 342 header files <em>instead</em> of <samp>/usr/local/include</samp>. 343 </p> 344 <p>You should specify <samp>--with-local-prefix</samp> <strong>only</strong> if your 345 site has a different convention (not <samp>/usr/local</samp>) for where to put 346 site-specific files. 347 </p> 348 <p>The default value for <samp>--with-local-prefix</samp> is <samp>/usr/local</samp> 349 regardless of the value of <samp>--prefix</samp>. Specifying 350 <samp>--prefix</samp> has no effect on which directory GCC searches for 351 local header files. This may seem counterintuitive, but actually it is 352 logical. 353 </p> 354 <p>The purpose of <samp>--prefix</samp> is to specify where to <em>install 355 GCC</em>. The local header files in <samp>/usr/local/include</samp>—if you put 356 any in that directory—are not part of GCC. They are part of other 357 programs—perhaps many others. (GCC installs its own header files in 358 another directory which is based on the <samp>--prefix</samp> value.) 359 </p> 360 <p>Both the local-prefix include directory and the GCC-prefix include 361 directory are part of GCC’s “system include” directories. Although these 362 two directories are not fixed, they need to be searched in the proper 363 order for the correct processing of the include_next directive. The 364 local-prefix include directory is searched before the GCC-prefix 365 include directory. Another characteristic of system include directories 366 is that pedantic warnings are turned off for headers in these directories. 367 </p> 368 <p>Some autoconf macros add <samp>-I <var>directory</var></samp> options to the 369 compiler command line, to ensure that directories containing installed 370 packages’ headers are searched. When <var>directory</var> is one of GCC’s 371 system include directories, GCC will ignore the option so that system 372 directories continue to be processed in the correct order. This 373 may result in a search order different from what was specified but the 374 directory will still be searched. 375 </p> 376 <p>GCC automatically searches for ordinary libraries using 377 <code>GCC_EXEC_PREFIX</code>. Thus, when the same installation prefix is 378 used for both GCC and packages, GCC will automatically search for 379 both headers and libraries. This provides a configuration that is 380 easy to use. GCC behaves in a manner similar to that when it is 381 installed as a system compiler in <samp>/usr</samp>. 382 </p> 383 <p>Sites that need to install multiple versions of GCC may not want to 384 use the above simple configuration. It is possible to use the 385 <samp>--program-prefix</samp>, <samp>--program-suffix</samp> and 386 <samp>--program-transform-name</samp> options to install multiple versions 387 into a single directory, but it may be simpler to use different prefixes 388 and the <samp>--with-local-prefix</samp> option to specify the location of the 389 site-specific files for each version. It will then be necessary for 390 users to specify explicitly the location of local site libraries 391 (e.g., with <code>LIBRARY_PATH</code>). 392 </p> 393 <p>The same value can be used for both <samp>--with-local-prefix</samp> and 394 <samp>--prefix</samp> provided it is not <samp>/usr</samp>. This can be used 395 to avoid the default search of <samp>/usr/local/include</samp>. 396 </p> 397 <p><strong>Do not</strong> specify <samp>/usr</samp> as the <samp>--with-local-prefix</samp>! 398 The directory you use for <samp>--with-local-prefix</samp> <strong>must not</strong> 399 contain any of the system’s standard header files. If it did contain 400 them, certain programs would be miscompiled (including GNU Emacs, on 401 certain targets), because this would override and nullify the header 402 file corrections made by the <code>fixincludes</code> script. 403 </p> 404 <p>Indications are that people who use this option use it based on mistaken 405 ideas of what it is for. People use it as if it specified where to 406 install part of GCC. Perhaps they make this assumption because 407 installing GCC creates the directory. 408 </p> 409 </dd> 410 <dt><code>--with-gcc-major-version-only</code></dt> 411 <dd><p>Specifies that GCC should use only the major number rather than 412 <var>major</var>.<var>minor</var>.<var>patchlevel</var> in filesystem paths. 413 </p> 414 </dd> 415 <dt><code>--with-native-system-header-dir=<var>dirname</var></code></dt> 416 <dd><p>Specifies that <var>dirname</var> is the directory that contains native system 417 header files, rather than <samp>/usr/include</samp>. This option is most useful 418 if you are creating a compiler that should be isolated from the system 419 as much as possible. It is most commonly used with the 420 <samp>--with-sysroot</samp> option and will cause GCC to search 421 <var>dirname</var> inside the system root specified by that option. 422 </p> 423 </dd> 424 <dt><code>--enable-shared[=<var>package</var>[,…]]</code></dt> 425 <dd><p>Build shared versions of libraries, if shared libraries are supported on 426 the target platform. Unlike GCC 2.95.x and earlier, shared libraries 427 are enabled by default on all platforms that support shared libraries. 428 </p> 429 <p>If a list of packages is given as an argument, build shared libraries 430 only for the listed packages. For other packages, only static libraries 431 will be built. Package names currently recognized in the GCC tree are 432 ‘<samp>libgcc</samp>’ (also known as ‘<samp>gcc</samp>’), ‘<samp>libstdc++</samp>’ (not 433 ‘<samp>libstdc++-v3</samp>’), ‘<samp>libffi</samp>’, ‘<samp>zlib</samp>’, ‘<samp>boehm-gc</samp>’, 434 ‘<samp>ada</samp>’, ‘<samp>libada</samp>’, ‘<samp>libgo</samp>’, and ‘<samp>libobjc</samp>’. 435 Note ‘<samp>libiberty</samp>’ does not support shared libraries at all. 436 </p> 437 <p>Use <samp>--disable-shared</samp> to build only static libraries. Note that 438 <samp>--disable-shared</samp> does not accept a list of package names as 439 argument, only <samp>--enable-shared</samp> does. 440 </p> 441 <p>Contrast with <samp>--enable-host-shared</samp>, which affects <em>host</em> 442 code. 443 </p> 444 </dd> 445 <dt><code>--enable-host-shared</code></dt> 446 <dd><p>Specify that the <em>host</em> code should be built into position-independent 447 machine code (with -fPIC), allowing it to be used within shared libraries, 448 but yielding a slightly slower compiler. 449 </p> 450 <p>This option is required when building the libgccjit.so library. 451 </p> 452 <p>Contrast with <samp>--enable-shared</samp>, which affects <em>target</em> 453 libraries. 454 </p> 455 </dd> 456 <dt><code><a name="with-gnu-as"></a>--with-gnu-as</code></dt> 457 <dd><p>Specify that the compiler should assume that the 458 assembler it finds is the GNU assembler. However, this does not modify 459 the rules to find an assembler and will result in confusion if the 460 assembler found is not actually the GNU assembler. (Confusion may also 461 result if the compiler finds the GNU assembler but has not been 462 configured with <samp>--with-gnu-as</samp>.) If you have more than one 463 assembler installed on your system, you may want to use this option in 464 connection with <samp>--with-as=<var>pathname</var></samp> or 465 <samp>--with-build-time-tools=<var>pathname</var></samp>. 466 </p> 467 <p>The following systems are the only ones where it makes a difference 468 whether you use the GNU assembler. On any other system, 469 <samp>--with-gnu-as</samp> has no effect. 470 </p> 471 <ul> 472 <li> ‘<samp>hppa1.0-<var>any</var>-<var>any</var></samp>’ 473 </li><li> ‘<samp>hppa1.1-<var>any</var>-<var>any</var></samp>’ 474 </li><li> ‘<samp>sparc-sun-solaris2.<var>any</var></samp>’ 475 </li><li> ‘<samp>sparc64-<var>any</var>-solaris2.<var>any</var></samp>’ 476 </li></ul> 477 478 </dd> 479 <dt><code><a name="with-as"></a>--with-as=<var>pathname</var></code></dt> 480 <dd><p>Specify that the compiler should use the assembler pointed to by 481 <var>pathname</var>, rather than the one found by the standard rules to find 482 an assembler, which are: 483 </p><ul> 484 <li> Unless GCC is being built with a cross compiler, check the 485 <samp><var>libexec</var>/gcc/<var>target</var>/<var>version</var></samp> directory. 486 <var>libexec</var> defaults to <samp><var>exec-prefix</var>/libexec</samp>; 487 <var>exec-prefix</var> defaults to <var>prefix</var>, which 488 defaults to <samp>/usr/local</samp> unless overridden by the 489 <samp>--prefix=<var>pathname</var></samp> switch described above. <var>target</var> 490 is the target system triple, such as ‘<samp>sparc-sun-solaris2.7</samp>’, and 491 <var>version</var> denotes the GCC version, such as 3.0. 492 493 </li><li> If the target system is the same that you are building on, check 494 operating system specific directories (e.g. <samp>/usr/ccs/bin</samp> on 495 Sun Solaris 2). 496 497 </li><li> Check in the <code>PATH</code> for a tool whose name is prefixed by the 498 target system triple. 499 500 </li><li> Check in the <code>PATH</code> for a tool whose name is not prefixed by the 501 target system triple, if the host and target system triple are 502 the same (in other words, we use a host tool if it can be used for 503 the target as well). 504 </li></ul> 505 506 <p>You may want to use <samp>--with-as</samp> if no assembler 507 is installed in the directories listed above, or if you have multiple 508 assemblers installed and want to choose one that is not found by the 509 above rules. 510 </p> 511 </dd> 512 <dt><code><a name="with-gnu-ld"></a>--with-gnu-ld</code></dt> 513 <dd><p>Same as <a href="#with-gnu-as"><samp>--with-gnu-as</samp></a> 514 but for the linker. 515 </p> 516 </dd> 517 <dt><code>--with-ld=<var>pathname</var></code></dt> 518 <dd><p>Same as <a href="#with-as"><samp>--with-as</samp></a> 519 but for the linker. 520 </p> 521 </dd> 522 <dt><code>--with-stabs</code></dt> 523 <dd><p>Specify that stabs debugging 524 information should be used instead of whatever format the host normally 525 uses. Normally GCC uses the same debug format as the host system. 526 </p> 527 </dd> 528 <dt><code>--with-tls=<var>dialect</var></code></dt> 529 <dd><p>Specify the default TLS dialect, for systems were there is a choice. 530 For ARM targets, possible values for <var>dialect</var> are <code>gnu</code> or 531 <code>gnu2</code>, which select between the original GNU dialect and the GNU TLS 532 descriptor-based dialect. 533 </p> 534 </dd> 535 <dt><code>--enable-multiarch</code></dt> 536 <dd><p>Specify whether to enable or disable multiarch support. The default is 537 to check for glibc start files in a multiarch location, and enable it 538 if the files are found. The auto detection is enabled for native builds, 539 and for cross builds configured with <samp>--with-sysroot</samp>, and without 540 <samp>--with-native-system-header-dir</samp>. 541 More documentation about multiarch can be found at 542 <a href="https://wiki.debian.org/Multiarch">https://wiki.debian.org/Multiarch</a>. 543 </p> 544 </dd> 545 <dt><code>--enable-sjlj-exceptions</code></dt> 546 <dd><p>Force use of the <code>setjmp</code>/<code>longjmp</code>-based scheme for exceptions. 547 ‘<samp>configure</samp>’ ordinarily picks the correct value based on the platform. 548 Only use this option if you are sure you need a different setting. 549 </p> 550 </dd> 551 <dt><code>--enable-vtable-verify</code></dt> 552 <dd><p>Specify whether to enable or disable the vtable verification feature. 553 Enabling this feature causes libstdc++ to be built with its virtual calls 554 in verifiable mode. This means that, when linked with libvtv, every 555 virtual call in libstdc++ will verify the vtable pointer through which the 556 call will be made before actually making the call. If not linked with libvtv, 557 the verifier will call stub functions (in libstdc++ itself) and do nothing. 558 If vtable verification is disabled, then libstdc++ is not built with its 559 virtual calls in verifiable mode at all. However the libvtv library will 560 still be built (see <samp>--disable-libvtv</samp> to turn off building libvtv). 561 <samp>--disable-vtable-verify</samp> is the default. 562 </p> 563 </dd> 564 <dt><code>--disable-multilib</code></dt> 565 <dd><p>Specify that multiple target 566 libraries to support different target variants, calling 567 conventions, etc. should not be built. The default is to build a 568 predefined set of them. 569 </p> 570 <p>Some targets provide finer-grained control over which multilibs are built 571 (e.g., <samp>--disable-softfloat</samp>): 572 </p><dl compact="compact"> 573 <dt><code>arm-*-*</code></dt> 574 <dd><p>fpu, 26bit, underscore, interwork, biendian, nofmult. 575 </p> 576 </dd> 577 <dt><code>m68*-*-*</code></dt> 578 <dd><p>softfloat, m68881, m68000, m68020. 579 </p> 580 </dd> 581 <dt><code>mips*-*-*</code></dt> 582 <dd><p>single-float, biendian, softfloat. 583 </p> 584 </dd> 585 <dt><code>powerpc*-*-*, rs6000*-*-*</code></dt> 586 <dd><p>aix64, pthread, softfloat, powercpu, powerpccpu, powerpcos, biendian, 587 sysv, aix. 588 </p> 589 </dd> 590 </dl> 591 592 </dd> 593 <dt><code>--with-multilib-list=<var>list</var></code></dt> 594 <dt><code>--without-multilib-list</code></dt> 595 <dd><p>Specify what multilibs to build. <var>list</var> is a comma separated list of 596 values, possibly consisting of a single value. Currently only implemented 597 for aarch64*-*-*, arm*-*-*, sh*-*-* and x86-64-*-linux*. The accepted 598 alues and meaning for each target is given below. 599 </p> 600 <dl compact="compact"> 601 <dt><code>aarch64*-*-*</code></dt> 602 <dd><p><var>list</var> is a comma separated list of <code>ilp32</code>, and <code>lp64</code> 603 to enable ILP32 and LP64 run-time libraries, respectively. If 604 <var>list</var> is empty, then there will be no multilibs and only the 605 default run-time library will be built. If <var>list</var> is 606 <code>default</code> or –with-multilib-list= is not specified, then the 607 default set of libraries is selected based on the value of 608 <samp>--target</samp>. 609 </p> 610 </dd> 611 <dt><code>arm*-*-*</code></dt> 612 <dd><p><var>list</var> is a comma separated list of <code>aprofile</code> and <code>rmprofile</code> 613 to build multilibs for A or R and M architecture profiles respectively. Note 614 that, due to some limitation of the current multilib framework, using the 615 combined <code>aprofile,rmprofile</code> multilibs selects in some cases a less 616 optimal multilib than when using the multilib profile for the architecture 617 targetted. The special value <code>default</code> is also accepted and is equivalent 618 to omitting the option, ie. only the default run-time library will be enabled. 619 </p> 620 <p>The table below gives the combination of ISAs, architectures, FPUs and 621 floating-point ABIs for which multilibs are built for each accepted value. 622 The union of these options is considered when specifying both <code>aprofile</code> 623 and <code>rmprofile</code>. 624 </p> 625 <table> 626 <tr><td width="15%">Option</td><td width="28%">aprofile</td><td width="30%">rmprofile</td></tr> 627 <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> 628 <tr><td width="15%">Architectures<br><br><br><br><br><br></td><td width="28%">default architecture<br> 629 <code>-march=armv7-a</code><br> 630 <code>-march=armv7ve</code><br> 631 <code>-march=armv8-a</code><br><br><br></td><td width="30%">default architecture<br> 632 <code>-march=armv6s-m</code><br> 633 <code>-march=armv7-m</code><br> 634 <code>-march=armv7e-m</code><br> 635 <code>-march=armv8-m.base</code><br> 636 <code>-march=armv8-m.main</code><br> 637 <code>-march=armv7</code></td></tr> 638 <tr><td width="15%">FPUs<br><br><br><br><br></td><td width="28%">none<br> 639 <code>-mfpu=vfpv3-d16</code><br> 640 <code>-mfpu=neon</code><br> 641 <code>-mfpu=vfpv4-d16</code><br> 642 <code>-mfpu=neon-vfpv4</code><br> 643 <code>-mfpu=neon-fp-armv8</code></td><td width="30%">none<br> 644 <code>-mfpu=vfpv3-d16</code><br> 645 <code>-mfpu=fpv4-sp-d16</code><br> 646 <code>-mfpu=fpv5-sp-d16</code><br> 647 <code>-mfpu=fpv5-d16</code><br></td></tr> 648 <tr><td width="15%">floating-point ABIs<br><br></td><td width="28%"><code>-mfloat-abi=soft</code><br> 649 <code>-mfloat-abi=softfp</code><br> 650 <code>-mfloat-abi=hard</code></td><td width="30%"><code>-mfloat-abi=soft</code><br> 651 <code>-mfloat-abi=softfp</code><br> 652 <code>-mfloat-abi=hard</code></td></tr> 653 </table> 654 655 </dd> 656 <dt><code>sh*-*-*</code></dt> 657 <dd><p><var>list</var> is a comma separated list of CPU names. These must be of the 658 form <code>sh*</code> or <code>m*</code> (in which case they match the compiler option 659 for that processor). The list should not contain any endian options - 660 these are handled by <samp>--with-endian</samp>. 661 </p> 662 <p>If <var>list</var> is empty, then there will be no multilibs for extra 663 processors. The multilib for the secondary endian remains enabled. 664 </p> 665 <p>As a special case, if an entry in the list starts with a <code>!</code> 666 (exclamation point), then it is added to the list of excluded multilibs. 667 Entries of this sort should be compatible with ‘<samp>MULTILIB_EXCLUDES</samp>’ 668 (once the leading <code>!</code> has been stripped). 669 </p> 670 <p>If <samp>--with-multilib-list</samp> is not given, then a default set of 671 multilibs is selected based on the value of <samp>--target</samp>. This is 672 usually the complete set of libraries, but some targets imply a more 673 specialized subset. 674 </p> 675 <p>Example 1: to configure a compiler for SH4A only, but supporting both 676 endians, with little endian being the default: 677 </p><div class="smallexample"> 678 <pre class="smallexample">--with-cpu=sh4a --with-endian=little,big --with-multilib-list= 679 </pre></div> 680 681 <p>Example 2: to configure a compiler for both SH4A and SH4AL-DSP, but with 682 only little endian SH4AL: 683 </p><div class="smallexample"> 684 <pre class="smallexample">--with-cpu=sh4a --with-endian=little,big \ 685 --with-multilib-list=sh4al,!mb/m4al 686 </pre></div> 687 688 </dd> 689 <dt><code>x86-64-*-linux*</code></dt> 690 <dd><p><var>list</var> is a comma separated list of <code>m32</code>, <code>m64</code> and 691 <code>mx32</code> to enable 32-bit, 64-bit and x32 run-time libraries, 692 respectively. If <var>list</var> is empty, then there will be no multilibs 693 and only the default run-time library will be enabled. 694 </p> 695 <p>If <samp>--with-multilib-list</samp> is not given, then only 32-bit and 696 64-bit run-time libraries will be enabled. 697 </p></dd> 698 </dl> 699 700 </dd> 701 <dt><code>--with-endian=<var>endians</var></code></dt> 702 <dd><p>Specify what endians to use. 703 Currently only implemented for sh*-*-*. 704 </p> 705 <p><var>endians</var> may be one of the following: 706 </p><dl compact="compact"> 707 <dt><code>big</code></dt> 708 <dd><p>Use big endian exclusively. 709 </p></dd> 710 <dt><code>little</code></dt> 711 <dd><p>Use little endian exclusively. 712 </p></dd> 713 <dt><code>big,little</code></dt> 714 <dd><p>Use big endian by default. Provide a multilib for little endian. 715 </p></dd> 716 <dt><code>little,big</code></dt> 717 <dd><p>Use little endian by default. Provide a multilib for big endian. 718 </p></dd> 719 </dl> 720 721 </dd> 722 <dt><code>--enable-threads</code></dt> 723 <dd><p>Specify that the target 724 supports threads. This affects the Objective-C compiler and runtime 725 library, and exception handling for other languages like C++. 726 On some systems, this is the default. 727 </p> 728 <p>In general, the best (and, in many cases, the only known) threading 729 model available will be configured for use. Beware that on some 730 systems, GCC has not been taught what threading models are generally 731 available for the system. In this case, <samp>--enable-threads</samp> is an 732 alias for <samp>--enable-threads=single</samp>. 733 </p> 734 </dd> 735 <dt><code>--disable-threads</code></dt> 736 <dd><p>Specify that threading support should be disabled for the system. 737 This is an alias for <samp>--enable-threads=single</samp>. 738 </p> 739 </dd> 740 <dt><code>--enable-threads=<var>lib</var></code></dt> 741 <dd><p>Specify that 742 <var>lib</var> is the thread support library. This affects the Objective-C 743 compiler and runtime library, and exception handling for other languages 744 like C++. The possibilities for <var>lib</var> are: 745 </p> 746 <dl compact="compact"> 747 <dt><code>aix</code></dt> 748 <dd><p>AIX thread support. 749 </p></dd> 750 <dt><code>dce</code></dt> 751 <dd><p>DCE thread support. 752 </p></dd> 753 <dt><code>lynx</code></dt> 754 <dd><p>LynxOS thread support. 755 </p></dd> 756 <dt><code>mipssde</code></dt> 757 <dd><p>MIPS SDE thread support. 758 </p></dd> 759 <dt><code>no</code></dt> 760 <dd><p>This is an alias for ‘<samp>single</samp>’. 761 </p></dd> 762 <dt><code>posix</code></dt> 763 <dd><p>Generic POSIX/Unix98 thread support. 764 </p></dd> 765 <dt><code>rtems</code></dt> 766 <dd><p>RTEMS thread support. 767 </p></dd> 768 <dt><code>single</code></dt> 769 <dd><p>Disable thread support, should work for all platforms. 770 </p></dd> 771 <dt><code>tpf</code></dt> 772 <dd><p>TPF thread support. 773 </p></dd> 774 <dt><code>vxworks</code></dt> 775 <dd><p>VxWorks thread support. 776 </p></dd> 777 <dt><code>win32</code></dt> 778 <dd><p>Microsoft Win32 API thread support. 779 </p></dd> 780 </dl> 781 782 </dd> 783 <dt><code>--enable-tls</code></dt> 784 <dd><p>Specify that the target supports TLS (Thread Local Storage). Usually 785 configure can correctly determine if TLS is supported. In cases where 786 it guesses incorrectly, TLS can be explicitly enabled or disabled with 787 <samp>--enable-tls</samp> or <samp>--disable-tls</samp>. This can happen if 788 the assembler supports TLS but the C library does not, or if the 789 assumptions made by the configure test are incorrect. 790 </p> 791 </dd> 792 <dt><code>--disable-tls</code></dt> 793 <dd><p>Specify that the target does not support TLS. 794 This is an alias for <samp>--enable-tls=no</samp>. 795 </p> 796 </dd> 797 <dt><code>--with-cpu=<var>cpu</var></code></dt> 798 <dt><code>--with-cpu-32=<var>cpu</var></code></dt> 799 <dt><code>--with-cpu-64=<var>cpu</var></code></dt> 800 <dd><p>Specify which cpu variant the compiler should generate code for by default. 801 <var>cpu</var> will be used as the default value of the <samp>-mcpu=</samp> switch. 802 This option is only supported on some targets, including ARC, ARM, i386, M68k, 803 PowerPC, and SPARC. It is mandatory for ARC. The <samp>--with-cpu-32</samp> and 804 <samp>--with-cpu-64</samp> options specify separate default CPUs for 805 32-bit and 64-bit modes; these options are only supported for i386, 806 x86-64, PowerPC, and SPARC. 807 </p> 808 </dd> 809 <dt><code>--with-schedule=<var>cpu</var></code></dt> 810 <dt><code>--with-arch=<var>cpu</var></code></dt> 811 <dt><code>--with-arch-32=<var>cpu</var></code></dt> 812 <dt><code>--with-arch-64=<var>cpu</var></code></dt> 813 <dt><code>--with-tune=<var>cpu</var></code></dt> 814 <dt><code>--with-tune-32=<var>cpu</var></code></dt> 815 <dt><code>--with-tune-64=<var>cpu</var></code></dt> 816 <dt><code>--with-abi=<var>abi</var></code></dt> 817 <dt><code>--with-fpu=<var>type</var></code></dt> 818 <dt><code>--with-float=<var>type</var></code></dt> 819 <dd><p>These configure options provide default values for the <samp>-mschedule=</samp>, 820 <samp>-march=</samp>, <samp>-mtune=</samp>, <samp>-mabi=</samp>, and <samp>-mfpu=</samp> 821 options and for <samp>-mhard-float</samp> or <samp>-msoft-float</samp>. As with 822 <samp>--with-cpu</samp>, which switches will be accepted and acceptable values 823 of the arguments depend on the target. 824 </p> 825 </dd> 826 <dt><code>--with-mode=<var>mode</var></code></dt> 827 <dd><p>Specify if the compiler should default to <samp>-marm</samp> or <samp>-mthumb</samp>. 828 This option is only supported on ARM targets. 829 </p> 830 </dd> 831 <dt><code>--with-stack-offset=<var>num</var></code></dt> 832 <dd><p>This option sets the default for the -mstack-offset=<var>num</var> option, 833 and will thus generally also control the setting of this option for 834 libraries. This option is only supported on Epiphany targets. 835 </p> 836 </dd> 837 <dt><code>--with-fpmath=<var>isa</var></code></dt> 838 <dd><p>This options sets <samp>-mfpmath=sse</samp> by default and specifies the default 839 ISA for floating-point arithmetics. You can select either ‘<samp>sse</samp>’ which 840 enables <samp>-msse2</samp> or ‘<samp>avx</samp>’ which enables <samp>-mavx</samp> by default. 841 This option is only supported on i386 and x86-64 targets. 842 </p> 843 </dd> 844 <dt><code>--with-fp-32=<var>mode</var></code></dt> 845 <dd><p>On MIPS targets, set the default value for the <samp>-mfp</samp> option when using 846 the o32 ABI. The possibilities for <var>mode</var> are: 847 </p><dl compact="compact"> 848 <dt><code>32</code></dt> 849 <dd><p>Use the o32 FP32 ABI extension, as with the <samp>-mfp32</samp> command-line 850 option. 851 </p></dd> 852 <dt><code>xx</code></dt> 853 <dd><p>Use the o32 FPXX ABI extension, as with the <samp>-mfpxx</samp> command-line 854 option. 855 </p></dd> 856 <dt><code>64</code></dt> 857 <dd><p>Use the o32 FP64 ABI extension, as with the <samp>-mfp64</samp> command-line 858 option. 859 </p></dd> 860 </dl> 861 <p>In the absence of this configuration option the default is to use the o32 862 FP32 ABI extension. 863 </p> 864 </dd> 865 <dt><code>--with-odd-spreg-32</code></dt> 866 <dd><p>On MIPS targets, set the <samp>-modd-spreg</samp> option by default when using 867 the o32 ABI. 868 </p> 869 </dd> 870 <dt><code>--without-odd-spreg-32</code></dt> 871 <dd><p>On MIPS targets, set the <samp>-mno-odd-spreg</samp> option by default when using 872 the o32 ABI. This is normally used in conjunction with 873 <samp>--with-fp-32=64</samp> in order to target the o32 FP64A ABI extension. 874 </p> 875 </dd> 876 <dt><code>--with-nan=<var>encoding</var></code></dt> 877 <dd><p>On MIPS targets, set the default encoding convention to use for the 878 special not-a-number (NaN) IEEE 754 floating-point data. The 879 possibilities for <var>encoding</var> are: 880 </p><dl compact="compact"> 881 <dt><code>legacy</code></dt> 882 <dd><p>Use the legacy encoding, as with the <samp>-mnan=legacy</samp> command-line 883 option. 884 </p></dd> 885 <dt><code>2008</code></dt> 886 <dd><p>Use the 754-2008 encoding, as with the <samp>-mnan=2008</samp> command-line 887 option. 888 </p></dd> 889 </dl> 890 <p>To use this configuration option you must have an assembler version 891 installed that supports the <samp>-mnan=</samp> command-line option too. 892 In the absence of this configuration option the default convention is 893 the legacy encoding, as when neither of the <samp>-mnan=2008</samp> and 894 <samp>-mnan=legacy</samp> command-line options has been used. 895 </p> 896 </dd> 897 <dt><code>--with-divide=<var>type</var></code></dt> 898 <dd><p>Specify how the compiler should generate code for checking for 899 division by zero. This option is only supported on the MIPS target. 900 The possibilities for <var>type</var> are: 901 </p><dl compact="compact"> 902 <dt><code>traps</code></dt> 903 <dd><p>Division by zero checks use conditional traps (this is the default on 904 systems that support conditional traps). 905 </p></dd> 906 <dt><code>breaks</code></dt> 907 <dd><p>Division by zero checks use the break instruction. 908 </p></dd> 909 </dl> 910 911 912 </dd> 913 <dt><code>--with-llsc</code></dt> 914 <dd><p>On MIPS targets, make <samp>-mllsc</samp> the default when no 915 <samp>-mno-llsc</samp> option is passed. This is the default for 916 Linux-based targets, as the kernel will emulate them if the ISA does 917 not provide them. 918 </p> 919 </dd> 920 <dt><code>--without-llsc</code></dt> 921 <dd><p>On MIPS targets, make <samp>-mno-llsc</samp> the default when no 922 <samp>-mllsc</samp> option is passed. 923 </p> 924 </dd> 925 <dt><code>--with-synci</code></dt> 926 <dd><p>On MIPS targets, make <samp>-msynci</samp> the default when no 927 <samp>-mno-synci</samp> option is passed. 928 </p> 929 </dd> 930 <dt><code>--without-synci</code></dt> 931 <dd><p>On MIPS targets, make <samp>-mno-synci</samp> the default when no 932 <samp>-msynci</samp> option is passed. This is the default. 933 </p> 934 </dd> 935 <dt><code>--with-lxc1-sxc1</code></dt> 936 <dd><p>On MIPS targets, make <samp>-mlxc1-sxc1</samp> the default when no 937 <samp>-mno-lxc1-sxc1</samp> option is passed. This is the default. 938 </p> 939 </dd> 940 <dt><code>--without-lxc1-sxc1</code></dt> 941 <dd><p>On MIPS targets, make <samp>-mno-lxc1-sxc1</samp> the default when no 942 <samp>-mlxc1-sxc1</samp> option is passed. The indexed load/store 943 instructions are not directly a problem but can lead to unexpected 944 behaviour when deployed in an application intended for a 32-bit address 945 space but run on a 64-bit processor. The issue is seen because all 946 known MIPS 64-bit Linux kernels execute o32 and n32 applications 947 with 64-bit addressing enabled which affects the overflow behaviour 948 of the indexed addressing mode. GCC will assume that ordinary 949 32-bit arithmetic overflow behaviour is the same whether performed 950 as an <code>addu</code> instruction or as part of the address calculation 951 in <code>lwxc1</code> type instructions. This assumption holds true in a 952 pure 32-bit environment and can hold true in a 64-bit environment if 953 the address space is accurately set to be 32-bit for o32 and n32. 954 </p> 955 </dd> 956 <dt><code>--with-madd4</code></dt> 957 <dd><p>On MIPS targets, make <samp>-mmadd4</samp> the default when no 958 <samp>-mno-madd4</samp> option is passed. This is the default. 959 </p> 960 </dd> 961 <dt><code>--without-madd4</code></dt> 962 <dd><p>On MIPS targets, make <samp>-mno-madd4</samp> the default when no 963 <samp>-mmadd4</samp> option is passed. The <code>madd4</code> instruction 964 family can be problematic when targeting a combination of cores that 965 implement these instructions differently. There are two known cores 966 that implement these as fused operations instead of unfused (where 967 unfused is normally expected). Disabling these instructions is the 968 only way to ensure compatible code is generated; this will incur 969 a performance penalty. 970 </p> 971 </dd> 972 <dt><code>--with-mips-plt</code></dt> 973 <dd><p>On MIPS targets, make use of copy relocations and PLTs. 974 These features are extensions to the traditional 975 SVR4-based MIPS ABIs and require support from GNU binutils 976 and the runtime C library. 977 </p> 978 </dd> 979 <dt><code>--enable-__cxa_atexit</code></dt> 980 <dd><p>Define if you want to use __cxa_atexit, rather than atexit, to 981 register C++ destructors for local statics and global objects. 982 This is essential for fully standards-compliant handling of 983 destructors, but requires __cxa_atexit in libc. This option is currently 984 only available on systems with GNU libc. When enabled, this will cause 985 <samp>-fuse-cxa-atexit</samp> to be passed by default. 986 </p> 987 </dd> 988 <dt><code>--enable-gnu-indirect-function</code></dt> 989 <dd><p>Define if you want to enable the <code>ifunc</code> attribute. This option is 990 currently only available on systems with GNU libc on certain targets. 991 </p> 992 </dd> 993 <dt><code>--enable-target-optspace</code></dt> 994 <dd><p>Specify that target 995 libraries should be optimized for code space instead of code speed. 996 This is the default for the m32r platform. 997 </p> 998 </dd> 999 <dt><code>--with-cpp-install-dir=<var>dirname</var></code></dt> 1000 <dd><p>Specify that the user visible <code>cpp</code> program should be installed 1001 in <samp><var>prefix</var>/<var>dirname</var>/cpp</samp>, in addition to <var>bindir</var>. 1002 </p> 1003 </dd> 1004 <dt><code>--enable-comdat</code></dt> 1005 <dd><p>Enable COMDAT group support. This is primarily used to override the 1006 automatically detected value. 1007 </p> 1008 </dd> 1009 <dt><code>--enable-initfini-array</code></dt> 1010 <dd><p>Force the use of sections <code>.init_array</code> and <code>.fini_array</code> 1011 (instead of <code>.init</code> and <code>.fini</code>) for constructors and 1012 destructors. Option <samp>--disable-initfini-array</samp> has the 1013 opposite effect. If neither option is specified, the configure script 1014 will try to guess whether the <code>.init_array</code> and 1015 <code>.fini_array</code> sections are supported and, if they are, use them. 1016 </p> 1017 </dd> 1018 <dt><code>--enable-link-mutex</code></dt> 1019 <dd><p>When building GCC, use a mutex to avoid linking the compilers for 1020 multiple languages at the same time, to avoid thrashing on build 1021 systems with limited free memory. The default is not to use such a mutex. 1022 </p> 1023 </dd> 1024 <dt><code>--enable-maintainer-mode</code></dt> 1025 <dd><p>The build rules that regenerate the Autoconf and Automake output files as 1026 well as the GCC master message catalog <samp>gcc.pot</samp> are normally 1027 disabled. This is because it can only be rebuilt if the complete source 1028 tree is present. If you have changed the sources and want to rebuild the 1029 catalog, configuring with <samp>--enable-maintainer-mode</samp> will enable 1030 this. Note that you need a recent version of the <code>gettext</code> tools 1031 to do so. 1032 </p> 1033 </dd> 1034 <dt><code>--disable-bootstrap</code></dt> 1035 <dd><p>For a native build, the default configuration is to perform 1036 a 3-stage bootstrap of the compiler when ‘<samp>make</samp>’ is invoked, 1037 testing that GCC can compile itself correctly. If you want to disable 1038 this process, you can configure with <samp>--disable-bootstrap</samp>. 1039 </p> 1040 </dd> 1041 <dt><code>--enable-bootstrap</code></dt> 1042 <dd><p>In special cases, you may want to perform a 3-stage build 1043 even if the target and host triplets are different. 1044 This is possible when the host can run code compiled for 1045 the target (e.g. host is i686-linux, target is i486-linux). 1046 Starting from GCC 4.2, to do this you have to configure explicitly 1047 with <samp>--enable-bootstrap</samp>. 1048 </p> 1049 </dd> 1050 <dt><code>--enable-generated-files-in-srcdir</code></dt> 1051 <dd><p>Neither the .c and .h files that are generated from Bison and flex nor the 1052 info manuals and man pages that are built from the .texi files are present 1053 in the SVN development tree. When building GCC from that development tree, 1054 or from one of our snapshots, those generated files are placed in your 1055 build directory, which allows for the source to be in a readonly 1056 directory. 1057 </p> 1058 <p>If you configure with <samp>--enable-generated-files-in-srcdir</samp> then those 1059 generated files will go into the source directory. This is mainly intended 1060 for generating release or prerelease tarballs of the GCC sources, since it 1061 is not a requirement that the users of source releases to have flex, Bison, 1062 or makeinfo. 1063 </p> 1064 </dd> 1065 <dt><code>--enable-version-specific-runtime-libs</code></dt> 1066 <dd><p>Specify 1067 that runtime libraries should be installed in the compiler specific 1068 subdirectory (<samp><var>libdir</var>/gcc</samp>) rather than the usual places. In 1069 addition, ‘<samp>libstdc++</samp>’’s include files will be installed into 1070 <samp><var>libdir</var></samp> unless you overruled it by using 1071 <samp>--with-gxx-include-dir=<var>dirname</var></samp>. Using this option is 1072 particularly useful if you intend to use several versions of GCC in 1073 parallel. This is currently supported by ‘<samp>libgfortran</samp>’, 1074 ‘<samp>libstdc++</samp>’, and ‘<samp>libobjc</samp>’. 1075 </p> 1076 </dd> 1077 <dt><code><a name="WithAixSoname"></a>--with-aix-soname=‘<samp>aix</samp>’, ‘<samp>svr4</samp>’ or ‘<samp>both</samp>’</code></dt> 1078 <dd><p>Traditional AIX shared library versioning (versioned <code>Shared Object</code> 1079 files as members of unversioned <code>Archive Library</code> files named 1080 ‘<samp>lib.a</samp>’) causes numerous headaches for package managers. However, 1081 <code>Import Files</code> as members of <code>Archive Library</code> files allow for 1082 <strong>filename-based versioning</strong> of shared libraries as seen on Linux/SVR4, 1083 where this is called the "SONAME". But as they prevent static linking, 1084 <code>Import Files</code> may be used with <code>Runtime Linking</code> only, where the 1085 linker does search for ‘<samp>libNAME.so</samp>’ before ‘<samp>libNAME.a</samp>’ library 1086 filenames with the ‘<samp>-lNAME</samp>’ linker flag. 1087 </p> 1088 <a name="AixLdCommand"></a><p>For detailed information please refer to the AIX 1089 <a href="https://www.ibm.com/support/knowledgecenter/search/%22the%20ld%20command%2C%20also%20called%20the%20linkage%20editor%20or%20binder%22">ld 1090 Command</a> reference. 1091 </p> 1092 <p>As long as shared library creation is enabled, upon: 1093 </p><dl compact="compact"> 1094 <dt><code>--with-aix-soname=aix</code></dt> 1095 <dt><code>--with-aix-soname=both</code></dt> 1096 <dd><p>A (traditional AIX) <code>Shared Archive Library</code> file is created: 1097 </p><ul> 1098 <li> using the ‘<samp>libNAME.a</samp>’ filename scheme 1099 </li><li> with the <code>Shared Object</code> file as archive member named 1100 ‘<samp>libNAME.so.V</samp>’ (except for ‘<samp>libgcc_s</samp>’, where the <code>Shared 1101 Object</code> file is named ‘<samp>shr.o</samp>’ for backwards compatibility), which 1102 <ul class="no-bullet"> 1103 <li>- is used for runtime loading from inside the ‘<samp>libNAME.a</samp>’ file 1104 </li><li>- is used for dynamic loading via 1105 <code>dlopen("libNAME.a(libNAME.so.V)", RTLD_MEMBER)</code> 1106 </li><li>- is used for shared linking 1107 </li><li>- is used for static linking, so no separate <code>Static Archive 1108 Library</code> file is needed 1109 </li></ul> 1110 </li></ul> 1111 </dd> 1112 <dt><code>--with-aix-soname=both</code></dt> 1113 <dt><code>--with-aix-soname=svr4</code></dt> 1114 <dd><p>A (second) <code>Shared Archive Library</code> file is created: 1115 </p><ul> 1116 <li> using the ‘<samp>libNAME.so.V</samp>’ filename scheme 1117 </li><li> with the <code>Shared Object</code> file as archive member named 1118 ‘<samp>shr.o</samp>’, which 1119 <ul class="no-bullet"> 1120 <li>- is created with the <code>-G linker flag</code> 1121 </li><li>- has the <code>F_LOADONLY</code> flag set 1122 </li><li>- is used for runtime loading from inside the ‘<samp>libNAME.so.V</samp>’ file 1123 </li><li>- is used for dynamic loading via <code>dlopen("libNAME.so.V(shr.o)", 1124 RTLD_MEMBER)</code> 1125 </li></ul> 1126 </li><li> with the <code>Import File</code> as archive member named ‘<samp>shr.imp</samp>’, 1127 which 1128 <ul class="no-bullet"> 1129 <li>- refers to ‘<samp>libNAME.so.V(shr.o)</samp>’ as the "SONAME", to be recorded 1130 in the <code>Loader Section</code> of subsequent binaries 1131 </li><li>- indicates whether ‘<samp>libNAME.so.V(shr.o)</samp>’ is 32 or 64 bit 1132 </li><li>- lists all the public symbols exported by ‘<samp>lib.so.V(shr.o)</samp>’, 1133 eventually decorated with the <code>‘<samp>weak</samp>’ Keyword</code> 1134 </li><li>- is necessary for shared linking against ‘<samp>lib.so.V(shr.o)</samp>’ 1135 </li></ul> 1136 </li></ul> 1137 <p>A symbolic link using the ‘<samp>libNAME.so</samp>’ filename scheme is created: 1138 </p><ul> 1139 <li> pointing to the ‘<samp>libNAME.so.V</samp>’ <code>Shared Archive Library</code> file 1140 </li><li> to permit the <code>ld Command</code> to find ‘<samp>lib.so.V(shr.imp)</samp>’ via 1141 the ‘<samp>-lNAME</samp>’ argument (requires <code>Runtime Linking</code> to be enabled) 1142 </li><li> to permit dynamic loading of ‘<samp>lib.so.V(shr.o)</samp>’ without the need 1143 to specify the version number via <code>dlopen("libNAME.so(shr.o)", 1144 RTLD_MEMBER)</code> 1145 </li></ul> 1146 </dd> 1147 </dl> 1148 1149 <p>As long as static library creation is enabled, upon: 1150 </p><dl compact="compact"> 1151 <dt><code>--with-aix-soname=svr4</code></dt> 1152 <dd><p>A <code>Static Archive Library</code> is created: 1153 </p><ul> 1154 <li> using the ‘<samp>libNAME.a</samp>’ filename scheme 1155 </li><li> with all the <code>Static Object</code> files as archive members, which 1156 <ul class="no-bullet"> 1157 <li>- are used for static linking 1158 </li></ul> 1159 </li></ul> 1160 </dd> 1161 </dl> 1162 1163 <p>While the aix-soname=‘<samp>svr4</samp>’ option does not create <code>Shared Object</code> 1164 files as members of unversioned <code>Archive Library</code> files any more, package 1165 managers still are responsible to 1166 <a href="./specific.html#TransferAixShobj">transfer</a> <code>Shared Object</code> files 1167 found as member of a previously installed unversioned <code>Archive Library</code> 1168 file into the newly installed <code>Archive Library</code> file with the same 1169 filename. 1170 </p> 1171 <p><em>WARNING:</em> Creating <code>Shared Object</code> files with <code>Runtime Linking</code> 1172 enabled may bloat the TOC, eventually leading to <code>TOC overflow</code> errors, 1173 requiring the use of either the <samp>-Wl,-bbigtoc</samp> linker flag (seen to 1174 break with the <code>GDB</code> debugger) or some of the TOC-related compiler flags, 1175 see “RS/6000 and PowerPC Options” in the main manual. 1176 </p> 1177 <p><samp>--with-aix-soname</samp> is currently supported by ‘<samp>libgcc_s</samp>’ only, so 1178 this option is still experimental and not for normal use yet. 1179 </p> 1180 <p>Default is the traditional behavior <samp>--with-aix-soname=‘<samp>aix</samp>’</samp>. 1181 </p> 1182 </dd> 1183 <dt><code>--enable-languages=<var>lang1</var>,<var>lang2</var>,…</code></dt> 1184 <dd><p>Specify that only a particular subset of compilers and 1185 their runtime libraries should be built. For a list of valid values for 1186 <var>langN</var> you can issue the following command in the 1187 <samp>gcc</samp> directory of your GCC source tree:<br> 1188 </p><div class="smallexample"> 1189 <pre class="smallexample">grep ^language= */config-lang.in 1190 </pre></div> 1191 <p>Currently, you can use any of the following: 1192 <code>all</code>, <code>default</code>, <code>ada</code>, <code>c</code>, <code>c++</code>, <code>fortran</code>, 1193 <code>go</code>, <code>jit</code>, <code>lto</code>, <code>objc</code>, <code>obj-c++</code>. 1194 Building the Ada compiler has special requirements, see below. 1195 If you do not pass this flag, or specify the option <code>default</code>, then the 1196 default languages available in the <samp>gcc</samp> sub-tree will be configured. 1197 Ada, Go, Jit, and Objective-C++ are not default languages. LTO is not a 1198 default language, but is built by default because <samp>--enable-lto</samp> is 1199 enabled by default. The other languages are default languages. If 1200 <code>all</code> is specified, then all available languages are built. An 1201 exception is <code>jit</code> language, which requires 1202 <samp>--enable-host-shared</samp> to be included with <code>all</code>. 1203 </p> 1204 </dd> 1205 <dt><code>--enable-stage1-languages=<var>lang1</var>,<var>lang2</var>,…</code></dt> 1206 <dd><p>Specify that a particular subset of compilers and their runtime 1207 libraries should be built with the system C compiler during stage 1 of 1208 the bootstrap process, rather than only in later stages with the 1209 bootstrapped C compiler. The list of valid values is the same as for 1210 <samp>--enable-languages</samp>, and the option <code>all</code> will select all 1211 of the languages enabled by <samp>--enable-languages</samp>. This option is 1212 primarily useful for GCC development; for instance, when a development 1213 version of the compiler cannot bootstrap due to compiler bugs, or when 1214 one is debugging front ends other than the C front end. When this 1215 option is used, one can then build the target libraries for the 1216 specified languages with the stage-1 compiler by using <code>make 1217 stage1-bubble all-target</code>, or run the testsuite on the stage-1 compiler 1218 for the specified languages using <code>make stage1-start check-gcc</code>. 1219 </p> 1220 </dd> 1221 <dt><code>--disable-libada</code></dt> 1222 <dd><p>Specify that the run-time libraries and tools used by GNAT should not 1223 be built. This can be useful for debugging, or for compatibility with 1224 previous Ada build procedures, when it was required to explicitly 1225 do a ‘<samp>make -C gcc gnatlib_and_tools</samp>’. 1226 </p> 1227 </dd> 1228 <dt><code>--disable-libsanitizer</code></dt> 1229 <dd><p>Specify that the run-time libraries for the various sanitizers should 1230 not be built. 1231 </p> 1232 </dd> 1233 <dt><code>--disable-libssp</code></dt> 1234 <dd><p>Specify that the run-time libraries for stack smashing protection 1235 should not be built or linked against. On many targets library support 1236 is provided by the C library instead. 1237 </p> 1238 </dd> 1239 <dt><code>--disable-libquadmath</code></dt> 1240 <dd><p>Specify that the GCC quad-precision math library should not be built. 1241 On some systems, the library is required to be linkable when building 1242 the Fortran front end, unless <samp>--disable-libquadmath-support</samp> 1243 is used. 1244 </p> 1245 </dd> 1246 <dt><code>--disable-libquadmath-support</code></dt> 1247 <dd><p>Specify that the Fortran front end and <code>libgfortran</code> do not add 1248 support for <code>libquadmath</code> on systems supporting it. 1249 </p> 1250 </dd> 1251 <dt><code>--disable-libgomp</code></dt> 1252 <dd><p>Specify that the GNU Offloading and Multi Processing Runtime Library 1253 should not be built. 1254 </p> 1255 </dd> 1256 <dt><code>--disable-libvtv</code></dt> 1257 <dd><p>Specify that the run-time libraries used by vtable verification 1258 should not be built. 1259 </p> 1260 </dd> 1261 <dt><code>--with-dwarf2</code></dt> 1262 <dd><p>Specify that the compiler should 1263 use DWARF 2 debugging information as the default. 1264 </p> 1265 </dd> 1266 <dt><code>--with-advance-toolchain=<var>at</var></code></dt> 1267 <dd><p>On 64-bit PowerPC Linux systems, configure the compiler to use the 1268 header files, library files, and the dynamic linker from the Advance 1269 Toolchain release <var>at</var> instead of the default versions that are 1270 provided by the Linux distribution. In general, this option is 1271 intended for the developers of GCC, and it is not intended for general 1272 use. 1273 </p> 1274 </dd> 1275 <dt><code>--enable-targets=all</code></dt> 1276 <dt><code>--enable-targets=<var>target_list</var></code></dt> 1277 <dd><p>Some GCC targets, e.g. powerpc64-linux, build bi-arch compilers. 1278 These are compilers that are able to generate either 64-bit or 32-bit 1279 code. Typically, the corresponding 32-bit target, e.g. 1280 powerpc-linux for powerpc64-linux, only generates 32-bit code. This 1281 option enables the 32-bit target to be a bi-arch compiler, which is 1282 useful when you want a bi-arch compiler that defaults to 32-bit, and 1283 you are building a bi-arch or multi-arch binutils in a combined tree. 1284 On mips-linux, this will build a tri-arch compiler (ABI o32/n32/64), 1285 defaulted to o32. 1286 Currently, this option only affects sparc-linux, powerpc-linux, x86-linux, 1287 mips-linux and s390-linux. 1288 </p> 1289 </dd> 1290 <dt><code>--enable-default-pie</code></dt> 1291 <dd><p>Turn on <samp>-fPIE</samp> and <samp>-pie</samp> by default. 1292 </p> 1293 </dd> 1294 <dt><code>--enable-secureplt</code></dt> 1295 <dd><p>This option enables <samp>-msecure-plt</samp> by default for powerpc-linux. 1296 See “RS/6000 and PowerPC Options” in the main manual 1297 </p> 1298 </dd> 1299 <dt><code>--enable-default-ssp</code></dt> 1300 <dd><p>Turn on <samp>-fstack-protector-strong</samp> by default. 1301 </p> 1302 </dd> 1303 <dt><code>--enable-cld</code></dt> 1304 <dd><p>This option enables <samp>-mcld</samp> by default for 32-bit x86 targets. 1305 See “i386 and x86-64 Options” in the main manual 1306 </p> 1307 </dd> 1308 <dt><code>--enable-win32-registry</code></dt> 1309 <dt><code>--enable-win32-registry=<var>key</var></code></dt> 1310 <dt><code>--disable-win32-registry</code></dt> 1311 <dd><p>The <samp>--enable-win32-registry</samp> option enables Microsoft Windows-hosted GCC 1312 to look up installations paths in the registry using the following key: 1313 </p> 1314 <div class="smallexample"> 1315 <pre class="smallexample"><code>HKEY_LOCAL_MACHINE\SOFTWARE\Free Software Foundation\<var>key</var></code> 1316 </pre></div> 1317 1318 <p><var>key</var> defaults to GCC version number, and can be overridden by the 1319 <samp>--enable-win32-registry=<var>key</var></samp> option. Vendors and distributors 1320 who use custom installers are encouraged to provide a different key, 1321 perhaps one comprised of vendor name and GCC version number, to 1322 avoid conflict with existing installations. This feature is enabled 1323 by default, and can be disabled by <samp>--disable-win32-registry</samp> 1324 option. This option has no effect on the other hosts. 1325 </p> 1326 </dd> 1327 <dt><code>--nfp</code></dt> 1328 <dd><p>Specify that the machine does not have a floating point unit. This 1329 option only applies to ‘<samp>m68k-sun-sunos<var>n</var></samp>’. On any other 1330 system, <samp>--nfp</samp> has no effect. 1331 </p> 1332 </dd> 1333 <dt><code>--enable-werror</code></dt> 1334 <dt><code>--disable-werror</code></dt> 1335 <dt><code>--enable-werror=yes</code></dt> 1336 <dt><code>--enable-werror=no</code></dt> 1337 <dd><p>When you specify this option, it controls whether certain files in the 1338 compiler are built with <samp>-Werror</samp> in bootstrap stage2 and later. 1339 If you don’t specify it, <samp>-Werror</samp> is turned on for the main 1340 development trunk. However it defaults to off for release branches and 1341 final releases. The specific files which get <samp>-Werror</samp> are 1342 controlled by the Makefiles. 1343 </p> 1344 </dd> 1345 <dt><code>--enable-checking</code></dt> 1346 <dt><code>--enable-checking=<var>list</var></code></dt> 1347 <dd><p>When you specify this option, the compiler is built to perform internal 1348 consistency checks of the requested complexity. This does not change the 1349 generated code, but adds error checking within the compiler. This will 1350 slow down the compiler and may only work properly if you are building 1351 the compiler with GCC. This is ‘<samp>yes,extra</samp>’ by default when building 1352 from SVN or snapshots, but ‘<samp>release</samp>’ for releases. The default 1353 for building the stage1 compiler is ‘<samp>yes</samp>’. More control 1354 over the checks may be had by specifying <var>list</var>. The categories of 1355 checks available are ‘<samp>yes</samp>’ (most common checks 1356 ‘<samp>assert,misc,tree,gc,rtlflag,runtime</samp>’), ‘<samp>no</samp>’ (no checks at 1357 all), ‘<samp>all</samp>’ (all but ‘<samp>valgrind</samp>’), ‘<samp>release</samp>’ (cheapest 1358 checks ‘<samp>assert,runtime</samp>’) or ‘<samp>none</samp>’ (same as ‘<samp>no</samp>’). 1359 Individual checks can be enabled with these flags ‘<samp>assert</samp>’, 1360 ‘<samp>df</samp>’, ‘<samp>fold</samp>’, ‘<samp>gc</samp>’, ‘<samp>gcac</samp>’, ‘<samp>misc</samp>’, ‘<samp>rtl</samp>’, 1361 ‘<samp>rtlflag</samp>’, ‘<samp>runtime</samp>’, ‘<samp>tree</samp>’, ‘<samp>extra</samp>’ and ‘<samp>valgrind</samp>’. 1362 ‘<samp>extra</samp>’ adds for ‘<samp>misc</samp>’ checking extra checks that might affect 1363 code generation and should therefore not differ between stage1 and later 1364 stages. 1365 </p> 1366 <p>The ‘<samp>valgrind</samp>’ check requires the external <code>valgrind</code> 1367 simulator, available from <a href="http://valgrind.org/">http://valgrind.org/</a>. The 1368 ‘<samp>df</samp>’, ‘<samp>rtl</samp>’, ‘<samp>gcac</samp>’ and ‘<samp>valgrind</samp>’ checks are very expensive. 1369 To disable all checking, ‘<samp>--disable-checking</samp>’ or 1370 ‘<samp>--enable-checking=none</samp>’ must be explicitly requested. Disabling 1371 assertions will make the compiler and runtime slightly faster but 1372 increase the risk of undetected internal errors causing wrong code to be 1373 generated. 1374 </p> 1375 </dd> 1376 <dt><code>--disable-stage1-checking</code></dt> 1377 <dt><code>--enable-stage1-checking</code></dt> 1378 <dt><code>--enable-stage1-checking=<var>list</var></code></dt> 1379 <dd><p>If no <samp>--enable-checking</samp> option is specified the stage1 1380 compiler will be built with ‘<samp>yes</samp>’ checking enabled, otherwise 1381 the stage1 checking flags are the same as specified by 1382 <samp>--enable-checking</samp>. To build the stage1 compiler with 1383 different checking options use <samp>--enable-stage1-checking</samp>. 1384 The list of checking options is the same as for <samp>--enable-checking</samp>. 1385 If your system is too slow or too small to bootstrap a released compiler 1386 with checking for stage1 enabled, you can use ‘<samp>--disable-stage1-checking</samp>’ 1387 to disable checking for the stage1 compiler. 1388 </p> 1389 </dd> 1390 <dt><code>--enable-coverage</code></dt> 1391 <dt><code>--enable-coverage=<var>level</var></code></dt> 1392 <dd><p>With this option, the compiler is built to collect self coverage 1393 information, every time it is run. This is for internal development 1394 purposes, and only works when the compiler is being built with gcc. The 1395 <var>level</var> argument controls whether the compiler is built optimized or 1396 not, values are ‘<samp>opt</samp>’ and ‘<samp>noopt</samp>’. For coverage analysis you 1397 want to disable optimization, for performance analysis you want to 1398 enable optimization. When coverage is enabled, the default level is 1399 without optimization. 1400 </p> 1401 </dd> 1402 <dt><code>--enable-gather-detailed-mem-stats</code></dt> 1403 <dd><p>When this option is specified more detailed information on memory 1404 allocation is gathered. This information is printed when using 1405 <samp>-fmem-report</samp>. 1406 </p> 1407 </dd> 1408 <dt><code>--enable-valgrind-annotations</code></dt> 1409 <dd><p>Mark selected memory related operations in the compiler when run under 1410 valgrind to suppress false positives. 1411 </p> 1412 </dd> 1413 <dt><code>--enable-nls</code></dt> 1414 <dt><code>--disable-nls</code></dt> 1415 <dd><p>The <samp>--enable-nls</samp> option enables Native Language Support (NLS), 1416 which lets GCC output diagnostics in languages other than American 1417 English. Native Language Support is enabled by default if not doing a 1418 canadian cross build. The <samp>--disable-nls</samp> option disables NLS. 1419 </p> 1420 </dd> 1421 <dt><code>--with-included-gettext</code></dt> 1422 <dd><p>If NLS is enabled, the <samp>--with-included-gettext</samp> option causes the build 1423 procedure to prefer its copy of GNU <code>gettext</code>. 1424 </p> 1425 </dd> 1426 <dt><code>--with-catgets</code></dt> 1427 <dd><p>If NLS is enabled, and if the host lacks <code>gettext</code> but has the 1428 inferior <code>catgets</code> interface, the GCC build procedure normally 1429 ignores <code>catgets</code> and instead uses GCC’s copy of the GNU 1430 <code>gettext</code> library. The <samp>--with-catgets</samp> option causes the 1431 build procedure to use the host’s <code>catgets</code> in this situation. 1432 </p> 1433 </dd> 1434 <dt><code>--with-libiconv-prefix=<var>dir</var></code></dt> 1435 <dd><p>Search for libiconv header files in <samp><var>dir</var>/include</samp> and 1436 libiconv library files in <samp><var>dir</var>/lib</samp>. 1437 </p> 1438 </dd> 1439 <dt><code>--enable-obsolete</code></dt> 1440 <dd><p>Enable configuration for an obsoleted system. If you attempt to 1441 configure GCC for a system (build, host, or target) which has been 1442 obsoleted, and you do not specify this flag, configure will halt with an 1443 error message. 1444 </p> 1445 <p>All support for systems which have been obsoleted in one release of GCC 1446 is removed entirely in the next major release, unless someone steps 1447 forward to maintain the port. 1448 </p> 1449 </dd> 1450 <dt><code>--enable-decimal-float</code></dt> 1451 <dt><code>--enable-decimal-float=yes</code></dt> 1452 <dt><code>--enable-decimal-float=no</code></dt> 1453 <dt><code>--enable-decimal-float=bid</code></dt> 1454 <dt><code>--enable-decimal-float=dpd</code></dt> 1455 <dt><code>--disable-decimal-float</code></dt> 1456 <dd><p>Enable (or disable) support for the C decimal floating point extension 1457 that is in the IEEE 754-2008 standard. This is enabled by default only 1458 on PowerPC, i386, and x86_64 GNU/Linux systems. Other systems may also 1459 support it, but require the user to specifically enable it. You can 1460 optionally control which decimal floating point format is used (either 1461 ‘<samp>bid</samp>’ or ‘<samp>dpd</samp>’). The ‘<samp>bid</samp>’ (binary integer decimal) 1462 format is default on i386 and x86_64 systems, and the ‘<samp>dpd</samp>’ 1463 (densely packed decimal) format is default on PowerPC systems. 1464 </p> 1465 </dd> 1466 <dt><code>--enable-fixed-point</code></dt> 1467 <dt><code>--disable-fixed-point</code></dt> 1468 <dd><p>Enable (or disable) support for C fixed-point arithmetic. 1469 This option is enabled by default for some targets (such as MIPS) which 1470 have hardware-support for fixed-point operations. On other targets, you 1471 may enable this option manually. 1472 </p> 1473 </dd> 1474 <dt><code>--with-long-double-128</code></dt> 1475 <dd><p>Specify if <code>long double</code> type should be 128-bit by default on selected 1476 GNU/Linux architectures. If using <code>--without-long-double-128</code>, 1477 <code>long double</code> will be by default 64-bit, the same as <code>double</code> type. 1478 When neither of these configure options are used, the default will be 1479 128-bit <code>long double</code> when built against GNU C Library 2.4 and later, 1480 64-bit <code>long double</code> otherwise. 1481 </p> 1482 </dd> 1483 <dt><code>--with-long-double-format=ibm</code></dt> 1484 <dt><code>--with-long-double-format=ieee</code></dt> 1485 <dd><p>Specify whether <code>long double</code> uses the IBM extended double format 1486 or the IEEE 128-bit floating point format on PowerPC Linux systems. 1487 This configuration switch will only work on little endian PowerPC 1488 Linux systems and on big endian 64-bit systems where the default cpu 1489 is at least power7 (i.e. <samp>--with-cpu=power7</samp>, 1490 <samp>--with-cpu=power8</samp>, or <samp>--with-cpu=power9</samp> is used). 1491 </p> 1492 <p>If you use the <samp>--with-long-double-64</samp> configuration option, 1493 the <samp>--with-long-double-format=ibm</samp> and 1494 <samp>--with-long-double-format=ieee</samp> options are ignored. 1495 </p> 1496 <p>The default <code>long double</code> format is to use IBM extended double. 1497 Until all of the libraries are converted to use IEEE 128-bit floating 1498 point, it is not recommended to use 1499 <samp>--with-long-double-format=ieee</samp>. 1500 </p> 1501 <p>On little endian PowerPC Linux systems, if you explicitly set the 1502 <code>long double</code> type, it will build multilibs to allow you to 1503 select either <code>long double</code> format, unless you disable multilibs 1504 with the <code>--disable-multilib</code> option. At present, 1505 <code>long double</code> multilibs are not built on big endian PowerPC Linux 1506 systems. If you are building multilibs, you will need to configure 1507 the compiler using the <samp>--with-system-zlib</samp> option. 1508 </p> 1509 <p>If you do not set the <code>long double</code> type explicitly, no multilibs 1510 will be generated. 1511 </p> 1512 </dd> 1513 <dt><code>--enable-fdpic</code></dt> 1514 <dd><p>On SH Linux systems, generate ELF FDPIC code. 1515 </p> 1516 </dd> 1517 <dt><code>--with-gmp=<var>pathname</var></code></dt> 1518 <dt><code>--with-gmp-include=<var>pathname</var></code></dt> 1519 <dt><code>--with-gmp-lib=<var>pathname</var></code></dt> 1520 <dt><code>--with-mpfr=<var>pathname</var></code></dt> 1521 <dt><code>--with-mpfr-include=<var>pathname</var></code></dt> 1522 <dt><code>--with-mpfr-lib=<var>pathname</var></code></dt> 1523 <dt><code>--with-mpc=<var>pathname</var></code></dt> 1524 <dt><code>--with-mpc-include=<var>pathname</var></code></dt> 1525 <dt><code>--with-mpc-lib=<var>pathname</var></code></dt> 1526 <dd><p>If you want to build GCC but do not have the GMP library, the MPFR 1527 library and/or the MPC library installed in a standard location and 1528 do not have their sources present in the GCC source tree then you 1529 can explicitly specify the directory where they are installed 1530 (‘<samp>--with-gmp=<var>gmpinstalldir</var></samp>’, 1531 ‘<samp>--with-mpfr=<var>mpfrinstalldir</var></samp>’, 1532 ‘<samp>--with-mpc=<var>mpcinstalldir</var></samp>’). The 1533 <samp>--with-gmp=<var>gmpinstalldir</var></samp> option is shorthand for 1534 <samp>--with-gmp-lib=<var>gmpinstalldir</var>/lib</samp> and 1535 <samp>--with-gmp-include=<var>gmpinstalldir</var>/include</samp>. Likewise the 1536 <samp>--with-mpfr=<var>mpfrinstalldir</var></samp> option is shorthand for 1537 <samp>--with-mpfr-lib=<var>mpfrinstalldir</var>/lib</samp> and 1538 <samp>--with-mpfr-include=<var>mpfrinstalldir</var>/include</samp>, also the 1539 <samp>--with-mpc=<var>mpcinstalldir</var></samp> option is shorthand for 1540 <samp>--with-mpc-lib=<var>mpcinstalldir</var>/lib</samp> and 1541 <samp>--with-mpc-include=<var>mpcinstalldir</var>/include</samp>. If these 1542 shorthand assumptions are not correct, you can use the explicit 1543 include and lib options directly. You might also need to ensure the 1544 shared libraries can be found by the dynamic linker when building and 1545 using GCC, for example by setting the runtime shared library path 1546 variable (<code>LD_LIBRARY_PATH</code> on GNU/Linux and Solaris systems). 1547 </p> 1548 <p>These flags are applicable to the host platform only. When building 1549 a cross compiler, they will not be used to configure target libraries. 1550 </p> 1551 </dd> 1552 <dt><code>--with-isl=<var>pathname</var></code></dt> 1553 <dt><code>--with-isl-include=<var>pathname</var></code></dt> 1554 <dt><code>--with-isl-lib=<var>pathname</var></code></dt> 1555 <dd><p>If you do not have the isl library installed in a standard location and you 1556 want to build GCC, you can explicitly specify the directory where it is 1557 installed (‘<samp>--with-isl=<var>islinstalldir</var></samp>’). The 1558 <samp>--with-isl=<var>islinstalldir</var></samp> option is shorthand for 1559 <samp>--with-isl-lib=<var>islinstalldir</var>/lib</samp> and 1560 <samp>--with-isl-include=<var>islinstalldir</var>/include</samp>. If this 1561 shorthand assumption is not correct, you can use the explicit 1562 include and lib options directly. 1563 </p> 1564 <p>These flags are applicable to the host platform only. When building 1565 a cross compiler, they will not be used to configure target libraries. 1566 </p> 1567 </dd> 1568 <dt><code>--with-stage1-ldflags=<var>flags</var></code></dt> 1569 <dd><p>This option may be used to set linker flags to be used when linking 1570 stage 1 of GCC. These are also used when linking GCC if configured with 1571 <samp>--disable-bootstrap</samp>. If <samp>--with-stage1-libs</samp> is not set to a 1572 value, then the default is ‘<samp>-static-libstdc++ -static-libgcc</samp>’, if 1573 supported. 1574 </p> 1575 </dd> 1576 <dt><code>--with-stage1-libs=<var>libs</var></code></dt> 1577 <dd><p>This option may be used to set libraries to be used when linking stage 1 1578 of GCC. These are also used when linking GCC if configured with 1579 <samp>--disable-bootstrap</samp>. 1580 </p> 1581 </dd> 1582 <dt><code>--with-boot-ldflags=<var>flags</var></code></dt> 1583 <dd><p>This option may be used to set linker flags to be used when linking 1584 stage 2 and later when bootstrapping GCC. If –with-boot-libs 1585 is not is set to a value, then the default is 1586 ‘<samp>-static-libstdc++ -static-libgcc</samp>’. 1587 </p> 1588 </dd> 1589 <dt><code>--with-boot-libs=<var>libs</var></code></dt> 1590 <dd><p>This option may be used to set libraries to be used when linking stage 2 1591 and later when bootstrapping GCC. 1592 </p> 1593 </dd> 1594 <dt><code>--with-debug-prefix-map=<var>map</var></code></dt> 1595 <dd><p>Convert source directory names using <samp>-fdebug-prefix-map</samp> when 1596 building runtime libraries. ‘<samp><var>map</var></samp>’ is a space-separated 1597 list of maps of the form ‘<samp><var>old</var>=<var>new</var></samp>’. 1598 </p> 1599 </dd> 1600 <dt><code>--enable-linker-build-id</code></dt> 1601 <dd><p>Tells GCC to pass <samp>--build-id</samp> option to the linker for all final 1602 links (links performed without the <samp>-r</samp> or <samp>--relocatable</samp> 1603 option), if the linker supports it. If you specify 1604 <samp>--enable-linker-build-id</samp>, but your linker does not 1605 support <samp>--build-id</samp> option, a warning is issued and the 1606 <samp>--enable-linker-build-id</samp> option is ignored. The default is off. 1607 </p> 1608 </dd> 1609 <dt><code>--with-linker-hash-style=<var>choice</var></code></dt> 1610 <dd><p>Tells GCC to pass <samp>--hash-style=<var>choice</var></samp> option to the 1611 linker for all final links. <var>choice</var> can be one of 1612 ‘<samp>sysv</samp>’, ‘<samp>gnu</samp>’, and ‘<samp>both</samp>’ where ‘<samp>sysv</samp>’ is the default. 1613 </p> 1614 </dd> 1615 <dt><code>--enable-gnu-unique-object</code></dt> 1616 <dt><code>--disable-gnu-unique-object</code></dt> 1617 <dd><p>Tells GCC to use the gnu_unique_object relocation for C++ template 1618 static data members and inline function local statics. Enabled by 1619 default for a toolchain with an assembler that accepts it and 1620 GLIBC 2.11 or above, otherwise disabled. 1621 </p> 1622 </dd> 1623 <dt><code>--with-diagnostics-color=<var>choice</var></code></dt> 1624 <dd><p>Tells GCC to use <var>choice</var> as the default for <samp>-fdiagnostics-color=</samp> 1625 option (if not used explicitly on the command line). <var>choice</var> 1626 can be one of ‘<samp>never</samp>’, ‘<samp>auto</samp>’, ‘<samp>always</samp>’, and ‘<samp>auto-if-env</samp>’ 1627 where ‘<samp>auto</samp>’ is the default. ‘<samp>auto-if-env</samp>’ means that 1628 <samp>-fdiagnostics-color=auto</samp> will be the default if <code>GCC_COLORS</code> 1629 is present and non-empty in the environment, and 1630 <samp>-fdiagnostics-color=never</samp> otherwise. 1631 </p> 1632 </dd> 1633 <dt><code>--enable-lto</code></dt> 1634 <dt><code>--disable-lto</code></dt> 1635 <dd><p>Enable support for link-time optimization (LTO). This is enabled by 1636 default, and may be disabled using <samp>--disable-lto</samp>. 1637 </p> 1638 </dd> 1639 <dt><code>--enable-linker-plugin-configure-flags=FLAGS</code></dt> 1640 <dt><code>--enable-linker-plugin-flags=FLAGS</code></dt> 1641 <dd><p>By default, linker plugins (such as the LTO plugin) are built for the 1642 host system architecture. For the case that the linker has a 1643 different (but run-time compatible) architecture, these flags can be 1644 specified to build plugins that are compatible to the linker. For 1645 example, if you are building GCC for a 64-bit x86_64 1646 (‘<samp>x86_64-unknown-linux-gnu</samp>’) host system, but have a 32-bit x86 1647 GNU/Linux (‘<samp>i686-pc-linux-gnu</samp>’) linker executable (which is 1648 executable on the former system), you can configure GCC as follows for 1649 getting compatible linker plugins: 1650 </p> 1651 <div class="smallexample"> 1652 <pre class="smallexample">% <var>srcdir</var>/configure \ 1653 --host=x86_64-unknown-linux-gnu \ 1654 --enable-linker-plugin-configure-flags=--host=i686-pc-linux-gnu \ 1655 --enable-linker-plugin-flags='CC=gcc\ -m32\ -Wl,-rpath,[...]/i686-pc-linux-gnu/lib' 1656 </pre></div> 1657 1658 </dd> 1659 <dt><code>--with-plugin-ld=<var>pathname</var></code></dt> 1660 <dd><p>Enable an alternate linker to be used at link-time optimization (LTO) 1661 link time when <samp>-fuse-linker-plugin</samp> is enabled. 1662 This linker should have plugin support such as gold starting with 1663 version 2.20 or GNU ld starting with version 2.21. 1664 See <samp>-fuse-linker-plugin</samp> for details. 1665 </p> 1666 </dd> 1667 <dt><code>--enable-canonical-system-headers</code></dt> 1668 <dt><code>--disable-canonical-system-headers</code></dt> 1669 <dd><p>Enable system header path canonicalization for <samp>libcpp</samp>. This can 1670 produce shorter header file paths in diagnostics and dependency output 1671 files, but these changed header paths may conflict with some compilation 1672 environments. Enabled by default, and may be disabled using 1673 <samp>--disable-canonical-system-headers</samp>. 1674 </p> 1675 </dd> 1676 <dt><code>--with-glibc-version=<var>major</var>.<var>minor</var></code></dt> 1677 <dd><p>Tell GCC that when the GNU C Library (glibc) is used on the target it 1678 will be version <var>major</var>.<var>minor</var> or later. Normally this can 1679 be detected from the C library’s header files, but this option may be 1680 needed when bootstrapping a cross toolchain without the header files 1681 available for building the initial bootstrap compiler. 1682 </p> 1683 <p>If GCC is configured with some multilibs that use glibc and some that 1684 do not, this option applies only to the multilibs that use glibc. 1685 However, such configurations may not work well as not all the relevant 1686 configuration in GCC is on a per-multilib basis. 1687 </p> 1688 </dd> 1689 <dt><code>--enable-as-accelerator-for=<var>target</var></code></dt> 1690 <dd><p>Build as offload target compiler. Specify offload host triple by <var>target</var>. 1691 </p> 1692 </dd> 1693 <dt><code>--enable-offload-targets=<var>target1</var>[=<var>path1</var>],…,<var>targetN</var>[=<var>pathN</var>]</code></dt> 1694 <dd><p>Enable offloading to targets <var>target1</var>, …, <var>targetN</var>. 1695 Offload compilers are expected to be already installed. Default search 1696 path for them is <samp><var>exec-prefix</var></samp>, but it can be changed by 1697 specifying paths <var>path1</var>, …, <var>pathN</var>. 1698 </p> 1699 <div class="smallexample"> 1700 <pre class="smallexample">% <var>srcdir</var>/configure \ 1701 --enable-offload-target=i686-unknown-linux-gnu=/path/to/i686/compiler,x86_64-pc-linux-gnu 1702 </pre></div> 1703 1704 <p>If ‘<samp>hsa</samp>’ is specified as one of the targets, the compiler will be 1705 built with support for HSA GPU accelerators. Because the same 1706 compiler will emit the accelerator code, no path should be specified. 1707 </p> 1708 </dd> 1709 <dt><code>--with-hsa-runtime=<var>pathname</var></code></dt> 1710 <dt><code>--with-hsa-runtime-include=<var>pathname</var></code></dt> 1711 <dt><code>--with-hsa-runtime-lib=<var>pathname</var></code></dt> 1712 <dd> 1713 <p>If you configure GCC with HSA offloading but do not have the HSA 1714 run-time library installed in a standard location then you can 1715 explicitly specify the directory where they are installed. The 1716 <samp>--with-hsa-runtime=<var>hsainstalldir</var></samp> option is a 1717 shorthand for 1718 <samp>--with-hsa-runtime-lib=<var>hsainstalldir</var>/lib</samp> and 1719 <samp>--with-hsa-runtime-include=<var>hsainstalldir</var>/include</samp>. 1720 </p> 1721 </dd> 1722 <dt><code>--enable-cet</code></dt> 1723 <dt><code>--disable-cet</code></dt> 1724 <dd><p>Enable building target run-time libraries with control-flow 1725 instrumentation, see <samp>-fcf-protection</samp> option. When 1726 <code>--enable-cet</code> is specified target libraries are configured 1727 to add <samp>-fcf-protection</samp> and, if needed, other target 1728 specific options to a set of building options. 1729 </p> 1730 <p>The option is disabled by default. When <code>--enable-cet=auto</code> 1731 is used, it is enabled on Linux/x86 if target binutils 1732 supports <code>Intel CET</code> instructions and disabled otherwise. 1733 In this case the target libraries are configured to get additional 1734 <samp>-fcf-protection</samp> option. 1735 </p></dd> 1736 </dl> 1737 1738 <a name="Cross-Compiler-Specific-Options"></a> 1739 <h4 class="subheading">Cross-Compiler-Specific Options</h4> 1740 <p>The following options only apply to building cross compilers. 1741 </p> 1742 <dl compact="compact"> 1743 <dt><code>--with-sysroot</code></dt> 1744 <dt><code>--with-sysroot=<var>dir</var></code></dt> 1745 <dd><p>Tells GCC to consider <var>dir</var> as the root of a tree that contains 1746 (a subset of) the root filesystem of the target operating system. 1747 Target system headers, libraries and run-time object files will be 1748 searched for in there. More specifically, this acts as if 1749 <samp>--sysroot=<var>dir</var></samp> was added to the default options of the built 1750 compiler. The specified directory is not copied into the 1751 install tree, unlike the options <samp>--with-headers</samp> and 1752 <samp>--with-libs</samp> that this option obsoletes. The default value, 1753 in case <samp>--with-sysroot</samp> is not given an argument, is 1754 <samp>${gcc_tooldir}/sys-root</samp>. If the specified directory is a 1755 subdirectory of <samp>${exec_prefix}</samp>, then it will be found relative to 1756 the GCC binaries if the installation tree is moved. 1757 </p> 1758 <p>This option affects the system root for the compiler used to build 1759 target libraries (which runs on the build system) and the compiler newly 1760 installed with <code>make install</code>; it does not affect the compiler which is 1761 used to build GCC itself. 1762 </p> 1763 <p>If you specify the <samp>--with-native-system-header-dir=<var>dirname</var></samp> 1764 option then the compiler will search that directory within <var>dirname</var> for 1765 native system headers rather than the default <samp>/usr/include</samp>. 1766 </p> 1767 </dd> 1768 <dt><code>--with-build-sysroot</code></dt> 1769 <dt><code>--with-build-sysroot=<var>dir</var></code></dt> 1770 <dd><p>Tells GCC to consider <var>dir</var> as the system root (see 1771 <samp>--with-sysroot</samp>) while building target libraries, instead of 1772 the directory specified with <samp>--with-sysroot</samp>. This option is 1773 only useful when you are already using <samp>--with-sysroot</samp>. You 1774 can use <samp>--with-build-sysroot</samp> when you are configuring with 1775 <samp>--prefix</samp> set to a directory that is different from the one in 1776 which you are installing GCC and your target libraries. 1777 </p> 1778 <p>This option affects the system root for the compiler used to build 1779 target libraries (which runs on the build system); it does not affect 1780 the compiler which is used to build GCC itself. 1781 </p> 1782 <p>If you specify the <samp>--with-native-system-header-dir=<var>dirname</var></samp> 1783 option then the compiler will search that directory within <var>dirname</var> for 1784 native system headers rather than the default <samp>/usr/include</samp>. 1785 </p> 1786 </dd> 1787 <dt><code>--with-headers</code></dt> 1788 <dt><code>--with-headers=<var>dir</var></code></dt> 1789 <dd><p>Deprecated in favor of <samp>--with-sysroot</samp>. 1790 Specifies that target headers are available when building a cross compiler. 1791 The <var>dir</var> argument specifies a directory which has the target include 1792 files. These include files will be copied into the <samp>gcc</samp> install 1793 directory. <em>This option with the <var>dir</var> argument is required</em> when 1794 building a cross compiler, if <samp><var>prefix</var>/<var>target</var>/sys-include</samp> 1795 doesn’t pre-exist. If <samp><var>prefix</var>/<var>target</var>/sys-include</samp> does 1796 pre-exist, the <var>dir</var> argument may be omitted. <code>fixincludes</code> 1797 will be run on these files to make them compatible with GCC. 1798 </p> 1799 </dd> 1800 <dt><code>--without-headers</code></dt> 1801 <dd><p>Tells GCC not use any target headers from a libc when building a cross 1802 compiler. When crossing to GNU/Linux, you need the headers so GCC 1803 can build the exception handling for libgcc. 1804 </p> 1805 </dd> 1806 <dt><code>--with-libs</code></dt> 1807 <dt><code>--with-libs="<var>dir1</var> <var>dir2</var> … <var>dirN</var>"</code></dt> 1808 <dd><p>Deprecated in favor of <samp>--with-sysroot</samp>. 1809 Specifies a list of directories which contain the target runtime 1810 libraries. These libraries will be copied into the <samp>gcc</samp> install 1811 directory. If the directory list is omitted, this option has no 1812 effect. 1813 </p> 1814 </dd> 1815 <dt><code>--with-newlib</code></dt> 1816 <dd><p>Specifies that ‘<samp>newlib</samp>’ is 1817 being used as the target C library. This causes <code>__eprintf</code> to be 1818 omitted from <samp>libgcc.a</samp> on the assumption that it will be provided by 1819 ‘<samp>newlib</samp>’. 1820 </p> 1821 </dd> 1822 <dt><code>--with-avrlibc</code></dt> 1823 <dd><p>Specifies that ‘<samp>AVR-Libc</samp>’ is 1824 being used as the target C library. This causes float support 1825 functions like <code>__addsf3</code> to be omitted from <samp>libgcc.a</samp> on 1826 the assumption that it will be provided by <samp>libm.a</samp>. For more 1827 technical details, cf. <a href="http://gcc.gnu.org/PR54461">PR54461</a>. 1828 This option is only supported for the AVR target. It is not supported for 1829 RTEMS configurations, which currently use newlib. The option is 1830 supported since version 4.7.2 and is the default in 4.8.0 and newer. 1831 </p> 1832 </dd> 1833 <dt><code>--with-nds32-lib=<var>library</var></code></dt> 1834 <dd><p>Specifies that <var>library</var> setting is used for building <samp>libgcc.a</samp>. 1835 Currently, the valid <var>library</var> is ‘<samp>newlib</samp>’ or ‘<samp>mculib</samp>’. 1836 This option is only supported for the NDS32 target. 1837 </p> 1838 </dd> 1839 <dt><code>--with-build-time-tools=<var>dir</var></code></dt> 1840 <dd><p>Specifies where to find the set of target tools (assembler, linker, etc.) 1841 that will be used while building GCC itself. This option can be useful 1842 if the directory layouts are different between the system you are building 1843 GCC on, and the system where you will deploy it. 1844 </p> 1845 <p>For example, on an ‘<samp>ia64-hp-hpux</samp>’ system, you may have the GNU 1846 assembler and linker in <samp>/usr/bin</samp>, and the native tools in a 1847 different path, and build a toolchain that expects to find the 1848 native tools in <samp>/usr/bin</samp>. 1849 </p> 1850 <p>When you use this option, you should ensure that <var>dir</var> includes 1851 <code>ar</code>, <code>as</code>, <code>ld</code>, <code>nm</code>, 1852 <code>ranlib</code> and <code>strip</code> if necessary, and possibly 1853 <code>objdump</code>. Otherwise, GCC may use an inconsistent set of 1854 tools. 1855 </p></dd> 1856 </dl> 1857 1858 <a name="Overriding-configure-test-results"></a> 1859 <h4 class="subsubheading">Overriding <code>configure</code> test results</h4> 1860 1861 <p>Sometimes, it might be necessary to override the result of some 1862 <code>configure</code> test, for example in order to ease porting to a new 1863 system or work around a bug in a test. The toplevel <code>configure</code> 1864 script provides three variables for this: 1865 </p> 1866 <dl compact="compact"> 1867 <dt><code>build_configargs</code></dt> 1868 <dd><a name="index-build_005fconfigargs"></a> 1869 <p>The contents of this variable is passed to all build <code>configure</code> 1870 scripts. 1871 </p> 1872 </dd> 1873 <dt><code>host_configargs</code></dt> 1874 <dd><a name="index-host_005fconfigargs"></a> 1875 <p>The contents of this variable is passed to all host <code>configure</code> 1876 scripts. 1877 </p> 1878 </dd> 1879 <dt><code>target_configargs</code></dt> 1880 <dd><a name="index-target_005fconfigargs"></a> 1881 <p>The contents of this variable is passed to all target <code>configure</code> 1882 scripts. 1883 </p> 1884 </dd> 1885 </dl> 1886 1887 <p>In order to avoid shell and <code>make</code> quoting issues for complex 1888 overrides, you can pass a setting for <code>CONFIG_SITE</code> and set 1889 variables in the site file. 1890 </p> 1891 <a name="Objective-C-Specific-Options"></a> 1892 <h4 class="subheading">Objective-C-Specific Options</h4> 1893 1894 <p>The following options apply to the build of the Objective-C runtime library. 1895 </p> 1896 <dl compact="compact"> 1897 <dt><code>--enable-objc-gc</code></dt> 1898 <dd><p>Specify that an additional variant of the GNU Objective-C runtime library 1899 is built, using an external build of the Boehm-Demers-Weiser garbage 1900 collector (<a href="http://www.hboehm.info/gc/">http://www.hboehm.info/gc/</a>). This library needs to be 1901 available for each multilib variant, unless configured with 1902 <samp>--enable-objc-gc=‘<samp>auto</samp>’</samp> in which case the build of the 1903 additional runtime library is skipped when not available and the build 1904 continues. 1905 </p> 1906 </dd> 1907 <dt><code>--with-target-bdw-gc=<var>list</var></code></dt> 1908 <dt><code>--with-target-bdw-gc-include=<var>list</var></code></dt> 1909 <dt><code>--with-target-bdw-gc-lib=<var>list</var></code></dt> 1910 <dd><p>Specify search directories for the garbage collector header files and 1911 libraries. <var>list</var> is a comma separated list of key value pairs of the 1912 form ‘<samp><var>multilibdir</var>=<var>path</var></samp>’, where the default multilib key 1913 is named as ‘<samp>.</samp>’ (dot), or is omitted (e.g. 1914 ‘<samp>--with-target-bdw-gc=/opt/bdw-gc,32=/opt-bdw-gc32</samp>’). 1915 </p> 1916 <p>The options <samp>--with-target-bdw-gc-include</samp> and 1917 <samp>--with-target-bdw-gc-lib</samp> must always be specified together 1918 for each multilib variant and they take precedence over 1919 <samp>--with-target-bdw-gc</samp>. If <samp>--with-target-bdw-gc-include</samp> 1920 is missing values for a multilib, then the value for the default 1921 multilib is used (e.g. ‘<samp>--with-target-bdw-gc-include=/opt/bdw-gc/include</samp>’ 1922 ‘<samp>--with-target-bdw-gc-lib=/opt/bdw-gc/lib64,32=/opt-bdw-gc/lib32</samp>’). 1923 If none of these options are specified, the library is assumed in 1924 default locations. 1925 </p></dd> 1926 </dl> 1927 1928 <hr /> 1929 <p> 1930 <p><a href="./index.html">Return to the GCC Installation page</a> 1931 </p> 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 </body> 1944 </html> 1945
Indexes created Thu Jul 16 00:26:27 UTC 2026