specific.html revision 1.17 1 1.10 mrg <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
2 1.10 mrg <html>
3 1.15 mrg <!-- Copyright (C) 1988-2022 Free Software Foundation, Inc.
4 1.1 mrg
5 1.10 mrg Permission is granted to copy, distribute and/or modify this document
6 1.3 skrll under the terms of the GNU Free Documentation License, Version 1.3 or
7 1.1 mrg any later version published by the Free Software Foundation; with no
8 1.1 mrg Invariant Sections, the Front-Cover texts being (a) (see below), and
9 1.1 mrg with the Back-Cover Texts being (b) (see below). A copy of the
10 1.10 mrg license is included in the section entitled "GNU
11 1.10 mrg Free Documentation License".
12 1.1 mrg
13 1.1 mrg (a) The FSF's Front-Cover Text is:
14 1.1 mrg
15 1.10 mrg A GNU Manual
16 1.1 mrg
17 1.1 mrg (b) The FSF's Back-Cover Text is:
18 1.1 mrg
19 1.10 mrg You have freedom to copy and modify this GNU Manual, like GNU
20 1.1 mrg software. Copies published by the Free Software Foundation raise
21 1.10 mrg funds for GNU development. -->
22 1.17 mrg <!-- Created by GNU Texinfo 6.5, http://www.gnu.org/software/texinfo/ -->
23 1.17 mrg <head>
24 1.17 mrg <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
25 1.12 mrg <title>Host/Target specific installation notes for GCC</title>
26 1.10 mrg
27 1.12 mrg <meta name="description" content="Host/Target specific installation notes for GCC">
28 1.12 mrg <meta name="keywords" content="Host/Target specific installation notes for GCC">
29 1.10 mrg <meta name="resource-type" content="document">
30 1.10 mrg <meta name="distribution" content="global">
31 1.10 mrg <meta name="Generator" content="makeinfo">
32 1.10 mrg <style type="text/css">
33 1.10 mrg <!--
34 1.10 mrg a.summary-letter {text-decoration: none}
35 1.10 mrg blockquote.indentedblock {margin-right: 0em}
36 1.17 mrg blockquote.smallindentedblock {margin-right: 0em; font-size: smaller}
37 1.17 mrg blockquote.smallquotation {font-size: smaller}
38 1.10 mrg div.display {margin-left: 3.2em}
39 1.10 mrg div.example {margin-left: 3.2em}
40 1.17 mrg div.lisp {margin-left: 3.2em}
41 1.17 mrg div.smalldisplay {margin-left: 3.2em}
42 1.17 mrg div.smallexample {margin-left: 3.2em}
43 1.17 mrg div.smalllisp {margin-left: 3.2em}
44 1.10 mrg kbd {font-style: oblique}
45 1.10 mrg pre.display {font-family: inherit}
46 1.10 mrg pre.format {font-family: inherit}
47 1.10 mrg pre.menu-comment {font-family: serif}
48 1.10 mrg pre.menu-preformatted {font-family: serif}
49 1.17 mrg pre.smalldisplay {font-family: inherit; font-size: smaller}
50 1.17 mrg pre.smallexample {font-size: smaller}
51 1.17 mrg pre.smallformat {font-family: inherit; font-size: smaller}
52 1.17 mrg pre.smalllisp {font-size: smaller}
53 1.10 mrg span.nolinebreak {white-space: nowrap}
54 1.10 mrg span.roman {font-family: initial; font-weight: normal}
55 1.10 mrg span.sansserif {font-family: sans-serif; font-weight: normal}
56 1.10 mrg ul.no-bullet {list-style: none}
57 1.10 mrg -->
58 1.10 mrg </style>
59 1.10 mrg
60 1.10 mrg
61 1.9 mrg </head>
62 1.8 mrg
63 1.10 mrg <body lang="en">
64 1.12 mrg <h1 class="settitle" align="center">Host/Target specific installation notes for GCC</h1>
65 1.8 mrg
66 1.8 mrg
67 1.6 mrg
68 1.8 mrg
69 1.8 mrg
70 1.8 mrg
71 1.8 mrg
72 1.8 mrg
73 1.8 mrg
74 1.8 mrg
75 1.8 mrg
76 1.8 mrg
77 1.8 mrg
78 1.8 mrg
79 1.8 mrg
80 1.8 mrg
81 1.8 mrg
82 1.8 mrg
83 1.8 mrg
84 1.8 mrg
85 1.8 mrg
86 1.8 mrg
87 1.8 mrg
88 1.8 mrg
89 1.17 mrg <a name="index-Specific"></a>
90 1.17 mrg <a name="index-Specific-installation-notes"></a>
91 1.17 mrg <a name="index-Target-specific-installation"></a>
92 1.17 mrg <a name="index-Host-specific-installation"></a>
93 1.17 mrg <a name="index-Target-specific-installation-notes"></a>
94 1.6 mrg
95 1.10 mrg <p>Please read this document carefully <em>before</em> installing the
96 1.10 mrg GNU Compiler Collection on your machine.
97 1.10 mrg </p>
98 1.10 mrg <p>Note that this list of install notes is <em>not</em> a list of supported
99 1.10 mrg hosts or targets. Not all supported hosts and targets are listed
100 1.10 mrg here, only the ones that require host-specific or target-specific
101 1.10 mrg information have to.
102 1.10 mrg </p>
103 1.10 mrg <ul>
104 1.10 mrg <li> <a href="#aarch64-x-x">aarch64*-*-*</a>
105 1.10 mrg </li><li> <a href="#alpha-x-x">alpha*-*-*</a>
106 1.15 mrg </li><li> <a href="#amdgcn-x-amdhsa">amdgcn-*-amdhsa</a>
107 1.12 mrg </li><li> <a href="#amd64-x-solaris2">amd64-*-solaris2*</a>
108 1.15 mrg </li><li> <a href="#arc-x-elf32">arc-*-elf32</a>
109 1.15 mrg </li><li> <a href="#arc-linux-uclibc">arc-linux-uclibc</a>
110 1.10 mrg </li><li> <a href="#arm-x-eabi">arm-*-eabi</a>
111 1.10 mrg </li><li> <a href="#avr">avr</a>
112 1.10 mrg </li><li> <a href="#bfin">Blackfin</a>
113 1.15 mrg </li><li> <a href="#cris">cris</a>
114 1.10 mrg </li><li> <a href="#dos">DOS</a>
115 1.15 mrg </li><li> <a href="#epiphany-x-elf">epiphany-*-elf</a>
116 1.15 mrg </li><li> <a href="#ft32-x-elf">ft32-*-elf</a>
117 1.10 mrg </li><li> <a href="#x-x-freebsd">*-*-freebsd*</a>
118 1.10 mrg </li><li> <a href="#h8300-hms">h8300-hms</a>
119 1.10 mrg </li><li> <a href="#hppa-hp-hpux">hppa*-hp-hpux*</a>
120 1.10 mrg </li><li> <a href="#hppa-hp-hpux10">hppa*-hp-hpux10</a>
121 1.10 mrg </li><li> <a href="#hppa-hp-hpux11">hppa*-hp-hpux11</a>
122 1.10 mrg </li><li> <a href="#x-x-linux-gnu">*-*-linux-gnu</a>
123 1.10 mrg </li><li> <a href="#ix86-x-linux">i?86-*-linux*</a>
124 1.12 mrg </li><li> <a href="#ix86-x-solaris2">i?86-*-solaris2*</a>
125 1.10 mrg </li><li> <a href="#ia64-x-linux">ia64-*-linux</a>
126 1.10 mrg </li><li> <a href="#ia64-x-hpux">ia64-*-hpux*</a>
127 1.10 mrg </li><li> <a href="#x-ibm-aix">*-ibm-aix*</a>
128 1.10 mrg </li><li> <a href="#iq2000-x-elf">iq2000-*-elf</a>
129 1.15 mrg </li><li> <a href="#loongarch">loongarch</a>
130 1.10 mrg </li><li> <a href="#lm32-x-elf">lm32-*-elf</a>
131 1.10 mrg </li><li> <a href="#lm32-x-uclinux">lm32-*-uclinux</a>
132 1.10 mrg </li><li> <a href="#m32c-x-elf">m32c-*-elf</a>
133 1.10 mrg </li><li> <a href="#m32r-x-elf">m32r-*-elf</a>
134 1.10 mrg </li><li> <a href="#m68k-x-x">m68k-*-*</a>
135 1.15 mrg </li><li> <a href="#m68k-x-uclinux">m68k-*-uclinux</a>
136 1.10 mrg </li><li> <a href="#microblaze-x-elf">microblaze-*-elf</a>
137 1.10 mrg </li><li> <a href="#mips-x-x">mips-*-*</a>
138 1.15 mrg </li><li> <a href="#moxie-x-elf">moxie-*-elf</a>
139 1.15 mrg </li><li> <a href="#msp430-x-elf">msp430-*-elf</a>
140 1.10 mrg </li><li> <a href="#nds32le-x-elf">nds32le-*-elf</a>
141 1.10 mrg </li><li> <a href="#nds32be-x-elf">nds32be-*-elf</a>
142 1.10 mrg </li><li> <a href="#nvptx-x-none">nvptx-*-none</a>
143 1.11 mrg </li><li> <a href="#or1k-x-elf">or1k-*-elf</a>
144 1.11 mrg </li><li> <a href="#or1k-x-linux">or1k-*-linux</a>
145 1.10 mrg </li><li> <a href="#powerpc-x-x">powerpc*-*-*</a>
146 1.10 mrg </li><li> <a href="#powerpc-x-darwin">powerpc-*-darwin*</a>
147 1.10 mrg </li><li> <a href="#powerpc-x-elf">powerpc-*-elf</a>
148 1.10 mrg </li><li> <a href="#powerpc-x-linux-gnu">powerpc*-*-linux-gnu*</a>
149 1.10 mrg </li><li> <a href="#powerpc-x-netbsd">powerpc-*-netbsd*</a>
150 1.10 mrg </li><li> <a href="#powerpc-x-eabisim">powerpc-*-eabisim</a>
151 1.10 mrg </li><li> <a href="#powerpc-x-eabi">powerpc-*-eabi</a>
152 1.10 mrg </li><li> <a href="#powerpcle-x-elf">powerpcle-*-elf</a>
153 1.10 mrg </li><li> <a href="#powerpcle-x-eabisim">powerpcle-*-eabisim</a>
154 1.10 mrg </li><li> <a href="#powerpcle-x-eabi">powerpcle-*-eabi</a>
155 1.10 mrg </li><li> <a href="#riscv32-x-elf">riscv32-*-elf</a>
156 1.10 mrg </li><li> <a href="#riscv32-x-linux">riscv32-*-linux</a>
157 1.10 mrg </li><li> <a href="#riscv64-x-elf">riscv64-*-elf</a>
158 1.10 mrg </li><li> <a href="#riscv64-x-linux">riscv64-*-linux</a>
159 1.15 mrg </li><li> <a href="#rl78-x-elf">rl78-*-elf</a>
160 1.15 mrg </li><li> <a href="#rx-x-elf">rx-*-elf</a>
161 1.10 mrg </li><li> <a href="#s390-x-linux">s390-*-linux*</a>
162 1.10 mrg </li><li> <a href="#s390x-x-linux">s390x-*-linux*</a>
163 1.10 mrg </li><li> <a href="#s390x-ibm-tpf">s390x-ibm-tpf*</a>
164 1.10 mrg </li><li> <a href="#x-x-solaris2">*-*-solaris2*</a>
165 1.10 mrg </li><li> <a href="#sparc-x-x">sparc*-*-*</a>
166 1.10 mrg </li><li> <a href="#sparc-sun-solaris2">sparc-sun-solaris2*</a>
167 1.10 mrg </li><li> <a href="#sparc-x-linux">sparc-*-linux*</a>
168 1.10 mrg </li><li> <a href="#sparc64-x-solaris2">sparc64-*-solaris2*</a>
169 1.10 mrg </li><li> <a href="#sparcv9-x-solaris2">sparcv9-*-solaris2*</a>
170 1.10 mrg </li><li> <a href="#c6x-x-x">c6x-*-*</a>
171 1.10 mrg </li><li> <a href="#tilegx-x-linux">tilegx-*-linux*</a>
172 1.10 mrg </li><li> <a href="#tilegxbe-x-linux">tilegxbe-*-linux*</a>
173 1.10 mrg </li><li> <a href="#tilepro-x-linux">tilepro-*-linux*</a>
174 1.10 mrg </li><li> <a href="#visium-x-elf">visium-*-elf</a>
175 1.10 mrg </li><li> <a href="#x-x-vxworks">*-*-vxworks*</a>
176 1.10 mrg </li><li> <a href="#x86-64-x-x">x86_64-*-*, amd64-*-*</a>
177 1.12 mrg </li><li> <a href="#x86-64-x-solaris2">x86_64-*-solaris2*</a>
178 1.10 mrg </li><li> <a href="#xtensa-x-elf">xtensa*-*-elf</a>
179 1.10 mrg </li><li> <a href="#xtensa-x-linux">xtensa*-*-linux*</a>
180 1.10 mrg </li><li> <a href="#windows">Microsoft Windows</a>
181 1.10 mrg </li><li> <a href="#x-x-cygwin">*-*-cygwin</a>
182 1.10 mrg </li><li> <a href="#x-x-mingw32">*-*-mingw32</a>
183 1.10 mrg </li><li> <a href="#os2">OS/2</a>
184 1.10 mrg </li><li> <a href="#older">Older systems</a>
185 1.10 mrg </li></ul>
186 1.10 mrg
187 1.10 mrg <ul>
188 1.10 mrg <li> <a href="#elf">all ELF targets</a> (SVR4, Solaris 2, etc.)
189 1.10 mrg </li></ul>
190 1.10 mrg
191 1.10 mrg
192 1.10 mrg <!-- -------- host/target specific issues start here ---------------- -->
193 1.10 mrg <hr />
194 1.17 mrg <a name="aarch64-x-x"></a><a name="aarch64*-*-*"></a>
195 1.17 mrg <h3 class="heading">aarch64*-*-*</h3>
196 1.10 mrg <p>Binutils pre 2.24 does not have support for selecting <samp>-mabi</samp> and
197 1.10 mrg does not support ILP32. If it is used to build GCC 4.9 or later, GCC will
198 1.10 mrg not support option <samp>-mabi=ilp32</samp>.
199 1.10 mrg </p>
200 1.10 mrg <p>To enable a workaround for the Cortex-A53 erratum number 835769 by default
201 1.10 mrg (for all CPUs regardless of -mcpu option given) at configure time use the
202 1.10 mrg <samp>--enable-fix-cortex-a53-835769</samp> option. This will enable the fix by
203 1.10 mrg default and can be explicitly disabled during compilation by passing the
204 1.10 mrg <samp>-mno-fix-cortex-a53-835769</samp> option. Conversely,
205 1.10 mrg <samp>--disable-fix-cortex-a53-835769</samp> will disable the workaround by
206 1.10 mrg default. The workaround is disabled by default if neither of
207 1.10 mrg <samp>--enable-fix-cortex-a53-835769</samp> or
208 1.10 mrg <samp>--disable-fix-cortex-a53-835769</samp> is given at configure time.
209 1.10 mrg </p>
210 1.10 mrg <p>To enable a workaround for the Cortex-A53 erratum number 843419 by default
211 1.10 mrg (for all CPUs regardless of -mcpu option given) at configure time use the
212 1.10 mrg <samp>--enable-fix-cortex-a53-843419</samp> option. This workaround is applied at
213 1.10 mrg link time. Enabling the workaround will cause GCC to pass the relevant option
214 1.10 mrg to the linker. It can be explicitly disabled during compilation by passing the
215 1.10 mrg <samp>-mno-fix-cortex-a53-843419</samp> option. Conversely,
216 1.10 mrg <samp>--disable-fix-cortex-a53-843419</samp> will disable the workaround by default.
217 1.10 mrg The workaround is disabled by default if neither of
218 1.10 mrg <samp>--enable-fix-cortex-a53-843419</samp> or
219 1.10 mrg <samp>--disable-fix-cortex-a53-843419</samp> is given at configure time.
220 1.10 mrg </p>
221 1.11 mrg <p>To enable Branch Target Identification Mechanism and Return Address Signing by
222 1.11 mrg default at configure time use the <samp>--enable-standard-branch-protection</samp>
223 1.11 mrg option. This is equivalent to having <samp>-mbranch-protection=standard</samp>
224 1.11 mrg during compilation. This can be explicitly disabled during compilation by
225 1.11 mrg passing the <samp>-mbranch-protection=none</samp> option which turns off all
226 1.11 mrg types of branch protections. Conversely,
227 1.11 mrg <samp>--disable-standard-branch-protection</samp> will disable both the
228 1.11 mrg protections by default. This mechanism is turned off by default if neither
229 1.11 mrg of the options are given at configure time.
230 1.11 mrg </p>
231 1.10 mrg <hr />
232 1.17 mrg <a name="alpha-x-x"></a><a name="alpha*-*-*"></a>
233 1.17 mrg <h3 class="heading">alpha*-*-*</h3>
234 1.10 mrg <p>This section contains general configuration information for all
235 1.10 mrg Alpha-based platforms using ELF. In addition to reading this
236 1.10 mrg section, please read all other sections that match your target.
237 1.10 mrg </p>
238 1.10 mrg <hr />
239 1.17 mrg <a name="amd64-x-solaris2"></a><a name="amd64-*-solaris2*"></a>
240 1.17 mrg <h3 class="heading">amd64-*-solaris2*</h3>
241 1.12 mrg <p>This is a synonym for ‘<samp>x86_64-*-solaris2*</samp>’.
242 1.10 mrg </p>
243 1.10 mrg <hr />
244 1.17 mrg <a name="amdgcn-x-amdhsa"></a><a name="amdgcn-*-amdhsa"></a>
245 1.17 mrg <h3 class="heading">amdgcn-*-amdhsa</h3>
246 1.11 mrg <p>AMD GCN GPU target.
247 1.11 mrg </p>
248 1.15 mrg <p>Instead of GNU Binutils, you will need to install LLVM 13.0.1, or later, and copy
249 1.12 mrg <samp>bin/llvm-mc</samp> to <samp>amdgcn-amdhsa/bin/as</samp>,
250 1.12 mrg <samp>bin/lld</samp> to <samp>amdgcn-amdhsa/bin/ld</samp>,
251 1.12 mrg <samp>bin/llvm-nm</samp> to <samp>amdgcn-amdhsa/bin/nm</samp>, and
252 1.12 mrg <samp>bin/llvm-ar</samp> to both <samp>bin/amdgcn-amdhsa-ar</samp> and
253 1.12 mrg <samp>bin/amdgcn-amdhsa-ranlib</samp>.
254 1.11 mrg </p>
255 1.15 mrg <p>Use Newlib (3.2.0, or newer).
256 1.11 mrg </p>
257 1.11 mrg <p>To run the binaries, install the HSA Runtime from the
258 1.11 mrg <a href="https://rocm.github.io">ROCm Platform</a>, and use
259 1.12 mrg <samp>libexec/gcc/amdhsa-amdhsa/<var>version</var>/gcn-run</samp> to launch them
260 1.11 mrg on the GPU.
261 1.11 mrg </p>
262 1.11 mrg <hr />
263 1.17 mrg <a name="arc-x-elf32"></a><a name="arc-*-elf32"></a>
264 1.17 mrg <h3 class="heading">arc-*-elf32</h3>
265 1.10 mrg
266 1.10 mrg <p>Use ‘<samp>configure --target=arc-elf32 --with-cpu=<var>cpu</var> --enable-languages="c,c++"</samp>’
267 1.10 mrg to configure GCC, with <var>cpu</var> being one of ‘<samp>arc600</samp>’, ‘<samp>arc601</samp>’,
268 1.10 mrg or ‘<samp>arc700</samp>’.
269 1.10 mrg </p>
270 1.10 mrg <hr />
271 1.17 mrg <a name="arc-linux-uclibc"></a><a name="arc-linux-uclibc-1"></a>
272 1.17 mrg <h3 class="heading">arc-linux-uclibc</h3>
273 1.10 mrg
274 1.10 mrg <p>Use ‘<samp>configure --target=arc-linux-uclibc --with-cpu=arc700 --enable-languages="c,c++"</samp>’ to configure GCC.
275 1.10 mrg </p>
276 1.10 mrg <hr />
277 1.17 mrg <a name="arm-x-eabi"></a><a name="arm-*-eabi"></a>
278 1.17 mrg <h3 class="heading">arm-*-eabi</h3>
279 1.9 mrg <p>ARM-family processors.
280 1.10 mrg </p>
281 1.10 mrg <p>Building the Ada frontend commonly fails (an infinite loop executing
282 1.7 mrg <code>xsinfo</code>) if the host compiler is GNAT 4.8. Host compilers built from the
283 1.7 mrg GNAT 4.6, 4.9 or 5 release branches are known to succeed.
284 1.10 mrg </p>
285 1.10 mrg <hr />
286 1.17 mrg <a name="avr"></a><a name="avr-1"></a>
287 1.17 mrg <h3 class="heading">avr</h3>
288 1.1 mrg <p>ATMEL AVR-family micro controllers. These are used in embedded
289 1.10 mrg applications. There are no standard Unix configurations.
290 1.1 mrg See “AVR Options” in the main manual
291 1.1 mrg for the list of supported MCU types.
292 1.10 mrg </p>
293 1.10 mrg <p>Use ‘<samp>configure --target=avr --enable-languages="c"</samp>’ to configure GCC.
294 1.10 mrg </p>
295 1.10 mrg <p>Further installation notes and other useful information about AVR tools
296 1.1 mrg can also be obtained from:
297 1.10 mrg </p>
298 1.10 mrg <ul>
299 1.10 mrg <li> <a href="http://www.nongnu.org/avr/">http://www.nongnu.org/avr/</a>
300 1.10 mrg </li><li> <a href="http://www.amelek.gda.pl/avr/">http://www.amelek.gda.pl/avr/</a>
301 1.10 mrg </li></ul>
302 1.10 mrg
303 1.10 mrg <p>The following error:
304 1.17 mrg </p><div class="smallexample">
305 1.17 mrg <pre class="smallexample">Error: register required
306 1.10 mrg </pre></div>
307 1.10 mrg
308 1.10 mrg <p>indicates that you should upgrade to a newer version of the binutils.
309 1.10 mrg </p>
310 1.10 mrg <hr />
311 1.17 mrg <a name="bfin"></a><a name="Blackfin"></a>
312 1.17 mrg <h3 class="heading">Blackfin</h3>
313 1.10 mrg <p>The Blackfin processor, an Analog Devices DSP.
314 1.1 mrg See “Blackfin Options” in the main manual
315 1.10 mrg </p>
316 1.10 mrg <p>More information, and a version of binutils with support for this processor,
317 1.12 mrg are available at <a href="https://sourceforge.net/projects/adi-toolchain/">https://sourceforge.net/projects/adi-toolchain/</a>.
318 1.10 mrg </p>
319 1.10 mrg <hr />
320 1.17 mrg <a name="cr16"></a><a name="CR16"></a>
321 1.17 mrg <h3 class="heading">CR16</h3>
322 1.5 mrg <p>The CR16 CompactRISC architecture is a 16-bit architecture. This
323 1.5 mrg architecture is used in embedded applications.
324 1.10 mrg </p>
325 1.3 skrll
326 1.10 mrg <p>See “CR16 Options” in the main manual for a list of CR16-specific options.
327 1.10 mrg </p>
328 1.10 mrg <p>Use ‘<samp>configure --target=cr16-elf --enable-languages=c,c++</samp>’ to configure
329 1.10 mrg GCC for building a CR16 elf cross-compiler.
330 1.10 mrg </p>
331 1.10 mrg <p>Use ‘<samp>configure --target=cr16-uclinux --enable-languages=c,c++</samp>’ to
332 1.10 mrg configure GCC for building a CR16 uclinux cross-compiler.
333 1.10 mrg </p>
334 1.10 mrg <hr />
335 1.17 mrg <a name="cris"></a><a name="CRIS"></a>
336 1.17 mrg <h3 class="heading">CRIS</h3>
337 1.15 mrg <p>CRIS is a CPU architecture in Axis Communications systems-on-a-chip, for
338 1.15 mrg example the ETRAX series. These are used in embedded applications.
339 1.10 mrg </p>
340 1.10 mrg <p>See “CRIS Options” in the main manual
341 1.1 mrg for a list of CRIS-specific options.
342 1.10 mrg </p>
343 1.15 mrg <p>Use ‘<samp>configure --target=cris-elf</samp>’ to configure GCC for building
344 1.15 mrg a cross-compiler for CRIS.
345 1.10 mrg <hr />
346 1.17 mrg <a name="dos"></a></p><a name="DOS"></a>
347 1.17 mrg <h3 class="heading">DOS</h3>
348 1.1 mrg <p>Please have a look at the <a href="binaries.html">binaries page</a>.
349 1.10 mrg </p>
350 1.10 mrg <p>You cannot install GCC by itself on MSDOS; it will not compile under
351 1.1 mrg any MSDOS compiler except itself. You need to get the complete
352 1.1 mrg compilation package DJGPP, which includes binaries as well as sources,
353 1.1 mrg and includes all the necessary compilation tools and libraries.
354 1.10 mrg </p>
355 1.10 mrg <hr />
356 1.17 mrg <a name="epiphany-x-elf"></a><a name="epiphany-*-elf"></a>
357 1.17 mrg <h3 class="heading">epiphany-*-elf</h3>
358 1.10 mrg <p>Adapteva Epiphany.
359 1.3 skrll This configuration is intended for embedded systems.
360 1.10 mrg </p>
361 1.10 mrg <hr />
362 1.17 mrg <a name="x-x-freebsd"></a><a name="g_t*-*-freebsd*"></a>
363 1.17 mrg <h3 class="heading">*-*-freebsd*</h3>
364 1.10 mrg <p>In order to better utilize FreeBSD base system functionality and match
365 1.3 skrll the configuration of the system compiler, GCC 4.5 and above as well as
366 1.3 skrll GCC 4.4 past 2010-06-20 leverage SSP support in libc (which is present
367 1.3 skrll on FreeBSD 7 or later) and the use of <code>__cxa_atexit</code> by default
368 1.3 skrll (on FreeBSD 6 or later). The use of <code>dl_iterate_phdr</code> inside
369 1.10 mrg <samp>libgcc_s.so.1</samp> and boehm-gc (on FreeBSD 7 or later) is enabled
370 1.3 skrll by GCC 4.5 and above.
371 1.10 mrg </p>
372 1.10 mrg <p>We support FreeBSD using the ELF file format with DWARF 2 debugging
373 1.10 mrg for all CPU architectures. You may use <samp>-gstabs</samp> instead of
374 1.10 mrg <samp>-g</samp>, if you really want the old debugging format. There are
375 1.1 mrg no known issues with mixing object files and libraries with different
376 1.1 mrg debugging formats. Otherwise, this release of GCC should now match
377 1.1 mrg more of the configuration used in the stock FreeBSD configuration of
378 1.10 mrg GCC. In particular, <samp>--enable-threads</samp> is now configured by
379 1.1 mrg default. However, as a general user, do not attempt to replace the
380 1.1 mrg system compiler with this release. Known to bootstrap and check with
381 1.1 mrg good results on FreeBSD 7.2-STABLE. In the past, known to bootstrap
382 1.1 mrg and check with good results on FreeBSD 3.0, 3.4, 4.0, 4.2, 4.3, 4.4,
383 1.1 mrg 4.5, 4.8, 4.9 and 5-CURRENT.
384 1.10 mrg </p>
385 1.10 mrg <p>The version of binutils installed in <samp>/usr/bin</samp> probably works
386 1.1 mrg with this release of GCC. Bootstrapping against the latest GNU
387 1.10 mrg binutils and/or the version found in <samp>/usr/ports/devel/binutils</samp> has
388 1.1 mrg been known to enable additional features and improve overall testsuite
389 1.9 mrg results. However, it is currently known that boehm-gc may not configure
390 1.9 mrg properly on FreeBSD prior to the FreeBSD 7.0 release with GNU binutils
391 1.9 mrg after 2.16.1.
392 1.10 mrg </p>
393 1.10 mrg <hr />
394 1.17 mrg <a name="ft32-x-elf"></a><a name="ft32-*-elf"></a>
395 1.17 mrg <h3 class="heading">ft32-*-elf</h3>
396 1.10 mrg <p>The FT32 processor.
397 1.7 mrg This configuration is intended for embedded systems.
398 1.10 mrg </p>
399 1.10 mrg <hr />
400 1.17 mrg <a name="h8300-hms"></a><a name="h8300-hms-1"></a>
401 1.17 mrg <h3 class="heading">h8300-hms</h3>
402 1.1 mrg <p>Renesas H8/300 series of processors.
403 1.10 mrg </p>
404 1.10 mrg <p>Please have a look at the <a href="binaries.html">binaries page</a>.
405 1.10 mrg </p>
406 1.10 mrg <p>The calling convention and structure layout has changed in release 2.6.
407 1.1 mrg All code must be recompiled. The calling convention now passes the
408 1.1 mrg first three arguments in function calls in registers. Structures are no
409 1.1 mrg longer a multiple of 2 bytes.
410 1.10 mrg </p>
411 1.10 mrg <hr />
412 1.17 mrg <a name="hppa-hp-hpux"></a><a name="hppa*-hp-hpux*"></a>
413 1.17 mrg <h3 class="heading">hppa*-hp-hpux*</h3>
414 1.1 mrg <p>Support for HP-UX version 9 and older was discontinued in GCC 3.4.
415 1.10 mrg </p>
416 1.10 mrg <p>We require using gas/binutils on all hppa platforms. Version 2.19 or
417 1.1 mrg later is recommended.
418 1.10 mrg </p>
419 1.10 mrg <p>It may be helpful to configure GCC with the
420 1.10 mrg <a href="./configure.html#with-gnu-as"><samp>--with-gnu-as</samp></a> and
421 1.10 mrg <samp>--with-as=…</samp> options to ensure that GCC can find GAS.
422 1.10 mrg </p>
423 1.10 mrg <p>The HP assembler should not be used with GCC. It is rarely tested and may
424 1.10 mrg not work. It shouldn’t be used with any languages other than C due to its
425 1.1 mrg many limitations.
426 1.10 mrg </p>
427 1.10 mrg <p>Specifically, <samp>-g</samp> does not work (HP-UX uses a peculiar debugging
428 1.1 mrg format which GCC does not know about). It also inserts timestamps
429 1.1 mrg into each object file it creates, causing the 3-stage comparison test to
430 1.1 mrg fail during a bootstrap. You should be able to continue by saying
431 1.10 mrg ‘<samp>make all-host all-target</samp>’ after getting the failure from ‘<samp>make</samp>’.
432 1.10 mrg </p>
433 1.10 mrg <p>Various GCC features are not supported. For example, it does not support weak
434 1.1 mrg symbols or alias definitions. As a result, explicit template instantiations
435 1.1 mrg are required when using C++. This makes it difficult if not impossible to
436 1.1 mrg build many C++ applications.
437 1.10 mrg </p>
438 1.10 mrg <p>There are two default scheduling models for instructions. These are
439 1.1 mrg PROCESSOR_7100LC and PROCESSOR_8000. They are selected from the pa-risc
440 1.10 mrg architecture specified for the target machine when configuring.
441 1.1 mrg PROCESSOR_8000 is the default. PROCESSOR_7100LC is selected when
442 1.10 mrg the target is a ‘<samp>hppa1*</samp>’ machine.
443 1.10 mrg </p>
444 1.10 mrg <p>The PROCESSOR_8000 model is not well suited to older processors. Thus,
445 1.1 mrg it is important to completely specify the machine architecture when
446 1.1 mrg configuring if you want a model other than PROCESSOR_8000. The macro
447 1.1 mrg TARGET_SCHED_DEFAULT can be defined in BOOT_CFLAGS if a different
448 1.1 mrg default scheduling model is desired.
449 1.10 mrg </p>
450 1.10 mrg <p>As of GCC 4.0, GCC uses the UNIX 95 namespace for HP-UX 10.10
451 1.10 mrg through 11.00, and the UNIX 98 namespace for HP-UX 11.11 and later.
452 1.1 mrg This namespace change might cause problems when bootstrapping with
453 1.1 mrg an earlier version of GCC or the HP compiler as essentially the same
454 1.1 mrg namespace is required for an entire build. This problem can be avoided
455 1.10 mrg in a number of ways. With HP cc, <code>UNIX_STD</code> can be set to ‘<samp>95</samp>’
456 1.10 mrg or ‘<samp>98</samp>’. Another way is to add an appropriate set of predefines
457 1.10 mrg to <code>CC</code>. The description for the <samp>munix=</samp> option contains
458 1.1 mrg a list of the predefines used with each standard.
459 1.10 mrg </p>
460 1.10 mrg <p>More specific information to ‘<samp>hppa*-hp-hpux*</samp>’ targets follows.
461 1.10 mrg </p>
462 1.10 mrg <hr />
463 1.17 mrg <a name="hppa-hp-hpux10"></a><a name="hppa*-hp-hpux10"></a>
464 1.17 mrg <h3 class="heading">hppa*-hp-hpux10</h3>
465 1.1 mrg <p>For hpux10.20, we <em>highly</em> recommend you pick up the latest sed patch
466 1.1 mrg <code>PHCO_19798</code> from HP.
467 1.10 mrg </p>
468 1.10 mrg <p>The C++ ABI has changed incompatibly in GCC 4.0. COMDAT subspaces are
469 1.1 mrg used for one-only code and data. This resolves many of the previous
470 1.1 mrg problems in using C++ on this target. However, the ABI is not compatible
471 1.1 mrg with the one implemented under HP-UX 11 using secondary definitions.
472 1.10 mrg </p>
473 1.10 mrg <hr />
474 1.17 mrg <a name="hppa-hp-hpux11"></a><a name="hppa*-hp-hpux11"></a>
475 1.17 mrg <h3 class="heading">hppa*-hp-hpux11</h3>
476 1.1 mrg <p>GCC 3.0 and up support HP-UX 11. GCC 2.95.x is not supported and cannot
477 1.1 mrg be used to compile GCC 3.0 and up.
478 1.10 mrg </p>
479 1.10 mrg <p>The libffi library haven’t been ported to 64-bit HP-UX and doesn’t build.
480 1.10 mrg </p>
481 1.10 mrg <p>Refer to <a href="binaries.html">binaries</a> for information about obtaining
482 1.1 mrg precompiled GCC binaries for HP-UX. Precompiled binaries must be obtained
483 1.9 mrg to build the Ada language as it cannot be bootstrapped using C. Ada is
484 1.1 mrg only available for the 32-bit PA-RISC runtime.
485 1.10 mrg </p>
486 1.10 mrg <p>Starting with GCC 3.4 an ISO C compiler is required to bootstrap. The
487 1.10 mrg bundled compiler supports only traditional C; you will need either HP’s
488 1.1 mrg unbundled compiler, or a binary distribution of GCC.
489 1.10 mrg </p>
490 1.10 mrg <p>It is possible to build GCC 3.3 starting with the bundled HP compiler,
491 1.1 mrg but the process requires several steps. GCC 3.3 can then be used to
492 1.9 mrg build later versions.
493 1.10 mrg </p>
494 1.10 mrg <p>There are several possible approaches to building the distribution.
495 1.1 mrg Binutils can be built first using the HP tools. Then, the GCC
496 1.1 mrg distribution can be built. The second approach is to build GCC
497 1.10 mrg first using the HP tools, then build binutils, then rebuild GCC.
498 1.1 mrg There have been problems with various binary distributions, so it
499 1.1 mrg is best not to start from a binary distribution.
500 1.10 mrg </p>
501 1.10 mrg <p>On 64-bit capable systems, there are two distinct targets. Different
502 1.1 mrg installation prefixes must be used if both are to be installed on
503 1.10 mrg the same system. The ‘<samp>hppa[1-2]*-hp-hpux11*</samp>’ target generates code
504 1.10 mrg for the 32-bit PA-RISC runtime architecture and uses the HP linker.
505 1.10 mrg The ‘<samp>hppa64-hp-hpux11*</samp>’ target generates 64-bit code for the
506 1.1 mrg PA-RISC 2.0 architecture.
507 1.10 mrg </p>
508 1.10 mrg <p>The script config.guess now selects the target type based on the compiler
509 1.10 mrg detected during configuration. You must define <code>PATH</code> or <code>CC</code> so
510 1.10 mrg that configure finds an appropriate compiler for the initial bootstrap.
511 1.10 mrg When <code>CC</code> is used, the definition should contain the options that are
512 1.10 mrg needed whenever <code>CC</code> is used.
513 1.10 mrg </p>
514 1.10 mrg <p>Specifically, options that determine the runtime architecture must be
515 1.10 mrg in <code>CC</code> to correctly select the target for the build. It is also
516 1.10 mrg convenient to place many other compiler options in <code>CC</code>. For example,
517 1.10 mrg <code>CC="cc -Ac +DA2.0W -Wp,-H16376 -D_CLASSIC_TYPES -D_HPUX_SOURCE"</code>
518 1.1 mrg can be used to bootstrap the GCC 3.3 branch with the HP compiler in
519 1.10 mrg 64-bit K&R/bundled mode. The <samp>+DA2.0W</samp> option will result in
520 1.10 mrg the automatic selection of the ‘<samp>hppa64-hp-hpux11*</samp>’ target. The
521 1.1 mrg macro definition table of cpp needs to be increased for a successful
522 1.1 mrg build with the HP compiler. _CLASSIC_TYPES and _HPUX_SOURCE need to
523 1.1 mrg be defined when building with the bundled compiler, or when using the
524 1.10 mrg <samp>-Ac</samp> option. These defines aren’t necessary with <samp>-Ae</samp>.
525 1.10 mrg </p>
526 1.10 mrg <p>It is best to explicitly configure the ‘<samp>hppa64-hp-hpux11*</samp>’ target
527 1.10 mrg with the <samp>--with-ld=…</samp> option. This overrides the standard
528 1.1 mrg search for ld. The two linkers supported on this target require different
529 1.1 mrg commands. The default linker is determined during configuration. As a
530 1.10 mrg result, it’s not possible to switch linkers in the middle of a GCC build.
531 1.1 mrg This has been reported to sometimes occur in unified builds of binutils
532 1.1 mrg and GCC.
533 1.10 mrg </p>
534 1.10 mrg <p>A recent linker patch must be installed for the correct operation of
535 1.1 mrg GCC 3.3 and later. <code>PHSS_26559</code> and <code>PHSS_24304</code> are the
536 1.1 mrg oldest linker patches that are known to work. They are for HP-UX
537 1.1 mrg 11.00 and 11.11, respectively. <code>PHSS_24303</code>, the companion to
538 1.10 mrg <code>PHSS_24304</code>, might be usable but it hasn’t been tested. These
539 1.1 mrg patches have been superseded. Consult the HP patch database to obtain
540 1.1 mrg the currently recommended linker patch for your system.
541 1.10 mrg </p>
542 1.10 mrg <p>The patches are necessary for the support of weak symbols on the
543 1.1 mrg 32-bit port, and for the running of initializers and finalizers. Weak
544 1.1 mrg symbols are implemented using SOM secondary definition symbols. Prior
545 1.10 mrg to HP-UX 11, there are bugs in the linker support for secondary symbols.
546 1.1 mrg The patches correct a problem of linker core dumps creating shared
547 1.1 mrg libraries containing secondary symbols, as well as various other
548 1.1 mrg linking issues involving secondary symbols.
549 1.10 mrg </p>
550 1.10 mrg <p>GCC 3.3 uses the ELF DT_INIT_ARRAY and DT_FINI_ARRAY capabilities to
551 1.1 mrg run initializers and finalizers on the 64-bit port. The 32-bit port
552 1.10 mrg uses the linker <samp>+init</samp> and <samp>+fini</samp> options for the same
553 1.1 mrg purpose. The patches correct various problems with the +init/+fini
554 1.1 mrg options, including program core dumps. Binutils 2.14 corrects a
555 1.10 mrg problem on the 64-bit port resulting from HP’s non-standard use of
556 1.1 mrg the .init and .fini sections for array initializers and finalizers.
557 1.10 mrg </p>
558 1.10 mrg <p>Although the HP and GNU linkers are both supported for the
559 1.10 mrg ‘<samp>hppa64-hp-hpux11*</samp>’ target, it is strongly recommended that the
560 1.1 mrg HP linker be used for link editing on this target.
561 1.10 mrg </p>
562 1.10 mrg <p>At this time, the GNU linker does not support the creation of long
563 1.9 mrg branch stubs. As a result, it cannot successfully link binaries
564 1.1 mrg containing branch offsets larger than 8 megabytes. In addition,
565 1.1 mrg there are problems linking shared libraries, linking executables
566 1.10 mrg with <samp>-static</samp>, and with dwarf2 unwind and exception support.
567 1.10 mrg It also doesn’t provide stubs for internal calls to global functions
568 1.9 mrg in shared libraries, so these calls cannot be overloaded.
569 1.10 mrg </p>
570 1.10 mrg <p>The HP dynamic loader does not support GNU symbol versioning, so symbol
571 1.1 mrg versioning is not supported. It may be necessary to disable symbol
572 1.10 mrg versioning with <samp>--disable-symvers</samp> when using GNU ld.
573 1.10 mrg </p>
574 1.10 mrg <p>POSIX threads are the default. The optional DCE thread library is not
575 1.10 mrg supported, so <samp>--enable-threads=dce</samp> does not work.
576 1.10 mrg </p>
577 1.10 mrg <hr />
578 1.17 mrg <a name="x-x-linux-gnu"></a><a name="g_t*-*-linux-gnu"></a>
579 1.17 mrg <h3 class="heading">*-*-linux-gnu</h3>
580 1.15 mrg <p>The <code>.init_array</code> and <code>.fini_array</code> sections are enabled
581 1.15 mrg unconditionally which requires at least glibc 2.1 and binutils 2.12.
582 1.15 mrg </p>
583 1.1 mrg <p>Versions of libstdc++-v3 starting with 3.2.1 require bug fixes present
584 1.1 mrg in glibc 2.2.5 and later. More information is available in the
585 1.1 mrg libstdc++-v3 documentation.
586 1.10 mrg </p>
587 1.10 mrg <hr />
588 1.17 mrg <a name="ix86-x-linux"></a><a name="i_003f86-*-linux*"></a>
589 1.17 mrg <h3 class="heading">i?86-*-linux*</h3>
590 1.10 mrg <p>As of GCC 3.3, binutils 2.13.1 or later is required for this platform.
591 1.15 mrg See <a href="https://gcc.gnu.org/PR10877">bug 10877</a> for more information.
592 1.10 mrg </p>
593 1.10 mrg <p>If you receive Signal 11 errors when building on GNU/Linux, then it is
594 1.1 mrg possible you have a hardware problem. Further information on this can be
595 1.15 mrg found on <a href="https://www.bitwizard.nl/sig11/">www.bitwizard.nl</a>.
596 1.10 mrg </p>
597 1.10 mrg <hr />
598 1.17 mrg <a name="ix86-x-solaris2"></a><a name="i_003f86-*-solaris2*"></a>
599 1.17 mrg <h3 class="heading">i?86-*-solaris2*</h3>
600 1.12 mrg <p>Use this for Solaris 11.3 or later on x86 and x86-64 systems. Starting
601 1.12 mrg with GCC 4.7, there is also a 64-bit ‘<samp>amd64-*-solaris2*</samp>’ or
602 1.12 mrg ‘<samp>x86_64-*-solaris2*</samp>’ configuration that corresponds to
603 1.10 mrg ‘<samp>sparcv9-sun-solaris2*</samp>’.
604 1.10 mrg </p>
605 1.10 mrg <p>It is recommended that you configure GCC to use the GNU assembler. The
606 1.12 mrg versions included in Solaris 11.3, from GNU binutils 2.23.1 or
607 1.12 mrg newer (available as <samp>/usr/bin/gas</samp> and
608 1.10 mrg <samp>/usr/gnu/bin/as</samp>), work fine. The current version, from GNU
609 1.12 mrg binutils 2.34, is known to work. Recent versions of the Solaris assembler in
610 1.12 mrg <samp>/usr/bin/as</samp> work almost as well, though.
611 1.12 mrg </p>
612 1.12 mrg <p>For linking, the Solaris linker is preferred. If you want to use the GNU
613 1.12 mrg linker instead, the version in Solaris 11.3, from GNU binutils 2.23.1 or
614 1.12 mrg newer (in <samp>/usr/gnu/bin/ld</samp> and <samp>/usr/bin/gld</samp>), works,
615 1.12 mrg as does the latest version, from GNU binutils 2.34.
616 1.10 mrg </p>
617 1.10 mrg <p>To use GNU <code>as</code>, configure with the options
618 1.12 mrg <samp>--with-gnu-as --with-as=/usr/gnu/bin/as</samp>. It may be necessary
619 1.10 mrg to configure with <samp>--without-gnu-ld --with-ld=/usr/ccs/bin/ld</samp> to
620 1.12 mrg guarantee use of Solaris <code>ld</code>.
621 1.10 mrg </p>
622 1.10 mrg <hr />
623 1.17 mrg <a name="ia64-x-linux"></a><a name="ia64-*-linux"></a>
624 1.17 mrg <h3 class="heading">ia64-*-linux</h3>
625 1.1 mrg <p>IA-64 processor (also known as IPF, or Itanium Processor Family)
626 1.1 mrg running GNU/Linux.
627 1.10 mrg </p>
628 1.10 mrg <p>If you are using the installed system libunwind library with
629 1.10 mrg <samp>--with-system-libunwind</samp>, then you must use libunwind 0.98 or
630 1.1 mrg later.
631 1.10 mrg </p>
632 1.10 mrg <hr />
633 1.17 mrg <a name="ia64-x-hpux"></a><a name="ia64-*-hpux*"></a>
634 1.17 mrg <h3 class="heading">ia64-*-hpux*</h3>
635 1.1 mrg <p>Building GCC on this target requires the GNU Assembler. The bundled HP
636 1.1 mrg assembler will not work. To prevent GCC from using the wrong assembler,
637 1.10 mrg the option <samp>--with-gnu-as</samp> may be necessary.
638 1.10 mrg </p>
639 1.10 mrg <p>The GCC libunwind library has not been ported to HPUX. This means that for
640 1.10 mrg GCC versions 3.2.3 and earlier, <samp>--enable-libunwind-exceptions</samp>
641 1.10 mrg is required to build GCC. For GCC 3.3 and later, this is the default.
642 1.10 mrg For gcc 3.4.3 and later, <samp>--enable-libunwind-exceptions</samp> is
643 1.1 mrg removed and the system libunwind library will always be used.
644 1.10 mrg </p>
645 1.10 mrg <hr />
646 1.10 mrg <!-- rs6000-ibm-aix*, powerpc-ibm-aix* -->
647 1.17 mrg <a name="x-ibm-aix"></a><a name="g_t*-ibm-aix*"></a>
648 1.17 mrg <h3 class="heading">*-ibm-aix*</h3>
649 1.10 mrg <p>Support for AIX version 3 and older was discontinued in GCC 3.4.
650 1.1 mrg Support for AIX version 4.2 and older was discontinued in GCC 4.5.
651 1.10 mrg </p>
652 1.10 mrg <p>“out of memory” bootstrap failures may indicate a problem with
653 1.1 mrg process resource limits (ulimit). Hard limits are configured in the
654 1.10 mrg <samp>/etc/security/limits</samp> system configuration file.
655 1.10 mrg </p>
656 1.10 mrg <p>GCC 4.9 and above require a C++ compiler for bootstrap. IBM VAC++ / xlC
657 1.7 mrg cannot bootstrap GCC. xlc can bootstrap an older version of GCC and
658 1.7 mrg G++ can bootstrap recent releases of GCC.
659 1.10 mrg </p>
660 1.10 mrg <p>GCC can bootstrap with recent versions of IBM XLC, but bootstrapping
661 1.1 mrg with an earlier release of GCC is recommended. Bootstrapping with XLC
662 1.1 mrg requires a larger data segment, which can be enabled through the
663 1.1 mrg <var>LDR_CNTRL</var> environment variable, e.g.,
664 1.10 mrg </p>
665 1.17 mrg <div class="smallexample">
666 1.17 mrg <pre class="smallexample">% LDR_CNTRL=MAXDATA=0x50000000
667 1.10 mrg % export LDR_CNTRL
668 1.10 mrg </pre></div>
669 1.1 mrg
670 1.10 mrg <p>One can start with a pre-compiled version of GCC to build from
671 1.10 mrg sources. One may delete GCC’s “fixed” header files when starting
672 1.1 mrg with a version of GCC built for an earlier release of AIX.
673 1.10 mrg </p>
674 1.10 mrg <p>To speed up the configuration phases of bootstrapping and installing GCC,
675 1.10 mrg one may use GNU Bash instead of AIX <code>/bin/sh</code>, e.g.,
676 1.10 mrg </p>
677 1.17 mrg <div class="smallexample">
678 1.17 mrg <pre class="smallexample">% CONFIG_SHELL=/opt/freeware/bin/bash
679 1.10 mrg % export CONFIG_SHELL
680 1.10 mrg </pre></div>
681 1.1 mrg
682 1.10 mrg <p>and then proceed as described in <a href="build.html">the build
683 1.10 mrg instructions</a>, where we strongly recommend specifying an absolute path
684 1.1 mrg to invoke <var>srcdir</var>/configure.
685 1.10 mrg </p>
686 1.10 mrg <p>Because GCC on AIX is built as a 32-bit executable by default,
687 1.1 mrg (although it can generate 64-bit programs) the GMP and MPFR libraries
688 1.1 mrg required by gfortran must be 32-bit libraries. Building GMP and MPFR
689 1.1 mrg as static archive libraries works better than shared libraries.
690 1.10 mrg </p>
691 1.10 mrg <p>Errors involving <code>alloca</code> when building GCC generally are due
692 1.1 mrg to an incorrect definition of <code>CC</code> in the Makefile or mixing files
693 1.1 mrg compiled with the native C compiler and GCC. During the stage1 phase of
694 1.10 mrg the build, the native AIX compiler <strong>must</strong> be invoked as <code>cc</code>
695 1.10 mrg (not <code>xlc</code>). Once <code>configure</code> has been informed of
696 1.10 mrg <code>xlc</code>, one needs to use ‘<samp>make distclean</samp>’ to remove the
697 1.10 mrg configure cache files and ensure that <code>CC</code> environment variable
698 1.10 mrg does not provide a definition that will confuse <code>configure</code>.
699 1.1 mrg If this error occurs during stage2 or later, then the problem most likely
700 1.1 mrg is the version of Make (see above).
701 1.10 mrg </p>
702 1.10 mrg <p>The native <code>as</code> and <code>ld</code> are recommended for
703 1.3 skrll bootstrapping on AIX. The GNU Assembler, GNU Linker, and GNU
704 1.3 skrll Binutils version 2.20 is the minimum level that supports bootstrap on
705 1.10 mrg AIX 5. The GNU Assembler has not been updated to support AIX 6 or
706 1.3 skrll AIX 7. The native AIX tools do interoperate with GCC.
707 1.10 mrg </p>
708 1.10 mrg <p>AIX 7.1 added partial support for DWARF debugging, but full support
709 1.7 mrg requires AIX 7.1 TL03 SP7 that supports additional DWARF sections and
710 1.7 mrg fixes a bug in the assembler. AIX 7.1 TL03 SP5 distributed a version
711 1.7 mrg of libm.a missing important symbols; a fix for IV77796 will be
712 1.7 mrg included in SP6.
713 1.10 mrg </p>
714 1.10 mrg <p>AIX 5.3 TL10, AIX 6.1 TL05 and AIX 7.1 TL00 introduced an AIX
715 1.3 skrll assembler change that sometimes produces corrupt assembly files
716 1.3 skrll causing AIX linker errors. The bug breaks GCC bootstrap on AIX and
717 1.3 skrll can cause compilation failures with existing GCC installations. An
718 1.3 skrll AIX iFix for AIX 5.3 is available (APAR IZ98385 for AIX 5.3 TL10, APAR
719 1.3 skrll IZ98477 for AIX 5.3 TL11 and IZ98134 for AIX 5.3 TL12). AIX 5.3 TL11 SP8,
720 1.3 skrll AIX 5.3 TL12 SP5, AIX 6.1 TL04 SP11, AIX 6.1 TL05 SP7, AIX 6.1 TL06 SP6,
721 1.3 skrll AIX 6.1 TL07 and AIX 7.1 TL01 should include the fix.
722 1.10 mrg </p>
723 1.10 mrg <p>Building <samp>libstdc++.a</samp> requires a fix for an AIX Assembler bug
724 1.1 mrg APAR IY26685 (AIX 4.3) or APAR IY25528 (AIX 5.1). It also requires a
725 1.1 mrg fix for another AIX Assembler bug and a co-dependent AIX Archiver fix
726 1.1 mrg referenced as APAR IY53606 (AIX 5.2) or as APAR IY54774 (AIX 5.1)
727 1.10 mrg </p>
728 1.17 mrg <a name="TransferAixShobj"></a><p>‘<samp>libstdc++</samp>’ in GCC 3.4 increments the major version number of the
729 1.10 mrg shared object and GCC installation places the <samp>libstdc++.a</samp>
730 1.1 mrg shared library in a common location which will overwrite the and GCC
731 1.1 mrg 3.3 version of the shared library. Applications either need to be
732 1.1 mrg re-linked against the new shared library or the GCC 3.1 and GCC 3.3
733 1.10 mrg versions of the ‘<samp>libstdc++</samp>’ shared object needs to be available
734 1.10 mrg to the AIX runtime loader. The GCC 3.1 ‘<samp>libstdc++.so.4</samp>’, if
735 1.10 mrg present, and GCC 3.3 ‘<samp>libstdc++.so.5</samp>’ shared objects can be
736 1.1 mrg installed for runtime dynamic loading using the following steps to set
737 1.10 mrg the ‘<samp>F_LOADONLY</samp>’ flag in the shared object for <em>each</em>
738 1.10 mrg multilib <samp>libstdc++.a</samp> installed:
739 1.10 mrg </p>
740 1.10 mrg <p>Extract the shared objects from the currently installed
741 1.10 mrg <samp>libstdc++.a</samp> archive:
742 1.17 mrg </p><div class="smallexample">
743 1.17 mrg <pre class="smallexample">% ar -x libstdc++.a libstdc++.so.4 libstdc++.so.5
744 1.10 mrg </pre></div>
745 1.1 mrg
746 1.10 mrg <p>Enable the ‘<samp>F_LOADONLY</samp>’ flag so that the shared object will be
747 1.1 mrg available for runtime dynamic loading, but not linking:
748 1.17 mrg </p><div class="smallexample">
749 1.17 mrg <pre class="smallexample">% strip -e libstdc++.so.4 libstdc++.so.5
750 1.10 mrg </pre></div>
751 1.10 mrg
752 1.10 mrg <p>Archive the runtime-only shared object in the GCC 3.4
753 1.10 mrg <samp>libstdc++.a</samp> archive:
754 1.17 mrg </p><div class="smallexample">
755 1.17 mrg <pre class="smallexample">% ar -q libstdc++.a libstdc++.so.4 libstdc++.so.5
756 1.10 mrg </pre></div>
757 1.10 mrg
758 1.10 mrg <p>Eventually, the
759 1.10 mrg <a href="./configure.html#WithAixSoname"><samp>--with-aix-soname=svr4</samp></a>
760 1.5 mrg configure option may drop the need for this procedure for libraries that
761 1.5 mrg support it.
762 1.10 mrg </p>
763 1.10 mrg <p>Linking executables and shared libraries may produce warnings of
764 1.1 mrg duplicate symbols. The assembly files generated by GCC for AIX always
765 1.1 mrg have included multiple symbol definitions for certain global variable
766 1.1 mrg and function declarations in the original program. The warnings should
767 1.1 mrg not prevent the linker from producing a correct library or runnable
768 1.1 mrg executable.
769 1.10 mrg </p>
770 1.10 mrg <p>AIX 4.3 utilizes a “large format” archive to support both 32-bit and
771 1.1 mrg 64-bit object modules. The routines provided in AIX 4.3.0 and AIX 4.3.1
772 1.10 mrg to parse archive libraries did not handle the new format correctly.
773 1.1 mrg These routines are used by GCC and result in error messages during
774 1.1 mrg linking such as “not a COFF file”. The version of the routines shipped
775 1.10 mrg with AIX 4.3.1 should work for a 32-bit environment. The <samp>-g</samp>
776 1.1 mrg option of the archive command may be used to create archives of 32-bit
777 1.1 mrg objects using the original “small format”. A correct version of the
778 1.1 mrg routines is shipped with AIX 4.3.2 and above.
779 1.10 mrg </p>
780 1.10 mrg <p>Some versions of the AIX binder (linker) can fail with a relocation
781 1.10 mrg overflow severe error when the <samp>-bbigtoc</samp> option is used to link
782 1.1 mrg GCC-produced object files into an executable that overflows the TOC. A fix
783 1.1 mrg for APAR IX75823 (OVERFLOW DURING LINK WHEN USING GCC AND -BBIGTOC) is
784 1.1 mrg available from IBM Customer Support and from its
785 1.15 mrg <a href="https://techsupport.services.ibm.com/">techsupport.services.ibm.com</a>
786 1.1 mrg website as PTF U455193.
787 1.10 mrg </p>
788 1.10 mrg <p>The AIX 4.3.2.1 linker (bos.rte.bind_cmds Level 4.3.2.1) will dump core
789 1.1 mrg with a segmentation fault when invoked by any version of GCC. A fix for
790 1.1 mrg APAR IX87327 is available from IBM Customer Support and from its
791 1.15 mrg <a href="https://techsupport.services.ibm.com/">techsupport.services.ibm.com</a>
792 1.1 mrg website as PTF U461879. This fix is incorporated in AIX 4.3.3 and above.
793 1.10 mrg </p>
794 1.10 mrg <p>The initial assembler shipped with AIX 4.3.0 generates incorrect object
795 1.1 mrg files. A fix for APAR IX74254 (64BIT DISASSEMBLED OUTPUT FROM COMPILER FAILS
796 1.1 mrg TO ASSEMBLE/BIND) is available from IBM Customer Support and from its
797 1.15 mrg <a href="https://techsupport.services.ibm.com/">techsupport.services.ibm.com</a>
798 1.1 mrg website as PTF U453956. This fix is incorporated in AIX 4.3.1 and above.
799 1.10 mrg </p>
800 1.10 mrg <p>AIX provides National Language Support (NLS). Compilers and assemblers
801 1.1 mrg use NLS to support locale-specific representations of various data
802 1.10 mrg formats including floating-point numbers (e.g., ‘<samp>.</samp>’ vs ‘<samp>,</samp>’ for
803 1.1 mrg separating decimal fractions). There have been problems reported where
804 1.1 mrg GCC does not produce the same floating-point formats that the assembler
805 1.10 mrg expects. If one encounters this problem, set the <code>LANG</code>
806 1.10 mrg environment variable to ‘<samp>C</samp>’ or ‘<samp>En_US</samp>’.
807 1.10 mrg </p>
808 1.10 mrg <p>A default can be specified with the <samp>-mcpu=<var>cpu_type</var></samp>
809 1.10 mrg switch and using the configure option <samp>--with-cpu-<var>cpu_type</var></samp>.
810 1.10 mrg </p>
811 1.10 mrg <hr />
812 1.17 mrg <a name="iq2000-x-elf"></a><a name="iq2000-*-elf"></a>
813 1.17 mrg <h3 class="heading">iq2000-*-elf</h3>
814 1.1 mrg <p>Vitesse IQ2000 processors. These are used in embedded
815 1.1 mrg applications. There are no standard Unix configurations.
816 1.10 mrg </p>
817 1.10 mrg <hr />
818 1.17 mrg <a name="lm32-x-elf"></a><a name="lm32-*-elf"></a>
819 1.17 mrg <h3 class="heading">lm32-*-elf</h3>
820 1.10 mrg <p>Lattice Mico32 processor.
821 1.1 mrg This configuration is intended for embedded systems.
822 1.10 mrg </p>
823 1.10 mrg <hr />
824 1.17 mrg <a name="lm32-x-uclinux"></a><a name="lm32-*-uclinux"></a>
825 1.17 mrg <h3 class="heading">lm32-*-uclinux</h3>
826 1.10 mrg <p>Lattice Mico32 processor.
827 1.1 mrg This configuration is intended for embedded systems running uClinux.
828 1.10 mrg </p>
829 1.10 mrg <hr />
830 1.17 mrg <a name="loongarch"></a><a name="LoongArch"></a>
831 1.17 mrg <h3 class="heading">LoongArch</h3>
832 1.15 mrg <p>LoongArch processor.
833 1.15 mrg The following LoongArch targets are available:
834 1.15 mrg </p><dl compact="compact">
835 1.17 mrg <dt><code>loongarch64-linux-gnu*</code></dt>
836 1.15 mrg <dd><p>LoongArch processor running GNU/Linux. This target triplet may be coupled
837 1.15 mrg with a small set of possible suffixes to identify their default ABI type:
838 1.15 mrg </p><dl compact="compact">
839 1.17 mrg <dt><code>f64</code></dt>
840 1.15 mrg <dd><p>Uses <code>lp64d/base</code> ABI by default.
841 1.15 mrg </p></dd>
842 1.17 mrg <dt><code>f32</code></dt>
843 1.15 mrg <dd><p>Uses <code>lp64f/base</code> ABI by default.
844 1.15 mrg </p></dd>
845 1.17 mrg <dt><code>sf</code></dt>
846 1.15 mrg <dd><p>Uses <code>lp64s/base</code> ABI by default.
847 1.15 mrg </p></dd>
848 1.15 mrg </dl>
849 1.15 mrg
850 1.15 mrg </dd>
851 1.17 mrg <dt><code>loongarch64-linux-gnu</code></dt>
852 1.15 mrg <dd><p>Same as <code>loongarch64-linux-gnuf64</code>, but may be used with
853 1.15 mrg <samp>--with-abi=*</samp> to configure the default ABI type.
854 1.15 mrg </p></dd>
855 1.15 mrg </dl>
856 1.15 mrg
857 1.15 mrg <p>More information about LoongArch can be found at
858 1.15 mrg <a href="https://github.com/loongson/LoongArch-Documentation">https://github.com/loongson/LoongArch-Documentation</a>.
859 1.15 mrg </p>
860 1.15 mrg <hr />
861 1.17 mrg <a name="m32c-x-elf"></a><a name="m32c-*-elf"></a>
862 1.17 mrg <h3 class="heading">m32c-*-elf</h3>
863 1.10 mrg <p>Renesas M32C processor.
864 1.1 mrg This configuration is intended for embedded systems.
865 1.10 mrg </p>
866 1.10 mrg <hr />
867 1.17 mrg <a name="m32r-x-elf"></a><a name="m32r-*-elf"></a>
868 1.17 mrg <h3 class="heading">m32r-*-elf</h3>
869 1.10 mrg <p>Renesas M32R processor.
870 1.1 mrg This configuration is intended for embedded systems.
871 1.10 mrg </p>
872 1.10 mrg <hr />
873 1.17 mrg <a name="m68k-x-x"></a><a name="m68k-*-*"></a>
874 1.17 mrg <h3 class="heading">m68k-*-*</h3>
875 1.1 mrg <p>By default,
876 1.10 mrg ‘<samp>m68k-*-elf*</samp>’, ‘<samp>m68k-*-rtems</samp>’, ‘<samp>m68k-*-uclinux</samp>’ and
877 1.10 mrg ‘<samp>m68k-*-linux</samp>’
878 1.1 mrg build libraries for both M680x0 and ColdFire processors. If you only
879 1.1 mrg need the M680x0 libraries, you can omit the ColdFire ones by passing
880 1.10 mrg <samp>--with-arch=m68k</samp> to <code>configure</code>. Alternatively, you
881 1.10 mrg can omit the M680x0 libraries by passing <samp>--with-arch=cf</samp> to
882 1.10 mrg <code>configure</code>. These targets default to 5206 or 5475 code as
883 1.1 mrg appropriate for the target system when
884 1.10 mrg configured with <samp>--with-arch=cf</samp> and 68020 code otherwise.
885 1.10 mrg </p>
886 1.10 mrg <p>The ‘<samp>m68k-*-netbsd</samp>’ and
887 1.10 mrg ‘<samp>m68k-*-openbsd</samp>’ targets also support the <samp>--with-arch</samp>
888 1.1 mrg option. They will generate ColdFire CFV4e code when configured with
889 1.10 mrg <samp>--with-arch=cf</samp> and 68020 code otherwise.
890 1.10 mrg </p>
891 1.10 mrg <p>You can override the default processors listed above by configuring
892 1.10 mrg with <samp>--with-cpu=<var>target</var></samp>. This <var>target</var> can either
893 1.10 mrg be a <samp>-mcpu</samp> argument or one of the following values:
894 1.10 mrg ‘<samp>m68000</samp>’, ‘<samp>m68010</samp>’, ‘<samp>m68020</samp>’, ‘<samp>m68030</samp>’,
895 1.10 mrg ‘<samp>m68040</samp>’, ‘<samp>m68060</samp>’, ‘<samp>m68020-40</samp>’ and ‘<samp>m68020-60</samp>’.
896 1.10 mrg </p>
897 1.10 mrg <p>GCC requires at least binutils version 2.17 on these targets.
898 1.10 mrg </p>
899 1.10 mrg <hr />
900 1.17 mrg <a name="m68k-x-uclinux"></a><a name="m68k-*-uclinux"></a>
901 1.17 mrg <h3 class="heading">m68k-*-uclinux</h3>
902 1.1 mrg <p>GCC 4.3 changed the uClinux configuration so that it uses the
903 1.10 mrg ‘<samp>m68k-linux-gnu</samp>’ ABI rather than the ‘<samp>m68k-elf</samp>’ ABI.
904 1.1 mrg It also added improved support for C++ and flat shared libraries,
905 1.3 skrll both of which were ABI changes.
906 1.10 mrg </p>
907 1.10 mrg <hr />
908 1.17 mrg <a name="microblaze-x-elf"></a><a name="microblaze-*-elf"></a>
909 1.17 mrg <h3 class="heading">microblaze-*-elf</h3>
910 1.10 mrg <p>Xilinx MicroBlaze processor.
911 1.1 mrg This configuration is intended for embedded systems.
912 1.10 mrg </p>
913 1.10 mrg <hr />
914 1.17 mrg <a name="mips-x-x"></a><a name="mips-*-*"></a>
915 1.17 mrg <h3 class="heading">mips-*-*</h3>
916 1.1 mrg <p>If on a MIPS system you get an error message saying “does not have gp
917 1.10 mrg sections for all it’s [sic] sectons [sic]”, don’t worry about it. This
918 1.1 mrg happens whenever you use GAS with the MIPS linker, but there is not
919 1.1 mrg really anything wrong, and it is okay to use the output file. You can
920 1.1 mrg stop such warnings by installing the GNU linker.
921 1.10 mrg </p>
922 1.10 mrg <p>It would be nice to extend GAS to produce the gp tables, but they are
923 1.1 mrg optional, and there should not be a warning about their absence.
924 1.10 mrg </p>
925 1.10 mrg <p>The libstdc++ atomic locking routines for MIPS targets requires MIPS II
926 1.1 mrg and later. A patch went in just after the GCC 3.3 release to
927 1.10 mrg make ‘<samp>mips*-*-*</samp>’ use the generic implementation instead. You can also
928 1.10 mrg configure for ‘<samp>mipsel-elf</samp>’ as a workaround. The
929 1.10 mrg ‘<samp>mips*-*-linux*</samp>’ target continues to use the MIPS II routines. More
930 1.1 mrg work on this is expected in future releases.
931 1.10 mrg </p>
932 1.1 mrg
933 1.10 mrg <p>The built-in <code>__sync_*</code> functions are available on MIPS II and
934 1.10 mrg later systems and others that support the ‘<samp>ll</samp>’, ‘<samp>sc</samp>’ and
935 1.10 mrg ‘<samp>sync</samp>’ instructions. This can be overridden by passing
936 1.10 mrg <samp>--with-llsc</samp> or <samp>--without-llsc</samp> when configuring GCC.
937 1.1 mrg Since the Linux kernel emulates these instructions if they are
938 1.10 mrg missing, the default for ‘<samp>mips*-*-linux*</samp>’ targets is
939 1.10 mrg <samp>--with-llsc</samp>. The <samp>--with-llsc</samp> and
940 1.10 mrg <samp>--without-llsc</samp> configure options may be overridden at compile
941 1.10 mrg time by passing the <samp>-mllsc</samp> or <samp>-mno-llsc</samp> options to
942 1.1 mrg the compiler.
943 1.10 mrg </p>
944 1.10 mrg <p>MIPS systems check for division by zero (unless
945 1.10 mrg <samp>-mno-check-zero-division</samp> is passed to the compiler) by
946 1.1 mrg generating either a conditional trap or a break instruction. Using
947 1.1 mrg trap results in smaller code, but is only supported on MIPS II and
948 1.1 mrg later. Also, some versions of the Linux kernel have a bug that
949 1.1 mrg prevents trap from generating the proper signal (<code>SIGFPE</code>). To enable
950 1.10 mrg the use of break, use the <samp>--with-divide=breaks</samp>
951 1.10 mrg <code>configure</code> option when configuring GCC. The default is to
952 1.1 mrg use traps on systems that support them.
953 1.10 mrg </p>
954 1.10 mrg <hr />
955 1.17 mrg <a name="moxie-x-elf"></a><a name="moxie-*-elf"></a>
956 1.17 mrg <h3 class="heading">moxie-*-elf</h3>
957 1.3 skrll <p>The moxie processor.
958 1.10 mrg </p>
959 1.10 mrg <hr />
960 1.17 mrg <a name="msp430-x-elf"></a><a name="msp430-*-elf*"></a>
961 1.17 mrg <h3 class="heading">msp430-*-elf*</h3>
962 1.10 mrg <p>TI MSP430 processor.
963 1.5 mrg This configuration is intended for embedded systems.
964 1.10 mrg </p>
965 1.12 mrg <p>‘<samp>msp430-*-elf</samp>’ is the standard configuration with most GCC
966 1.12 mrg features enabled by default.
967 1.12 mrg </p>
968 1.12 mrg <p>‘<samp>msp430-*-elfbare</samp>’ is tuned for a bare-metal environment, and disables
969 1.12 mrg features related to shared libraries and other functionality not used for
970 1.12 mrg this device. This reduces code and data usage of the GCC libraries, resulting
971 1.12 mrg in a minimal run-time environment by default.
972 1.12 mrg </p>
973 1.12 mrg <p>Features disabled by default include:
974 1.12 mrg </p><ul>
975 1.12 mrg <li> transactional memory
976 1.12 mrg </li><li> __cxa_atexit
977 1.12 mrg </li></ul>
978 1.12 mrg
979 1.10 mrg <hr />
980 1.17 mrg <a name="nds32le-x-elf"></a><a name="nds32le-*-elf"></a>
981 1.17 mrg <h3 class="heading">nds32le-*-elf</h3>
982 1.5 mrg <p>Andes NDS32 target in little endian mode.
983 1.10 mrg </p>
984 1.10 mrg <hr />
985 1.17 mrg <a name="nds32be-x-elf"></a><a name="nds32be-*-elf"></a>
986 1.17 mrg <h3 class="heading">nds32be-*-elf</h3>
987 1.5 mrg <p>Andes NDS32 target in big endian mode.
988 1.10 mrg </p>
989 1.10 mrg <hr />
990 1.17 mrg <a name="nvptx-x-none"></a><a name="nvptx-*-none"></a>
991 1.17 mrg <h3 class="heading">nvptx-*-none</h3>
992 1.5 mrg <p>Nvidia PTX target.
993 1.10 mrg </p>
994 1.10 mrg <p>Instead of GNU binutils, you will need to install
995 1.10 mrg <a href="https://github.com/MentorEmbedded/nvptx-tools/">nvptx-tools</a>.
996 1.5 mrg Tell GCC where to find it:
997 1.10 mrg <samp>--with-build-time-tools=[install-nvptx-tools]/nvptx-none/bin</samp>.
998 1.10 mrg </p>
999 1.15 mrg <p>You will need newlib 3.1.0 or later. It can be
1000 1.10 mrg automatically built together with GCC. For this, add a symbolic link
1001 1.10 mrg to nvptx-newlib’s <samp>newlib</samp> directory to the directory containing
1002 1.10 mrg the GCC sources.
1003 1.10 mrg </p>
1004 1.10 mrg <p>Use the <samp>--disable-sjlj-exceptions</samp> and
1005 1.10 mrg <samp>--enable-newlib-io-long-long</samp> options when configuring.
1006 1.10 mrg </p>
1007 1.10 mrg <hr />
1008 1.17 mrg <a name="or1k-x-elf"></a><a name="or1k-*-elf"></a>
1009 1.17 mrg <h3 class="heading">or1k-*-elf</h3>
1010 1.11 mrg <p>The OpenRISC 1000 32-bit processor with delay slots.
1011 1.11 mrg This configuration is intended for embedded systems.
1012 1.11 mrg </p>
1013 1.11 mrg <hr />
1014 1.17 mrg <a name="or1k-x-linux"></a><a name="or1k-*-linux"></a>
1015 1.17 mrg <h3 class="heading">or1k-*-linux</h3>
1016 1.11 mrg <p>The OpenRISC 1000 32-bit processor with delay slots.
1017 1.11 mrg </p>
1018 1.11 mrg <hr />
1019 1.17 mrg <a name="powerpc-x-x"></a><a name="powerpc-*-*"></a>
1020 1.17 mrg <h3 class="heading">powerpc-*-*</h3>
1021 1.10 mrg <p>You can specify a default version for the <samp>-mcpu=<var>cpu_type</var></samp>
1022 1.10 mrg switch by using the configure option <samp>--with-cpu-<var>cpu_type</var></samp>.
1023 1.10 mrg </p>
1024 1.12 mrg <p>You will need GNU binutils 2.20 or newer.
1025 1.10 mrg </p>
1026 1.10 mrg <hr />
1027 1.17 mrg <a name="powerpc-x-darwin"></a><a name="powerpc-*-darwin*"></a>
1028 1.17 mrg <h3 class="heading">powerpc-*-darwin*</h3>
1029 1.1 mrg <p>PowerPC running Darwin (Mac OS X kernel).
1030 1.10 mrg </p>
1031 1.10 mrg <p>Pre-installed versions of Mac OS X may not include any developer tools,
1032 1.1 mrg meaning that you will not be able to build GCC from source. Tool
1033 1.1 mrg binaries are available at
1034 1.9 mrg <a href="https://opensource.apple.com">https://opensource.apple.com</a>.
1035 1.10 mrg </p>
1036 1.10 mrg <p>This version of GCC requires at least cctools-590.36. The
1037 1.1 mrg cctools-590.36 package referenced from
1038 1.15 mrg <a href="https://gcc.gnu.org/ml/gcc/2006-03/msg00507.html">https://gcc.gnu.org/ml/gcc/2006-03/msg00507.html</a> will not work
1039 1.1 mrg on systems older than 10.3.9 (aka darwin7.9.0).
1040 1.10 mrg </p>
1041 1.10 mrg <hr />
1042 1.17 mrg <a name="powerpc-x-elf"></a><a name="powerpc-*-elf"></a>
1043 1.17 mrg <h3 class="heading">powerpc-*-elf</h3>
1044 1.1 mrg <p>PowerPC system in big endian mode, running System V.4.
1045 1.10 mrg </p>
1046 1.10 mrg <hr />
1047 1.17 mrg <a name="powerpc-x-linux-gnu"></a><a name="powerpc*-*-linux-gnu*"></a>
1048 1.17 mrg <h3 class="heading">powerpc*-*-linux-gnu*</h3>
1049 1.1 mrg <p>PowerPC system in big endian mode running Linux.
1050 1.10 mrg </p>
1051 1.10 mrg <hr />
1052 1.17 mrg <a name="powerpc-x-netbsd"></a><a name="powerpc-*-netbsd*"></a>
1053 1.17 mrg <h3 class="heading">powerpc-*-netbsd*</h3>
1054 1.1 mrg <p>PowerPC system in big endian mode running NetBSD.
1055 1.10 mrg </p>
1056 1.10 mrg <hr />
1057 1.17 mrg <a name="powerpc-x-eabisim"></a><a name="powerpc-*-eabisim"></a>
1058 1.17 mrg <h3 class="heading">powerpc-*-eabisim</h3>
1059 1.1 mrg <p>Embedded PowerPC system in big endian mode for use in running under the
1060 1.1 mrg PSIM simulator.
1061 1.10 mrg </p>
1062 1.10 mrg <hr />
1063 1.17 mrg <a name="powerpc-x-eabi"></a><a name="powerpc-*-eabi"></a>
1064 1.17 mrg <h3 class="heading">powerpc-*-eabi</h3>
1065 1.1 mrg <p>Embedded PowerPC system in big endian mode.
1066 1.10 mrg </p>
1067 1.10 mrg <hr />
1068 1.17 mrg <a name="powerpcle-x-elf"></a><a name="powerpcle-*-elf"></a>
1069 1.17 mrg <h3 class="heading">powerpcle-*-elf</h3>
1070 1.1 mrg <p>PowerPC system in little endian mode, running System V.4.
1071 1.10 mrg </p>
1072 1.10 mrg <hr />
1073 1.17 mrg <a name="powerpcle-x-eabisim"></a><a name="powerpcle-*-eabisim"></a>
1074 1.17 mrg <h3 class="heading">powerpcle-*-eabisim</h3>
1075 1.1 mrg <p>Embedded PowerPC system in little endian mode for use in running under
1076 1.1 mrg the PSIM simulator.
1077 1.10 mrg </p>
1078 1.10 mrg <hr />
1079 1.17 mrg <a name="powerpcle-x-eabi"></a><a name="powerpcle-*-eabi"></a>
1080 1.17 mrg <h3 class="heading">powerpcle-*-eabi</h3>
1081 1.1 mrg <p>Embedded PowerPC system in little endian mode.
1082 1.10 mrg </p>
1083 1.10 mrg <hr />
1084 1.17 mrg <a name="rl78-x-elf"></a><a name="rl78-*-elf"></a>
1085 1.17 mrg <h3 class="heading">rl78-*-elf</h3>
1086 1.10 mrg <p>The Renesas RL78 processor.
1087 1.10 mrg This configuration is intended for embedded systems.
1088 1.10 mrg </p>
1089 1.10 mrg <hr />
1090 1.17 mrg <a name="riscv32-x-elf"></a><a name="riscv32-*-elf"></a>
1091 1.17 mrg <h3 class="heading">riscv32-*-elf</h3>
1092 1.10 mrg <p>The RISC-V RV32 instruction set.
1093 1.3 skrll This configuration is intended for embedded systems.
1094 1.12 mrg This (and all other RISC-V) targets require the binutils 2.30 release.
1095 1.10 mrg </p>
1096 1.10 mrg <hr />
1097 1.17 mrg <a name="riscv32-x-linux"></a><a name="riscv32-*-linux"></a>
1098 1.17 mrg <h3 class="heading">riscv32-*-linux</h3>
1099 1.10 mrg <p>The RISC-V RV32 instruction set running GNU/Linux.
1100 1.12 mrg This (and all other RISC-V) targets require the binutils 2.30 release.
1101 1.10 mrg </p>
1102 1.10 mrg <hr />
1103 1.17 mrg <a name="riscv64-x-elf"></a><a name="riscv64-*-elf"></a>
1104 1.17 mrg <h3 class="heading">riscv64-*-elf</h3>
1105 1.10 mrg <p>The RISC-V RV64 instruction set.
1106 1.10 mrg This configuration is intended for embedded systems.
1107 1.12 mrg This (and all other RISC-V) targets require the binutils 2.30 release.
1108 1.10 mrg </p>
1109 1.10 mrg <hr />
1110 1.17 mrg <a name="riscv64-x-linux"></a><a name="riscv64-*-linux"></a>
1111 1.17 mrg <h3 class="heading">riscv64-*-linux</h3>
1112 1.10 mrg <p>The RISC-V RV64 instruction set running GNU/Linux.
1113 1.12 mrg This (and all other RISC-V) targets require the binutils 2.30 release.
1114 1.10 mrg </p>
1115 1.10 mrg <hr />
1116 1.17 mrg <a name="rx-x-elf"></a><a name="rx-*-elf"></a>
1117 1.17 mrg <h3 class="heading">rx-*-elf</h3>
1118 1.9 mrg <p>The Renesas RX processor.
1119 1.10 mrg </p>
1120 1.10 mrg <hr />
1121 1.17 mrg <a name="s390-x-linux"></a><a name="s390-*-linux*"></a>
1122 1.17 mrg <h3 class="heading">s390-*-linux*</h3>
1123 1.1 mrg <p>S/390 system running GNU/Linux for S/390.
1124 1.10 mrg </p>
1125 1.10 mrg <hr />
1126 1.17 mrg <a name="s390x-x-linux"></a><a name="s390x-*-linux*"></a>
1127 1.17 mrg <h3 class="heading">s390x-*-linux*</h3>
1128 1.1 mrg <p>zSeries system (64-bit) running GNU/Linux for zSeries.
1129 1.10 mrg </p>
1130 1.10 mrg <hr />
1131 1.17 mrg <a name="s390x-ibm-tpf"></a><a name="s390x-ibm-tpf*"></a>
1132 1.17 mrg <h3 class="heading">s390x-ibm-tpf*</h3>
1133 1.1 mrg <p>zSeries system (64-bit) running TPF. This platform is
1134 1.1 mrg supported as cross-compilation target only.
1135 1.10 mrg </p>
1136 1.10 mrg <hr />
1137 1.17 mrg <a name="x-x-solaris2"></a><a name="g_t*-*-solaris2*"></a>
1138 1.17 mrg <h3 class="heading">*-*-solaris2*</h3>
1139 1.12 mrg <p>Support for Solaris 10 has been removed in GCC 10. Support for Solaris
1140 1.12 mrg 9 has been removed in GCC 5. Support for Solaris 8 has been removed in
1141 1.12 mrg GCC 4.8. Support for Solaris 7 has been removed in GCC 4.6.
1142 1.12 mrg </p>
1143 1.12 mrg <p>Solaris 11.3 provides GCC 4.5.2, 4.7.3, and 4.8.2 as
1144 1.12 mrg <code>/usr/gcc/4.5/bin/gcc</code> or similar. Newer Solaris versions
1145 1.12 mrg provide one or more of GCC 5, 7, and 9. Alternatively,
1146 1.1 mrg you can install a pre-built GCC to bootstrap and install GCC. See the
1147 1.1 mrg <a href="binaries.html">binaries page</a> for details.
1148 1.10 mrg </p>
1149 1.10 mrg <p>The Solaris 2 <code>/bin/sh</code> will often fail to configure
1150 1.12 mrg ‘<samp>libstdc++-v3</samp>’. We therefore recommend using the
1151 1.9 mrg following initial sequence of commands
1152 1.10 mrg </p>
1153 1.17 mrg <div class="smallexample">
1154 1.17 mrg <pre class="smallexample">% CONFIG_SHELL=/bin/ksh
1155 1.10 mrg % export CONFIG_SHELL
1156 1.10 mrg </pre></div>
1157 1.9 mrg
1158 1.10 mrg <p>and proceed as described in <a href="configure.html">the configure instructions</a>.
1159 1.1 mrg In addition we strongly recommend specifying an absolute path to invoke
1160 1.10 mrg <code><var>srcdir</var>/configure</code>.
1161 1.10 mrg </p>
1162 1.12 mrg <p>In Solaris 11, you need to check for <code>system/header</code>,
1163 1.12 mrg <code>system/linker</code>, and <code>developer/assembler</code> packages.
1164 1.10 mrg </p>
1165 1.10 mrg <p>Trying to use the linker and other tools in
1166 1.10 mrg <samp>/usr/ucb</samp> to install GCC has been observed to cause trouble.
1167 1.1 mrg For example, the linker may hang indefinitely. The fix is to remove
1168 1.10 mrg <samp>/usr/ucb</samp> from your <code>PATH</code>.
1169 1.10 mrg </p>
1170 1.12 mrg <p>The build process works more smoothly with the legacy Solaris tools so, if you
1171 1.10 mrg have <samp>/usr/xpg4/bin</samp> in your <code>PATH</code>, we recommend that you place
1172 1.10 mrg <samp>/usr/bin</samp> before <samp>/usr/xpg4/bin</samp> for the duration of the build.
1173 1.10 mrg </p>
1174 1.10 mrg <p>We recommend the use of the Solaris assembler or the GNU assembler, in
1175 1.10 mrg conjunction with the Solaris linker. The GNU <code>as</code>
1176 1.12 mrg versions included in Solaris 11.3,
1177 1.12 mrg from GNU binutils 2.23.1 or newer (in <samp>/usr/bin/gas</samp> and
1178 1.10 mrg <samp>/usr/gnu/bin/as</samp>), are known to work.
1179 1.12 mrg The current version, from GNU binutils 2.34,
1180 1.10 mrg is known to work as well. Note that your mileage may vary
1181 1.7 mrg if you use a combination of the GNU tools and the Solaris tools: while the
1182 1.12 mrg combination GNU <code>as</code> + Solaris <code>ld</code> should reasonably work,
1183 1.12 mrg the reverse combination Solaris <code>as</code> + GNU <code>ld</code> may fail to
1184 1.10 mrg build or cause memory corruption at runtime in some cases for C++ programs.
1185 1.12 mrg GNU <code>ld</code> usually works as well. Again, the current
1186 1.12 mrg version (2.34) is known to work, but generally lacks platform specific
1187 1.10 mrg features, so better stay with Solaris <code>ld</code>. To use the LTO linker
1188 1.10 mrg plugin (<samp>-fuse-linker-plugin</samp>) with GNU <code>ld</code>, GNU
1189 1.10 mrg binutils <em>must</em> be configured with <samp>--enable-largefile</samp>.
1190 1.10 mrg </p>
1191 1.10 mrg <p>To enable symbol versioning in ‘<samp>libstdc++</samp>’ with the Solaris linker,
1192 1.10 mrg you need to have any version of GNU <code>c++filt</code>, which is part of
1193 1.10 mrg GNU binutils. ‘<samp>libstdc++</samp>’ symbol versioning will be disabled if no
1194 1.10 mrg appropriate version is found. Solaris <code>c++filt</code> from the Solaris
1195 1.7 mrg Studio compilers does <em>not</em> work.
1196 1.10 mrg </p>
1197 1.15 mrg <p>In order to build the GNU D compiler, GDC, a working ‘<samp>libphobos</samp>’ is
1198 1.15 mrg needed. That library wasn’t built by default in GCC 9–11 on SPARC, or
1199 1.15 mrg on x86 when the Solaris assembler is used, but can be enabled by
1200 1.15 mrg configuring with <samp>--enable-libphobos</samp>. Also, GDC 9.4.0 is
1201 1.15 mrg required on x86, while GDC 9.3.0 is known to work on SPARC.
1202 1.15 mrg </p>
1203 1.12 mrg <p>The versions of the GNU Multiple Precision Library (GMP), the MPFR
1204 1.12 mrg library and the MPC library bundled with Solaris 11.3 and later are
1205 1.12 mrg usually recent enough to match GCC’s requirements. There are two
1206 1.12 mrg caveats:
1207 1.10 mrg </p>
1208 1.12 mrg <ul>
1209 1.12 mrg <li> While the version of the GMP library in Solaris 11.3 works with GCC, you
1210 1.12 mrg need to configure with <samp>--with-gmp-include=/usr/include/gmp</samp>.
1211 1.12 mrg
1212 1.12 mrg </li><li> The version of the MPFR libary included in Solaris 11.3 is too old; you
1213 1.12 mrg need to provide a more recent one.
1214 1.15 mrg
1215 1.12 mrg </li></ul>
1216 1.12 mrg
1217 1.10 mrg <hr />
1218 1.17 mrg <a name="sparc-x-x"></a><a name="sparc*-*-*"></a>
1219 1.17 mrg <h3 class="heading">sparc*-*-*</h3>
1220 1.3 skrll <p>This section contains general configuration information for all
1221 1.3 skrll SPARC-based platforms. In addition to reading this section, please
1222 1.3 skrll read all other sections that match your target.
1223 1.10 mrg </p>
1224 1.10 mrg <p>Newer versions of the GNU Multiple Precision Library (GMP), the MPFR
1225 1.3 skrll library and the MPC library are known to be miscompiled by earlier
1226 1.3 skrll versions of GCC on these platforms. We therefore recommend the use
1227 1.3 skrll of the exact versions of these libraries listed as minimal versions
1228 1.3 skrll in <a href="prerequisites.html">the prerequisites</a>.
1229 1.10 mrg </p>
1230 1.10 mrg <hr />
1231 1.17 mrg <a name="sparc-sun-solaris2"></a><a name="sparc-sun-solaris2*"></a>
1232 1.17 mrg <h3 class="heading">sparc-sun-solaris2*</h3>
1233 1.1 mrg <p>When GCC is configured to use GNU binutils 2.14 or later, the binaries
1234 1.12 mrg produced are smaller than the ones produced using Solaris native tools;
1235 1.1 mrg this difference is quite significant for binaries containing debugging
1236 1.1 mrg information.
1237 1.10 mrg </p>
1238 1.10 mrg <p>Starting with Solaris 7, the operating system is capable of executing
1239 1.1 mrg 64-bit SPARC V9 binaries. GCC 3.1 and later properly supports
1240 1.10 mrg this; the <samp>-m64</samp> option enables 64-bit code generation.
1241 1.1 mrg However, if all you want is code tuned for the UltraSPARC CPU, you
1242 1.10 mrg should try the <samp>-mtune=ultrasparc</samp> option instead, which produces
1243 1.1 mrg code that, unlike full 64-bit code, can still run on non-UltraSPARC
1244 1.1 mrg machines.
1245 1.10 mrg </p>
1246 1.10 mrg <p>When configuring the GNU Multiple Precision Library (GMP), the MPFR
1247 1.3 skrll library or the MPC library on a Solaris 7 or later system, the canonical
1248 1.10 mrg target triplet must be specified as the <code>build</code> parameter on the
1249 1.10 mrg configure line. This target triplet can be obtained by invoking <code>./config.guess</code> in the toplevel source directory of GCC (and
1250 1.12 mrg not that of GMP or MPFR or MPC). For example on a Solaris 11 system:
1251 1.10 mrg </p>
1252 1.17 mrg <div class="smallexample">
1253 1.17 mrg <pre class="smallexample">% ./configure --build=sparc-sun-solaris2.11 --prefix=xxx
1254 1.10 mrg </pre></div>
1255 1.10 mrg
1256 1.10 mrg <hr />
1257 1.17 mrg <a name="sparc-x-linux"></a><a name="sparc-*-linux*"></a>
1258 1.17 mrg <h3 class="heading">sparc-*-linux*</h3>
1259 1.10 mrg
1260 1.10 mrg <hr />
1261 1.17 mrg <a name="sparc64-x-solaris2"></a><a name="sparc64-*-solaris2*"></a>
1262 1.17 mrg <h3 class="heading">sparc64-*-solaris2*</h3>
1263 1.12 mrg <p>When configuring a 64-bit-default GCC on Solaris/SPARC, you must use a
1264 1.12 mrg build compiler that generates 64-bit code, either by default or by
1265 1.12 mrg specifying ‘<samp>CC='gcc -m64' CXX='gcc-m64'</samp>’ to <code>configure</code>.
1266 1.12 mrg Additionally, you <em>must</em> pass <samp>--build=sparc64-sun-solaris2.11</samp>
1267 1.12 mrg or <samp>--build=sparcv9-sun-solaris2.11</samp> because <samp>config.guess</samp>
1268 1.12 mrg misdetects this situation, which can cause build failures.
1269 1.12 mrg </p>
1270 1.3 skrll <p>When configuring the GNU Multiple Precision Library (GMP), the MPFR
1271 1.3 skrll library or the MPC library, the canonical target triplet must be specified
1272 1.10 mrg as the <code>build</code> parameter on the configure line. For example
1273 1.12 mrg on a Solaris 11 system:
1274 1.10 mrg </p>
1275 1.17 mrg <div class="smallexample">
1276 1.17 mrg <pre class="smallexample">% ./configure --build=sparc64-sun-solaris2.11 --prefix=xxx
1277 1.10 mrg </pre></div>
1278 1.10 mrg
1279 1.10 mrg <hr />
1280 1.17 mrg <a name="sparcv9-x-solaris2"></a><a name="sparcv9-*-solaris2*"></a>
1281 1.17 mrg <h3 class="heading">sparcv9-*-solaris2*</h3>
1282 1.10 mrg <p>This is a synonym for ‘<samp>sparc64-*-solaris2*</samp>’.
1283 1.10 mrg </p>
1284 1.10 mrg <hr />
1285 1.17 mrg <a name="c6x-x-x"></a><a name="c6x-*-*"></a>
1286 1.17 mrg <h3 class="heading">c6x-*-*</h3>
1287 1.3 skrll <p>The C6X family of processors. This port requires binutils-2.22 or newer.
1288 1.10 mrg </p>
1289 1.10 mrg <hr />
1290 1.17 mrg <a name="tilegx-*-linux"></a><a name="tilegx-*-linux*"></a>
1291 1.17 mrg <h3 class="heading">tilegx-*-linux*</h3>
1292 1.5 mrg <p>The TILE-Gx processor in little endian mode, running GNU/Linux. This
1293 1.5 mrg port requires binutils-2.22 or newer.
1294 1.10 mrg </p>
1295 1.10 mrg <hr />
1296 1.17 mrg <a name="tilegxbe-*-linux"></a><a name="tilegxbe-*-linux*"></a>
1297 1.17 mrg <h3 class="heading">tilegxbe-*-linux*</h3>
1298 1.5 mrg <p>The TILE-Gx processor in big endian mode, running GNU/Linux. This
1299 1.5 mrg port requires binutils-2.23 or newer.
1300 1.10 mrg </p>
1301 1.10 mrg <hr />
1302 1.17 mrg <a name="tilepro-*-linux"></a><a name="tilepro-*-linux*"></a>
1303 1.17 mrg <h3 class="heading">tilepro-*-linux*</h3>
1304 1.3 skrll <p>The TILEPro processor running GNU/Linux. This port requires
1305 1.3 skrll binutils-2.22 or newer.
1306 1.10 mrg </p>
1307 1.10 mrg <hr />
1308 1.17 mrg <a name="visium-x-elf"></a><a name="visium-*-elf"></a>
1309 1.17 mrg <h3 class="heading">visium-*-elf</h3>
1310 1.10 mrg <p>CDS VISIUMcore processor.
1311 1.5 mrg This configuration is intended for embedded systems.
1312 1.10 mrg </p>
1313 1.10 mrg <hr />
1314 1.17 mrg <a name="x-x-vxworks"></a><a name="g_t*-*-vxworks*"></a>
1315 1.17 mrg <h3 class="heading">*-*-vxworks*</h3>
1316 1.1 mrg <p>Support for VxWorks is in flux. At present GCC supports <em>only</em> the
1317 1.10 mrg very recent VxWorks 5.5 (aka Tornado 2.2) release, and only on PowerPC.
1318 1.10 mrg We welcome patches for other architectures supported by VxWorks 5.5.
1319 1.1 mrg Support for VxWorks AE would also be welcome; we believe this is merely
1320 1.1 mrg a matter of writing an appropriate “configlette” (see below). We are
1321 1.1 mrg not interested in supporting older, a.out or COFF-based, versions of
1322 1.1 mrg VxWorks in GCC 3.
1323 1.10 mrg </p>
1324 1.10 mrg <p>VxWorks comes with an older version of GCC installed in
1325 1.10 mrg <samp><var>$WIND_BASE</var>/host</samp>; we recommend you do not overwrite it.
1326 1.10 mrg Choose an installation <var>prefix</var> entirely outside <var>$WIND_BASE</var>.
1327 1.10 mrg Before running <code>configure</code>, create the directories <samp><var>prefix</var></samp>
1328 1.10 mrg and <samp><var>prefix</var>/bin</samp>. Link or copy the appropriate assembler,
1329 1.10 mrg linker, etc. into <samp><var>prefix</var>/bin</samp>, and set your <var>PATH</var> to
1330 1.10 mrg include that directory while running both <code>configure</code> and
1331 1.10 mrg <code>make</code>.
1332 1.10 mrg </p>
1333 1.10 mrg <p>You must give <code>configure</code> the
1334 1.10 mrg <samp>--with-headers=<var>$WIND_BASE</var>/target/h</samp> switch so that it can
1335 1.1 mrg find the VxWorks system headers. Since VxWorks is a cross compilation
1336 1.10 mrg target only, you must also specify <samp>--target=<var>target</var></samp>.
1337 1.10 mrg <code>configure</code> will attempt to create the directory
1338 1.10 mrg <samp><var>prefix</var>/<var>target</var>/sys-include</samp> and copy files into it;
1339 1.10 mrg make sure the user running <code>configure</code> has sufficient privilege
1340 1.1 mrg to do so.
1341 1.10 mrg </p>
1342 1.10 mrg <p>GCC’s exception handling runtime requires a special “configlette”
1343 1.10 mrg module, <samp>contrib/gthr_supp_vxw_5x.c</samp>. Follow the instructions in
1344 1.1 mrg that file to add the module to your kernel build. (Future versions of
1345 1.1 mrg VxWorks will incorporate this module.)
1346 1.10 mrg </p>
1347 1.10 mrg <hr />
1348 1.17 mrg <a name="x86-64-x-x"></a><a name="x86_005f64-*-*_002c-amd64-*-*"></a>
1349 1.17 mrg <h3 class="heading">x86_64-*-*, amd64-*-*</h3>
1350 1.1 mrg <p>GCC supports the x86-64 architecture implemented by the AMD64 processor
1351 1.10 mrg (amd64-*-* is an alias for x86_64-*-*) on GNU/Linux, FreeBSD and NetBSD.
1352 1.1 mrg On GNU/Linux the default is a bi-arch compiler which is able to generate
1353 1.10 mrg both 64-bit x86-64 and 32-bit x86 code (via the <samp>-m32</samp> switch).
1354 1.10 mrg </p>
1355 1.10 mrg <hr />
1356 1.17 mrg <a name="x86-64-x-solaris2"></a><a name="x86_005f64-*-solaris2*"></a>
1357 1.17 mrg <h3 class="heading">x86_64-*-solaris2*</h3>
1358 1.3 skrll <p>GCC also supports the x86-64 architecture implemented by the AMD64
1359 1.10 mrg processor (‘<samp>amd64-*-*</samp>’ is an alias for ‘<samp>x86_64-*-*</samp>’) on
1360 1.3 skrll Solaris 10 or later. Unlike other systems, without special options a
1361 1.3 skrll bi-arch compiler is built which generates 32-bit code by default, but
1362 1.10 mrg can generate 64-bit x86-64 code with the <samp>-m64</samp> switch. Since
1363 1.7 mrg GCC 4.7, there is also a configuration that defaults to 64-bit code, but
1364 1.10 mrg can generate 32-bit code with <samp>-m32</samp>. To configure and build
1365 1.10 mrg this way, you have to provide all support libraries like <samp>libgmp</samp>
1366 1.12 mrg as 64-bit code, configure with <samp>--target=x86_64-pc-solaris2.11</samp>
1367 1.10 mrg and ‘<samp>CC=gcc -m64</samp>’.
1368 1.10 mrg </p>
1369 1.10 mrg <hr />
1370 1.17 mrg <a name="xtensa-x-elf"></a><a name="xtensa*-*-elf"></a>
1371 1.17 mrg <h3 class="heading">xtensa*-*-elf</h3>
1372 1.1 mrg <p>This target is intended for embedded Xtensa systems using the
1373 1.10 mrg ‘<samp>newlib</samp>’ C library. It uses ELF but does not support shared
1374 1.1 mrg objects. Designed-defined instructions specified via the
1375 1.1 mrg Tensilica Instruction Extension (TIE) language are only supported
1376 1.1 mrg through inline assembly.
1377 1.10 mrg </p>
1378 1.10 mrg <p>The Xtensa configuration information must be specified prior to
1379 1.10 mrg building GCC. The <samp>include/xtensa-config.h</samp> header
1380 1.1 mrg file contains the configuration information. If you created your
1381 1.1 mrg own Xtensa configuration with the Xtensa Processor Generator, the
1382 1.1 mrg downloaded files include a customized copy of this header file,
1383 1.1 mrg which you can use to replace the default header file.
1384 1.10 mrg </p>
1385 1.10 mrg <hr />
1386 1.17 mrg <a name="xtensa-x-linux"></a><a name="xtensa*-*-linux*"></a>
1387 1.17 mrg <h3 class="heading">xtensa*-*-linux*</h3>
1388 1.1 mrg <p>This target is for Xtensa systems running GNU/Linux. It supports ELF
1389 1.1 mrg shared objects and the GNU C library (glibc). It also generates
1390 1.1 mrg position-independent code (PIC) regardless of whether the
1391 1.10 mrg <samp>-fpic</samp> or <samp>-fPIC</samp> options are used. In other
1392 1.1 mrg respects, this target is the same as the
1393 1.10 mrg <a href="#xtensa*-*-elf">‘<samp>xtensa*-*-elf</samp>’</a> target.
1394 1.10 mrg </p>
1395 1.10 mrg <hr />
1396 1.17 mrg <a name="windows"></a><a name="Microsoft-Windows"></a>
1397 1.17 mrg <h3 class="heading">Microsoft Windows</h3>
1398 1.1 mrg
1399 1.17 mrg <a name="Intel-16-bit-versions"></a>
1400 1.17 mrg <h4 class="subheading">Intel 16-bit versions</h4>
1401 1.1 mrg <p>The 16-bit versions of Microsoft Windows, such as Windows 3.1, are not
1402 1.1 mrg supported.
1403 1.10 mrg </p>
1404 1.10 mrg <p>However, the 32-bit port has limited support for Microsoft
1405 1.1 mrg Windows 3.11 in the Win32s environment, as a target only. See below.
1406 1.10 mrg </p>
1407 1.17 mrg <a name="Intel-32-bit-versions"></a>
1408 1.17 mrg <h4 class="subheading">Intel 32-bit versions</h4>
1409 1.1 mrg <p>The 32-bit versions of Windows, including Windows 95, Windows NT, Windows
1410 1.1 mrg XP, and Windows Vista, are supported by several different target
1411 1.1 mrg platforms. These targets differ in which Windows subsystem they target
1412 1.1 mrg and which C libraries are used.
1413 1.10 mrg </p>
1414 1.10 mrg <ul>
1415 1.10 mrg <li> Cygwin <a href="#x-x-cygwin">*-*-cygwin</a>: Cygwin provides a user-space
1416 1.10 mrg Linux API emulation layer in the Win32 subsystem.
1417 1.10 mrg </li><li> MinGW <a href="#x-x-mingw32">*-*-mingw32</a>: MinGW is a native GCC port for
1418 1.10 mrg the Win32 subsystem that provides a subset of POSIX.
1419 1.10 mrg </li><li> MKS i386-pc-mks: NuTCracker from MKS. See
1420 1.10 mrg <a href="https://www.mkssoftware.com">https://www.mkssoftware.com</a> for more information.
1421 1.10 mrg </li></ul>
1422 1.1 mrg
1423 1.17 mrg <a name="Intel-64-bit-versions"></a>
1424 1.17 mrg <h4 class="subheading">Intel 64-bit versions</h4>
1425 1.1 mrg <p>GCC contains support for x86-64 using the mingw-w64
1426 1.15 mrg runtime library, available from <a href="https://www.mingw-w64.org/downloads/">https://www.mingw-w64.org/downloads/</a>.
1427 1.1 mrg This library should be used with the target triple x86_64-pc-mingw32.
1428 1.10 mrg </p>
1429 1.17 mrg <a name="Windows-CE"></a>
1430 1.17 mrg <h4 class="subheading">Windows CE</h4>
1431 1.3 skrll <p>Windows CE is supported as a target only on Hitachi
1432 1.1 mrg SuperH (sh-wince-pe), and MIPS (mips-wince-pe).
1433 1.10 mrg </p>
1434 1.17 mrg <a name="Other-Windows-Platforms"></a>
1435 1.17 mrg <h4 class="subheading">Other Windows Platforms</h4>
1436 1.1 mrg <p>GCC no longer supports Windows NT on the Alpha or PowerPC.
1437 1.10 mrg </p>
1438 1.10 mrg <p>GCC no longer supports the Windows POSIX subsystem. However, it does
1439 1.1 mrg support the Interix subsystem. See above.
1440 1.10 mrg </p>
1441 1.10 mrg <p>Old target names including *-*-winnt and *-*-windowsnt are no longer used.
1442 1.10 mrg </p>
1443 1.10 mrg <p>PW32 (i386-pc-pw32) support was never completed, and the project seems to
1444 1.1 mrg be inactive. See <a href="http://pw32.sourceforge.net/">http://pw32.sourceforge.net/</a> for more information.
1445 1.10 mrg </p>
1446 1.10 mrg <p>UWIN support has been removed due to a lack of maintenance.
1447 1.10 mrg </p>
1448 1.10 mrg <hr />
1449 1.17 mrg <a name="x-x-cygwin"></a><a name="g_t*-*-cygwin"></a>
1450 1.17 mrg <h3 class="heading">*-*-cygwin</h3>
1451 1.1 mrg <p>Ports of GCC are included with the
1452 1.1 mrg <a href="http://www.cygwin.com/">Cygwin environment</a>.
1453 1.10 mrg </p>
1454 1.10 mrg <p>GCC will build under Cygwin without modification; it does not build
1455 1.10 mrg with Microsoft’s C++ compiler and there are no plans to make it do so.
1456 1.10 mrg </p>
1457 1.10 mrg <p>The Cygwin native compiler can be configured to target any 32-bit x86
1458 1.1 mrg cpu architecture desired; the default is i686-pc-cygwin. It should be
1459 1.1 mrg used with as up-to-date a version of binutils as possible; use either
1460 1.1 mrg the latest official GNU binutils release in the Cygwin distribution,
1461 1.1 mrg or version 2.20 or above if building your own.
1462 1.10 mrg </p>
1463 1.10 mrg <hr />
1464 1.17 mrg <a name="x-x-mingw32"></a><a name="g_t*-*-mingw32"></a>
1465 1.17 mrg <h3 class="heading">*-*-mingw32</h3>
1466 1.10 mrg <p>GCC will build with and support only MinGW runtime 3.12 and later.
1467 1.1 mrg Earlier versions of headers are incompatible with the new default semantics
1468 1.1 mrg of <code>extern inline</code> in <code>-std=c99</code> and <code>-std=gnu99</code> modes.
1469 1.10 mrg </p>
1470 1.15 mrg <p>To support emitting DWARF debugging info you need to use GNU binutils
1471 1.15 mrg version 2.16 or above containing support for the <code>.secrel32</code>
1472 1.15 mrg assembler pseudo-op.
1473 1.15 mrg </p>
1474 1.10 mrg <hr />
1475 1.17 mrg <a name="older"></a><a name="Older-systems"></a>
1476 1.17 mrg <h3 class="heading">Older systems</h3>
1477 1.1 mrg <p>GCC contains support files for many older (1980s and early
1478 1.1 mrg 1990s) Unix variants. For the most part, support for these systems
1479 1.1 mrg has not been deliberately removed, but it has not been maintained for
1480 1.1 mrg several years and may suffer from bitrot.
1481 1.10 mrg </p>
1482 1.10 mrg <p>Starting with GCC 3.1, each release has a list of “obsoleted” systems.
1483 1.1 mrg Support for these systems is still present in that release, but
1484 1.10 mrg <code>configure</code> will fail unless the <samp>--enable-obsolete</samp>
1485 1.1 mrg option is given. Unless a maintainer steps forward, support for these
1486 1.1 mrg systems will be removed from the next release of GCC.
1487 1.10 mrg </p>
1488 1.10 mrg <p>Support for old systems as hosts for GCC can cause problems if the
1489 1.1 mrg workarounds for compiler, library and operating system bugs affect the
1490 1.1 mrg cleanliness or maintainability of the rest of GCC. In some cases, to
1491 1.1 mrg bring GCC up on such a system, if still possible with current GCC, may
1492 1.1 mrg require first installing an old version of GCC which did work on that
1493 1.1 mrg system, and using it to compile a more recent GCC, to avoid bugs in the
1494 1.1 mrg vendor compiler. Old releases of GCC 1 and GCC 2 are available in the
1495 1.10 mrg <samp>old-releases</samp> directory on the <a href="../mirrors.html">GCC mirror
1496 1.10 mrg sites</a>. Header bugs may generally be avoided using
1497 1.10 mrg <code>fixincludes</code>, but bugs or deficiencies in libraries and the
1498 1.1 mrg operating system may still cause problems.
1499 1.10 mrg </p>
1500 1.10 mrg <p>Support for older systems as targets for cross-compilation is less
1501 1.1 mrg problematic than support for them as hosts for GCC; if an enthusiast
1502 1.1 mrg wishes to make such a target work again (including resurrecting any of
1503 1.1 mrg the targets that never worked with GCC 2, starting from the last
1504 1.1 mrg version before they were removed), patches
1505 1.1 mrg <a href="../contribute.html">following the usual requirements</a> would be
1506 1.1 mrg likely to be accepted, since they should not affect the support for more
1507 1.1 mrg modern targets.
1508 1.10 mrg </p>
1509 1.10 mrg <p>For some systems, old versions of GNU binutils may also be useful,
1510 1.10 mrg and are available from <samp>pub/binutils/old-releases</samp> on
1511 1.9 mrg <a href="https://sourceware.org/mirrors.html">sourceware.org mirror sites</a>.
1512 1.10 mrg </p>
1513 1.10 mrg <p>Some of the information on specific systems above relates to
1514 1.1 mrg such older systems, but much of the information
1515 1.1 mrg about GCC on such systems (which may no longer be applicable to
1516 1.1 mrg current GCC) is to be found in the GCC texinfo manual.
1517 1.10 mrg </p>
1518 1.10 mrg <hr />
1519 1.17 mrg <a name="elf"></a><a name="all-ELF-targets-_0028SVR4_002c-Solaris-2_002c-etc_002e_0029"></a>
1520 1.17 mrg <h3 class="heading">all ELF targets (SVR4, Solaris 2, etc.)</h3>
1521 1.1 mrg <p>C++ support is significantly better on ELF targets if you use the
1522 1.1 mrg <a href="./configure.html#with-gnu-ld">GNU linker</a>; duplicate copies of
1523 1.1 mrg inlines, vtables and template instantiations will be discarded
1524 1.1 mrg automatically.
1525 1.10 mrg </p>
1526 1.1 mrg
1527 1.10 mrg <hr />
1528 1.10 mrg <p>
1529 1.1 mrg <p><a href="./index.html">Return to the GCC Installation page</a>
1530 1.10 mrg </p>
1531 1.10 mrg
1532 1.10 mrg
1533 1.10 mrg
1534 1.10 mrg
1535 1.1 mrg
1536 1.10 mrg </body>
1537 1.10 mrg </html>
1538