1 |
38 |
julius |
\input texinfo.tex @c -*-texinfo-*-
|
2 |
|
|
@c @ifnothtml
|
3 |
|
|
@c %**start of header
|
4 |
|
|
@setfilename gccinstall.info
|
5 |
|
|
@settitle Installing GCC
|
6 |
|
|
@setchapternewpage odd
|
7 |
|
|
@c %**end of header
|
8 |
|
|
@c @end ifnothtml
|
9 |
|
|
|
10 |
|
|
@include gcc-common.texi
|
11 |
|
|
|
12 |
|
|
@c Specify title for specific html page
|
13 |
|
|
@ifset indexhtml
|
14 |
|
|
@settitle Installing GCC
|
15 |
|
|
@end ifset
|
16 |
|
|
@ifset specifichtml
|
17 |
|
|
@settitle Host/Target specific installation notes for GCC
|
18 |
|
|
@end ifset
|
19 |
|
|
@ifset prerequisiteshtml
|
20 |
|
|
@settitle Prerequisites for GCC
|
21 |
|
|
@end ifset
|
22 |
|
|
@ifset downloadhtml
|
23 |
|
|
@settitle Downloading GCC
|
24 |
|
|
@end ifset
|
25 |
|
|
@ifset configurehtml
|
26 |
|
|
@settitle Installing GCC: Configuration
|
27 |
|
|
@end ifset
|
28 |
|
|
@ifset buildhtml
|
29 |
|
|
@settitle Installing GCC: Building
|
30 |
|
|
@end ifset
|
31 |
|
|
@ifset testhtml
|
32 |
|
|
@settitle Installing GCC: Testing
|
33 |
|
|
@end ifset
|
34 |
|
|
@ifset finalinstallhtml
|
35 |
|
|
@settitle Installing GCC: Final installation
|
36 |
|
|
@end ifset
|
37 |
|
|
@ifset binarieshtml
|
38 |
|
|
@settitle Installing GCC: Binaries
|
39 |
|
|
@end ifset
|
40 |
|
|
@ifset oldhtml
|
41 |
|
|
@settitle Installing GCC: Old documentation
|
42 |
|
|
@end ifset
|
43 |
|
|
@ifset gfdlhtml
|
44 |
|
|
@settitle Installing GCC: GNU Free Documentation License
|
45 |
|
|
@end ifset
|
46 |
|
|
|
47 |
|
|
@c Copyright (C) 1988, 1989, 1992, 1993, 1994, 1995, 1996, 1997, 1998,
|
48 |
|
|
@c 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
|
49 |
|
|
@c *** Converted to texinfo by Dean Wakerley, dean@wakerley.com
|
50 |
|
|
|
51 |
|
|
@c IMPORTANT: whenever you modify this file, run `install.texi2html' to
|
52 |
|
|
@c test the generation of HTML documents for the gcc.gnu.org web pages.
|
53 |
|
|
@c
|
54 |
|
|
@c Do not use @footnote{} in this file as it breaks install.texi2html!
|
55 |
|
|
|
56 |
|
|
@c Include everything if we're not making html
|
57 |
|
|
@ifnothtml
|
58 |
|
|
@set indexhtml
|
59 |
|
|
@set specifichtml
|
60 |
|
|
@set prerequisiteshtml
|
61 |
|
|
@set downloadhtml
|
62 |
|
|
@set configurehtml
|
63 |
|
|
@set buildhtml
|
64 |
|
|
@set testhtml
|
65 |
|
|
@set finalinstallhtml
|
66 |
|
|
@set binarieshtml
|
67 |
|
|
@set oldhtml
|
68 |
|
|
@set gfdlhtml
|
69 |
|
|
@end ifnothtml
|
70 |
|
|
|
71 |
|
|
@c Part 2 Summary Description and Copyright
|
72 |
|
|
@copying
|
73 |
|
|
Copyright @copyright{} 1988, 1989, 1992, 1993, 1994, 1995, 1996, 1997, 1998,
|
74 |
|
|
1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006 Free Software Foundation, Inc.
|
75 |
|
|
@sp 1
|
76 |
|
|
Permission is granted to copy, distribute and/or modify this document
|
77 |
|
|
under the terms of the GNU Free Documentation License, Version 1.2 or
|
78 |
|
|
any later version published by the Free Software Foundation; with no
|
79 |
|
|
Invariant Sections, the Front-Cover texts being (a) (see below), and
|
80 |
|
|
with the Back-Cover Texts being (b) (see below). A copy of the
|
81 |
|
|
license is included in the section entitled ``@uref{./gfdl.html,,GNU
|
82 |
|
|
Free Documentation License}''.
|
83 |
|
|
|
84 |
|
|
(a) The FSF's Front-Cover Text is:
|
85 |
|
|
|
86 |
|
|
A GNU Manual
|
87 |
|
|
|
88 |
|
|
(b) The FSF's Back-Cover Text is:
|
89 |
|
|
|
90 |
|
|
You have freedom to copy and modify this GNU Manual, like GNU
|
91 |
|
|
software. Copies published by the Free Software Foundation raise
|
92 |
|
|
funds for GNU development.
|
93 |
|
|
@end copying
|
94 |
|
|
@ifinfo
|
95 |
|
|
@insertcopying
|
96 |
|
|
@end ifinfo
|
97 |
|
|
@dircategory Software development
|
98 |
|
|
@direntry
|
99 |
|
|
* gccinstall: (gccinstall). Installing the GNU Compiler Collection.
|
100 |
|
|
@end direntry
|
101 |
|
|
|
102 |
|
|
@c Part 3 Titlepage and Copyright
|
103 |
|
|
@titlepage
|
104 |
|
|
@title Installing GCC
|
105 |
|
|
@versionsubtitle
|
106 |
|
|
|
107 |
|
|
@c The following two commands start the copyright page.
|
108 |
|
|
@page
|
109 |
|
|
@vskip 0pt plus 1filll
|
110 |
|
|
@insertcopying
|
111 |
|
|
@end titlepage
|
112 |
|
|
|
113 |
|
|
@c Part 4 Top node, Master Menu, and/or Table of Contents
|
114 |
|
|
@ifinfo
|
115 |
|
|
@node Top, , , (dir)
|
116 |
|
|
@comment node-name, next, Previous, up
|
117 |
|
|
|
118 |
|
|
@menu
|
119 |
|
|
* Installing GCC:: This document describes the generic installation
|
120 |
|
|
procedure for GCC as well as detailing some target
|
121 |
|
|
specific installation instructions.
|
122 |
|
|
|
123 |
|
|
* Specific:: Host/target specific installation notes for GCC.
|
124 |
|
|
* Binaries:: Where to get pre-compiled binaries.
|
125 |
|
|
|
126 |
|
|
* Old:: Old installation documentation.
|
127 |
|
|
|
128 |
|
|
* GNU Free Documentation License:: How you can copy and share this manual.
|
129 |
|
|
* Concept Index:: This index has two entries.
|
130 |
|
|
@end menu
|
131 |
|
|
@end ifinfo
|
132 |
|
|
|
133 |
|
|
@iftex
|
134 |
|
|
@contents
|
135 |
|
|
@end iftex
|
136 |
|
|
|
137 |
|
|
@c Part 5 The Body of the Document
|
138 |
|
|
@c ***Installing GCC**********************************************************
|
139 |
|
|
@ifnothtml
|
140 |
|
|
@comment node-name, next, previous, up
|
141 |
|
|
@node Installing GCC, Binaries, , Top
|
142 |
|
|
@end ifnothtml
|
143 |
|
|
@ifset indexhtml
|
144 |
|
|
@ifnothtml
|
145 |
|
|
@chapter Installing GCC
|
146 |
|
|
@end ifnothtml
|
147 |
|
|
|
148 |
|
|
The latest version of this document is always available at
|
149 |
|
|
@uref{http://gcc.gnu.org/install/,,http://gcc.gnu.org/install/}.
|
150 |
|
|
|
151 |
|
|
This document describes the generic installation procedure for GCC as well
|
152 |
|
|
as detailing some target specific installation instructions.
|
153 |
|
|
|
154 |
|
|
GCC includes several components that previously were separate distributions
|
155 |
|
|
with their own installation instructions. This document supersedes all
|
156 |
|
|
package specific installation instructions.
|
157 |
|
|
|
158 |
|
|
@emph{Before} starting the build/install procedure please check the
|
159 |
|
|
@ifnothtml
|
160 |
|
|
@ref{Specific, host/target specific installation notes}.
|
161 |
|
|
@end ifnothtml
|
162 |
|
|
@ifhtml
|
163 |
|
|
@uref{specific.html,,host/target specific installation notes}.
|
164 |
|
|
@end ifhtml
|
165 |
|
|
We recommend you browse the entire generic installation instructions before
|
166 |
|
|
you proceed.
|
167 |
|
|
|
168 |
|
|
Lists of successful builds for released versions of GCC are
|
169 |
|
|
available at @uref{http://gcc.gnu.org/buildstat.html}.
|
170 |
|
|
These lists are updated as new information becomes available.
|
171 |
|
|
|
172 |
|
|
The installation procedure itself is broken into five steps.
|
173 |
|
|
|
174 |
|
|
@ifinfo
|
175 |
|
|
@menu
|
176 |
|
|
* Prerequisites::
|
177 |
|
|
* Downloading the source::
|
178 |
|
|
* Configuration::
|
179 |
|
|
* Building::
|
180 |
|
|
* Testing:: (optional)
|
181 |
|
|
* Final install::
|
182 |
|
|
@end menu
|
183 |
|
|
@end ifinfo
|
184 |
|
|
@ifhtml
|
185 |
|
|
@enumerate
|
186 |
|
|
@item
|
187 |
|
|
@uref{prerequisites.html,,Prerequisites}
|
188 |
|
|
@item
|
189 |
|
|
@uref{download.html,,Downloading the source}
|
190 |
|
|
@item
|
191 |
|
|
@uref{configure.html,,Configuration}
|
192 |
|
|
@item
|
193 |
|
|
@uref{build.html,,Building}
|
194 |
|
|
@item
|
195 |
|
|
@uref{test.html,,Testing} (optional)
|
196 |
|
|
@item
|
197 |
|
|
@uref{finalinstall.html,,Final install}
|
198 |
|
|
@end enumerate
|
199 |
|
|
@end ifhtml
|
200 |
|
|
|
201 |
|
|
Please note that GCC does not support @samp{make uninstall} and probably
|
202 |
|
|
won't do so in the near future as this would open a can of worms. Instead,
|
203 |
|
|
we suggest that you install GCC into a directory of its own and simply
|
204 |
|
|
remove that directory when you do not need that specific version of GCC
|
205 |
|
|
any longer, and, if shared libraries are installed there as well, no
|
206 |
|
|
more binaries exist that use them.
|
207 |
|
|
|
208 |
|
|
@ifhtml
|
209 |
|
|
There are also some @uref{old.html,,old installation instructions},
|
210 |
|
|
which are mostly obsolete but still contain some information which has
|
211 |
|
|
not yet been merged into the main part of this manual.
|
212 |
|
|
@end ifhtml
|
213 |
|
|
|
214 |
|
|
@html
|
215 |
|
|
<hr />
|
216 |
|
|
<p>
|
217 |
|
|
@end html
|
218 |
|
|
@ifhtml
|
219 |
|
|
@uref{./index.html,,Return to the GCC Installation page}
|
220 |
|
|
|
221 |
|
|
@insertcopying
|
222 |
|
|
@end ifhtml
|
223 |
|
|
@end ifset
|
224 |
|
|
|
225 |
|
|
@c ***Prerequisites**************************************************
|
226 |
|
|
@ifnothtml
|
227 |
|
|
@comment node-name, next, previous, up
|
228 |
|
|
@node Prerequisites, Downloading the source, , Installing GCC
|
229 |
|
|
@end ifnothtml
|
230 |
|
|
@ifset prerequisiteshtml
|
231 |
|
|
@ifnothtml
|
232 |
|
|
@chapter Prerequisites
|
233 |
|
|
@end ifnothtml
|
234 |
|
|
@cindex Prerequisites
|
235 |
|
|
|
236 |
|
|
GCC requires that various tools and packages be available for use in the
|
237 |
|
|
build procedure. Modifying GCC sources requires additional tools
|
238 |
|
|
described below.
|
239 |
|
|
|
240 |
|
|
@heading Tools/packages necessary for building GCC
|
241 |
|
|
@table @asis
|
242 |
|
|
@item ISO C90 compiler
|
243 |
|
|
Necessary to bootstrap GCC, although versions of GCC prior
|
244 |
|
|
to 3.4 also allow bootstrapping with a traditional (K&R) C compiler.
|
245 |
|
|
|
246 |
|
|
To build all languages in a cross-compiler or other configuration where
|
247 |
|
|
3-stage bootstrap is not performed, you need to start with an existing
|
248 |
|
|
GCC binary (version 2.95 or later) because source code for language
|
249 |
|
|
frontends other than C might use GCC extensions.
|
250 |
|
|
|
251 |
|
|
@item GNAT
|
252 |
|
|
|
253 |
|
|
In order to build the Ada compiler (GNAT) you must already have GNAT
|
254 |
|
|
installed because portions of the Ada frontend are written in Ada (with
|
255 |
|
|
GNAT extensions.) Refer to the Ada installation instructions for more
|
256 |
|
|
specific information.
|
257 |
|
|
|
258 |
|
|
@item A ``working'' POSIX compatible shell, or GNU bash
|
259 |
|
|
|
260 |
|
|
Necessary when running @command{configure} because some
|
261 |
|
|
@command{/bin/sh} shells have bugs and may crash when configuring the
|
262 |
|
|
target libraries. In other cases, @command{/bin/sh} or @command{ksh}
|
263 |
|
|
have disastrous corner-case performance problems. This
|
264 |
|
|
can cause target @command{configure} runs to literally take days to
|
265 |
|
|
complete in some cases.
|
266 |
|
|
|
267 |
|
|
So on some platforms @command{/bin/ksh} is sufficient, on others it
|
268 |
|
|
isn't. See the host/target specific instructions for your platform, or
|
269 |
|
|
use @command{bash} to be sure. Then set @env{CONFIG_SHELL} in your
|
270 |
|
|
environment to your ``good'' shell prior to running
|
271 |
|
|
@command{configure}/@command{make}.
|
272 |
|
|
|
273 |
|
|
@command{zsh} is not a fully compliant POSIX shell and will not
|
274 |
|
|
work when configuring GCC@.
|
275 |
|
|
|
276 |
|
|
@item GNU binutils
|
277 |
|
|
|
278 |
|
|
Necessary in some circumstances, optional in others. See the
|
279 |
|
|
host/target specific instructions for your platform for the exact
|
280 |
|
|
requirements.
|
281 |
|
|
|
282 |
|
|
@item gzip version 1.2.4 (or later) or
|
283 |
|
|
@itemx bzip2 version 1.0.2 (or later)
|
284 |
|
|
|
285 |
|
|
Necessary to uncompress GCC @command{tar} files when source code is
|
286 |
|
|
obtained via FTP mirror sites.
|
287 |
|
|
|
288 |
|
|
@item GNU make version 3.79.1 (or later)
|
289 |
|
|
|
290 |
|
|
You must have GNU make installed to build GCC@.
|
291 |
|
|
|
292 |
|
|
@item GNU tar version 1.14 (or later)
|
293 |
|
|
|
294 |
|
|
Necessary (only on some platforms) to untar the source code. Many
|
295 |
|
|
systems' @command{tar} programs will also work, only try GNU
|
296 |
|
|
@command{tar} if you have problems.
|
297 |
|
|
|
298 |
|
|
@item GNU Multiple Precision Library (GMP) version 4.1 (or later)
|
299 |
|
|
|
300 |
|
|
Necessary to build the Fortran frontend. If you do not have it
|
301 |
|
|
installed in your library search path, you will have to configure with
|
302 |
|
|
the @option{--with-gmp} configure option. See also
|
303 |
|
|
@option{--with-gmp-lib} and @option{--with-gmp-include}.
|
304 |
|
|
|
305 |
|
|
@item MPFR Library version 2.2.1 (or later)
|
306 |
|
|
|
307 |
|
|
Necessary to build the Fortran frontend. It can be downloaded from
|
308 |
|
|
@uref{http://www.mpfr.org/}. The version of MPFR that is bundled with
|
309 |
|
|
GMP 4.1.x contains numerous bugs. Although GNU Fortran will appear
|
310 |
|
|
to function with the buggy versions of MPFR, there are a few GNU Fortran
|
311 |
|
|
bugs that will not be fixed when using this version. It is strongly
|
312 |
|
|
recommended to upgrade to the recommended version of MPFR.
|
313 |
|
|
|
314 |
|
|
The @option{--with-mpfr} configure option should be used if your MPFR
|
315 |
|
|
Library is not installed in your default library search path. See
|
316 |
|
|
also @option{--with-mpfr-lib} and @option{--with-mpfr-include}.
|
317 |
|
|
|
318 |
|
|
@item @command{jar}, or InfoZIP (@command{zip} and @command{unzip})
|
319 |
|
|
|
320 |
|
|
Necessary to build libgcj, the GCJ runtime.
|
321 |
|
|
|
322 |
|
|
@end table
|
323 |
|
|
|
324 |
|
|
|
325 |
|
|
@heading Tools/packages necessary for modifying GCC
|
326 |
|
|
@table @asis
|
327 |
|
|
@item autoconf versions 2.13 and 2.59
|
328 |
|
|
@itemx GNU m4 version 1.4 (or later)
|
329 |
|
|
|
330 |
|
|
Necessary when modifying @file{configure.ac}, @file{aclocal.m4}, etc.@:
|
331 |
|
|
to regenerate @file{configure} and @file{config.in} files. Most
|
332 |
|
|
directories require autoconf 2.59 (exactly), but the toplevel
|
333 |
|
|
still requires autoconf 2.13 (exactly).
|
334 |
|
|
|
335 |
|
|
@item automake version 1.9.6
|
336 |
|
|
|
337 |
|
|
Necessary when modifying a @file{Makefile.am} file to regenerate its
|
338 |
|
|
associated @file{Makefile.in}.
|
339 |
|
|
|
340 |
|
|
Much of GCC does not use automake, so directly edit the @file{Makefile.in}
|
341 |
|
|
file. Specifically this applies to the @file{gcc}, @file{intl},
|
342 |
|
|
@file{libcpp}, @file{libiberty}, @file{libobjc} directories as well
|
343 |
|
|
as any of their subdirectories.
|
344 |
|
|
|
345 |
|
|
For directories that use automake, GCC requires the latest release in
|
346 |
|
|
the 1.9.x series, which is currently 1.9.6. When regenerating a directory
|
347 |
|
|
to a newer version, please update all the directories using an older 1.9.x
|
348 |
|
|
to the latest released version.
|
349 |
|
|
|
350 |
|
|
@item gettext version 0.14.5 (or later)
|
351 |
|
|
|
352 |
|
|
Needed to regenerate @file{gcc.pot}.
|
353 |
|
|
|
354 |
|
|
@item gperf version 2.7.2 (or later)
|
355 |
|
|
|
356 |
|
|
Necessary when modifying @command{gperf} input files, e.g.@:
|
357 |
|
|
@file{gcc/cp/cfns.gperf} to regenerate its associated header file, e.g.@:
|
358 |
|
|
@file{gcc/cp/cfns.h}.
|
359 |
|
|
|
360 |
|
|
@item DejaGnu 1.4.4
|
361 |
|
|
@itemx Expect
|
362 |
|
|
@itemx Tcl
|
363 |
|
|
|
364 |
|
|
Necessary to run the GCC testsuite; see the section on testing for details.
|
365 |
|
|
|
366 |
|
|
@item autogen version 5.5.4 (or later) and
|
367 |
|
|
@itemx guile version 1.4.1 (or later)
|
368 |
|
|
|
369 |
|
|
Necessary to regenerate @file{fixinc/fixincl.x} from
|
370 |
|
|
@file{fixinc/inclhack.def} and @file{fixinc/*.tpl}.
|
371 |
|
|
|
372 |
|
|
Necessary to run @samp{make check} for @file{fixinc}.
|
373 |
|
|
|
374 |
|
|
Necessary to regenerate the top level @file{Makefile.in} file from
|
375 |
|
|
@file{Makefile.tpl} and @file{Makefile.def}.
|
376 |
|
|
|
377 |
|
|
@item GNU Bison version 1.28 (or later)
|
378 |
|
|
Berkeley @command{yacc} (@command{byacc}) is also reported to work other
|
379 |
|
|
than for GCJ.
|
380 |
|
|
|
381 |
|
|
Necessary when modifying @file{*.y} files.
|
382 |
|
|
|
383 |
|
|
Necessary to build GCC during development because the generated output
|
384 |
|
|
files are not included in the SVN repository. They are included in
|
385 |
|
|
releases.
|
386 |
|
|
|
387 |
|
|
@item Flex version 2.5.4 (or later)
|
388 |
|
|
|
389 |
|
|
Necessary when modifying @file{*.l} files.
|
390 |
|
|
|
391 |
|
|
Necessary to build GCC during development because the generated output
|
392 |
|
|
files are not included in the SVN repository. They are included in
|
393 |
|
|
releases.
|
394 |
|
|
|
395 |
|
|
@item Texinfo version 4.4 (or later)
|
396 |
|
|
|
397 |
|
|
Necessary for running @command{makeinfo} when modifying @file{*.texi}
|
398 |
|
|
files to test your changes.
|
399 |
|
|
|
400 |
|
|
Necessary for running @command{make dvi} or @command{make pdf} to
|
401 |
|
|
create printable documentation in DVI or PDF format. Texinfo version
|
402 |
|
|
4.8 or later is required for @command{make pdf}.
|
403 |
|
|
|
404 |
|
|
Necessary to build GCC documentation during development because the
|
405 |
|
|
generated output files are not included in the SVN repository. They are
|
406 |
|
|
included in releases.
|
407 |
|
|
|
408 |
|
|
@item @TeX{} (any working version)
|
409 |
|
|
|
410 |
|
|
Necessary for running @command{texi2dvi} and @command{texi2pdf}, which
|
411 |
|
|
are used when running @command{make dvi} or @command{make pdf} to create
|
412 |
|
|
DVI or PDF files, respectively.
|
413 |
|
|
|
414 |
|
|
@item SVN (any version)
|
415 |
|
|
@itemx SSH (any version)
|
416 |
|
|
|
417 |
|
|
Necessary to access the SVN repository. Public releases and weekly
|
418 |
|
|
snapshots of the development sources are also available via FTP@.
|
419 |
|
|
|
420 |
|
|
@item Perl version 5.6.1 (or later)
|
421 |
|
|
|
422 |
|
|
Necessary when regenerating @file{Makefile} dependencies in libiberty.
|
423 |
|
|
Necessary when regenerating @file{libiberty/functions.texi}.
|
424 |
|
|
Necessary when generating manpages from Texinfo manuals.
|
425 |
|
|
Necessary when targetting Darwin, building libstdc++,
|
426 |
|
|
and not using @option{--disable-symvers}.
|
427 |
|
|
Used by various scripts to generate some files included in SVN (mainly
|
428 |
|
|
Unicode-related and rarely changing) from source tables.
|
429 |
|
|
|
430 |
|
|
@item GNU diffutils version 2.7 (or later)
|
431 |
|
|
|
432 |
|
|
Useful when submitting patches for the GCC source code.
|
433 |
|
|
|
434 |
|
|
@item patch version 2.5.4 (or later)
|
435 |
|
|
|
436 |
|
|
Necessary when applying patches, created with @command{diff}, to one's
|
437 |
|
|
own sources.
|
438 |
|
|
|
439 |
|
|
@end table
|
440 |
|
|
|
441 |
|
|
@html
|
442 |
|
|
<hr />
|
443 |
|
|
<p>
|
444 |
|
|
@end html
|
445 |
|
|
@ifhtml
|
446 |
|
|
@uref{./index.html,,Return to the GCC Installation page}
|
447 |
|
|
@end ifhtml
|
448 |
|
|
@end ifset
|
449 |
|
|
|
450 |
|
|
@c ***Downloading the source**************************************************
|
451 |
|
|
@ifnothtml
|
452 |
|
|
@comment node-name, next, previous, up
|
453 |
|
|
@node Downloading the source, Configuration, Prerequisites, Installing GCC
|
454 |
|
|
@end ifnothtml
|
455 |
|
|
@ifset downloadhtml
|
456 |
|
|
@ifnothtml
|
457 |
|
|
@chapter Downloading GCC
|
458 |
|
|
@end ifnothtml
|
459 |
|
|
@cindex Downloading GCC
|
460 |
|
|
@cindex Downloading the Source
|
461 |
|
|
|
462 |
|
|
GCC is distributed via @uref{http://gcc.gnu.org/svn.html,,SVN} and FTP
|
463 |
|
|
tarballs compressed with @command{gzip} or
|
464 |
|
|
@command{bzip2}. It is possible to download a full distribution or specific
|
465 |
|
|
components.
|
466 |
|
|
|
467 |
|
|
Please refer to the @uref{http://gcc.gnu.org/releases.html,,releases web page}
|
468 |
|
|
for information on how to obtain GCC@.
|
469 |
|
|
|
470 |
|
|
The full distribution includes the C, C++, Objective-C, Fortran, Java,
|
471 |
|
|
and Ada (in the case of GCC 3.1 and later) compilers. The full
|
472 |
|
|
distribution also includes runtime libraries for C++, Objective-C,
|
473 |
|
|
Fortran, and Java. In GCC 3.0 and later versions, the GNU compiler
|
474 |
|
|
testsuites are also included in the full distribution.
|
475 |
|
|
|
476 |
|
|
If you choose to download specific components, you must download the core
|
477 |
|
|
GCC distribution plus any language specific distributions you wish to
|
478 |
|
|
use. The core distribution includes the C language front end as well as the
|
479 |
|
|
shared components. Each language has a tarball which includes the language
|
480 |
|
|
front end as well as the language runtime (when appropriate).
|
481 |
|
|
|
482 |
|
|
Unpack the core distribution as well as any language specific
|
483 |
|
|
distributions in the same directory.
|
484 |
|
|
|
485 |
|
|
If you also intend to build binutils (either to upgrade an existing
|
486 |
|
|
installation or for use in place of the corresponding tools of your
|
487 |
|
|
OS), unpack the binutils distribution either in the same directory or
|
488 |
|
|
a separate one. In the latter case, add symbolic links to any
|
489 |
|
|
components of the binutils you intend to build alongside the compiler
|
490 |
|
|
(@file{bfd}, @file{binutils}, @file{gas}, @file{gprof}, @file{ld},
|
491 |
|
|
@file{opcodes}, @dots{}) to the directory containing the GCC sources.
|
492 |
|
|
|
493 |
|
|
@html
|
494 |
|
|
<hr />
|
495 |
|
|
<p>
|
496 |
|
|
@end html
|
497 |
|
|
@ifhtml
|
498 |
|
|
@uref{./index.html,,Return to the GCC Installation page}
|
499 |
|
|
@end ifhtml
|
500 |
|
|
@end ifset
|
501 |
|
|
|
502 |
|
|
@c ***Configuration***********************************************************
|
503 |
|
|
@ifnothtml
|
504 |
|
|
@comment node-name, next, previous, up
|
505 |
|
|
@node Configuration, Building, Downloading the source, Installing GCC
|
506 |
|
|
@end ifnothtml
|
507 |
|
|
@ifset configurehtml
|
508 |
|
|
@ifnothtml
|
509 |
|
|
@chapter Installing GCC: Configuration
|
510 |
|
|
@end ifnothtml
|
511 |
|
|
@cindex Configuration
|
512 |
|
|
@cindex Installing GCC: Configuration
|
513 |
|
|
|
514 |
|
|
Like most GNU software, GCC must be configured before it can be built.
|
515 |
|
|
This document describes the recommended configuration procedure
|
516 |
|
|
for both native and cross targets.
|
517 |
|
|
|
518 |
|
|
We use @var{srcdir} to refer to the toplevel source directory for
|
519 |
|
|
GCC; we use @var{objdir} to refer to the toplevel build/object directory.
|
520 |
|
|
|
521 |
|
|
If you obtained the sources via SVN, @var{srcdir} must refer to the top
|
522 |
|
|
@file{gcc} directory, the one where the @file{MAINTAINERS} can be found,
|
523 |
|
|
and not its @file{gcc} subdirectory, otherwise the build will fail.
|
524 |
|
|
|
525 |
|
|
If either @var{srcdir} or @var{objdir} is located on an automounted NFS
|
526 |
|
|
file system, the shell's built-in @command{pwd} command will return
|
527 |
|
|
temporary pathnames. Using these can lead to various sorts of build
|
528 |
|
|
problems. To avoid this issue, set the @env{PWDCMD} environment
|
529 |
|
|
variable to an automounter-aware @command{pwd} command, e.g.,
|
530 |
|
|
@command{pawd} or @samp{amq -w}, during the configuration and build
|
531 |
|
|
phases.
|
532 |
|
|
|
533 |
|
|
First, we @strong{highly} recommend that GCC be built into a
|
534 |
|
|
separate directory than the sources which does @strong{not} reside
|
535 |
|
|
within the source tree. This is how we generally build GCC; building
|
536 |
|
|
where @var{srcdir} == @var{objdir} should still work, but doesn't
|
537 |
|
|
get extensive testing; building where @var{objdir} is a subdirectory
|
538 |
|
|
of @var{srcdir} is unsupported.
|
539 |
|
|
|
540 |
|
|
If you have previously built GCC in the same directory for a
|
541 |
|
|
different target machine, do @samp{make distclean} to delete all files
|
542 |
|
|
that might be invalid. One of the files this deletes is @file{Makefile};
|
543 |
|
|
if @samp{make distclean} complains that @file{Makefile} does not exist
|
544 |
|
|
or issues a message like ``don't know how to make distclean'' it probably
|
545 |
|
|
means that the directory is already suitably clean. However, with the
|
546 |
|
|
recommended method of building in a separate @var{objdir}, you should
|
547 |
|
|
simply use a different @var{objdir} for each target.
|
548 |
|
|
|
549 |
|
|
Second, when configuring a native system, either @command{cc} or
|
550 |
|
|
@command{gcc} must be in your path or you must set @env{CC} in
|
551 |
|
|
your environment before running configure. Otherwise the configuration
|
552 |
|
|
scripts may fail.
|
553 |
|
|
|
554 |
|
|
@ignore
|
555 |
|
|
Note that the bootstrap compiler and the resulting GCC must be link
|
556 |
|
|
compatible, else the bootstrap will fail with linker errors about
|
557 |
|
|
incompatible object file formats. Several multilibed targets are
|
558 |
|
|
affected by this requirement, see
|
559 |
|
|
@ifnothtml
|
560 |
|
|
@ref{Specific, host/target specific installation notes}.
|
561 |
|
|
@end ifnothtml
|
562 |
|
|
@ifhtml
|
563 |
|
|
@uref{specific.html,,host/target specific installation notes}.
|
564 |
|
|
@end ifhtml
|
565 |
|
|
@end ignore
|
566 |
|
|
|
567 |
|
|
To configure GCC:
|
568 |
|
|
|
569 |
|
|
@smallexample
|
570 |
|
|
% mkdir @var{objdir}
|
571 |
|
|
% cd @var{objdir}
|
572 |
|
|
% @var{srcdir}/configure [@var{options}] [@var{target}]
|
573 |
|
|
@end smallexample
|
574 |
|
|
|
575 |
|
|
|
576 |
|
|
@heading Target specification
|
577 |
|
|
@itemize @bullet
|
578 |
|
|
@item
|
579 |
|
|
GCC has code to correctly determine the correct value for @var{target}
|
580 |
|
|
for nearly all native systems. Therefore, we highly recommend you not
|
581 |
|
|
provide a configure target when configuring a native compiler.
|
582 |
|
|
|
583 |
|
|
@item
|
584 |
|
|
@var{target} must be specified as @option{--target=@var{target}}
|
585 |
|
|
when configuring a cross compiler; examples of valid targets would be
|
586 |
|
|
m68k-coff, sh-elf, etc.
|
587 |
|
|
|
588 |
|
|
@item
|
589 |
|
|
Specifying just @var{target} instead of @option{--target=@var{target}}
|
590 |
|
|
implies that the host defaults to @var{target}.
|
591 |
|
|
@end itemize
|
592 |
|
|
|
593 |
|
|
|
594 |
|
|
@heading Options specification
|
595 |
|
|
|
596 |
|
|
Use @var{options} to override several configure time options for
|
597 |
|
|
GCC@. A list of supported @var{options} follows; @samp{configure
|
598 |
|
|
--help} may list other options, but those not listed below may not
|
599 |
|
|
work and should not normally be used.
|
600 |
|
|
|
601 |
|
|
Note that each @option{--enable} option has a corresponding
|
602 |
|
|
@option{--disable} option and that each @option{--with} option has a
|
603 |
|
|
corresponding @option{--without} option.
|
604 |
|
|
|
605 |
|
|
@table @code
|
606 |
|
|
@item --prefix=@var{dirname}
|
607 |
|
|
Specify the toplevel installation
|
608 |
|
|
directory. This is the recommended way to install the tools into a directory
|
609 |
|
|
other than the default. The toplevel installation directory defaults to
|
610 |
|
|
@file{/usr/local}.
|
611 |
|
|
|
612 |
|
|
We @strong{highly} recommend against @var{dirname} being the same or a
|
613 |
|
|
subdirectory of @var{objdir} or vice versa. If specifying a directory
|
614 |
|
|
beneath a user's home directory tree, some shells will not expand
|
615 |
|
|
@var{dirname} correctly if it contains the @samp{~} metacharacter; use
|
616 |
|
|
@env{$HOME} instead.
|
617 |
|
|
|
618 |
|
|
The following standard @command{autoconf} options are supported. Normally you
|
619 |
|
|
should not need to use these options.
|
620 |
|
|
@table @code
|
621 |
|
|
@item --exec-prefix=@var{dirname}
|
622 |
|
|
Specify the toplevel installation directory for architecture-dependent
|
623 |
|
|
files. The default is @file{@var{prefix}}.
|
624 |
|
|
|
625 |
|
|
@item --bindir=@var{dirname}
|
626 |
|
|
Specify the installation directory for the executables called by users
|
627 |
|
|
(such as @command{gcc} and @command{g++}). The default is
|
628 |
|
|
@file{@var{exec-prefix}/bin}.
|
629 |
|
|
|
630 |
|
|
@item --libdir=@var{dirname}
|
631 |
|
|
Specify the installation directory for object code libraries and
|
632 |
|
|
internal data files of GCC@. The default is @file{@var{exec-prefix}/lib}.
|
633 |
|
|
|
634 |
|
|
@item --libexecdir=@var{dirname}
|
635 |
|
|
Specify the installation directory for internal executables of GCC@.
|
636 |
|
|
The default is @file{@var{exec-prefix}/libexec}.
|
637 |
|
|
|
638 |
|
|
@item --with-slibdir=@var{dirname}
|
639 |
|
|
Specify the installation directory for the shared libgcc library. The
|
640 |
|
|
default is @file{@var{libdir}}.
|
641 |
|
|
|
642 |
|
|
@item --infodir=@var{dirname}
|
643 |
|
|
Specify the installation directory for documentation in info format.
|
644 |
|
|
The default is @file{@var{prefix}/info}.
|
645 |
|
|
|
646 |
|
|
@item --datadir=@var{dirname}
|
647 |
|
|
Specify the installation directory for some architecture-independent
|
648 |
|
|
data files referenced by GCC@. The default is @file{@var{prefix}/share}.
|
649 |
|
|
|
650 |
|
|
@item --mandir=@var{dirname}
|
651 |
|
|
Specify the installation directory for manual pages. The default is
|
652 |
|
|
@file{@var{prefix}/man}. (Note that the manual pages are only extracts from
|
653 |
|
|
the full GCC manuals, which are provided in Texinfo format. The manpages
|
654 |
|
|
are derived by an automatic conversion process from parts of the full
|
655 |
|
|
manual.)
|
656 |
|
|
|
657 |
|
|
@item --with-gxx-include-dir=@var{dirname}
|
658 |
|
|
Specify
|
659 |
|
|
the installation directory for G++ header files. The default is
|
660 |
|
|
@file{@var{prefix}/include/c++/@var{version}}.
|
661 |
|
|
|
662 |
|
|
@end table
|
663 |
|
|
|
664 |
|
|
@item --program-prefix=@var{prefix}
|
665 |
|
|
GCC supports some transformations of the names of its programs when
|
666 |
|
|
installing them. This option prepends @var{prefix} to the names of
|
667 |
|
|
programs to install in @var{bindir} (see above). For example, specifying
|
668 |
|
|
@option{--program-prefix=foo-} would result in @samp{gcc}
|
669 |
|
|
being installed as @file{/usr/local/bin/foo-gcc}.
|
670 |
|
|
|
671 |
|
|
@item --program-suffix=@var{suffix}
|
672 |
|
|
Appends @var{suffix} to the names of programs to install in @var{bindir}
|
673 |
|
|
(see above). For example, specifying @option{--program-suffix=-3.1}
|
674 |
|
|
would result in @samp{gcc} being installed as
|
675 |
|
|
@file{/usr/local/bin/gcc-3.1}.
|
676 |
|
|
|
677 |
|
|
@item --program-transform-name=@var{pattern}
|
678 |
|
|
Applies the @samp{sed} script @var{pattern} to be applied to the names
|
679 |
|
|
of programs to install in @var{bindir} (see above). @var{pattern} has to
|
680 |
|
|
consist of one or more basic @samp{sed} editing commands, separated by
|
681 |
|
|
semicolons. For example, if you want the @samp{gcc} program name to be
|
682 |
|
|
transformed to the installed program @file{/usr/local/bin/myowngcc} and
|
683 |
|
|
the @samp{g++} program name to be transformed to
|
684 |
|
|
@file{/usr/local/bin/gspecial++} without changing other program names,
|
685 |
|
|
you could use the pattern
|
686 |
|
|
@option{--program-transform-name='s/^gcc$/myowngcc/; s/^g++$/gspecial++/'}
|
687 |
|
|
to achieve this effect.
|
688 |
|
|
|
689 |
|
|
All three options can be combined and used together, resulting in more
|
690 |
|
|
complex conversion patterns. As a basic rule, @var{prefix} (and
|
691 |
|
|
@var{suffix}) are prepended (appended) before further transformations
|
692 |
|
|
can happen with a special transformation script @var{pattern}.
|
693 |
|
|
|
694 |
|
|
As currently implemented, this option only takes effect for native
|
695 |
|
|
builds; cross compiler binaries' names are not transformed even when a
|
696 |
|
|
transformation is explicitly asked for by one of these options.
|
697 |
|
|
|
698 |
|
|
For native builds, some of the installed programs are also installed
|
699 |
|
|
with the target alias in front of their name, as in
|
700 |
|
|
@samp{i686-pc-linux-gnu-gcc}. All of the above transformations happen
|
701 |
|
|
before the target alias is prepended to the name---so, specifying
|
702 |
|
|
@option{--program-prefix=foo-} and @option{program-suffix=-3.1}, the
|
703 |
|
|
resulting binary would be installed as
|
704 |
|
|
@file{/usr/local/bin/i686-pc-linux-gnu-foo-gcc-3.1}.
|
705 |
|
|
|
706 |
|
|
As a last shortcoming, none of the installed Ada programs are
|
707 |
|
|
transformed yet, which will be fixed in some time.
|
708 |
|
|
|
709 |
|
|
@item --with-local-prefix=@var{dirname}
|
710 |
|
|
Specify the
|
711 |
|
|
installation directory for local include files. The default is
|
712 |
|
|
@file{/usr/local}. Specify this option if you want the compiler to
|
713 |
|
|
search directory @file{@var{dirname}/include} for locally installed
|
714 |
|
|
header files @emph{instead} of @file{/usr/local/include}.
|
715 |
|
|
|
716 |
|
|
You should specify @option{--with-local-prefix} @strong{only} if your
|
717 |
|
|
site has a different convention (not @file{/usr/local}) for where to put
|
718 |
|
|
site-specific files.
|
719 |
|
|
|
720 |
|
|
The default value for @option{--with-local-prefix} is @file{/usr/local}
|
721 |
|
|
regardless of the value of @option{--prefix}. Specifying
|
722 |
|
|
@option{--prefix} has no effect on which directory GCC searches for
|
723 |
|
|
local header files. This may seem counterintuitive, but actually it is
|
724 |
|
|
logical.
|
725 |
|
|
|
726 |
|
|
The purpose of @option{--prefix} is to specify where to @emph{install
|
727 |
|
|
GCC}. The local header files in @file{/usr/local/include}---if you put
|
728 |
|
|
any in that directory---are not part of GCC@. They are part of other
|
729 |
|
|
programs---perhaps many others. (GCC installs its own header files in
|
730 |
|
|
another directory which is based on the @option{--prefix} value.)
|
731 |
|
|
|
732 |
|
|
Both the local-prefix include directory and the GCC-prefix include
|
733 |
|
|
directory are part of GCC's ``system include'' directories. Although these
|
734 |
|
|
two directories are not fixed, they need to be searched in the proper
|
735 |
|
|
order for the correct processing of the include_next directive. The
|
736 |
|
|
local-prefix include directory is searched before the GCC-prefix
|
737 |
|
|
include directory. Another characteristic of system include directories
|
738 |
|
|
is that pedantic warnings are turned off for headers in these directories.
|
739 |
|
|
|
740 |
|
|
Some autoconf macros add @option{-I @var{directory}} options to the
|
741 |
|
|
compiler command line, to ensure that directories containing installed
|
742 |
|
|
packages' headers are searched. When @var{directory} is one of GCC's
|
743 |
|
|
system include directories, GCC will ignore the option so that system
|
744 |
|
|
directories continue to be processed in the correct order. This
|
745 |
|
|
may result in a search order different from what was specified but the
|
746 |
|
|
directory will still be searched.
|
747 |
|
|
|
748 |
|
|
GCC automatically searches for ordinary libraries using
|
749 |
|
|
@env{GCC_EXEC_PREFIX}. Thus, when the same installation prefix is
|
750 |
|
|
used for both GCC and packages, GCC will automatically search for
|
751 |
|
|
both headers and libraries. This provides a configuration that is
|
752 |
|
|
easy to use. GCC behaves in a manner similar to that when it is
|
753 |
|
|
installed as a system compiler in @file{/usr}.
|
754 |
|
|
|
755 |
|
|
Sites that need to install multiple versions of GCC may not want to
|
756 |
|
|
use the above simple configuration. It is possible to use the
|
757 |
|
|
@option{--program-prefix}, @option{--program-suffix} and
|
758 |
|
|
@option{--program-transform-name} options to install multiple versions
|
759 |
|
|
into a single directory, but it may be simpler to use different prefixes
|
760 |
|
|
and the @option{--with-local-prefix} option to specify the location of the
|
761 |
|
|
site-specific files for each version. It will then be necessary for
|
762 |
|
|
users to specify explicitly the location of local site libraries
|
763 |
|
|
(e.g., with @env{LIBRARY_PATH}).
|
764 |
|
|
|
765 |
|
|
The same value can be used for both @option{--with-local-prefix} and
|
766 |
|
|
@option{--prefix} provided it is not @file{/usr}. This can be used
|
767 |
|
|
to avoid the default search of @file{/usr/local/include}.
|
768 |
|
|
|
769 |
|
|
@strong{Do not} specify @file{/usr} as the @option{--with-local-prefix}!
|
770 |
|
|
The directory you use for @option{--with-local-prefix} @strong{must not}
|
771 |
|
|
contain any of the system's standard header files. If it did contain
|
772 |
|
|
them, certain programs would be miscompiled (including GNU Emacs, on
|
773 |
|
|
certain targets), because this would override and nullify the header
|
774 |
|
|
file corrections made by the @command{fixincludes} script.
|
775 |
|
|
|
776 |
|
|
Indications are that people who use this option use it based on mistaken
|
777 |
|
|
ideas of what it is for. People use it as if it specified where to
|
778 |
|
|
install part of GCC@. Perhaps they make this assumption because
|
779 |
|
|
installing GCC creates the directory.
|
780 |
|
|
|
781 |
|
|
@item --enable-shared[=@var{package}[,@dots{}]]
|
782 |
|
|
Build shared versions of libraries, if shared libraries are supported on
|
783 |
|
|
the target platform. Unlike GCC 2.95.x and earlier, shared libraries
|
784 |
|
|
are enabled by default on all platforms that support shared libraries.
|
785 |
|
|
|
786 |
|
|
If a list of packages is given as an argument, build shared libraries
|
787 |
|
|
only for the listed packages. For other packages, only static libraries
|
788 |
|
|
will be built. Package names currently recognized in the GCC tree are
|
789 |
|
|
@samp{libgcc} (also known as @samp{gcc}), @samp{libstdc++} (not
|
790 |
|
|
@samp{libstdc++-v3}), @samp{libffi}, @samp{zlib}, @samp{boehm-gc},
|
791 |
|
|
@samp{ada}, @samp{libada}, @samp{libjava} and @samp{libobjc}.
|
792 |
|
|
Note @samp{libiberty} does not support shared libraries at all.
|
793 |
|
|
|
794 |
|
|
Use @option{--disable-shared} to build only static libraries. Note that
|
795 |
|
|
@option{--disable-shared} does not accept a list of package names as
|
796 |
|
|
argument, only @option{--enable-shared} does.
|
797 |
|
|
|
798 |
|
|
@item @anchor{with-gnu-as}--with-gnu-as
|
799 |
|
|
Specify that the compiler should assume that the
|
800 |
|
|
assembler it finds is the GNU assembler. However, this does not modify
|
801 |
|
|
the rules to find an assembler and will result in confusion if the
|
802 |
|
|
assembler found is not actually the GNU assembler. (Confusion may also
|
803 |
|
|
result if the compiler finds the GNU assembler but has not been
|
804 |
|
|
configured with @option{--with-gnu-as}.) If you have more than one
|
805 |
|
|
assembler installed on your system, you may want to use this option in
|
806 |
|
|
connection with @option{--with-as=@var{pathname}} or
|
807 |
|
|
@option{--with-build-time-tools=@var{pathname}}.
|
808 |
|
|
|
809 |
|
|
The following systems are the only ones where it makes a difference
|
810 |
|
|
whether you use the GNU assembler. On any other system,
|
811 |
|
|
@option{--with-gnu-as} has no effect.
|
812 |
|
|
|
813 |
|
|
@itemize @bullet
|
814 |
|
|
@item @samp{hppa1.0-@var{any}-@var{any}}
|
815 |
|
|
@item @samp{hppa1.1-@var{any}-@var{any}}
|
816 |
|
|
@item @samp{i386-@var{any}-sysv}
|
817 |
|
|
@item @samp{m68k-bull-sysv}
|
818 |
|
|
@item @samp{m68k-hp-hpux}
|
819 |
|
|
@item @samp{m68000-hp-hpux}
|
820 |
|
|
@item @samp{m68000-att-sysv}
|
821 |
|
|
@item @samp{sparc-sun-solaris2.@var{any}}
|
822 |
|
|
@item @samp{sparc64-@var{any}-solaris2.@var{any}}
|
823 |
|
|
@end itemize
|
824 |
|
|
|
825 |
|
|
On the systems listed above (except for the HP-PA, the SPARC, for ISC on
|
826 |
|
|
the 386, if you use the GNU assembler, you should also use the GNU linker
|
827 |
|
|
(and specify @option{--with-gnu-ld}).
|
828 |
|
|
|
829 |
|
|
@item @anchor{with-as}--with-as=@var{pathname}
|
830 |
|
|
Specify that the compiler should use the assembler pointed to by
|
831 |
|
|
@var{pathname}, rather than the one found by the standard rules to find
|
832 |
|
|
an assembler, which are:
|
833 |
|
|
@itemize @bullet
|
834 |
|
|
@item
|
835 |
|
|
Unless GCC is being built with a cross compiler, check the
|
836 |
|
|
@file{@var{libexec}/gcc/@var{target}/@var{version}} directory.
|
837 |
|
|
@var{libexec} defaults to @file{@var{exec-prefix}/libexec};
|
838 |
|
|
@var{exec-prefix} defaults to @var{prefix}, which
|
839 |
|
|
defaults to @file{/usr/local} unless overridden by the
|
840 |
|
|
@option{--prefix=@var{pathname}} switch described above. @var{target}
|
841 |
|
|
is the target system triple, such as @samp{sparc-sun-solaris2.7}, and
|
842 |
|
|
@var{version} denotes the GCC version, such as 3.0.
|
843 |
|
|
|
844 |
|
|
@item
|
845 |
|
|
If the target system is the same that you are building on, check
|
846 |
|
|
operating system specific directories (e.g.@: @file{/usr/ccs/bin} on
|
847 |
|
|
Sun Solaris 2).
|
848 |
|
|
|
849 |
|
|
@item
|
850 |
|
|
Check in the @env{PATH} for a tool whose name is prefixed by the
|
851 |
|
|
target system triple.
|
852 |
|
|
|
853 |
|
|
@item
|
854 |
|
|
Check in the @env{PATH} for a tool whose name is not prefixed by the
|
855 |
|
|
target system triple, if the host and target system triple are
|
856 |
|
|
the same (in other words, we use a host tool if it can be used for
|
857 |
|
|
the target as well).
|
858 |
|
|
@end itemize
|
859 |
|
|
|
860 |
|
|
You may want to use @option{--with-as} if no assembler
|
861 |
|
|
is installed in the directories listed above, or if you have multiple
|
862 |
|
|
assemblers installed and want to choose one that is not found by the
|
863 |
|
|
above rules.
|
864 |
|
|
|
865 |
|
|
@item @anchor{with-gnu-ld}--with-gnu-ld
|
866 |
|
|
Same as @uref{#with-gnu-as,,@option{--with-gnu-as}}
|
867 |
|
|
but for the linker.
|
868 |
|
|
|
869 |
|
|
@item --with-ld=@var{pathname}
|
870 |
|
|
Same as @uref{#with-as,,@option{--with-as}}
|
871 |
|
|
but for the linker.
|
872 |
|
|
|
873 |
|
|
@item --with-stabs
|
874 |
|
|
Specify that stabs debugging
|
875 |
|
|
information should be used instead of whatever format the host normally
|
876 |
|
|
uses. Normally GCC uses the same debug format as the host system.
|
877 |
|
|
|
878 |
|
|
On MIPS based systems and on Alphas, you must specify whether you want
|
879 |
|
|
GCC to create the normal ECOFF debugging format, or to use BSD-style
|
880 |
|
|
stabs passed through the ECOFF symbol table. The normal ECOFF debug
|
881 |
|
|
format cannot fully handle languages other than C@. BSD stabs format can
|
882 |
|
|
handle other languages, but it only works with the GNU debugger GDB@.
|
883 |
|
|
|
884 |
|
|
Normally, GCC uses the ECOFF debugging format by default; if you
|
885 |
|
|
prefer BSD stabs, specify @option{--with-stabs} when you configure GCC@.
|
886 |
|
|
|
887 |
|
|
No matter which default you choose when you configure GCC, the user
|
888 |
|
|
can use the @option{-gcoff} and @option{-gstabs+} options to specify explicitly
|
889 |
|
|
the debug format for a particular compilation.
|
890 |
|
|
|
891 |
|
|
@option{--with-stabs} is meaningful on the ISC system on the 386, also, if
|
892 |
|
|
@option{--with-gas} is used. It selects use of stabs debugging
|
893 |
|
|
information embedded in COFF output. This kind of debugging information
|
894 |
|
|
supports C++ well; ordinary COFF debugging information does not.
|
895 |
|
|
|
896 |
|
|
@option{--with-stabs} is also meaningful on 386 systems running SVR4. It
|
897 |
|
|
selects use of stabs debugging information embedded in ELF output. The
|
898 |
|
|
C++ compiler currently (2.6.0) does not support the DWARF debugging
|
899 |
|
|
information normally used on 386 SVR4 platforms; stabs provide a
|
900 |
|
|
workable alternative. This requires gas and gdb, as the normal SVR4
|
901 |
|
|
tools can not generate or interpret stabs.
|
902 |
|
|
|
903 |
|
|
@item --disable-multilib
|
904 |
|
|
Specify that multiple target
|
905 |
|
|
libraries to support different target variants, calling
|
906 |
|
|
conventions, etc.@: should not be built. The default is to build a
|
907 |
|
|
predefined set of them.
|
908 |
|
|
|
909 |
|
|
Some targets provide finer-grained control over which multilibs are built
|
910 |
|
|
(e.g., @option{--disable-softfloat}):
|
911 |
|
|
@table @code
|
912 |
|
|
@item arc-*-elf*
|
913 |
|
|
biendian.
|
914 |
|
|
|
915 |
|
|
@item arm-*-*
|
916 |
|
|
fpu, 26bit, underscore, interwork, biendian, nofmult.
|
917 |
|
|
|
918 |
|
|
@item m68*-*-*
|
919 |
|
|
softfloat, m68881, m68000, m68020.
|
920 |
|
|
|
921 |
|
|
@item mips*-*-*
|
922 |
|
|
single-float, biendian, softfloat.
|
923 |
|
|
|
924 |
|
|
@item powerpc*-*-*, rs6000*-*-*
|
925 |
|
|
aix64, pthread, softfloat, powercpu, powerpccpu, powerpcos, biendian,
|
926 |
|
|
sysv, aix.
|
927 |
|
|
|
928 |
|
|
@end table
|
929 |
|
|
|
930 |
|
|
@item --enable-threads
|
931 |
|
|
Specify that the target
|
932 |
|
|
supports threads. This affects the Objective-C compiler and runtime
|
933 |
|
|
library, and exception handling for other languages like C++ and Java.
|
934 |
|
|
On some systems, this is the default.
|
935 |
|
|
|
936 |
|
|
In general, the best (and, in many cases, the only known) threading
|
937 |
|
|
model available will be configured for use. Beware that on some
|
938 |
|
|
systems, GCC has not been taught what threading models are generally
|
939 |
|
|
available for the system. In this case, @option{--enable-threads} is an
|
940 |
|
|
alias for @option{--enable-threads=single}.
|
941 |
|
|
|
942 |
|
|
@item --disable-threads
|
943 |
|
|
Specify that threading support should be disabled for the system.
|
944 |
|
|
This is an alias for @option{--enable-threads=single}.
|
945 |
|
|
|
946 |
|
|
@item --enable-threads=@var{lib}
|
947 |
|
|
Specify that
|
948 |
|
|
@var{lib} is the thread support library. This affects the Objective-C
|
949 |
|
|
compiler and runtime library, and exception handling for other languages
|
950 |
|
|
like C++ and Java. The possibilities for @var{lib} are:
|
951 |
|
|
|
952 |
|
|
@table @code
|
953 |
|
|
@item aix
|
954 |
|
|
AIX thread support.
|
955 |
|
|
@item dce
|
956 |
|
|
DCE thread support.
|
957 |
|
|
@item gnat
|
958 |
|
|
Ada tasking support. For non-Ada programs, this setting is equivalent
|
959 |
|
|
to @samp{single}. When used in conjunction with the Ada run time, it
|
960 |
|
|
causes GCC to use the same thread primitives as Ada uses. This option
|
961 |
|
|
is necessary when using both Ada and the back end exception handling,
|
962 |
|
|
which is the default for most Ada targets.
|
963 |
|
|
@item mach
|
964 |
|
|
Generic MACH thread support, known to work on NeXTSTEP@. (Please note
|
965 |
|
|
that the file needed to support this configuration, @file{gthr-mach.h}, is
|
966 |
|
|
missing and thus this setting will cause a known bootstrap failure.)
|
967 |
|
|
@item no
|
968 |
|
|
This is an alias for @samp{single}.
|
969 |
|
|
@item posix
|
970 |
|
|
Generic POSIX/Unix98 thread support.
|
971 |
|
|
@item posix95
|
972 |
|
|
Generic POSIX/Unix95 thread support.
|
973 |
|
|
@item rtems
|
974 |
|
|
RTEMS thread support.
|
975 |
|
|
@item single
|
976 |
|
|
Disable thread support, should work for all platforms.
|
977 |
|
|
@item solaris
|
978 |
|
|
Sun Solaris 2 thread support.
|
979 |
|
|
@item vxworks
|
980 |
|
|
VxWorks thread support.
|
981 |
|
|
@item win32
|
982 |
|
|
Microsoft Win32 API thread support.
|
983 |
|
|
@item nks
|
984 |
|
|
Novell Kernel Services thread support.
|
985 |
|
|
@end table
|
986 |
|
|
|
987 |
|
|
@item --enable-tls
|
988 |
|
|
Specify that the target supports TLS (Thread Local Storage). Usually
|
989 |
|
|
configure can correctly determine if TLS is supported. In cases where
|
990 |
|
|
it guesses incorrectly, TLS can be explicitly enabled or disabled with
|
991 |
|
|
@option{--enable-tls} or @option{--disable-tls}. This can happen if
|
992 |
|
|
the assembler supports TLS but the C library does not, or if the
|
993 |
|
|
assumptions made by the configure test are incorrect.
|
994 |
|
|
|
995 |
|
|
@item --disable-tls
|
996 |
|
|
Specify that the target does not support TLS.
|
997 |
|
|
This is an alias for @option{--enable-tls=no}.
|
998 |
|
|
|
999 |
|
|
@item --with-cpu=@var{cpu}
|
1000 |
|
|
Specify which cpu variant the compiler should generate code for by default.
|
1001 |
|
|
@var{cpu} will be used as the default value of the @option{-mcpu=} switch.
|
1002 |
|
|
This option is only supported on some targets, including ARM, i386, PowerPC,
|
1003 |
|
|
and SPARC@.
|
1004 |
|
|
|
1005 |
|
|
@item --with-schedule=@var{cpu}
|
1006 |
|
|
@itemx --with-arch=@var{cpu}
|
1007 |
|
|
@itemx --with-tune=@var{cpu}
|
1008 |
|
|
@itemx --with-abi=@var{abi}
|
1009 |
|
|
@itemx --with-fpu=@var{type}
|
1010 |
|
|
@itemx --with-float=@var{type}
|
1011 |
|
|
These configure options provide default values for the @option{-mschedule=},
|
1012 |
|
|
@option{-march=}, @option{-mtune=}, @option{-mabi=}, and @option{-mfpu=}
|
1013 |
|
|
options and for @option{-mhard-float} or @option{-msoft-float}. As with
|
1014 |
|
|
@option{--with-cpu}, which switches will be accepted and acceptable values
|
1015 |
|
|
of the arguments depend on the target.
|
1016 |
|
|
|
1017 |
|
|
@item --with-mode=@var{mode}
|
1018 |
|
|
Specify if the compiler should default to @option{-marm} or @option{-mthumb}.
|
1019 |
|
|
This option is only supported on ARM targets.
|
1020 |
|
|
|
1021 |
|
|
@item --with-divide=@var{type}
|
1022 |
|
|
Specify how the compiler should generate code for checking for
|
1023 |
|
|
division by zero. This option is only supported on the MIPS target.
|
1024 |
|
|
The possibilities for @var{type} are:
|
1025 |
|
|
@table @code
|
1026 |
|
|
@item traps
|
1027 |
|
|
Division by zero checks use conditional traps (this is the default on
|
1028 |
|
|
systems that support conditional traps).
|
1029 |
|
|
@item breaks
|
1030 |
|
|
Division by zero checks use the break instruction.
|
1031 |
|
|
@end table
|
1032 |
|
|
|
1033 |
|
|
@item --enable-__cxa_atexit
|
1034 |
|
|
Define if you want to use __cxa_atexit, rather than atexit, to
|
1035 |
|
|
register C++ destructors for local statics and global objects.
|
1036 |
|
|
This is essential for fully standards-compliant handling of
|
1037 |
|
|
destructors, but requires __cxa_atexit in libc. This option is currently
|
1038 |
|
|
only available on systems with GNU libc. When enabled, this will cause
|
1039 |
|
|
@option{-fuse-cxa-exit} to be passed by default.
|
1040 |
|
|
|
1041 |
|
|
@item --enable-target-optspace
|
1042 |
|
|
Specify that target
|
1043 |
|
|
libraries should be optimized for code space instead of code speed.
|
1044 |
|
|
This is the default for the m32r platform.
|
1045 |
|
|
|
1046 |
|
|
@item --disable-cpp
|
1047 |
|
|
Specify that a user visible @command{cpp} program should not be installed.
|
1048 |
|
|
|
1049 |
|
|
@item --with-cpp-install-dir=@var{dirname}
|
1050 |
|
|
Specify that the user visible @command{cpp} program should be installed
|
1051 |
|
|
in @file{@var{prefix}/@var{dirname}/cpp}, in addition to @var{bindir}.
|
1052 |
|
|
|
1053 |
|
|
@item --enable-initfini-array
|
1054 |
|
|
Force the use of sections @code{.init_array} and @code{.fini_array}
|
1055 |
|
|
(instead of @code{.init} and @code{.fini}) for constructors and
|
1056 |
|
|
destructors. Option @option{--disable-initfini-array} has the
|
1057 |
|
|
opposite effect. If neither option is specified, the configure script
|
1058 |
|
|
will try to guess whether the @code{.init_array} and
|
1059 |
|
|
@code{.fini_array} sections are supported and, if they are, use them.
|
1060 |
|
|
|
1061 |
|
|
@item --enable-maintainer-mode
|
1062 |
|
|
The build rules that
|
1063 |
|
|
regenerate the GCC master message catalog @file{gcc.pot} are normally
|
1064 |
|
|
disabled. This is because it can only be rebuilt if the complete source
|
1065 |
|
|
tree is present. If you have changed the sources and want to rebuild the
|
1066 |
|
|
catalog, configuring with @option{--enable-maintainer-mode} will enable
|
1067 |
|
|
this. Note that you need a recent version of the @code{gettext} tools
|
1068 |
|
|
to do so.
|
1069 |
|
|
|
1070 |
|
|
@item --disable-bootstrap
|
1071 |
|
|
For a native build, the default configuration is to perform
|
1072 |
|
|
a 3-stage bootstrap of the compiler when @samp{make} is invoked,
|
1073 |
|
|
testing that GCC can compile itself correctly. If you want to disable
|
1074 |
|
|
this process, you can configure with @option{--disable-bootstrap}.
|
1075 |
|
|
|
1076 |
|
|
@item --enable-bootstrap
|
1077 |
|
|
In special cases, you may want to perform a 3-stage build
|
1078 |
|
|
even if the target and host triplets are different.
|
1079 |
|
|
This could happen when the host can run code compiled for
|
1080 |
|
|
the target (e.g.@: host is i686-linux, target is i486-linux).
|
1081 |
|
|
Starting from GCC 4.2, to do this you have to configure explicitly
|
1082 |
|
|
with @option{--enable-bootstrap}.
|
1083 |
|
|
|
1084 |
|
|
@item --enable-generated-files-in-srcdir
|
1085 |
|
|
Neither the .c and .h files that are generated from Bison and flex nor the
|
1086 |
|
|
info manuals and man pages that are built from the .texi files are present
|
1087 |
|
|
in the SVN development tree. When building GCC from that development tree,
|
1088 |
|
|
or from one of our snapshots, those generated files are placed in your
|
1089 |
|
|
build directory, which allows for the source to be in a readonly
|
1090 |
|
|
directory.
|
1091 |
|
|
|
1092 |
|
|
If you configure with @option{--enable-generated-files-in-srcdir} then those
|
1093 |
|
|
generated files will go into the source directory. This is mainly intended
|
1094 |
|
|
for generating release or prerelease tarballs of the GCC sources, since it
|
1095 |
|
|
is not a requirement that the users of source releases to have flex, Bison,
|
1096 |
|
|
or makeinfo.
|
1097 |
|
|
|
1098 |
|
|
@item --enable-version-specific-runtime-libs
|
1099 |
|
|
Specify
|
1100 |
|
|
that runtime libraries should be installed in the compiler specific
|
1101 |
|
|
subdirectory (@file{@var{libdir}/gcc}) rather than the usual places. In
|
1102 |
|
|
addition, @samp{libstdc++}'s include files will be installed into
|
1103 |
|
|
@file{@var{libdir}} unless you overruled it by using
|
1104 |
|
|
@option{--with-gxx-include-dir=@var{dirname}}. Using this option is
|
1105 |
|
|
particularly useful if you intend to use several versions of GCC in
|
1106 |
|
|
parallel. This is currently supported by @samp{libgfortran},
|
1107 |
|
|
@samp{libjava}, @samp{libmudflap}, @samp{libstdc++}, and @samp{libobjc}.
|
1108 |
|
|
|
1109 |
|
|
@item --with-java-home=@var{dirname}
|
1110 |
|
|
This @samp{libjava} option overrides the default value of the
|
1111 |
|
|
@samp{java.home} system property. It is also used to set
|
1112 |
|
|
@samp{sun.boot.class.path} to @file{@var{dirname}/lib/rt.jar}. By
|
1113 |
|
|
default @samp{java.home} is set to @file{@var{prefix}} and
|
1114 |
|
|
@samp{sun.boot.class.path} to
|
1115 |
|
|
@file{@var{datadir}/java/libgcj-@var{version}.jar}.
|
1116 |
|
|
|
1117 |
|
|
@item --enable-languages=@var{lang1},@var{lang2},@dots{}
|
1118 |
|
|
Specify that only a particular subset of compilers and
|
1119 |
|
|
their runtime libraries should be built. For a list of valid values for
|
1120 |
|
|
@var{langN} you can issue the following command in the
|
1121 |
|
|
@file{gcc} directory of your GCC source tree:@*
|
1122 |
|
|
@smallexample
|
1123 |
|
|
grep language= */config-lang.in
|
1124 |
|
|
@end smallexample
|
1125 |
|
|
Currently, you can use any of the following:
|
1126 |
|
|
@code{all}, @code{ada}, @code{c}, @code{c++}, @code{fortran}, @code{java},
|
1127 |
|
|
@code{objc}, @code{obj-c++}, @code{treelang}.
|
1128 |
|
|
Building the Ada compiler has special requirements, see below.
|
1129 |
|
|
If you do not pass this flag, or specify the option @code{all}, then all
|
1130 |
|
|
default languages available in the @file{gcc} sub-tree will be configured.
|
1131 |
|
|
Ada, Objective-C++, and treelang are not default languages; the rest are.
|
1132 |
|
|
Re-defining @code{LANGUAGES} when calling @samp{make} @strong{does not}
|
1133 |
|
|
work anymore, as those language sub-directories might not have been
|
1134 |
|
|
configured!
|
1135 |
|
|
|
1136 |
|
|
@item --disable-libada
|
1137 |
|
|
Specify that the run-time libraries and tools used by GNAT should not
|
1138 |
|
|
be built. This can be useful for debugging, or for compatibility with
|
1139 |
|
|
previous Ada build procedures, when it was required to explicitly
|
1140 |
|
|
do a @samp{make -C gcc gnatlib_and_tools}.
|
1141 |
|
|
|
1142 |
|
|
@item --disable-libssp
|
1143 |
|
|
Specify that the run-time libraries for stack smashing protection
|
1144 |
|
|
should not be built.
|
1145 |
|
|
|
1146 |
|
|
@item --disable-libgomp
|
1147 |
|
|
Specify that the run-time libraries used by GOMP should not be built.
|
1148 |
|
|
|
1149 |
|
|
@item --with-dwarf2
|
1150 |
|
|
Specify that the compiler should
|
1151 |
|
|
use DWARF 2 debugging information as the default.
|
1152 |
|
|
|
1153 |
|
|
@item --enable-targets=all
|
1154 |
|
|
@itemx --enable-targets=@var{target_list}
|
1155 |
|
|
Some GCC targets, e.g.@: powerpc64-linux, build bi-arch compilers.
|
1156 |
|
|
These are compilers that are able to generate either 64-bit or 32-bit
|
1157 |
|
|
code. Typically, the corresponding 32-bit target, e.g.@:
|
1158 |
|
|
powerpc-linux for powerpc64-linux, only generates 32-bit code. This
|
1159 |
|
|
option enables the 32-bit target to be a bi-arch compiler, which is
|
1160 |
|
|
useful when you want a bi-arch compiler that defaults to 32-bit, and
|
1161 |
|
|
you are building a bi-arch or multi-arch binutils in a combined tree.
|
1162 |
|
|
Currently, this option only affects powerpc-linux.
|
1163 |
|
|
|
1164 |
|
|
@item --enable-secureplt
|
1165 |
|
|
This option enables @option{-msecure-plt} by default for powerpc-linux.
|
1166 |
|
|
@ifnothtml
|
1167 |
|
|
@xref{RS/6000 and PowerPC Options,, RS/6000 and PowerPC Options, gcc,
|
1168 |
|
|
Using the GNU Compiler Collection (GCC)},
|
1169 |
|
|
@end ifnothtml
|
1170 |
|
|
@ifhtml
|
1171 |
|
|
See ``RS/6000 and PowerPC Options'' in the main manual
|
1172 |
|
|
@end ifhtml
|
1173 |
|
|
|
1174 |
|
|
@item --enable-win32-registry
|
1175 |
|
|
@itemx --enable-win32-registry=@var{key}
|
1176 |
|
|
@itemx --disable-win32-registry
|
1177 |
|
|
The @option{--enable-win32-registry} option enables Microsoft Windows-hosted GCC
|
1178 |
|
|
to look up installations paths in the registry using the following key:
|
1179 |
|
|
|
1180 |
|
|
@smallexample
|
1181 |
|
|
@code{HKEY_LOCAL_MACHINE\SOFTWARE\Free Software Foundation\@var{key}}
|
1182 |
|
|
@end smallexample
|
1183 |
|
|
|
1184 |
|
|
@var{key} defaults to GCC version number, and can be overridden by the
|
1185 |
|
|
@option{--enable-win32-registry=@var{key}} option. Vendors and distributors
|
1186 |
|
|
who use custom installers are encouraged to provide a different key,
|
1187 |
|
|
perhaps one comprised of vendor name and GCC version number, to
|
1188 |
|
|
avoid conflict with existing installations. This feature is enabled
|
1189 |
|
|
by default, and can be disabled by @option{--disable-win32-registry}
|
1190 |
|
|
option. This option has no effect on the other hosts.
|
1191 |
|
|
|
1192 |
|
|
@item --nfp
|
1193 |
|
|
Specify that the machine does not have a floating point unit. This
|
1194 |
|
|
option only applies to @samp{m68k-sun-sunos@var{n}}. On any other
|
1195 |
|
|
system, @option{--nfp} has no effect.
|
1196 |
|
|
|
1197 |
|
|
@item --enable-werror
|
1198 |
|
|
@itemx --disable-werror
|
1199 |
|
|
@itemx --enable-werror=yes
|
1200 |
|
|
@itemx --enable-werror=no
|
1201 |
|
|
When you specify this option, it controls whether certain files in the
|
1202 |
|
|
compiler are built with @option{-Werror} in bootstrap stage2 and later.
|
1203 |
|
|
If you don't specify it, @option{-Werror} is turned on for the main
|
1204 |
|
|
development trunk. However it defaults to off for release branches and
|
1205 |
|
|
final releases. The specific files which get @option{-Werror} are
|
1206 |
|
|
controlled by the Makefiles.
|
1207 |
|
|
|
1208 |
|
|
@item --enable-checking
|
1209 |
|
|
@itemx --enable-checking=@var{list}
|
1210 |
|
|
When you specify this option, the compiler is built to perform internal
|
1211 |
|
|
consistency checks of the requested complexity. This does not change the
|
1212 |
|
|
generated code, but adds error checking within the compiler. This will
|
1213 |
|
|
slow down the compiler and may only work properly if you are building
|
1214 |
|
|
the compiler with GCC@. This is @samp{yes} by default when building
|
1215 |
|
|
from SVN or snapshots, but @samp{release} for releases. More control
|
1216 |
|
|
over the checks may be had by specifying @var{list}. The categories of
|
1217 |
|
|
checks available are @samp{yes} (most common checks
|
1218 |
|
|
@samp{assert,misc,tree,gc,rtlflag,runtime}), @samp{no} (no checks at
|
1219 |
|
|
all), @samp{all} (all but @samp{valgrind}), @samp{release} (cheapest
|
1220 |
|
|
checks @samp{assert,runtime}) or @samp{none} (same as @samp{no}).
|
1221 |
|
|
Individual checks can be enabled with these flags @samp{assert},
|
1222 |
|
|
@samp{fold}, @samp{gc}, @samp{gcac} @samp{misc}, @samp{rtl},
|
1223 |
|
|
@samp{rtlflag}, @samp{runtime}, @samp{tree}, and @samp{valgrind}.
|
1224 |
|
|
|
1225 |
|
|
The @samp{valgrind} check requires the external @command{valgrind}
|
1226 |
|
|
simulator, available from @uref{http://valgrind.org/}. The
|
1227 |
|
|
@samp{rtl}, @samp{gcac} and @samp{valgrind} checks are very expensive.
|
1228 |
|
|
To disable all checking, @samp{--disable-checking} or
|
1229 |
|
|
@samp{--enable-checking=none} must be explicitly requested. Disabling
|
1230 |
|
|
assertions will make the compiler and runtime slightly faster but
|
1231 |
|
|
increase the risk of undetected internal errors causing wrong code to be
|
1232 |
|
|
generated.
|
1233 |
|
|
|
1234 |
|
|
@item --enable-coverage
|
1235 |
|
|
@itemx --enable-coverage=@var{level}
|
1236 |
|
|
With this option, the compiler is built to collect self coverage
|
1237 |
|
|
information, every time it is run. This is for internal development
|
1238 |
|
|
purposes, and only works when the compiler is being built with gcc. The
|
1239 |
|
|
@var{level} argument controls whether the compiler is built optimized or
|
1240 |
|
|
not, values are @samp{opt} and @samp{noopt}. For coverage analysis you
|
1241 |
|
|
want to disable optimization, for performance analysis you want to
|
1242 |
|
|
enable optimization. When coverage is enabled, the default level is
|
1243 |
|
|
without optimization.
|
1244 |
|
|
|
1245 |
|
|
@item --enable-gather-detailed-mem-stats
|
1246 |
|
|
When this option is specified more detailed information on memory
|
1247 |
|
|
allocation is gathered. This information is printed when using
|
1248 |
|
|
@option{-fmem-report}.
|
1249 |
|
|
|
1250 |
|
|
@item --with-gc
|
1251 |
|
|
@itemx --with-gc=@var{choice}
|
1252 |
|
|
With this option you can specify the garbage collector implementation
|
1253 |
|
|
used during the compilation process. @var{choice} can be one of
|
1254 |
|
|
@samp{page} and @samp{zone}, where @samp{page} is the default.
|
1255 |
|
|
|
1256 |
|
|
@item --enable-nls
|
1257 |
|
|
@itemx --disable-nls
|
1258 |
|
|
The @option{--enable-nls} option enables Native Language Support (NLS),
|
1259 |
|
|
which lets GCC output diagnostics in languages other than American
|
1260 |
|
|
English. Native Language Support is enabled by default if not doing a
|
1261 |
|
|
canadian cross build. The @option{--disable-nls} option disables NLS@.
|
1262 |
|
|
|
1263 |
|
|
@item --with-included-gettext
|
1264 |
|
|
If NLS is enabled, the @option{--with-included-gettext} option causes the build
|
1265 |
|
|
procedure to prefer its copy of GNU @command{gettext}.
|
1266 |
|
|
|
1267 |
|
|
@item --with-catgets
|
1268 |
|
|
If NLS is enabled, and if the host lacks @code{gettext} but has the
|
1269 |
|
|
inferior @code{catgets} interface, the GCC build procedure normally
|
1270 |
|
|
ignores @code{catgets} and instead uses GCC's copy of the GNU
|
1271 |
|
|
@code{gettext} library. The @option{--with-catgets} option causes the
|
1272 |
|
|
build procedure to use the host's @code{catgets} in this situation.
|
1273 |
|
|
|
1274 |
|
|
@item --with-libiconv-prefix=@var{dir}
|
1275 |
|
|
Search for libiconv header files in @file{@var{dir}/include} and
|
1276 |
|
|
libiconv library files in @file{@var{dir}/lib}.
|
1277 |
|
|
|
1278 |
|
|
@item --enable-obsolete
|
1279 |
|
|
Enable configuration for an obsoleted system. If you attempt to
|
1280 |
|
|
configure GCC for a system (build, host, or target) which has been
|
1281 |
|
|
obsoleted, and you do not specify this flag, configure will halt with an
|
1282 |
|
|
error message.
|
1283 |
|
|
|
1284 |
|
|
All support for systems which have been obsoleted in one release of GCC
|
1285 |
|
|
is removed entirely in the next major release, unless someone steps
|
1286 |
|
|
forward to maintain the port.
|
1287 |
|
|
|
1288 |
|
|
@item --enable-decimal-float
|
1289 |
|
|
@itemx --disable-decimal-float
|
1290 |
|
|
Enable (or disable) support for the C decimal floating point
|
1291 |
|
|
extension. This is enabled by default only on PowerPC GNU/Linux
|
1292 |
|
|
systems. Other systems may also support it, but require the user to
|
1293 |
|
|
specifically enable it.
|
1294 |
|
|
|
1295 |
|
|
@item --with-long-double-128
|
1296 |
|
|
Specify if @code{long double} type should be 128-bit by default on selected
|
1297 |
|
|
GNU/Linux architectures. If using @code{--without-long-double-128},
|
1298 |
|
|
@code{long double} will be by default 64-bit, the same as @code{double} type.
|
1299 |
|
|
When neither of these configure options are used, the default will be
|
1300 |
|
|
128-bit @code{long double} when built against GNU C Library 2.4 and later,
|
1301 |
|
|
64-bit @code{long double} otherwise.
|
1302 |
|
|
|
1303 |
|
|
@end table
|
1304 |
|
|
|
1305 |
|
|
@subheading Cross-Compiler-Specific Options
|
1306 |
|
|
The following options only apply to building cross compilers.
|
1307 |
|
|
@table @code
|
1308 |
|
|
@item --with-sysroot
|
1309 |
|
|
@itemx --with-sysroot=@var{dir}
|
1310 |
|
|
Tells GCC to consider @var{dir} as the root of a tree that contains a
|
1311 |
|
|
(subset of) the root filesystem of the target operating system.
|
1312 |
|
|
Target system headers, libraries and run-time object files will be
|
1313 |
|
|
searched in there. The specified directory is not copied into the
|
1314 |
|
|
install tree, unlike the options @option{--with-headers} and
|
1315 |
|
|
@option{--with-libs} that this option obsoletes. The default value,
|
1316 |
|
|
in case @option{--with-sysroot} is not given an argument, is
|
1317 |
|
|
@option{$@{gcc_tooldir@}/sys-root}. If the specified directory is a
|
1318 |
|
|
subdirectory of @option{$@{exec_prefix@}}, then it will be found relative to
|
1319 |
|
|
the GCC binaries if the installation tree is moved.
|
1320 |
|
|
|
1321 |
|
|
@item --with-build-sysroot
|
1322 |
|
|
@itemx --with-build-sysroot=@var{dir}
|
1323 |
|
|
Tells GCC to consider @var{dir} as the system root (see
|
1324 |
|
|
@option{--with-sysroot}) while building target libraries, instead of
|
1325 |
|
|
the directory specified with @option{--with-sysroot}. This option is
|
1326 |
|
|
only useful when you are already using @option{--with-sysroot}. You
|
1327 |
|
|
can use @option{--with-build-sysroot} when you are configuring with
|
1328 |
|
|
@option{--prefix} set to a directory that is different from the one in
|
1329 |
|
|
which you are installing GCC and your target libraries.
|
1330 |
|
|
|
1331 |
|
|
This option affects the system root for the compiler used to build
|
1332 |
|
|
target libraries (which runs on the build system); it does not affect
|
1333 |
|
|
the compiler which is used to build GCC itself.
|
1334 |
|
|
|
1335 |
|
|
@item --with-headers
|
1336 |
|
|
@itemx --with-headers=@var{dir}
|
1337 |
|
|
Deprecated in favor of @option{--with-sysroot}.
|
1338 |
|
|
Specifies that target headers are available when building a cross compiler.
|
1339 |
|
|
The @var{dir} argument specifies a directory which has the target include
|
1340 |
|
|
files. These include files will be copied into the @file{gcc} install
|
1341 |
|
|
directory. @emph{This option with the @var{dir} argument is required} when
|
1342 |
|
|
building a cross compiler, if @file{@var{prefix}/@var{target}/sys-include}
|
1343 |
|
|
doesn't pre-exist. If @file{@var{prefix}/@var{target}/sys-include} does
|
1344 |
|
|
pre-exist, the @var{dir} argument may be omitted. @command{fixincludes}
|
1345 |
|
|
will be run on these files to make them compatible with GCC@.
|
1346 |
|
|
|
1347 |
|
|
@item --without-headers
|
1348 |
|
|
Tells GCC not use any target headers from a libc when building a cross
|
1349 |
|
|
compiler. When crossing to GNU/Linux, you need the headers so GCC
|
1350 |
|
|
can build the exception handling for libgcc.
|
1351 |
|
|
|
1352 |
|
|
@item --with-libs
|
1353 |
|
|
@itemx --with-libs=``@var{dir1} @var{dir2} @dots{} @var{dirN}''
|
1354 |
|
|
Deprecated in favor of @option{--with-sysroot}.
|
1355 |
|
|
Specifies a list of directories which contain the target runtime
|
1356 |
|
|
libraries. These libraries will be copied into the @file{gcc} install
|
1357 |
|
|
directory. If the directory list is omitted, this option has no
|
1358 |
|
|
effect.
|
1359 |
|
|
|
1360 |
|
|
@item --with-newlib
|
1361 |
|
|
Specifies that @samp{newlib} is
|
1362 |
|
|
being used as the target C library. This causes @code{__eprintf} to be
|
1363 |
|
|
omitted from @file{libgcc.a} on the assumption that it will be provided by
|
1364 |
|
|
@samp{newlib}.
|
1365 |
|
|
|
1366 |
|
|
@item --with-build-time-tools=@var{dir}
|
1367 |
|
|
Specifies where to find the set of target tools (assembler, linker, etc.)
|
1368 |
|
|
that will be used while building GCC itself. This option can be useful
|
1369 |
|
|
if the directory layouts are different between the system you are building
|
1370 |
|
|
GCC on, and the system where you will deploy it.
|
1371 |
|
|
|
1372 |
|
|
For example, on a @option{ia64-hp-hpux} system, you may have the GNU
|
1373 |
|
|
assembler and linker in @file{/usr/bin}, and the native tools in a
|
1374 |
|
|
different path, and build a toolchain that expects to find the
|
1375 |
|
|
native tools in @file{/usr/bin}.
|
1376 |
|
|
|
1377 |
|
|
When you use this option, you should ensure that @var{dir} includes
|
1378 |
|
|
@command{ar}, @command{as}, @command{ld}, @command{nm},
|
1379 |
|
|
@command{ranlib} and @command{strip} if necessary, and possibly
|
1380 |
|
|
@command{objdump}. Otherwise, GCC may use an inconsistent set of
|
1381 |
|
|
tools.
|
1382 |
|
|
@end table
|
1383 |
|
|
|
1384 |
|
|
@subheading Fortran-Specific Options
|
1385 |
|
|
|
1386 |
|
|
The following options apply to the build of the Fortran front end.
|
1387 |
|
|
|
1388 |
|
|
@table @code
|
1389 |
|
|
|
1390 |
|
|
@item --with-gmp=@var{pathname}
|
1391 |
|
|
@itemx --with-gmp-include=@var{pathname}
|
1392 |
|
|
@itemx --with-gmp-lib=@var{pathname}
|
1393 |
|
|
@itemx --with-mpfr=@var{pathname}
|
1394 |
|
|
@itemx --with-mpfr-include=@var{pathname}
|
1395 |
|
|
@itemx --with-mpfr-lib=@var{pathname}
|
1396 |
|
|
If you do not have GMP (the GNU Multiple Precision library) and the
|
1397 |
|
|
MPFR Libraries installed in a standard location and you want to build
|
1398 |
|
|
the Fortran front-end, you can explicitly specify the directory where
|
1399 |
|
|
they are installed (@samp{--with-gmp=@var{gmpinstalldir}},
|
1400 |
|
|
@samp{--with-mpfr=@var{mpfrinstalldir}}). The
|
1401 |
|
|
@option{--with-gmp=@var{gmpinstalldir}} option is shorthand for
|
1402 |
|
|
@option{--with-gmp-lib=@var{gmpinstalldir}/lib} and
|
1403 |
|
|
@option{--with-gmp-include=@var{gmpinstalldir}/include}. Likewise the
|
1404 |
|
|
@option{--with-mpfr=@var{mpfrinstalldir}} option is shorthand for
|
1405 |
|
|
@option{--with-mpfr-lib=@var{mpfrinstalldir}/lib} and
|
1406 |
|
|
@option{--with-mpfr-include=@var{mpfrinstalldir}/include}. If these
|
1407 |
|
|
shorthand assumptions are not correct, you can use the explicit
|
1408 |
|
|
include and lib options directly.
|
1409 |
|
|
|
1410 |
|
|
@end table
|
1411 |
|
|
|
1412 |
|
|
@subheading Java-Specific Options
|
1413 |
|
|
|
1414 |
|
|
The following option applies to the build of the Java front end.
|
1415 |
|
|
|
1416 |
|
|
@table @code
|
1417 |
|
|
@item --disable-libgcj
|
1418 |
|
|
Specify that the run-time libraries
|
1419 |
|
|
used by GCJ should not be built. This is useful in case you intend
|
1420 |
|
|
to use GCJ with some other run-time, or you're going to install it
|
1421 |
|
|
separately, or it just happens not to build on your particular
|
1422 |
|
|
machine. In general, if the Java front end is enabled, the GCJ
|
1423 |
|
|
libraries will be enabled too, unless they're known to not work on
|
1424 |
|
|
the target platform. If GCJ is enabled but @samp{libgcj} isn't built, you
|
1425 |
|
|
may need to port it; in this case, before modifying the top-level
|
1426 |
|
|
@file{configure.in} so that @samp{libgcj} is enabled by default on this platform,
|
1427 |
|
|
you may use @option{--enable-libgcj} to override the default.
|
1428 |
|
|
|
1429 |
|
|
@end table
|
1430 |
|
|
|
1431 |
|
|
The following options apply to building @samp{libgcj}.
|
1432 |
|
|
|
1433 |
|
|
@subsubheading General Options
|
1434 |
|
|
|
1435 |
|
|
@table @code
|
1436 |
|
|
@item --disable-getenv-properties
|
1437 |
|
|
Don't set system properties from @env{GCJ_PROPERTIES}.
|
1438 |
|
|
|
1439 |
|
|
@item --enable-hash-synchronization
|
1440 |
|
|
Use a global hash table for monitor locks. Ordinarily,
|
1441 |
|
|
@samp{libgcj}'s @samp{configure} script automatically makes
|
1442 |
|
|
the correct choice for this option for your platform. Only use
|
1443 |
|
|
this if you know you need the library to be configured differently.
|
1444 |
|
|
|
1445 |
|
|
@item --enable-interpreter
|
1446 |
|
|
Enable the Java interpreter. The interpreter is automatically
|
1447 |
|
|
enabled by default on all platforms that support it. This option
|
1448 |
|
|
is really only useful if you want to disable the interpreter
|
1449 |
|
|
(using @option{--disable-interpreter}).
|
1450 |
|
|
|
1451 |
|
|
@item --disable-java-net
|
1452 |
|
|
Disable java.net. This disables the native part of java.net only,
|
1453 |
|
|
using non-functional stubs for native method implementations.
|
1454 |
|
|
|
1455 |
|
|
@item --disable-jvmpi
|
1456 |
|
|
Disable JVMPI support.
|
1457 |
|
|
|
1458 |
|
|
@item --with-ecos
|
1459 |
|
|
Enable runtime eCos target support.
|
1460 |
|
|
|
1461 |
|
|
@item --without-libffi
|
1462 |
|
|
Don't use @samp{libffi}. This will disable the interpreter and JNI
|
1463 |
|
|
support as well, as these require @samp{libffi} to work.
|
1464 |
|
|
|
1465 |
|
|
@item --enable-libgcj-debug
|
1466 |
|
|
Enable runtime debugging code.
|
1467 |
|
|
|
1468 |
|
|
@item --enable-libgcj-multifile
|
1469 |
|
|
If specified, causes all @file{.java} source files to be
|
1470 |
|
|
compiled into @file{.class} files in one invocation of
|
1471 |
|
|
@samp{gcj}. This can speed up build time, but is more
|
1472 |
|
|
resource-intensive. If this option is unspecified or
|
1473 |
|
|
disabled, @samp{gcj} is invoked once for each @file{.java}
|
1474 |
|
|
file to compile into a @file{.class} file.
|
1475 |
|
|
|
1476 |
|
|
@item --with-libiconv-prefix=DIR
|
1477 |
|
|
Search for libiconv in @file{DIR/include} and @file{DIR/lib}.
|
1478 |
|
|
|
1479 |
|
|
@item --enable-sjlj-exceptions
|
1480 |
|
|
Force use of the @code{setjmp}/@code{longjmp}-based scheme for exceptions.
|
1481 |
|
|
@samp{configure} ordinarily picks the correct value based on the platform.
|
1482 |
|
|
Only use this option if you are sure you need a different setting.
|
1483 |
|
|
|
1484 |
|
|
@item --with-system-zlib
|
1485 |
|
|
Use installed @samp{zlib} rather than that included with GCC@.
|
1486 |
|
|
|
1487 |
|
|
@item --with-win32-nlsapi=ansi, unicows or unicode
|
1488 |
|
|
Indicates how MinGW @samp{libgcj} translates between UNICODE
|
1489 |
|
|
characters and the Win32 API@.
|
1490 |
|
|
@table @code
|
1491 |
|
|
@item ansi
|
1492 |
|
|
Use the single-byte @code{char} and the Win32 A functions natively,
|
1493 |
|
|
translating to and from UNICODE when using these functions. If
|
1494 |
|
|
unspecified, this is the default.
|
1495 |
|
|
|
1496 |
|
|
@item unicows
|
1497 |
|
|
Use the @code{WCHAR} and Win32 W functions natively. Adds
|
1498 |
|
|
@code{-lunicows} to @file{libgcj.spec} to link with @samp{libunicows}.
|
1499 |
|
|
@file{unicows.dll} needs to be deployed on Microsoft Windows 9X machines
|
1500 |
|
|
running built executables. @file{libunicows.a}, an open-source
|
1501 |
|
|
import library around Microsoft's @code{unicows.dll}, is obtained from
|
1502 |
|
|
@uref{http://libunicows.sourceforge.net/}, which also gives details
|
1503 |
|
|
on getting @file{unicows.dll} from Microsoft.
|
1504 |
|
|
|
1505 |
|
|
@item unicode
|
1506 |
|
|
Use the @code{WCHAR} and Win32 W functions natively. Does @emph{not}
|
1507 |
|
|
add @code{-lunicows} to @file{libgcj.spec}. The built executables will
|
1508 |
|
|
only run on Microsoft Windows NT and above.
|
1509 |
|
|
@end table
|
1510 |
|
|
@end table
|
1511 |
|
|
|
1512 |
|
|
@subsubheading AWT-Specific Options
|
1513 |
|
|
|
1514 |
|
|
@table @code
|
1515 |
|
|
@item --with-x
|
1516 |
|
|
Use the X Window System.
|
1517 |
|
|
|
1518 |
|
|
@item --enable-java-awt=PEER(S)
|
1519 |
|
|
Specifies the AWT peer library or libraries to build alongside
|
1520 |
|
|
@samp{libgcj}. If this option is unspecified or disabled, AWT
|
1521 |
|
|
will be non-functional. Current valid values are @option{gtk} and
|
1522 |
|
|
@option{xlib}. Multiple libraries should be separated by a
|
1523 |
|
|
comma (i.e.@: @option{--enable-java-awt=gtk,xlib}).
|
1524 |
|
|
|
1525 |
|
|
@item --enable-gtk-cairo
|
1526 |
|
|
Build the cairo Graphics2D implementation on GTK@.
|
1527 |
|
|
|
1528 |
|
|
@item --enable-java-gc=TYPE
|
1529 |
|
|
Choose garbage collector. Defaults to @option{boehm} if unspecified.
|
1530 |
|
|
|
1531 |
|
|
@item --disable-gtktest
|
1532 |
|
|
Do not try to compile and run a test GTK+ program.
|
1533 |
|
|
|
1534 |
|
|
@item --disable-glibtest
|
1535 |
|
|
Do not try to compile and run a test GLIB program.
|
1536 |
|
|
|
1537 |
|
|
@item --with-libart-prefix=PFX
|
1538 |
|
|
Prefix where libart is installed (optional).
|
1539 |
|
|
|
1540 |
|
|
@item --with-libart-exec-prefix=PFX
|
1541 |
|
|
Exec prefix where libart is installed (optional).
|
1542 |
|
|
|
1543 |
|
|
@item --disable-libarttest
|
1544 |
|
|
Do not try to compile and run a test libart program.
|
1545 |
|
|
|
1546 |
|
|
@end table
|
1547 |
|
|
|
1548 |
|
|
@html
|
1549 |
|
|
<hr />
|
1550 |
|
|
<p>
|
1551 |
|
|
@end html
|
1552 |
|
|
@ifhtml
|
1553 |
|
|
@uref{./index.html,,Return to the GCC Installation page}
|
1554 |
|
|
@end ifhtml
|
1555 |
|
|
@end ifset
|
1556 |
|
|
|
1557 |
|
|
@c ***Building****************************************************************
|
1558 |
|
|
@ifnothtml
|
1559 |
|
|
@comment node-name, next, previous, up
|
1560 |
|
|
@node Building, Testing, Configuration, Installing GCC
|
1561 |
|
|
@end ifnothtml
|
1562 |
|
|
@ifset buildhtml
|
1563 |
|
|
@ifnothtml
|
1564 |
|
|
@chapter Building
|
1565 |
|
|
@end ifnothtml
|
1566 |
|
|
@cindex Installing GCC: Building
|
1567 |
|
|
|
1568 |
|
|
Now that GCC is configured, you are ready to build the compiler and
|
1569 |
|
|
runtime libraries.
|
1570 |
|
|
|
1571 |
|
|
Some commands executed when making the compiler may fail (return a
|
1572 |
|
|
nonzero status) and be ignored by @command{make}. These failures, which
|
1573 |
|
|
are often due to files that were not found, are expected, and can safely
|
1574 |
|
|
be ignored.
|
1575 |
|
|
|
1576 |
|
|
It is normal to have compiler warnings when compiling certain files.
|
1577 |
|
|
Unless you are a GCC developer, you can generally ignore these warnings
|
1578 |
|
|
unless they cause compilation to fail. Developers should attempt to fix
|
1579 |
|
|
any warnings encountered, however they can temporarily continue past
|
1580 |
|
|
warnings-as-errors by specifying the configure flag
|
1581 |
|
|
@option{--disable-werror}.
|
1582 |
|
|
|
1583 |
|
|
On certain old systems, defining certain environment variables such as
|
1584 |
|
|
@env{CC} can interfere with the functioning of @command{make}.
|
1585 |
|
|
|
1586 |
|
|
If you encounter seemingly strange errors when trying to build the
|
1587 |
|
|
compiler in a directory other than the source directory, it could be
|
1588 |
|
|
because you have previously configured the compiler in the source
|
1589 |
|
|
directory. Make sure you have done all the necessary preparations.
|
1590 |
|
|
|
1591 |
|
|
If you build GCC on a BSD system using a directory stored in an old System
|
1592 |
|
|
V file system, problems may occur in running @command{fixincludes} if the
|
1593 |
|
|
System V file system doesn't support symbolic links. These problems
|
1594 |
|
|
result in a failure to fix the declaration of @code{size_t} in
|
1595 |
|
|
@file{sys/types.h}. If you find that @code{size_t} is a signed type and
|
1596 |
|
|
that type mismatches occur, this could be the cause.
|
1597 |
|
|
|
1598 |
|
|
The solution is not to use such a directory for building GCC@.
|
1599 |
|
|
|
1600 |
|
|
When building from SVN or snapshots, or if you modify parser sources,
|
1601 |
|
|
you need the Bison parser generator installed. If you do not modify
|
1602 |
|
|
parser sources, releases contain the Bison-generated files and you do
|
1603 |
|
|
not need Bison installed to build them.
|
1604 |
|
|
|
1605 |
|
|
When building from SVN or snapshots, or if you modify Texinfo
|
1606 |
|
|
documentation, you need version 4.4 or later of Texinfo installed if you
|
1607 |
|
|
want Info documentation to be regenerated. Releases contain Info
|
1608 |
|
|
documentation pre-built for the unmodified documentation in the release.
|
1609 |
|
|
|
1610 |
|
|
@section Building a native compiler
|
1611 |
|
|
|
1612 |
|
|
For a native build, the default configuration is to perform
|
1613 |
|
|
a 3-stage bootstrap of the compiler when @samp{make} is invoked.
|
1614 |
|
|
This will build the entire GCC system and ensure that it compiles
|
1615 |
|
|
itself correctly. It can be disabled with the @option{--disable-bootstrap}
|
1616 |
|
|
parameter to @samp{configure}, but bootstrapping is suggested because
|
1617 |
|
|
the compiler will be tested more completely and could also have
|
1618 |
|
|
better performance.
|
1619 |
|
|
|
1620 |
|
|
The bootstrapping process will complete the following steps:
|
1621 |
|
|
|
1622 |
|
|
@itemize @bullet
|
1623 |
|
|
@item
|
1624 |
|
|
Build tools necessary to build the compiler.
|
1625 |
|
|
|
1626 |
|
|
@item
|
1627 |
|
|
Perform a 3-stage bootstrap of the compiler. This includes building
|
1628 |
|
|
three times the target tools for use by the compiler such as binutils
|
1629 |
|
|
(bfd, binutils, gas, gprof, ld, and opcodes) if they have been
|
1630 |
|
|
individually linked or moved into the top level GCC source tree before
|
1631 |
|
|
configuring.
|
1632 |
|
|
|
1633 |
|
|
@item
|
1634 |
|
|
Perform a comparison test of the stage2 and stage3 compilers.
|
1635 |
|
|
|
1636 |
|
|
@item
|
1637 |
|
|
Build runtime libraries using the stage3 compiler from the previous step.
|
1638 |
|
|
|
1639 |
|
|
@end itemize
|
1640 |
|
|
|
1641 |
|
|
If you are short on disk space you might consider @samp{make
|
1642 |
|
|
bootstrap-lean} instead. The sequence of compilation is the
|
1643 |
|
|
same described above, but object files from the stage1 and
|
1644 |
|
|
stage2 of the 3-stage bootstrap of the compiler are deleted as
|
1645 |
|
|
soon as they are no longer needed.
|
1646 |
|
|
|
1647 |
|
|
If you want to save additional space during the bootstrap and in
|
1648 |
|
|
the final installation as well, you can build the compiler binaries
|
1649 |
|
|
without debugging information as in the following example. This will save
|
1650 |
|
|
roughly 40% of disk space both for the bootstrap and the final installation.
|
1651 |
|
|
(Libraries will still contain debugging information.)
|
1652 |
|
|
|
1653 |
|
|
@smallexample
|
1654 |
|
|
make CFLAGS='-O' LIBCFLAGS='-g -O2' \
|
1655 |
|
|
LIBCXXFLAGS='-g -O2 -fno-implicit-templates' bootstrap
|
1656 |
|
|
@end smallexample
|
1657 |
|
|
|
1658 |
|
|
If you wish to use non-default GCC flags when compiling the stage2 and
|
1659 |
|
|
stage3 compilers, set @code{BOOT_CFLAGS} on the command line when doing
|
1660 |
|
|
@samp{make}. Non-default optimization flags are less well
|
1661 |
|
|
tested here than the default of @samp{-g -O2}, but should still work.
|
1662 |
|
|
In a few cases, you may find that you need to specify special flags such
|
1663 |
|
|
as @option{-msoft-float} here to complete the bootstrap; or, if the
|
1664 |
|
|
native compiler miscompiles the stage1 compiler, you may need to work
|
1665 |
|
|
around this, by choosing @code{BOOT_CFLAGS} to avoid the parts of the
|
1666 |
|
|
stage1 compiler that were miscompiled, or by using @samp{make
|
1667 |
|
|
bootstrap4} to increase the number of stages of bootstrap.
|
1668 |
|
|
|
1669 |
|
|
Note that using non-standard @code{CFLAGS} can cause bootstrap to fail
|
1670 |
|
|
if these trigger a warning with the new compiler. For example using
|
1671 |
|
|
@samp{-O2 -g -mcpu=i686} on @code{i686-pc-linux-gnu} will cause bootstrap
|
1672 |
|
|
failure as @option{-mcpu=} is deprecated in 3.4.0 and above.
|
1673 |
|
|
|
1674 |
|
|
|
1675 |
|
|
If you used the flag @option{--enable-languages=@dots{}} to restrict
|
1676 |
|
|
the compilers to be built, only those you've actually enabled will be
|
1677 |
|
|
built. This will of course only build those runtime libraries, for
|
1678 |
|
|
which the particular compiler has been built. Please note,
|
1679 |
|
|
that re-defining @env{LANGUAGES} when calling @samp{make}
|
1680 |
|
|
@strong{does not} work anymore!
|
1681 |
|
|
|
1682 |
|
|
If the comparison of stage2 and stage3 fails, this normally indicates
|
1683 |
|
|
that the stage2 compiler has compiled GCC incorrectly, and is therefore
|
1684 |
|
|
a potentially serious bug which you should investigate and report. (On
|
1685 |
|
|
a few systems, meaningful comparison of object files is impossible; they
|
1686 |
|
|
always appear ``different''. If you encounter this problem, you will
|
1687 |
|
|
need to disable comparison in the @file{Makefile}.)
|
1688 |
|
|
|
1689 |
|
|
If you do not want to bootstrap your compiler, you can configure with
|
1690 |
|
|
@option{--disable-bootstrap}. In particular cases, you may want to
|
1691 |
|
|
bootstrap your compiler even if the target system is not the same as
|
1692 |
|
|
the one you are building on: for example, you could build a
|
1693 |
|
|
@code{powerpc-unknown-linux-gnu} toolchain on a
|
1694 |
|
|
@code{powerpc64-unknown-linux-gnu} host. In this case, pass
|
1695 |
|
|
@option{--enable-bootstrap} to the configure script.
|
1696 |
|
|
|
1697 |
|
|
|
1698 |
|
|
@section Building a cross compiler
|
1699 |
|
|
|
1700 |
|
|
When building a cross compiler, it is not generally possible to do a
|
1701 |
|
|
3-stage bootstrap of the compiler. This makes for an interesting problem
|
1702 |
|
|
as parts of GCC can only be built with GCC@.
|
1703 |
|
|
|
1704 |
|
|
To build a cross compiler, we first recommend building and installing a
|
1705 |
|
|
native compiler. You can then use the native GCC compiler to build the
|
1706 |
|
|
cross compiler. The installed native compiler needs to be GCC version
|
1707 |
|
|
2.95 or later.
|
1708 |
|
|
|
1709 |
|
|
Assuming you have already installed a native copy of GCC and configured
|
1710 |
|
|
your cross compiler, issue the command @command{make}, which performs the
|
1711 |
|
|
following steps:
|
1712 |
|
|
|
1713 |
|
|
@itemize @bullet
|
1714 |
|
|
@item
|
1715 |
|
|
Build host tools necessary to build the compiler.
|
1716 |
|
|
|
1717 |
|
|
@item
|
1718 |
|
|
Build target tools for use by the compiler such as binutils (bfd,
|
1719 |
|
|
binutils, gas, gprof, ld, and opcodes)
|
1720 |
|
|
if they have been individually linked or moved into the top level GCC source
|
1721 |
|
|
tree before configuring.
|
1722 |
|
|
|
1723 |
|
|
@item
|
1724 |
|
|
Build the compiler (single stage only).
|
1725 |
|
|
|
1726 |
|
|
@item
|
1727 |
|
|
Build runtime libraries using the compiler from the previous step.
|
1728 |
|
|
@end itemize
|
1729 |
|
|
|
1730 |
|
|
Note that if an error occurs in any step the make process will exit.
|
1731 |
|
|
|
1732 |
|
|
If you are not building GNU binutils in the same source tree as GCC,
|
1733 |
|
|
you will need a cross-assembler and cross-linker installed before
|
1734 |
|
|
configuring GCC@. Put them in the directory
|
1735 |
|
|
@file{@var{prefix}/@var{target}/bin}. Here is a table of the tools
|
1736 |
|
|
you should put in this directory:
|
1737 |
|
|
|
1738 |
|
|
@table @file
|
1739 |
|
|
@item as
|
1740 |
|
|
This should be the cross-assembler.
|
1741 |
|
|
|
1742 |
|
|
@item ld
|
1743 |
|
|
This should be the cross-linker.
|
1744 |
|
|
|
1745 |
|
|
@item ar
|
1746 |
|
|
This should be the cross-archiver: a program which can manipulate
|
1747 |
|
|
archive files (linker libraries) in the target machine's format.
|
1748 |
|
|
|
1749 |
|
|
@item ranlib
|
1750 |
|
|
This should be a program to construct a symbol table in an archive file.
|
1751 |
|
|
@end table
|
1752 |
|
|
|
1753 |
|
|
The installation of GCC will find these programs in that directory,
|
1754 |
|
|
and copy or link them to the proper place to for the cross-compiler to
|
1755 |
|
|
find them when run later.
|
1756 |
|
|
|
1757 |
|
|
The easiest way to provide these files is to build the Binutils package.
|
1758 |
|
|
Configure it with the same @option{--host} and @option{--target}
|
1759 |
|
|
options that you use for configuring GCC, then build and install
|
1760 |
|
|
them. They install their executables automatically into the proper
|
1761 |
|
|
directory. Alas, they do not support all the targets that GCC
|
1762 |
|
|
supports.
|
1763 |
|
|
|
1764 |
|
|
If you are not building a C library in the same source tree as GCC,
|
1765 |
|
|
you should also provide the target libraries and headers before
|
1766 |
|
|
configuring GCC, specifying the directories with
|
1767 |
|
|
@option{--with-sysroot} or @option{--with-headers} and
|
1768 |
|
|
@option{--with-libs}. Many targets also require ``start files'' such
|
1769 |
|
|
as @file{crt0.o} and
|
1770 |
|
|
@file{crtn.o} which are linked into each executable. There may be several
|
1771 |
|
|
alternatives for @file{crt0.o}, for use with profiling or other
|
1772 |
|
|
compilation options. Check your target's definition of
|
1773 |
|
|
@code{STARTFILE_SPEC} to find out what start files it uses.
|
1774 |
|
|
|
1775 |
|
|
@section Building in parallel
|
1776 |
|
|
|
1777 |
|
|
GNU Make 3.79 and above, which is necessary to build GCC, support
|
1778 |
|
|
building in parallel. To activate this, you can use @samp{make -j 2}
|
1779 |
|
|
instead of @samp{make}. You can also specify a bigger number, and
|
1780 |
|
|
in most cases using a value greater than the number of processors in
|
1781 |
|
|
your machine will result in fewer and shorter I/O latency hits, thus
|
1782 |
|
|
improving overall throughput; this is especially true for slow drives
|
1783 |
|
|
and network filesystems.
|
1784 |
|
|
|
1785 |
|
|
@section Building the Ada compiler
|
1786 |
|
|
|
1787 |
|
|
In order to build GNAT, the Ada compiler, you need a working GNAT
|
1788 |
|
|
compiler (GNAT version 3.14 or later, or GCC version 3.1 or later).
|
1789 |
|
|
This includes GNAT tools such as @command{gnatmake} and
|
1790 |
|
|
@command{gnatlink}, since the Ada front end is written in Ada and
|
1791 |
|
|
uses some GNAT-specific extensions.
|
1792 |
|
|
|
1793 |
|
|
In order to build a cross compiler, it is suggested to install
|
1794 |
|
|
the new compiler as native first, and then use it to build the cross
|
1795 |
|
|
compiler.
|
1796 |
|
|
|
1797 |
|
|
@command{configure} does not test whether the GNAT installation works
|
1798 |
|
|
and has a sufficiently recent version; if too old a GNAT version is
|
1799 |
|
|
installed, the build will fail unless @option{--enable-languages} is
|
1800 |
|
|
used to disable building the Ada front end.
|
1801 |
|
|
|
1802 |
|
|
@section Building with profile feedback
|
1803 |
|
|
|
1804 |
|
|
It is possible to use profile feedback to optimize the compiler itself. This
|
1805 |
|
|
should result in a faster compiler binary. Experiments done on x86 using gcc
|
1806 |
|
|
3.3 showed approximately 7 percent speedup on compiling C programs. To
|
1807 |
|
|
bootstrap the compiler with profile feedback, use @code{make profiledbootstrap}.
|
1808 |
|
|
|
1809 |
|
|
When @samp{make profiledbootstrap} is run, it will first build a @code{stage1}
|
1810 |
|
|
compiler. This compiler is used to build a @code{stageprofile} compiler
|
1811 |
|
|
instrumented to collect execution counts of instruction and branch
|
1812 |
|
|
probabilities. Then runtime libraries are compiled with profile collected.
|
1813 |
|
|
Finally a @code{stagefeedback} compiler is built using the information collected.
|
1814 |
|
|
|
1815 |
|
|
Unlike standard bootstrap, several additional restrictions apply. The
|
1816 |
|
|
compiler used to build @code{stage1} needs to support a 64-bit integral type.
|
1817 |
|
|
It is recommended to only use GCC for this. Also parallel make is currently
|
1818 |
|
|
not supported since collisions in profile collecting may occur.
|
1819 |
|
|
|
1820 |
|
|
@html
|
1821 |
|
|
<hr />
|
1822 |
|
|
<p>
|
1823 |
|
|
@end html
|
1824 |
|
|
@ifhtml
|
1825 |
|
|
@uref{./index.html,,Return to the GCC Installation page}
|
1826 |
|
|
@end ifhtml
|
1827 |
|
|
@end ifset
|
1828 |
|
|
|
1829 |
|
|
@c ***Testing*****************************************************************
|
1830 |
|
|
@ifnothtml
|
1831 |
|
|
@comment node-name, next, previous, up
|
1832 |
|
|
@node Testing, Final install, Building, Installing GCC
|
1833 |
|
|
@end ifnothtml
|
1834 |
|
|
@ifset testhtml
|
1835 |
|
|
@ifnothtml
|
1836 |
|
|
@chapter Installing GCC: Testing
|
1837 |
|
|
@end ifnothtml
|
1838 |
|
|
@cindex Testing
|
1839 |
|
|
@cindex Installing GCC: Testing
|
1840 |
|
|
@cindex Testsuite
|
1841 |
|
|
|
1842 |
|
|
Before you install GCC, we encourage you to run the testsuites and to
|
1843 |
|
|
compare your results with results from a similar configuration that have
|
1844 |
|
|
been submitted to the
|
1845 |
|
|
@uref{http://gcc.gnu.org/ml/gcc-testresults/,,gcc-testresults mailing list}.
|
1846 |
|
|
Some of these archived results are linked from the build status lists
|
1847 |
|
|
at @uref{http://gcc.gnu.org/buildstat.html}, although not everyone who
|
1848 |
|
|
reports a successful build runs the testsuites and submits the results.
|
1849 |
|
|
This step is optional and may require you to download additional software,
|
1850 |
|
|
but it can give you confidence in your new GCC installation or point out
|
1851 |
|
|
problems before you install and start using your new GCC@.
|
1852 |
|
|
|
1853 |
|
|
First, you must have @uref{download.html,,downloaded the testsuites}.
|
1854 |
|
|
These are part of the full distribution, but if you downloaded the
|
1855 |
|
|
``core'' compiler plus any front ends, you must download the testsuites
|
1856 |
|
|
separately.
|
1857 |
|
|
|
1858 |
|
|
Second, you must have the testing tools installed. This includes
|
1859 |
|
|
@uref{http://www.gnu.org/software/dejagnu/,,DejaGnu}, Tcl, and Expect;
|
1860 |
|
|
the DejaGnu site has links to these.
|
1861 |
|
|
|
1862 |
|
|
If the directories where @command{runtest} and @command{expect} were
|
1863 |
|
|
installed are not in the @env{PATH}, you may need to set the following
|
1864 |
|
|
environment variables appropriately, as in the following example (which
|
1865 |
|
|
assumes that DejaGnu has been installed under @file{/usr/local}):
|
1866 |
|
|
|
1867 |
|
|
@smallexample
|
1868 |
|
|
TCL_LIBRARY = /usr/local/share/tcl8.0
|
1869 |
|
|
DEJAGNULIBS = /usr/local/share/dejagnu
|
1870 |
|
|
@end smallexample
|
1871 |
|
|
|
1872 |
|
|
(On systems such as Cygwin, these paths are required to be actual
|
1873 |
|
|
paths, not mounts or links; presumably this is due to some lack of
|
1874 |
|
|
portability in the DejaGnu code.)
|
1875 |
|
|
|
1876 |
|
|
|
1877 |
|
|
Finally, you can run the testsuite (which may take a long time):
|
1878 |
|
|
@smallexample
|
1879 |
|
|
cd @var{objdir}; make -k check
|
1880 |
|
|
@end smallexample
|
1881 |
|
|
|
1882 |
|
|
This will test various components of GCC, such as compiler
|
1883 |
|
|
front ends and runtime libraries. While running the testsuite, DejaGnu
|
1884 |
|
|
might emit some harmless messages resembling
|
1885 |
|
|
@samp{WARNING: Couldn't find the global config file.} or
|
1886 |
|
|
@samp{WARNING: Couldn't find tool init file} that can be ignored.
|
1887 |
|
|
|
1888 |
|
|
@section How can you run the testsuite on selected tests?
|
1889 |
|
|
|
1890 |
|
|
In order to run sets of tests selectively, there are targets
|
1891 |
|
|
@samp{make check-gcc} and @samp{make check-g++}
|
1892 |
|
|
in the @file{gcc} subdirectory of the object directory. You can also
|
1893 |
|
|
just run @samp{make check} in a subdirectory of the object directory.
|
1894 |
|
|
|
1895 |
|
|
|
1896 |
|
|
A more selective way to just run all @command{gcc} execute tests in the
|
1897 |
|
|
testsuite is to use
|
1898 |
|
|
|
1899 |
|
|
@smallexample
|
1900 |
|
|
make check-gcc RUNTESTFLAGS="execute.exp @var{other-options}"
|
1901 |
|
|
@end smallexample
|
1902 |
|
|
|
1903 |
|
|
Likewise, in order to run only the @command{g++} ``old-deja'' tests in
|
1904 |
|
|
the testsuite with filenames matching @samp{9805*}, you would use
|
1905 |
|
|
|
1906 |
|
|
@smallexample
|
1907 |
|
|
make check-g++ RUNTESTFLAGS="old-deja.exp=9805* @var{other-options}"
|
1908 |
|
|
@end smallexample
|
1909 |
|
|
|
1910 |
|
|
The @file{*.exp} files are located in the testsuite directories of the GCC
|
1911 |
|
|
source, the most important ones being @file{compile.exp},
|
1912 |
|
|
@file{execute.exp}, @file{dg.exp} and @file{old-deja.exp}.
|
1913 |
|
|
To get a list of the possible @file{*.exp} files, pipe the
|
1914 |
|
|
output of @samp{make check} into a file and look at the
|
1915 |
|
|
@samp{Running @dots{} .exp} lines.
|
1916 |
|
|
|
1917 |
|
|
@section Passing options and running multiple testsuites
|
1918 |
|
|
|
1919 |
|
|
You can pass multiple options to the testsuite using the
|
1920 |
|
|
@samp{--target_board} option of DejaGNU, either passed as part of
|
1921 |
|
|
@samp{RUNTESTFLAGS}, or directly to @command{runtest} if you prefer to
|
1922 |
|
|
work outside the makefiles. For example,
|
1923 |
|
|
|
1924 |
|
|
@smallexample
|
1925 |
|
|
make check-g++ RUNTESTFLAGS="--target_board=unix/-O3/-fno-strength-reduce"
|
1926 |
|
|
@end smallexample
|
1927 |
|
|
|
1928 |
|
|
will run the standard @command{g++} testsuites (``unix'' is the target name
|
1929 |
|
|
for a standard native testsuite situation), passing
|
1930 |
|
|
@samp{-O3 -fno-strength-reduce} to the compiler on every test, i.e.,
|
1931 |
|
|
slashes separate options.
|
1932 |
|
|
|
1933 |
|
|
You can run the testsuites multiple times using combinations of options
|
1934 |
|
|
with a syntax similar to the brace expansion of popular shells:
|
1935 |
|
|
|
1936 |
|
|
@smallexample
|
1937 |
|
|
@dots{}"--target_board=arm-sim/@{-mhard-float,-msoft-float@}@{-O1,-O2,-O3,@}"
|
1938 |
|
|
@end smallexample
|
1939 |
|
|
|
1940 |
|
|
(Note the empty option caused by the trailing comma in the final group.)
|
1941 |
|
|
The following will run each testsuite eight times using the @samp{arm-sim}
|
1942 |
|
|
target, as if you had specified all possible combinations yourself:
|
1943 |
|
|
|
1944 |
|
|
@smallexample
|
1945 |
|
|
--target_board=arm-sim/-mhard-float/-O1
|
1946 |
|
|
--target_board=arm-sim/-mhard-float/-O2
|
1947 |
|
|
--target_board=arm-sim/-mhard-float/-O3
|
1948 |
|
|
--target_board=arm-sim/-mhard-float
|
1949 |
|
|
--target_board=arm-sim/-msoft-float/-O1
|
1950 |
|
|
--target_board=arm-sim/-msoft-float/-O2
|
1951 |
|
|
--target_board=arm-sim/-msoft-float/-O3
|
1952 |
|
|
--target_board=arm-sim/-msoft-float
|
1953 |
|
|
@end smallexample
|
1954 |
|
|
|
1955 |
|
|
They can be combined as many times as you wish, in arbitrary ways. This
|
1956 |
|
|
list:
|
1957 |
|
|
|
1958 |
|
|
@smallexample
|
1959 |
|
|
@dots{}"--target_board=unix/-Wextra@{-O3,-fno-strength-reduce@}@{-fomit-frame-pointer,@}"
|
1960 |
|
|
@end smallexample
|
1961 |
|
|
|
1962 |
|
|
will generate four combinations, all involving @samp{-Wextra}.
|
1963 |
|
|
|
1964 |
|
|
The disadvantage to this method is that the testsuites are run in serial,
|
1965 |
|
|
which is a waste on multiprocessor systems. For users with GNU Make and
|
1966 |
|
|
a shell which performs brace expansion, you can run the testsuites in
|
1967 |
|
|
parallel by having the shell perform the combinations and @command{make}
|
1968 |
|
|
do the parallel runs. Instead of using @samp{--target_board}, use a
|
1969 |
|
|
special makefile target:
|
1970 |
|
|
|
1971 |
|
|
@smallexample
|
1972 |
|
|
make -j@var{N} check-@var{testsuite}//@var{test-target}/@var{option1}/@var{option2}/@dots{}
|
1973 |
|
|
@end smallexample
|
1974 |
|
|
|
1975 |
|
|
For example,
|
1976 |
|
|
|
1977 |
|
|
@smallexample
|
1978 |
|
|
make -j3 check-gcc//sh-hms-sim/@{-m1,-m2,-m3,-m3e,-m4@}/@{,-nofpu@}
|
1979 |
|
|
@end smallexample
|
1980 |
|
|
|
1981 |
|
|
will run three concurrent ``make-gcc'' testsuites, eventually testing all
|
1982 |
|
|
ten combinations as described above. Note that this is currently only
|
1983 |
|
|
supported in the @file{gcc} subdirectory. (To see how this works, try
|
1984 |
|
|
typing @command{echo} before the example given here.)
|
1985 |
|
|
|
1986 |
|
|
|
1987 |
|
|
@section Additional testing for Java Class Libraries
|
1988 |
|
|
|
1989 |
|
|
The Java runtime tests can be executed via @samp{make check}
|
1990 |
|
|
in the @file{@var{target}/libjava/testsuite} directory in
|
1991 |
|
|
the build tree.
|
1992 |
|
|
|
1993 |
|
|
The @uref{http://sourceware.org/mauve/,,Mauve Project} provides
|
1994 |
|
|
a suite of tests for the Java Class Libraries. This suite can be run
|
1995 |
|
|
as part of libgcj testing by placing the Mauve tree within the libjava
|
1996 |
|
|
testsuite at @file{libjava/testsuite/libjava.mauve/mauve}, or by
|
1997 |
|
|
specifying the location of that tree when invoking @samp{make}, as in
|
1998 |
|
|
@samp{make MAUVEDIR=~/mauve check}.
|
1999 |
|
|
|
2000 |
|
|
@uref{http://sourceware.org/mauve/jacks.html,,Jacks}
|
2001 |
|
|
is a free testsuite that tests Java compiler front ends. This suite
|
2002 |
|
|
can be run as part of libgcj testing by placing the Jacks tree within
|
2003 |
|
|
the libjava testsuite at @file{libjava/testsuite/libjava.jacks/jacks}.
|
2004 |
|
|
|
2005 |
|
|
@section How to interpret test results
|
2006 |
|
|
|
2007 |
|
|
The result of running the testsuite are various @file{*.sum} and @file{*.log}
|
2008 |
|
|
files in the testsuite subdirectories. The @file{*.log} files contain a
|
2009 |
|
|
detailed log of the compiler invocations and the corresponding
|
2010 |
|
|
results, the @file{*.sum} files summarize the results. These summaries
|
2011 |
|
|
contain status codes for all tests:
|
2012 |
|
|
|
2013 |
|
|
@itemize @bullet
|
2014 |
|
|
@item
|
2015 |
|
|
PASS: the test passed as expected
|
2016 |
|
|
@item
|
2017 |
|
|
XPASS: the test unexpectedly passed
|
2018 |
|
|
@item
|
2019 |
|
|
FAIL: the test unexpectedly failed
|
2020 |
|
|
@item
|
2021 |
|
|
XFAIL: the test failed as expected
|
2022 |
|
|
@item
|
2023 |
|
|
UNSUPPORTED: the test is not supported on this platform
|
2024 |
|
|
@item
|
2025 |
|
|
ERROR: the testsuite detected an error
|
2026 |
|
|
@item
|
2027 |
|
|
WARNING: the testsuite detected a possible problem
|
2028 |
|
|
@end itemize
|
2029 |
|
|
|
2030 |
|
|
It is normal for some tests to report unexpected failures. At the
|
2031 |
|
|
current time the testing harness does not allow fine grained control
|
2032 |
|
|
over whether or not a test is expected to fail. This problem should
|
2033 |
|
|
be fixed in future releases.
|
2034 |
|
|
|
2035 |
|
|
|
2036 |
|
|
@section Submitting test results
|
2037 |
|
|
|
2038 |
|
|
If you want to report the results to the GCC project, use the
|
2039 |
|
|
@file{contrib/test_summary} shell script. Start it in the @var{objdir} with
|
2040 |
|
|
|
2041 |
|
|
@smallexample
|
2042 |
|
|
@var{srcdir}/contrib/test_summary -p your_commentary.txt \
|
2043 |
|
|
-m gcc-testresults@@gcc.gnu.org |sh
|
2044 |
|
|
@end smallexample
|
2045 |
|
|
|
2046 |
|
|
This script uses the @command{Mail} program to send the results, so
|
2047 |
|
|
make sure it is in your @env{PATH}. The file @file{your_commentary.txt} is
|
2048 |
|
|
prepended to the testsuite summary and should contain any special
|
2049 |
|
|
remarks you have on your results or your build environment. Please
|
2050 |
|
|
do not edit the testsuite result block or the subject line, as these
|
2051 |
|
|
messages may be automatically processed.
|
2052 |
|
|
|
2053 |
|
|
@html
|
2054 |
|
|
<hr />
|
2055 |
|
|
<p>
|
2056 |
|
|
@end html
|
2057 |
|
|
@ifhtml
|
2058 |
|
|
@uref{./index.html,,Return to the GCC Installation page}
|
2059 |
|
|
@end ifhtml
|
2060 |
|
|
@end ifset
|
2061 |
|
|
|
2062 |
|
|
@c ***Final install***********************************************************
|
2063 |
|
|
@ifnothtml
|
2064 |
|
|
@comment node-name, next, previous, up
|
2065 |
|
|
@node Final install, , Testing, Installing GCC
|
2066 |
|
|
@end ifnothtml
|
2067 |
|
|
@ifset finalinstallhtml
|
2068 |
|
|
@ifnothtml
|
2069 |
|
|
@chapter Installing GCC: Final installation
|
2070 |
|
|
@end ifnothtml
|
2071 |
|
|
|
2072 |
|
|
Now that GCC has been built (and optionally tested), you can install it with
|
2073 |
|
|
@smallexample
|
2074 |
|
|
cd @var{objdir}; make install
|
2075 |
|
|
@end smallexample
|
2076 |
|
|
|
2077 |
|
|
We strongly recommend to install into a target directory where there is
|
2078 |
|
|
no previous version of GCC present.
|
2079 |
|
|
|
2080 |
|
|
That step completes the installation of GCC; user level binaries can
|
2081 |
|
|
be found in @file{@var{prefix}/bin} where @var{prefix} is the value
|
2082 |
|
|
you specified with the @option{--prefix} to configure (or
|
2083 |
|
|
@file{/usr/local} by default). (If you specified @option{--bindir},
|
2084 |
|
|
that directory will be used instead; otherwise, if you specified
|
2085 |
|
|
@option{--exec-prefix}, @file{@var{exec-prefix}/bin} will be used.)
|
2086 |
|
|
Headers for the C++ and Java libraries are installed in
|
2087 |
|
|
@file{@var{prefix}/include}; libraries in @file{@var{libdir}}
|
2088 |
|
|
(normally @file{@var{prefix}/lib}); internal parts of the compiler in
|
2089 |
|
|
@file{@var{libdir}/gcc} and @file{@var{libexecdir}/gcc}; documentation
|
2090 |
|
|
in info format in @file{@var{infodir}} (normally
|
2091 |
|
|
@file{@var{prefix}/info}).
|
2092 |
|
|
|
2093 |
|
|
When installing cross-compilers, GCC's executables
|
2094 |
|
|
are not only installed into @file{@var{bindir}}, that
|
2095 |
|
|
is, @file{@var{exec-prefix}/bin}, but additionally into
|
2096 |
|
|
@file{@var{exec-prefix}/@var{target-alias}/bin}, if that directory
|
2097 |
|
|
exists. Typically, such @dfn{tooldirs} hold target-specific
|
2098 |
|
|
binutils, including assembler and linker.
|
2099 |
|
|
|
2100 |
|
|
Installation into a temporary staging area or into a @command{chroot}
|
2101 |
|
|
jail can be achieved with the command
|
2102 |
|
|
|
2103 |
|
|
@smallexample
|
2104 |
|
|
make DESTDIR=@var{path-to-rootdir} install
|
2105 |
|
|
@end smallexample
|
2106 |
|
|
|
2107 |
|
|
@noindent where @var{path-to-rootdir} is the absolute path of
|
2108 |
|
|
a directory relative to which all installation paths will be
|
2109 |
|
|
interpreted. Note that the directory specified by @code{DESTDIR}
|
2110 |
|
|
need not exist yet; it will be created if necessary.
|
2111 |
|
|
|
2112 |
|
|
There is a subtle point with tooldirs and @code{DESTDIR}:
|
2113 |
|
|
If you relocate a cross-compiler installation with
|
2114 |
|
|
e.g.@: @samp{DESTDIR=@var{rootdir}}, then the directory
|
2115 |
|
|
@file{@var{rootdir}/@var{exec-prefix}/@var{target-alias}/bin} will
|
2116 |
|
|
be filled with duplicated GCC executables only if it already exists,
|
2117 |
|
|
it will not be created otherwise. This is regarded as a feature,
|
2118 |
|
|
not as a bug, because it gives slightly more control to the packagers
|
2119 |
|
|
using the @code{DESTDIR} feature.
|
2120 |
|
|
|
2121 |
|
|
If you are bootstrapping a released version of GCC then please
|
2122 |
|
|
quickly review the build status page for your release, available from
|
2123 |
|
|
@uref{http://gcc.gnu.org/buildstat.html}.
|
2124 |
|
|
If your system is not listed for the version of GCC that you built,
|
2125 |
|
|
send a note to
|
2126 |
|
|
@email{gcc@@gcc.gnu.org} indicating
|
2127 |
|
|
that you successfully built and installed GCC@.
|
2128 |
|
|
Include the following information:
|
2129 |
|
|
|
2130 |
|
|
@itemize @bullet
|
2131 |
|
|
@item
|
2132 |
|
|
Output from running @file{@var{srcdir}/config.guess}. Do not send
|
2133 |
|
|
that file itself, just the one-line output from running it.
|
2134 |
|
|
|
2135 |
|
|
@item
|
2136 |
|
|
The output of @samp{gcc -v} for your newly installed @command{gcc}.
|
2137 |
|
|
This tells us which version of GCC you built and the options you passed to
|
2138 |
|
|
configure.
|
2139 |
|
|
|
2140 |
|
|
@item
|
2141 |
|
|
Whether you enabled all languages or a subset of them. If you used a
|
2142 |
|
|
full distribution then this information is part of the configure
|
2143 |
|
|
options in the output of @samp{gcc -v}, but if you downloaded the
|
2144 |
|
|
``core'' compiler plus additional front ends then it isn't apparent
|
2145 |
|
|
which ones you built unless you tell us about it.
|
2146 |
|
|
|
2147 |
|
|
@item
|
2148 |
|
|
If the build was for GNU/Linux, also include:
|
2149 |
|
|
@itemize @bullet
|
2150 |
|
|
@item
|
2151 |
|
|
The distribution name and version (e.g., Red Hat 7.1 or Debian 2.2.3);
|
2152 |
|
|
this information should be available from @file{/etc/issue}.
|
2153 |
|
|
|
2154 |
|
|
@item
|
2155 |
|
|
The version of the Linux kernel, available from @samp{uname --version}
|
2156 |
|
|
or @samp{uname -a}.
|
2157 |
|
|
|
2158 |
|
|
@item
|
2159 |
|
|
The version of glibc you used; for RPM-based systems like Red Hat,
|
2160 |
|
|
Mandrake, and SuSE type @samp{rpm -q glibc} to get the glibc version,
|
2161 |
|
|
and on systems like Debian and Progeny use @samp{dpkg -l libc6}.
|
2162 |
|
|
@end itemize
|
2163 |
|
|
For other systems, you can include similar information if you think it is
|
2164 |
|
|
relevant.
|
2165 |
|
|
|
2166 |
|
|
@item
|
2167 |
|
|
Any other information that you think would be useful to people building
|
2168 |
|
|
GCC on the same configuration. The new entry in the build status list
|
2169 |
|
|
will include a link to the archived copy of your message.
|
2170 |
|
|
@end itemize
|
2171 |
|
|
|
2172 |
|
|
We'd also like to know if the
|
2173 |
|
|
@ifnothtml
|
2174 |
|
|
@ref{Specific, host/target specific installation notes}
|
2175 |
|
|
@end ifnothtml
|
2176 |
|
|
@ifhtml
|
2177 |
|
|
@uref{specific.html,,host/target specific installation notes}
|
2178 |
|
|
@end ifhtml
|
2179 |
|
|
didn't include your host/target information or if that information is
|
2180 |
|
|
incomplete or out of date. Send a note to
|
2181 |
|
|
@email{gcc@@gcc.gnu.org} detailing how the information should be changed.
|
2182 |
|
|
|
2183 |
|
|
If you find a bug, please report it following the
|
2184 |
|
|
@uref{../bugs.html,,bug reporting guidelines}.
|
2185 |
|
|
|
2186 |
|
|
If you want to print the GCC manuals, do @samp{cd @var{objdir}; make
|
2187 |
|
|
dvi}. You will need to have @command{texi2dvi} (version at least 4.4)
|
2188 |
|
|
and @TeX{} installed. This creates a number of @file{.dvi} files in
|
2189 |
|
|
subdirectories of @file{@var{objdir}}; these may be converted for
|
2190 |
|
|
printing with programs such as @command{dvips}. Alternately, by using
|
2191 |
|
|
@samp{make pdf} in place of @samp{make dvi}, you can create documentation
|
2192 |
|
|
in the form of @file{.pdf} files; this requires @command{texi2pdf}, which
|
2193 |
|
|
is included with Texinfo version 4.8 and later. You can also
|
2194 |
|
|
@uref{http://www.gnu.org/order/order.html,,buy printed manuals from the
|
2195 |
|
|
Free Software Foundation}, though such manuals may not be for the most
|
2196 |
|
|
recent version of GCC@.
|
2197 |
|
|
|
2198 |
|
|
If you would like to generate online HTML documentation, do @samp{cd
|
2199 |
|
|
@var{objdir}; make html} and HTML will be generated for the gcc manuals in
|
2200 |
|
|
@file{@var{objdir}/gcc/HTML}.
|
2201 |
|
|
|
2202 |
|
|
@html
|
2203 |
|
|
<hr />
|
2204 |
|
|
<p>
|
2205 |
|
|
@end html
|
2206 |
|
|
@ifhtml
|
2207 |
|
|
@uref{./index.html,,Return to the GCC Installation page}
|
2208 |
|
|
@end ifhtml
|
2209 |
|
|
@end ifset
|
2210 |
|
|
|
2211 |
|
|
@c ***Binaries****************************************************************
|
2212 |
|
|
@ifnothtml
|
2213 |
|
|
@comment node-name, next, previous, up
|
2214 |
|
|
@node Binaries, Specific, Installing GCC, Top
|
2215 |
|
|
@end ifnothtml
|
2216 |
|
|
@ifset binarieshtml
|
2217 |
|
|
@ifnothtml
|
2218 |
|
|
@chapter Installing GCC: Binaries
|
2219 |
|
|
@end ifnothtml
|
2220 |
|
|
@cindex Binaries
|
2221 |
|
|
@cindex Installing GCC: Binaries
|
2222 |
|
|
|
2223 |
|
|
We are often asked about pre-compiled versions of GCC@. While we cannot
|
2224 |
|
|
provide these for all platforms, below you'll find links to binaries for
|
2225 |
|
|
various platforms where creating them by yourself is not easy due to various
|
2226 |
|
|
reasons.
|
2227 |
|
|
|
2228 |
|
|
Please note that we did not create these binaries, nor do we
|
2229 |
|
|
support them. If you have any problems installing them, please
|
2230 |
|
|
contact their makers.
|
2231 |
|
|
|
2232 |
|
|
@itemize
|
2233 |
|
|
@item
|
2234 |
|
|
AIX:
|
2235 |
|
|
@itemize
|
2236 |
|
|
@item
|
2237 |
|
|
@uref{http://www.bullfreeware.com,,Bull's Freeware and Shareware Archive for AIX};
|
2238 |
|
|
|
2239 |
|
|
@item
|
2240 |
|
|
@uref{http://aixpdslib.seas.ucla.edu,,UCLA Software Library for AIX}.
|
2241 |
|
|
@end itemize
|
2242 |
|
|
|
2243 |
|
|
@item
|
2244 |
|
|
DOS---@uref{http://www.delorie.com/djgpp/,,DJGPP}.
|
2245 |
|
|
|
2246 |
|
|
@item
|
2247 |
|
|
Renesas H8/300[HS]---@uref{http://h8300-hms.sourceforge.net/,,GNU
|
2248 |
|
|
Development Tools for the Renesas H8/300[HS] Series}.
|
2249 |
|
|
|
2250 |
|
|
@item
|
2251 |
|
|
HP-UX:
|
2252 |
|
|
@itemize
|
2253 |
|
|
@item
|
2254 |
|
|
@uref{http://hpux.cs.utah.edu/,,HP-UX Porting Center};
|
2255 |
|
|
|
2256 |
|
|
@item
|
2257 |
|
|
@uref{ftp://sunsite.informatik.rwth-aachen.de/pub/packages/gcc_hpux/,,Binaries for HP-UX 11.00 at Aachen University of Technology}.
|
2258 |
|
|
@end itemize
|
2259 |
|
|
|
2260 |
|
|
@item
|
2261 |
|
|
Motorola 68HC11/68HC12---@uref{http://www.gnu-m68hc11.org,,GNU
|
2262 |
|
|
Development Tools for the Motorola 68HC11/68HC12}.
|
2263 |
|
|
|
2264 |
|
|
@item
|
2265 |
|
|
@uref{http://www.sco.com/skunkware/devtools/index.html#gcc,,SCO
|
2266 |
|
|
OpenServer/Unixware}.
|
2267 |
|
|
|
2268 |
|
|
@item
|
2269 |
|
|
Solaris 2 (SPARC, Intel)---@uref{http://www.sunfreeware.com/,,Sunfreeware}.
|
2270 |
|
|
|
2271 |
|
|
@item
|
2272 |
|
|
SGI---@uref{http://freeware.sgi.com/,,SGI Freeware}.
|
2273 |
|
|
|
2274 |
|
|
@item
|
2275 |
|
|
Microsoft Windows:
|
2276 |
|
|
@itemize
|
2277 |
|
|
@item
|
2278 |
|
|
The @uref{http://sourceware.org/cygwin/,,Cygwin} project;
|
2279 |
|
|
@item
|
2280 |
|
|
The @uref{http://www.mingw.org/,,MinGW} project.
|
2281 |
|
|
@end itemize
|
2282 |
|
|
|
2283 |
|
|
@item
|
2284 |
|
|
@uref{ftp://ftp.thewrittenword.com/packages/by-name/,,The
|
2285 |
|
|
Written Word} offers binaries for
|
2286 |
|
|
AIX 4.3.2.
|
2287 |
|
|
IRIX 6.5,
|
2288 |
|
|
Digital UNIX 4.0D and 5.1,
|
2289 |
|
|
GNU/Linux (i386),
|
2290 |
|
|
HP-UX 10.20, 11.00, and 11.11, and
|
2291 |
|
|
Solaris/SPARC 2.5.1, 2.6, 7, 8, and 9.
|
2292 |
|
|
|
2293 |
|
|
@item
|
2294 |
|
|
@uref{http://www.openpkg.org/,,OpenPKG} offers binaries for quite a
|
2295 |
|
|
number of platforms.
|
2296 |
|
|
|
2297 |
|
|
@item
|
2298 |
|
|
The @uref{http://gcc.gnu.org/wiki/GFortranBinaries,,GFortran Wiki} has
|
2299 |
|
|
links to GNU Fortran binaries for several platforms.
|
2300 |
|
|
@end itemize
|
2301 |
|
|
|
2302 |
|
|
In addition to those specific offerings, you can get a binary
|
2303 |
|
|
distribution CD-ROM from the
|
2304 |
|
|
@uref{http://www.gnu.org/order/order.html,,Free Software Foundation}.
|
2305 |
|
|
It contains binaries for a number of platforms, and
|
2306 |
|
|
includes not only GCC, but other stuff as well. The current CD does
|
2307 |
|
|
not contain the latest version of GCC, but it should allow
|
2308 |
|
|
bootstrapping the compiler. An updated version of that disk is in the
|
2309 |
|
|
works.
|
2310 |
|
|
|
2311 |
|
|
@html
|
2312 |
|
|
<hr />
|
2313 |
|
|
<p>
|
2314 |
|
|
@end html
|
2315 |
|
|
@ifhtml
|
2316 |
|
|
@uref{./index.html,,Return to the GCC Installation page}
|
2317 |
|
|
@end ifhtml
|
2318 |
|
|
@end ifset
|
2319 |
|
|
|
2320 |
|
|
@c ***Specific****************************************************************
|
2321 |
|
|
@ifnothtml
|
2322 |
|
|
@comment node-name, next, previous, up
|
2323 |
|
|
@node Specific, Old, Binaries, Top
|
2324 |
|
|
@end ifnothtml
|
2325 |
|
|
@ifset specifichtml
|
2326 |
|
|
@ifnothtml
|
2327 |
|
|
@chapter Host/target specific installation notes for GCC
|
2328 |
|
|
@end ifnothtml
|
2329 |
|
|
@cindex Specific
|
2330 |
|
|
@cindex Specific installation notes
|
2331 |
|
|
@cindex Target specific installation
|
2332 |
|
|
@cindex Host specific installation
|
2333 |
|
|
@cindex Target specific installation notes
|
2334 |
|
|
|
2335 |
|
|
Please read this document carefully @emph{before} installing the
|
2336 |
|
|
GNU Compiler Collection on your machine.
|
2337 |
|
|
|
2338 |
|
|
Note that this list of install notes is @emph{not} a list of supported
|
2339 |
|
|
hosts or targets. Not all supported hosts and targets are listed
|
2340 |
|
|
here, only the ones that require host-specific or target-specific
|
2341 |
|
|
information are.
|
2342 |
|
|
|
2343 |
|
|
@ifhtml
|
2344 |
|
|
@itemize
|
2345 |
|
|
@item
|
2346 |
|
|
@uref{#alpha-x-x,,alpha*-*-*}
|
2347 |
|
|
@item
|
2348 |
|
|
@uref{#alpha-dec-osf,,alpha*-dec-osf*}
|
2349 |
|
|
@item
|
2350 |
|
|
@uref{#alphaev5-cray-unicosmk,,alphaev5-cray-unicosmk*}
|
2351 |
|
|
@item
|
2352 |
|
|
@uref{#arc-x-elf,,arc-*-elf}
|
2353 |
|
|
@item
|
2354 |
|
|
@uref{#arm-x-elf,,arm-*-elf}
|
2355 |
|
|
@uref{#arm-x-coff,,arm-*-coff}
|
2356 |
|
|
@uref{#arm-x-aout,,arm-*-aout}
|
2357 |
|
|
@item
|
2358 |
|
|
@uref{#xscale-x-x,,xscale-*-*}
|
2359 |
|
|
@item
|
2360 |
|
|
@uref{#avr,,avr}
|
2361 |
|
|
@item
|
2362 |
|
|
@uref{#bfin,,Blackfin}
|
2363 |
|
|
@item
|
2364 |
|
|
@uref{#c4x,,c4x}
|
2365 |
|
|
@item
|
2366 |
|
|
@uref{#dos,,DOS}
|
2367 |
|
|
@item
|
2368 |
|
|
@uref{#x-x-freebsd,,*-*-freebsd*}
|
2369 |
|
|
@item
|
2370 |
|
|
@uref{#h8300-hms,,h8300-hms}
|
2371 |
|
|
@item
|
2372 |
|
|
@uref{#hppa-hp-hpux,,hppa*-hp-hpux*}
|
2373 |
|
|
@item
|
2374 |
|
|
@uref{#hppa-hp-hpux10,,hppa*-hp-hpux10}
|
2375 |
|
|
@item
|
2376 |
|
|
@uref{#hppa-hp-hpux11,,hppa*-hp-hpux11}
|
2377 |
|
|
@item
|
2378 |
|
|
@uref{#x-x-linux-gnu,,*-*-linux-gnu}
|
2379 |
|
|
@item
|
2380 |
|
|
@uref{#ix86-x-linuxaout,,i?86-*-linux*aout}
|
2381 |
|
|
@item
|
2382 |
|
|
@uref{#ix86-x-linux,,i?86-*-linux*}
|
2383 |
|
|
@item
|
2384 |
|
|
@uref{#ix86-x-sco32v5,,i?86-*-sco3.2v5*}
|
2385 |
|
|
@item
|
2386 |
|
|
@uref{#ix86-x-solaris210,,i?86-*-solaris2.10}
|
2387 |
|
|
@item
|
2388 |
|
|
@uref{#ix86-x-udk,,i?86-*-udk}
|
2389 |
|
|
@item
|
2390 |
|
|
@uref{#ia64-x-linux,,ia64-*-linux}
|
2391 |
|
|
@item
|
2392 |
|
|
@uref{#ia64-x-hpux,,ia64-*-hpux*}
|
2393 |
|
|
@item
|
2394 |
|
|
@uref{#x-ibm-aix,,*-ibm-aix*}
|
2395 |
|
|
@item
|
2396 |
|
|
@uref{#iq2000-x-elf,,iq2000-*-elf}
|
2397 |
|
|
@item
|
2398 |
|
|
@uref{#m32c-x-elf,,m32c-*-elf}
|
2399 |
|
|
@item
|
2400 |
|
|
@uref{#m32r-x-elf,,m32r-*-elf}
|
2401 |
|
|
@item
|
2402 |
|
|
@uref{#m6811-elf,,m6811-elf}
|
2403 |
|
|
@item
|
2404 |
|
|
@uref{#m6812-elf,,m6812-elf}
|
2405 |
|
|
@item
|
2406 |
|
|
@uref{#m68k-hp-hpux,,m68k-hp-hpux}
|
2407 |
|
|
@item
|
2408 |
|
|
@uref{#mips-x-x,,mips-*-*}
|
2409 |
|
|
@item
|
2410 |
|
|
@uref{#mips-sgi-irix5,,mips-sgi-irix5}
|
2411 |
|
|
@item
|
2412 |
|
|
@uref{#mips-sgi-irix6,,mips-sgi-irix6}
|
2413 |
|
|
@item
|
2414 |
|
|
@uref{#powerpc-x-x,,powerpc*-*-*, powerpc-*-sysv4}
|
2415 |
|
|
@item
|
2416 |
|
|
@uref{#powerpc-x-darwin,,powerpc-*-darwin*}
|
2417 |
|
|
@item
|
2418 |
|
|
@uref{#powerpc-x-elf,,powerpc-*-elf, powerpc-*-sysv4}
|
2419 |
|
|
@item
|
2420 |
|
|
@uref{#powerpc-x-linux-gnu,,powerpc*-*-linux-gnu*}
|
2421 |
|
|
@item
|
2422 |
|
|
@uref{#powerpc-x-netbsd,,powerpc-*-netbsd*}
|
2423 |
|
|
@item
|
2424 |
|
|
@uref{#powerpc-x-eabisim,,powerpc-*-eabisim}
|
2425 |
|
|
@item
|
2426 |
|
|
@uref{#powerpc-x-eabi,,powerpc-*-eabi}
|
2427 |
|
|
@item
|
2428 |
|
|
@uref{#powerpcle-x-elf,,powerpcle-*-elf, powerpcle-*-sysv4}
|
2429 |
|
|
@item
|
2430 |
|
|
@uref{#powerpcle-x-eabisim,,powerpcle-*-eabisim}
|
2431 |
|
|
@item
|
2432 |
|
|
@uref{#powerpcle-x-eabi,,powerpcle-*-eabi}
|
2433 |
|
|
@item
|
2434 |
|
|
@uref{#s390-x-linux,,s390-*-linux*}
|
2435 |
|
|
@item
|
2436 |
|
|
@uref{#s390x-x-linux,,s390x-*-linux*}
|
2437 |
|
|
@item
|
2438 |
|
|
@uref{#s390x-ibm-tpf,,s390x-ibm-tpf*}
|
2439 |
|
|
@item
|
2440 |
|
|
@uref{#x-x-solaris2,,*-*-solaris2*}
|
2441 |
|
|
@item
|
2442 |
|
|
@uref{#sparc-sun-solaris2,,sparc-sun-solaris2*}
|
2443 |
|
|
@item
|
2444 |
|
|
@uref{#sparc-sun-solaris27,,sparc-sun-solaris2.7}
|
2445 |
|
|
@item
|
2446 |
|
|
@uref{#sparc-x-linux,,sparc-*-linux*}
|
2447 |
|
|
@item
|
2448 |
|
|
@uref{#sparc64-x-solaris2,,sparc64-*-solaris2*}
|
2449 |
|
|
@item
|
2450 |
|
|
@uref{#sparcv9-x-solaris2,,sparcv9-*-solaris2*}
|
2451 |
|
|
@item
|
2452 |
|
|
@uref{#x-x-sysv,,*-*-sysv*}
|
2453 |
|
|
@item
|
2454 |
|
|
@uref{#vax-dec-ultrix,,vax-dec-ultrix}
|
2455 |
|
|
@item
|
2456 |
|
|
@uref{#x-x-vxworks,,*-*-vxworks*}
|
2457 |
|
|
@item
|
2458 |
|
|
@uref{#x86-64-x-x,,x86_64-*-*, amd64-*-*}
|
2459 |
|
|
@item
|
2460 |
|
|
@uref{#xtensa-x-elf,,xtensa-*-elf}
|
2461 |
|
|
@item
|
2462 |
|
|
@uref{#xtensa-x-linux,,xtensa-*-linux*}
|
2463 |
|
|
@item
|
2464 |
|
|
@uref{#windows,,Microsoft Windows}
|
2465 |
|
|
@item
|
2466 |
|
|
@uref{#os2,,OS/2}
|
2467 |
|
|
@item
|
2468 |
|
|
@uref{#older,,Older systems}
|
2469 |
|
|
@end itemize
|
2470 |
|
|
|
2471 |
|
|
@itemize
|
2472 |
|
|
@item
|
2473 |
|
|
@uref{#elf,,all ELF targets} (SVR4, Solaris 2, etc.)
|
2474 |
|
|
@end itemize
|
2475 |
|
|
@end ifhtml
|
2476 |
|
|
|
2477 |
|
|
|
2478 |
|
|
@html
|
2479 |
|
|
<!-- -------- host/target specific issues start here ---------------- -->
|
2480 |
|
|
<hr />
|
2481 |
|
|
@end html
|
2482 |
|
|
@heading @anchor{alpha-x-x}alpha*-*-*
|
2483 |
|
|
|
2484 |
|
|
This section contains general configuration information for all
|
2485 |
|
|
alpha-based platforms using ELF (in particular, ignore this section for
|
2486 |
|
|
DEC OSF/1, Digital UNIX and Tru64 UNIX)@. In addition to reading this
|
2487 |
|
|
section, please read all other sections that match your target.
|
2488 |
|
|
|
2489 |
|
|
We require binutils 2.11.2 or newer.
|
2490 |
|
|
Previous binutils releases had a number of problems with DWARF 2
|
2491 |
|
|
debugging information, not the least of which is incorrect linking of
|
2492 |
|
|
shared libraries.
|
2493 |
|
|
|
2494 |
|
|
@html
|
2495 |
|
|
<hr />
|
2496 |
|
|
@end html
|
2497 |
|
|
@heading @anchor{alpha-dec-osf}alpha*-dec-osf*
|
2498 |
|
|
Systems using processors that implement the DEC Alpha architecture and
|
2499 |
|
|
are running the DEC/Compaq Unix (DEC OSF/1, Digital UNIX, or Compaq
|
2500 |
|
|
Tru64 UNIX) operating system, for example the DEC Alpha AXP systems.
|
2501 |
|
|
|
2502 |
|
|
As of GCC 3.2, versions before @code{alpha*-dec-osf4} are no longer
|
2503 |
|
|
supported. (These are the versions which identify themselves as DEC
|
2504 |
|
|
OSF/1.)
|
2505 |
|
|
|
2506 |
|
|
In Digital Unix V4.0, virtual memory exhausted bootstrap failures
|
2507 |
|
|
may be fixed by configuring with @option{--with-gc=simple},
|
2508 |
|
|
reconfiguring Kernel Virtual Memory and Swap parameters
|
2509 |
|
|
per the @command{/usr/sbin/sys_check} Tuning Suggestions,
|
2510 |
|
|
or applying the patch in
|
2511 |
|
|
@uref{http://gcc.gnu.org/ml/gcc/2002-08/msg00822.html}.
|
2512 |
|
|
|
2513 |
|
|
In Tru64 UNIX V5.1, Compaq introduced a new assembler that does not
|
2514 |
|
|
currently (2001-06-13) work with @command{mips-tfile}. As a workaround,
|
2515 |
|
|
we need to use the old assembler, invoked via the barely documented
|
2516 |
|
|
@option{-oldas} option. To bootstrap GCC, you either need to use the
|
2517 |
|
|
Compaq C Compiler:
|
2518 |
|
|
|
2519 |
|
|
@smallexample
|
2520 |
|
|
% CC=cc @var{srcdir}/configure [@var{options}] [@var{target}]
|
2521 |
|
|
@end smallexample
|
2522 |
|
|
|
2523 |
|
|
or you can use a copy of GCC 2.95.3 or higher built on Tru64 UNIX V4.0:
|
2524 |
|
|
|
2525 |
|
|
@smallexample
|
2526 |
|
|
% CC=gcc -Wa,-oldas @var{srcdir}/configure [@var{options}] [@var{target}]
|
2527 |
|
|
@end smallexample
|
2528 |
|
|
|
2529 |
|
|
As of GNU binutils 2.11.2, neither GNU @command{as} nor GNU @command{ld}
|
2530 |
|
|
are supported on Tru64 UNIX, so you must not configure GCC with
|
2531 |
|
|
@option{--with-gnu-as} or @option{--with-gnu-ld}.
|
2532 |
|
|
|
2533 |
|
|
GCC writes a @samp{.verstamp} directive to the assembler output file
|
2534 |
|
|
unless it is built as a cross-compiler. It gets the version to use from
|
2535 |
|
|
the system header file @file{/usr/include/stamp.h}. If you install a
|
2536 |
|
|
new version of DEC Unix, you should rebuild GCC to pick up the new version
|
2537 |
|
|
stamp.
|
2538 |
|
|
|
2539 |
|
|
Note that since the Alpha is a 64-bit architecture, cross-compilers from
|
2540 |
|
|
32-bit machines will not generate code as efficient as that generated
|
2541 |
|
|
when the compiler is running on a 64-bit machine because many
|
2542 |
|
|
optimizations that depend on being able to represent a word on the
|
2543 |
|
|
target in an integral value on the host cannot be performed. Building
|
2544 |
|
|
cross-compilers on the Alpha for 32-bit machines has only been tested in
|
2545 |
|
|
a few cases and may not work properly.
|
2546 |
|
|
|
2547 |
|
|
@samp{make compare} may fail on old versions of DEC Unix unless you add
|
2548 |
|
|
@option{-save-temps} to @code{CFLAGS}. On these systems, the name of the
|
2549 |
|
|
assembler input file is stored in the object file, and that makes
|
2550 |
|
|
comparison fail if it differs between the @code{stage1} and
|
2551 |
|
|
@code{stage2} compilations. The option @option{-save-temps} forces a
|
2552 |
|
|
fixed name to be used for the assembler input file, instead of a
|
2553 |
|
|
randomly chosen name in @file{/tmp}. Do not add @option{-save-temps}
|
2554 |
|
|
unless the comparisons fail without that option. If you add
|
2555 |
|
|
@option{-save-temps}, you will have to manually delete the @samp{.i} and
|
2556 |
|
|
@samp{.s} files after each series of compilations.
|
2557 |
|
|
|
2558 |
|
|
GCC now supports both the native (ECOFF) debugging format used by DBX
|
2559 |
|
|
and GDB and an encapsulated STABS format for use only with GDB@. See the
|
2560 |
|
|
discussion of the @option{--with-stabs} option of @file{configure} above
|
2561 |
|
|
for more information on these formats and how to select them.
|
2562 |
|
|
|
2563 |
|
|
There is a bug in DEC's assembler that produces incorrect line numbers
|
2564 |
|
|
for ECOFF format when the @samp{.align} directive is used. To work
|
2565 |
|
|
around this problem, GCC will not emit such alignment directives
|
2566 |
|
|
while writing ECOFF format debugging information even if optimization is
|
2567 |
|
|
being performed. Unfortunately, this has the very undesirable
|
2568 |
|
|
side-effect that code addresses when @option{-O} is specified are
|
2569 |
|
|
different depending on whether or not @option{-g} is also specified.
|
2570 |
|
|
|
2571 |
|
|
To avoid this behavior, specify @option{-gstabs+} and use GDB instead of
|
2572 |
|
|
DBX@. DEC is now aware of this problem with the assembler and hopes to
|
2573 |
|
|
provide a fix shortly.
|
2574 |
|
|
|
2575 |
|
|
@html
|
2576 |
|
|
<hr />
|
2577 |
|
|
@end html
|
2578 |
|
|
@heading @anchor{alphaev5-cray-unicosmk}alphaev5-cray-unicosmk*
|
2579 |
|
|
Cray T3E systems running Unicos/Mk.
|
2580 |
|
|
|
2581 |
|
|
This port is incomplete and has many known bugs. We hope to improve the
|
2582 |
|
|
support for this target soon. Currently, only the C front end is supported,
|
2583 |
|
|
and it is not possible to build parallel applications. Cray modules are not
|
2584 |
|
|
supported; in particular, Craylibs are assumed to be in
|
2585 |
|
|
@file{/opt/ctl/craylibs/craylibs}.
|
2586 |
|
|
|
2587 |
|
|
On this platform, you need to tell GCC where to find the assembler and
|
2588 |
|
|
the linker. The simplest way to do so is by providing @option{--with-as}
|
2589 |
|
|
and @option{--with-ld} to @file{configure}, e.g.@:
|
2590 |
|
|
|
2591 |
|
|
@smallexample
|
2592 |
|
|
configure --with-as=/opt/ctl/bin/cam --with-ld=/opt/ctl/bin/cld \
|
2593 |
|
|
--enable-languages=c
|
2594 |
|
|
@end smallexample
|
2595 |
|
|
|
2596 |
|
|
The comparison test at the end of the bootstrapping process fails on Unicos/Mk
|
2597 |
|
|
because the assembler inserts timestamps into object files. You should
|
2598 |
|
|
be able to work around this by doing @samp{make all} after getting this
|
2599 |
|
|
failure.
|
2600 |
|
|
|
2601 |
|
|
@html
|
2602 |
|
|
<hr />
|
2603 |
|
|
@end html
|
2604 |
|
|
@heading @anchor{arc-x-elf}arc-*-elf
|
2605 |
|
|
Argonaut ARC processor.
|
2606 |
|
|
This configuration is intended for embedded systems.
|
2607 |
|
|
|
2608 |
|
|
@html
|
2609 |
|
|
<hr />
|
2610 |
|
|
@end html
|
2611 |
|
|
@heading @anchor{arm-x-elf}arm-*-elf
|
2612 |
|
|
@heading @anchor{xscale-x-x}xscale-*-*
|
2613 |
|
|
ARM-family processors. Subtargets that use the ELF object format
|
2614 |
|
|
require GNU binutils 2.13 or newer. Such subtargets include:
|
2615 |
|
|
@code{arm-*-freebsd}, @code{arm-*-netbsdelf}, @code{arm-*-*linux},
|
2616 |
|
|
@code{arm-*-rtems} and @code{arm-*-kaos}.
|
2617 |
|
|
|
2618 |
|
|
@html
|
2619 |
|
|
<hr />
|
2620 |
|
|
@end html
|
2621 |
|
|
@heading @anchor{arm-x-coff}arm-*-coff
|
2622 |
|
|
ARM-family processors. Note that there are two different varieties
|
2623 |
|
|
of PE format subtarget supported: @code{arm-wince-pe} and
|
2624 |
|
|
@code{arm-pe} as well as a standard COFF target @code{arm-*-coff}.
|
2625 |
|
|
|
2626 |
|
|
@html
|
2627 |
|
|
<hr />
|
2628 |
|
|
@end html
|
2629 |
|
|
@heading @anchor{arm-x-aout}arm-*-aout
|
2630 |
|
|
ARM-family processors. These targets support the AOUT file format:
|
2631 |
|
|
@code{arm-*-aout}, @code{arm-*-netbsd}.
|
2632 |
|
|
|
2633 |
|
|
@html
|
2634 |
|
|
<hr />
|
2635 |
|
|
@end html
|
2636 |
|
|
@heading @anchor{avr}avr
|
2637 |
|
|
|
2638 |
|
|
ATMEL AVR-family micro controllers. These are used in embedded
|
2639 |
|
|
applications. There are no standard Unix configurations.
|
2640 |
|
|
@ifnothtml
|
2641 |
|
|
@xref{AVR Options,, AVR Options, gcc, Using the GNU Compiler
|
2642 |
|
|
Collection (GCC)},
|
2643 |
|
|
@end ifnothtml
|
2644 |
|
|
@ifhtml
|
2645 |
|
|
See ``AVR Options'' in the main manual
|
2646 |
|
|
@end ifhtml
|
2647 |
|
|
for the list of supported MCU types.
|
2648 |
|
|
|
2649 |
|
|
Use @samp{configure --target=avr --enable-languages="c"} to configure GCC@.
|
2650 |
|
|
|
2651 |
|
|
Further installation notes and other useful information about AVR tools
|
2652 |
|
|
can also be obtained from:
|
2653 |
|
|
|
2654 |
|
|
@itemize @bullet
|
2655 |
|
|
@item
|
2656 |
|
|
@uref{http://www.nongnu.org/avr/,,http://www.nongnu.org/avr/}
|
2657 |
|
|
@item
|
2658 |
|
|
@uref{http://home.overta.ru/users/denisc/,,http://home.overta.ru/users/denisc/}
|
2659 |
|
|
@item
|
2660 |
|
|
@uref{http://www.amelek.gda.pl/avr/,,http://www.amelek.gda.pl/avr/}
|
2661 |
|
|
@end itemize
|
2662 |
|
|
|
2663 |
|
|
We @emph{strongly} recommend using binutils 2.13 or newer.
|
2664 |
|
|
|
2665 |
|
|
The following error:
|
2666 |
|
|
@smallexample
|
2667 |
|
|
Error: register required
|
2668 |
|
|
@end smallexample
|
2669 |
|
|
|
2670 |
|
|
indicates that you should upgrade to a newer version of the binutils.
|
2671 |
|
|
|
2672 |
|
|
@html
|
2673 |
|
|
<hr />
|
2674 |
|
|
@end html
|
2675 |
|
|
@heading @anchor{bfin}Blackfin
|
2676 |
|
|
|
2677 |
|
|
The Blackfin processor, an Analog Devices DSP.
|
2678 |
|
|
@ifnothtml
|
2679 |
|
|
@xref{Blackfin Options,, Blackfin Options, gcc, Using the GNU Compiler
|
2680 |
|
|
Collection (GCC)},
|
2681 |
|
|
@end ifnothtml
|
2682 |
|
|
@ifhtml
|
2683 |
|
|
See ``Blackfin Options'' in the main manual
|
2684 |
|
|
@end ifhtml
|
2685 |
|
|
|
2686 |
|
|
More information, and a version of binutils with support for this processor,
|
2687 |
|
|
is available at @uref{http://blackfin.uclinux.org}
|
2688 |
|
|
|
2689 |
|
|
@html
|
2690 |
|
|
<hr />
|
2691 |
|
|
@end html
|
2692 |
|
|
@heading @anchor{c4x}c4x
|
2693 |
|
|
|
2694 |
|
|
Texas Instruments TMS320C3x and TMS320C4x Floating Point Digital Signal
|
2695 |
|
|
Processors. These are used in embedded applications. There are no
|
2696 |
|
|
standard Unix configurations.
|
2697 |
|
|
@ifnothtml
|
2698 |
|
|
@xref{TMS320C3x/C4x Options,, TMS320C3x/C4x Options, gcc, Using the
|
2699 |
|
|
GNU Compiler Collection (GCC)},
|
2700 |
|
|
@end ifnothtml
|
2701 |
|
|
@ifhtml
|
2702 |
|
|
See ``TMS320C3x/C4x Options'' in the main manual
|
2703 |
|
|
@end ifhtml
|
2704 |
|
|
for the list of supported MCU types.
|
2705 |
|
|
|
2706 |
|
|
GCC can be configured as a cross compiler for both the C3x and C4x
|
2707 |
|
|
architectures on the same system. Use @samp{configure --target=c4x
|
2708 |
|
|
--enable-languages="c,c++"} to configure.
|
2709 |
|
|
|
2710 |
|
|
|
2711 |
|
|
Further installation notes and other useful information about C4x tools
|
2712 |
|
|
can also be obtained from:
|
2713 |
|
|
|
2714 |
|
|
@itemize @bullet
|
2715 |
|
|
@item
|
2716 |
|
|
@uref{http://www.elec.canterbury.ac.nz/c4x/,,http://www.elec.canterbury.ac.nz/c4x/}
|
2717 |
|
|
@end itemize
|
2718 |
|
|
|
2719 |
|
|
@html
|
2720 |
|
|
<hr />
|
2721 |
|
|
@end html
|
2722 |
|
|
@heading @anchor{cris}CRIS
|
2723 |
|
|
|
2724 |
|
|
CRIS is the CPU architecture in Axis Communications ETRAX system-on-a-chip
|
2725 |
|
|
series. These are used in embedded applications.
|
2726 |
|
|
|
2727 |
|
|
@ifnothtml
|
2728 |
|
|
@xref{CRIS Options,, CRIS Options, gcc, Using the GNU Compiler
|
2729 |
|
|
Collection (GCC)},
|
2730 |
|
|
@end ifnothtml
|
2731 |
|
|
@ifhtml
|
2732 |
|
|
See ``CRIS Options'' in the main manual
|
2733 |
|
|
@end ifhtml
|
2734 |
|
|
for a list of CRIS-specific options.
|
2735 |
|
|
|
2736 |
|
|
There are a few different CRIS targets:
|
2737 |
|
|
@table @code
|
2738 |
|
|
@item cris-axis-aout
|
2739 |
|
|
Old target. Includes a multilib for the @samp{elinux} a.out-based
|
2740 |
|
|
target. No multilibs for newer architecture variants.
|
2741 |
|
|
@item cris-axis-elf
|
2742 |
|
|
Mainly for monolithic embedded systems. Includes a multilib for the
|
2743 |
|
|
@samp{v10} core used in @samp{ETRAX 100 LX}.
|
2744 |
|
|
@item cris-axis-linux-gnu
|
2745 |
|
|
A GNU/Linux port for the CRIS architecture, currently targeting
|
2746 |
|
|
@samp{ETRAX 100 LX} by default.
|
2747 |
|
|
@end table
|
2748 |
|
|
|
2749 |
|
|
For @code{cris-axis-aout} and @code{cris-axis-elf} you need binutils 2.11
|
2750 |
|
|
or newer. For @code{cris-axis-linux-gnu} you need binutils 2.12 or newer.
|
2751 |
|
|
|
2752 |
|
|
Pre-packaged tools can be obtained from
|
2753 |
|
|
@uref{ftp://ftp.axis.com/pub/axis/tools/cris/compiler-kit/}. More
|
2754 |
|
|
information about this platform is available at
|
2755 |
|
|
@uref{http://developer.axis.com/}.
|
2756 |
|
|
|
2757 |
|
|
@html
|
2758 |
|
|
<hr />
|
2759 |
|
|
@end html
|
2760 |
|
|
@heading @anchor{crx}CRX
|
2761 |
|
|
|
2762 |
|
|
The CRX CompactRISC architecture is a low-power 32-bit architecture with
|
2763 |
|
|
fast context switching and architectural extensibility features.
|
2764 |
|
|
|
2765 |
|
|
@ifnothtml
|
2766 |
|
|
@xref{CRX Options,, CRX Options, gcc, Using and Porting the GNU Compiler
|
2767 |
|
|
Collection (GCC)},
|
2768 |
|
|
@end ifnothtml
|
2769 |
|
|
|
2770 |
|
|
@ifhtml
|
2771 |
|
|
See ``CRX Options'' in the main manual for a list of CRX-specific options.
|
2772 |
|
|
@end ifhtml
|
2773 |
|
|
|
2774 |
|
|
Use @samp{configure --target=crx-elf --enable-languages=c,c++} to configure
|
2775 |
|
|
GCC@ for building a CRX cross-compiler. The option @samp{--target=crx-elf}
|
2776 |
|
|
is also used to build the @samp{newlib} C library for CRX.
|
2777 |
|
|
|
2778 |
|
|
It is also possible to build libstdc++-v3 for the CRX architecture. This
|
2779 |
|
|
needs to be done in a separate step with the following configure settings:
|
2780 |
|
|
@samp{gcc/libstdc++-v3/configure --host=crx-elf --with-newlib
|
2781 |
|
|
--enable-sjlj-exceptions --enable-cxx-flags='-fexceptions -frtti'}
|
2782 |
|
|
|
2783 |
|
|
@html
|
2784 |
|
|
<hr />
|
2785 |
|
|
@end html
|
2786 |
|
|
@heading @anchor{dos}DOS
|
2787 |
|
|
|
2788 |
|
|
Please have a look at the @uref{binaries.html,,binaries page}.
|
2789 |
|
|
|
2790 |
|
|
You cannot install GCC by itself on MSDOS; it will not compile under
|
2791 |
|
|
any MSDOS compiler except itself. You need to get the complete
|
2792 |
|
|
compilation package DJGPP, which includes binaries as well as sources,
|
2793 |
|
|
and includes all the necessary compilation tools and libraries.
|
2794 |
|
|
|
2795 |
|
|
@html
|
2796 |
|
|
<hr />
|
2797 |
|
|
@end html
|
2798 |
|
|
@heading @anchor{x-x-freebsd}*-*-freebsd*
|
2799 |
|
|
|
2800 |
|
|
The version of binutils installed in @file{/usr/bin} probably works with
|
2801 |
|
|
this release of GCC@. However, on FreeBSD 4, bootstrapping against the
|
2802 |
|
|
latest FSF binutils is known to improve overall testsuite results; and,
|
2803 |
|
|
on FreeBSD/alpha, using binutils 2.14 or later is required to build libjava.
|
2804 |
|
|
|
2805 |
|
|
Support for FreeBSD 1 was discontinued in GCC 3.2.
|
2806 |
|
|
|
2807 |
|
|
Support for FreeBSD 2 will be discontinued after GCC 3.4. The
|
2808 |
|
|
following was true for GCC 3.1 but the current status is unknown.
|
2809 |
|
|
For FreeBSD 2 or any mutant a.out versions of FreeBSD 3: All
|
2810 |
|
|
configuration support and files as shipped with GCC 2.95 are still in
|
2811 |
|
|
place. FreeBSD 2.2.7 has been known to bootstrap completely; however,
|
2812 |
|
|
it is unknown which version of binutils was used (it is assumed that it
|
2813 |
|
|
was the system copy in @file{/usr/bin}) and C++ EH failures were noted.
|
2814 |
|
|
|
2815 |
|
|
For FreeBSD using the ELF file format: DWARF 2 debugging is now the
|
2816 |
|
|
default for all CPU architectures. It had been the default on
|
2817 |
|
|
FreeBSD/alpha since its inception. You may use @option{-gstabs} instead
|
2818 |
|
|
of @option{-g}, if you really want the old debugging format. There are
|
2819 |
|
|
no known issues with mixing object files and libraries with different
|
2820 |
|
|
debugging formats. Otherwise, this release of GCC should now match more
|
2821 |
|
|
of the configuration used in the stock FreeBSD configuration of GCC@. In
|
2822 |
|
|
particular, @option{--enable-threads} is now configured by default.
|
2823 |
|
|
However, as a general user, do not attempt to replace the system
|
2824 |
|
|
compiler with this release. Known to bootstrap and check with good
|
2825 |
|
|
results on FreeBSD 4.9-STABLE and 5-CURRENT@. In the past, known to
|
2826 |
|
|
bootstrap and check with good results on FreeBSD 3.0, 3.4, 4.0, 4.2,
|
2827 |
|
|
4.3, 4.4, 4.5, 4.8-STABLE@.
|
2828 |
|
|
|
2829 |
|
|
In principle, @option{--enable-threads} is now compatible with
|
2830 |
|
|
@option{--enable-libgcj} on FreeBSD@. However, it has only been built
|
2831 |
|
|
and tested on @samp{i386-*-freebsd[45]} and @samp{alpha-*-freebsd[45]}.
|
2832 |
|
|
The static
|
2833 |
|
|
library may be incorrectly built (symbols are missing at link time).
|
2834 |
|
|
There is a rare timing-based startup hang (probably involves an
|
2835 |
|
|
assumption about the thread library). Multi-threaded boehm-gc (required for
|
2836 |
|
|
libjava) exposes severe threaded signal-handling bugs on FreeBSD before
|
2837 |
|
|
4.5-RELEASE@. Other CPU architectures
|
2838 |
|
|
supported by FreeBSD will require additional configuration tuning in, at
|
2839 |
|
|
the very least, both boehm-gc and libffi.
|
2840 |
|
|
|
2841 |
|
|
Shared @file{libgcc_s.so} is now built and installed by default.
|
2842 |
|
|
|
2843 |
|
|
@html
|
2844 |
|
|
<hr />
|
2845 |
|
|
@end html
|
2846 |
|
|
@heading @anchor{h8300-hms}h8300-hms
|
2847 |
|
|
Renesas H8/300 series of processors.
|
2848 |
|
|
|
2849 |
|
|
Please have a look at the @uref{binaries.html,,binaries page}.
|
2850 |
|
|
|
2851 |
|
|
The calling convention and structure layout has changed in release 2.6.
|
2852 |
|
|
All code must be recompiled. The calling convention now passes the
|
2853 |
|
|
first three arguments in function calls in registers. Structures are no
|
2854 |
|
|
longer a multiple of 2 bytes.
|
2855 |
|
|
|
2856 |
|
|
@html
|
2857 |
|
|
<hr />
|
2858 |
|
|
@end html
|
2859 |
|
|
@heading @anchor{hppa-hp-hpux}hppa*-hp-hpux*
|
2860 |
|
|
Support for HP-UX version 9 and older was discontinued in GCC 3.4.
|
2861 |
|
|
|
2862 |
|
|
We require using gas/binutils on all hppa platforms;
|
2863 |
|
|
you may encounter a variety of problems if you try to use the HP assembler.
|
2864 |
|
|
|
2865 |
|
|
Specifically, @option{-g} does not work on HP-UX (since that system
|
2866 |
|
|
uses a peculiar debugging format which GCC does not know about), unless
|
2867 |
|
|
you use GAS and GDB@. It may be helpful to configure GCC with the
|
2868 |
|
|
@uref{./configure.html#with-gnu-as,,@option{--with-gnu-as}} and
|
2869 |
|
|
@option{--with-as=@dots{}} options to ensure that GCC can find GAS@.
|
2870 |
|
|
|
2871 |
|
|
If you wish to use the pa-risc 2.0 architecture support with a 32-bit
|
2872 |
|
|
runtime, you must use gas/binutils 2.11 or newer.
|
2873 |
|
|
|
2874 |
|
|
There are two default scheduling models for instructions. These are
|
2875 |
|
|
PROCESSOR_7100LC and PROCESSOR_8000. They are selected from the pa-risc
|
2876 |
|
|
architecture specified for the target machine when configuring.
|
2877 |
|
|
PROCESSOR_8000 is the default. PROCESSOR_7100LC is selected when
|
2878 |
|
|
the target is a @samp{hppa1*} machine.
|
2879 |
|
|
|
2880 |
|
|
The PROCESSOR_8000 model is not well suited to older processors. Thus,
|
2881 |
|
|
it is important to completely specify the machine architecture when
|
2882 |
|
|
configuring if you want a model other than PROCESSOR_8000. The macro
|
2883 |
|
|
TARGET_SCHED_DEFAULT can be defined in BOOT_CFLAGS if a different
|
2884 |
|
|
default scheduling model is desired.
|
2885 |
|
|
|
2886 |
|
|
As of GCC 4.0, GCC uses the UNIX 95 namespace for HP-UX 10.10
|
2887 |
|
|
through 11.00, and the UNIX 98 namespace for HP-UX 11.11 and later.
|
2888 |
|
|
This namespace change might cause problems when bootstrapping with
|
2889 |
|
|
an earlier version of GCC or the HP compiler as essentially the same
|
2890 |
|
|
namespace is required for an entire build. This problem can be avoided
|
2891 |
|
|
in a number of ways. With HP cc, @env{UNIX_STD} can be set to @samp{95}
|
2892 |
|
|
or @samp{98}. Another way is to add an appropriate set of predefines
|
2893 |
|
|
to @env{CC}. The description for the @option{munix=} option contains
|
2894 |
|
|
a list of the predefines used with each standard.
|
2895 |
|
|
|
2896 |
|
|
As of GCC 4.1, @env{DWARF2} exception handling is available on HP-UX.
|
2897 |
|
|
It is now the default. This exposed a bug in the handling of data
|
2898 |
|
|
relocations in the GAS assembler. The handling of 64-bit data relocations
|
2899 |
|
|
was seriously broken, affecting debugging and exception support on all
|
2900 |
|
|
@samp{hppa64-*-*} targets. Under some circumstances, 32-bit data relocations
|
2901 |
|
|
could also be handled incorrectly. This problem is fixed in GAS version
|
2902 |
|
|
2.16.91 20051125.
|
2903 |
|
|
|
2904 |
|
|
GCC versions prior to 4.1 incorrectly passed and returned complex
|
2905 |
|
|
values. They are now passed in the same manner as aggregates.
|
2906 |
|
|
|
2907 |
|
|
More specific information to @samp{hppa*-hp-hpux*} targets follows.
|
2908 |
|
|
|
2909 |
|
|
@html
|
2910 |
|
|
<hr />
|
2911 |
|
|
@end html
|
2912 |
|
|
@heading @anchor{hppa-hp-hpux10}hppa*-hp-hpux10
|
2913 |
|
|
|
2914 |
|
|
For hpux10.20, we @emph{highly} recommend you pick up the latest sed patch
|
2915 |
|
|
@code{PHCO_19798} from HP@. HP has two sites which provide patches free of
|
2916 |
|
|
charge:
|
2917 |
|
|
|
2918 |
|
|
@itemize @bullet
|
2919 |
|
|
@item
|
2920 |
|
|
@html
|
2921 |
|
|
<a href="http://us.itrc.hp.com/service/home/home.do">US, Canada, Asia-Pacific, and
|
2922 |
|
|
Latin-America</a>
|
2923 |
|
|
@end html
|
2924 |
|
|
@ifnothtml
|
2925 |
|
|
@uref{http://us.itrc.hp.com/service/home/home.do,,} US, Canada, Asia-Pacific,
|
2926 |
|
|
and Latin-America.
|
2927 |
|
|
@end ifnothtml
|
2928 |
|
|
@item
|
2929 |
|
|
@uref{http://europe.itrc.hp.com/service/home/home.do,,} Europe.
|
2930 |
|
|
@end itemize
|
2931 |
|
|
|
2932 |
|
|
The HP assembler on these systems has some problems. Most notably the
|
2933 |
|
|
assembler inserts timestamps into each object file it creates, causing
|
2934 |
|
|
the 3-stage comparison test to fail during a bootstrap.
|
2935 |
|
|
You should be able to continue by saying @samp{make all-host all-target}
|
2936 |
|
|
after getting the failure from @samp{make}.
|
2937 |
|
|
|
2938 |
|
|
GCC 4.0 requires CVS binutils as of April 28, 2004 or later. Earlier
|
2939 |
|
|
versions require binutils 2.8 or later.
|
2940 |
|
|
|
2941 |
|
|
The C++ ABI has changed incompatibly in GCC 4.0. COMDAT subspaces are
|
2942 |
|
|
used for one-only code and data. This resolves many of the previous
|
2943 |
|
|
problems in using C++ on this target. However, the ABI is not compatible
|
2944 |
|
|
with the one implemented under HP-UX 11 using secondary definitions.
|
2945 |
|
|
|
2946 |
|
|
@html
|
2947 |
|
|
<hr />
|
2948 |
|
|
@end html
|
2949 |
|
|
@heading @anchor{hppa-hp-hpux11}hppa*-hp-hpux11
|
2950 |
|
|
|
2951 |
|
|
GCC 3.0 and up support HP-UX 11. GCC 2.95.x is not supported and cannot
|
2952 |
|
|
be used to compile GCC 3.0 and up.
|
2953 |
|
|
|
2954 |
|
|
Refer to @uref{binaries.html,,binaries} for information about obtaining
|
2955 |
|
|
precompiled GCC binaries for HP-UX@. Precompiled binaries must be obtained
|
2956 |
|
|
to build the Ada language as it can't be bootstrapped using C@. Ada is
|
2957 |
|
|
only available for the 32-bit PA-RISC runtime. The libffi and libjava
|
2958 |
|
|
haven't been ported to HP-UX and don't build.
|
2959 |
|
|
|
2960 |
|
|
Starting with GCC 3.4 an ISO C compiler is required to bootstrap. The
|
2961 |
|
|
bundled compiler supports only traditional C; you will need either HP's
|
2962 |
|
|
unbundled compiler, or a binary distribution of GCC@.
|
2963 |
|
|
|
2964 |
|
|
It is possible to build GCC 3.3 starting with the bundled HP compiler,
|
2965 |
|
|
but the process requires several steps. GCC 3.3 can then be used to
|
2966 |
|
|
build later versions. The fastjar program contains ISO C code and
|
2967 |
|
|
can't be built with the HP bundled compiler. This problem can be
|
2968 |
|
|
avoided by not building the Java language. For example, use the
|
2969 |
|
|
@option{--enable-languages="c,c++,f77,objc"} option in your configure
|
2970 |
|
|
command.
|
2971 |
|
|
|
2972 |
|
|
There are several possible approaches to building the distribution.
|
2973 |
|
|
Binutils can be built first using the HP tools. Then, the GCC
|
2974 |
|
|
distribution can be built. The second approach is to build GCC
|
2975 |
|
|
first using the HP tools, then build binutils, then rebuild GCC@.
|
2976 |
|
|
There have been problems with various binary distributions, so it
|
2977 |
|
|
is best not to start from a binary distribution.
|
2978 |
|
|
|
2979 |
|
|
On 64-bit capable systems, there are two distinct targets. Different
|
2980 |
|
|
installation prefixes must be used if both are to be installed on
|
2981 |
|
|
the same system. The @samp{hppa[1-2]*-hp-hpux11*} target generates code
|
2982 |
|
|
for the 32-bit PA-RISC runtime architecture and uses the HP linker.
|
2983 |
|
|
The @samp{hppa64-hp-hpux11*} target generates 64-bit code for the
|
2984 |
|
|
PA-RISC 2.0 architecture. The HP and GNU linkers are both supported
|
2985 |
|
|
for this target.
|
2986 |
|
|
|
2987 |
|
|
The script config.guess now selects the target type based on the compiler
|
2988 |
|
|
detected during configuration. You must define @env{PATH} or @env{CC} so
|
2989 |
|
|
that configure finds an appropriate compiler for the initial bootstrap.
|
2990 |
|
|
When @env{CC} is used, the definition should contain the options that are
|
2991 |
|
|
needed whenever @env{CC} is used.
|
2992 |
|
|
|
2993 |
|
|
Specifically, options that determine the runtime architecture must be
|
2994 |
|
|
in @env{CC} to correctly select the target for the build. It is also
|
2995 |
|
|
convenient to place many other compiler options in @env{CC}. For example,
|
2996 |
|
|
@env{CC="cc -Ac +DA2.0W -Wp,-H16376 -D_CLASSIC_TYPES -D_HPUX_SOURCE"}
|
2997 |
|
|
can be used to bootstrap the GCC 3.3 branch with the HP compiler in
|
2998 |
|
|
64-bit K&R/bundled mode. The @option{+DA2.0W} option will result in
|
2999 |
|
|
the automatic selection of the @samp{hppa64-hp-hpux11*} target. The
|
3000 |
|
|
macro definition table of cpp needs to be increased for a successful
|
3001 |
|
|
build with the HP compiler. _CLASSIC_TYPES and _HPUX_SOURCE need to
|
3002 |
|
|
be defined when building with the bundled compiler, or when using the
|
3003 |
|
|
@option{-Ac} option. These defines aren't necessary with @option{-Ae}.
|
3004 |
|
|
|
3005 |
|
|
It is best to explicitly configure the @samp{hppa64-hp-hpux11*} target
|
3006 |
|
|
with the @option{--with-ld=@dots{}} option. This overrides the standard
|
3007 |
|
|
search for ld. The two linkers supported on this target require different
|
3008 |
|
|
commands. The default linker is determined during configuration. As a
|
3009 |
|
|
result, it's not possible to switch linkers in the middle of a GCC build.
|
3010 |
|
|
This has been been reported to sometimes occur in unified builds of
|
3011 |
|
|
binutils and GCC@.
|
3012 |
|
|
|
3013 |
|
|
GCC 3.0 through 3.2 require binutils 2.11 or above. GCC 3.3 through
|
3014 |
|
|
GCC 4.0 require binutils 2.14 or later.
|
3015 |
|
|
|
3016 |
|
|
Although the HP assembler can be used for an initial build, it shouldn't
|
3017 |
|
|
be used with any languages other than C and perhaps Fortran due to its
|
3018 |
|
|
many limitations. For example, it does not support weak symbols or alias
|
3019 |
|
|
definitions. As a result, explicit template instantiations are required
|
3020 |
|
|
when using C++. This makes it difficult if not impossible to build many
|
3021 |
|
|
C++ applications. You can't generate debugging information when using
|
3022 |
|
|
the HP assembler. Finally, bootstrapping fails in the final
|
3023 |
|
|
comparison of object modules due to the time stamps that it inserts into
|
3024 |
|
|
the modules. The bootstrap can be continued from this point with
|
3025 |
|
|
@samp{make all-host all-target}.
|
3026 |
|
|
|
3027 |
|
|
A recent linker patch must be installed for the correct operation of
|
3028 |
|
|
GCC 3.3 and later. @code{PHSS_26559} and @code{PHSS_24304} are the
|
3029 |
|
|
oldest linker patches that are known to work. They are for HP-UX
|
3030 |
|
|
11.00 and 11.11, respectively. @code{PHSS_24303}, the companion to
|
3031 |
|
|
@code{PHSS_24304}, might be usable but it hasn't been tested. These
|
3032 |
|
|
patches have been superseded. Consult the HP patch database to obtain
|
3033 |
|
|
the currently recommended linker patch for your system.
|
3034 |
|
|
|
3035 |
|
|
The patches are necessary for the support of weak symbols on the
|
3036 |
|
|
32-bit port, and for the running of initializers and finalizers. Weak
|
3037 |
|
|
symbols are implemented using SOM secondary definition symbols. Prior
|
3038 |
|
|
to HP-UX 11, there are bugs in the linker support for secondary symbols.
|
3039 |
|
|
The patches correct a problem of linker core dumps creating shared
|
3040 |
|
|
libraries containing secondary symbols, as well as various other
|
3041 |
|
|
linking issues involving secondary symbols.
|
3042 |
|
|
|
3043 |
|
|
GCC 3.3 uses the ELF DT_INIT_ARRAY and DT_FINI_ARRAY capabilities to
|
3044 |
|
|
run initializers and finalizers on the 64-bit port. The 32-bit port
|
3045 |
|
|
uses the linker @option{+init} and @option{+fini} options for the same
|
3046 |
|
|
purpose. The patches correct various problems with the +init/+fini
|
3047 |
|
|
options, including program core dumps. Binutils 2.14 corrects a
|
3048 |
|
|
problem on the 64-bit port resulting from HP's non-standard use of
|
3049 |
|
|
the .init and .fini sections for array initializers and finalizers.
|
3050 |
|
|
|
3051 |
|
|
There are a number of issues to consider in selecting which linker to
|
3052 |
|
|
use with the 64-bit port. The GNU 64-bit linker can only create dynamic
|
3053 |
|
|
binaries. The @option{-static} option causes linking with archive
|
3054 |
|
|
libraries but doesn't produce a truly static binary. Dynamic binaries
|
3055 |
|
|
still require final binding by the dynamic loader to resolve a set of
|
3056 |
|
|
dynamic-loader-defined symbols. The default behavior of the HP linker
|
3057 |
|
|
is the same as the GNU linker. However, it can generate true 64-bit
|
3058 |
|
|
static binaries using the @option{+compat} option.
|
3059 |
|
|
|
3060 |
|
|
The HP 64-bit linker doesn't support linkonce semantics. As a
|
3061 |
|
|
result, C++ programs have many more sections than they should.
|
3062 |
|
|
|
3063 |
|
|
The GNU 64-bit linker has some issues with shared library support
|
3064 |
|
|
and exceptions. As a result, we only support libgcc in archive
|
3065 |
|
|
format. For similar reasons, dwarf2 unwind and exception support
|
3066 |
|
|
are disabled. The GNU linker also has problems creating binaries
|
3067 |
|
|
with @option{-static}. It doesn't provide stubs for internal
|
3068 |
|
|
calls to global functions in shared libraries, so these calls
|
3069 |
|
|
can't be overloaded.
|
3070 |
|
|
|
3071 |
|
|
Thread support is not implemented in GCC 3.0 through 3.2, so the
|
3072 |
|
|
@option{--enable-threads} configure option does not work. In 3.3
|
3073 |
|
|
and later, POSIX threads are supported. The optional DCE thread
|
3074 |
|
|
library is not supported.
|
3075 |
|
|
|
3076 |
|
|
This port still is undergoing significant development.
|
3077 |
|
|
|
3078 |
|
|
@html
|
3079 |
|
|
<hr />
|
3080 |
|
|
@end html
|
3081 |
|
|
@heading @anchor{x-x-linux-gnu}*-*-linux-gnu
|
3082 |
|
|
|
3083 |
|
|
Versions of libstdc++-v3 starting with 3.2.1 require bugfixes present
|
3084 |
|
|
in glibc 2.2.5 and later. More information is available in the
|
3085 |
|
|
libstdc++-v3 documentation.
|
3086 |
|
|
|
3087 |
|
|
@html
|
3088 |
|
|
<hr />
|
3089 |
|
|
@end html
|
3090 |
|
|
@heading @anchor{ix86-x-linuxaout}i?86-*-linux*aout
|
3091 |
|
|
Use this configuration to generate @file{a.out} binaries on Linux-based
|
3092 |
|
|
GNU systems. This configuration is being superseded.
|
3093 |
|
|
|
3094 |
|
|
@html
|
3095 |
|
|
<hr />
|
3096 |
|
|
@end html
|
3097 |
|
|
@heading @anchor{ix86-x-linux}i?86-*-linux*
|
3098 |
|
|
|
3099 |
|
|
As of GCC 3.3, binutils 2.13.1 or later is required for this platform.
|
3100 |
|
|
See @uref{http://gcc.gnu.org/PR10877,,bug 10877} for more information.
|
3101 |
|
|
|
3102 |
|
|
If you receive Signal 11 errors when building on GNU/Linux, then it is
|
3103 |
|
|
possible you have a hardware problem. Further information on this can be
|
3104 |
|
|
found on @uref{http://www.bitwizard.nl/sig11/,,www.bitwizard.nl}.
|
3105 |
|
|
|
3106 |
|
|
@html
|
3107 |
|
|
<hr />
|
3108 |
|
|
@end html
|
3109 |
|
|
@heading @anchor{ix86-x-sco32v5}i?86-*-sco3.2v5*
|
3110 |
|
|
Use this for the SCO OpenServer Release 5 family of operating systems.
|
3111 |
|
|
|
3112 |
|
|
Unlike earlier versions of GCC, the ability to generate COFF with this
|
3113 |
|
|
target is no longer provided.
|
3114 |
|
|
|
3115 |
|
|
Earlier versions of GCC emitted DWARF 1 when generating ELF to allow
|
3116 |
|
|
the system debugger to be used. That support was too burdensome to
|
3117 |
|
|
maintain. GCC now emits only DWARF 2 for this target. This means you
|
3118 |
|
|
may use either the UDK debugger or GDB to debug programs built by this
|
3119 |
|
|
version of GCC@.
|
3120 |
|
|
|
3121 |
|
|
GCC is now only supported on releases 5.0.4 and later, and requires that
|
3122 |
|
|
you install Support Level Supplement OSS646B or later, and Support Level
|
3123 |
|
|
Supplement OSS631C or later. If you are using release 5.0.7 of
|
3124 |
|
|
OpenServer, you must have at least the first maintenance pack installed
|
3125 |
|
|
(this includes the relevant portions of OSS646). OSS646, also known as
|
3126 |
|
|
the ``Execution Environment Update'', provides updated link editors and
|
3127 |
|
|
assemblers, as well as updated standard C and math libraries. The C
|
3128 |
|
|
startup modules are also updated to support the System V gABI draft, and
|
3129 |
|
|
GCC relies on that behavior. OSS631 provides a collection of commonly
|
3130 |
|
|
used open source libraries, some of which GCC depends on (such as GNU
|
3131 |
|
|
gettext and zlib). SCO OpenServer Release 5.0.7 has all of this built
|
3132 |
|
|
in by default, but OSS631C and later also apply to that release. Please
|
3133 |
|
|
visit
|
3134 |
|
|
@uref{ftp://ftp.sco.com/pub/openserver5,,ftp://ftp.sco.com/pub/openserver5}
|
3135 |
|
|
for the latest versions of these (and other potentially useful)
|
3136 |
|
|
supplements.
|
3137 |
|
|
|
3138 |
|
|
Although there is support for using the native assembler, it is
|
3139 |
|
|
recommended that you configure GCC to use the GNU assembler. You do
|
3140 |
|
|
this by using the flags
|
3141 |
|
|
@uref{./configure.html#with-gnu-as,,@option{--with-gnu-as}}. You should
|
3142 |
|
|
use a modern version of GNU binutils. Version 2.13.2.1 was used for all
|
3143 |
|
|
testing. In general, only the @option{--with-gnu-as} option is tested.
|
3144 |
|
|
A modern bintuils (as well as a plethora of other development related
|
3145 |
|
|
GNU utilities) can be found in Support Level Supplement OSS658A, the
|
3146 |
|
|
``GNU Development Tools'' package. See the SCO web and ftp sites for details.
|
3147 |
|
|
That package also contains the currently ``officially supported'' version of
|
3148 |
|
|
GCC, version 2.95.3. It is useful for bootstrapping this version.
|
3149 |
|
|
|
3150 |
|
|
@html
|
3151 |
|
|
<hr />
|
3152 |
|
|
@end html
|
3153 |
|
|
@heading @anchor{ix86-x-solaris210}i?86-*-solaris2.10
|
3154 |
|
|
Use this for Solaris 10 or later on x86 and x86-64 systems. This
|
3155 |
|
|
configuration is supported by GCC 4.0 and later versions only.
|
3156 |
|
|
|
3157 |
|
|
It is recommended that you configure GCC to use the GNU assembler in
|
3158 |
|
|
@file{/usr/sfw/bin/gas} but the Sun linker, using the options
|
3159 |
|
|
@option{--with-gnu-as --with-as=/usr/sfw/bin/gas --without-gnu-ld
|
3160 |
|
|
--with-ld=/usr/ccs/bin/ld}.
|
3161 |
|
|
|
3162 |
|
|
@html
|
3163 |
|
|
<hr />
|
3164 |
|
|
@end html
|
3165 |
|
|
@heading @anchor{ix86-x-udk}i?86-*-udk
|
3166 |
|
|
|
3167 |
|
|
This target emulates the SCO Universal Development Kit and requires that
|
3168 |
|
|
package be installed. (If it is installed, you will have a
|
3169 |
|
|
@file{/udk/usr/ccs/bin/cc} file present.) It's very much like the
|
3170 |
|
|
@samp{i?86-*-unixware7*} target
|
3171 |
|
|
but is meant to be used when hosting on a system where UDK isn't the
|
3172 |
|
|
default compiler such as OpenServer 5 or Unixware 2. This target will
|
3173 |
|
|
generate binaries that will run on OpenServer, Unixware 2, or Unixware 7,
|
3174 |
|
|
with the same warnings and caveats as the SCO UDK@.
|
3175 |
|
|
|
3176 |
|
|
This target is a little tricky to build because we have to distinguish
|
3177 |
|
|
it from the native tools (so it gets headers, startups, and libraries
|
3178 |
|
|
from the right place) while making the tools not think we're actually
|
3179 |
|
|
building a cross compiler. The easiest way to do this is with a configure
|
3180 |
|
|
command like this:
|
3181 |
|
|
|
3182 |
|
|
@smallexample
|
3183 |
|
|
CC=/udk/usr/ccs/bin/cc @var{/your/path/to}/gcc/configure \
|
3184 |
|
|
--host=i686-pc-udk --target=i686-pc-udk --program-prefix=udk-
|
3185 |
|
|
@end smallexample
|
3186 |
|
|
|
3187 |
|
|
@emph{You should substitute @samp{i686} in the above command with the appropriate
|
3188 |
|
|
processor for your host.}
|
3189 |
|
|
|
3190 |
|
|
After the usual @samp{make} and
|
3191 |
|
|
@samp{make install}, you can then access the UDK-targeted GCC
|
3192 |
|
|
tools by adding @command{udk-} before the commonly known name. For
|
3193 |
|
|
example, to invoke the C compiler, you would use @command{udk-gcc}.
|
3194 |
|
|
They will coexist peacefully with any native-target GCC tools you may
|
3195 |
|
|
have installed.
|
3196 |
|
|
|
3197 |
|
|
|
3198 |
|
|
@html
|
3199 |
|
|
<hr />
|
3200 |
|
|
@end html
|
3201 |
|
|
@heading @anchor{ia64-x-linux}ia64-*-linux
|
3202 |
|
|
IA-64 processor (also known as IPF, or Itanium Processor Family)
|
3203 |
|
|
running GNU/Linux.
|
3204 |
|
|
|
3205 |
|
|
If you are using the installed system libunwind library with
|
3206 |
|
|
@option{--with-system-libunwind}, then you must use libunwind 0.98 or
|
3207 |
|
|
later.
|
3208 |
|
|
|
3209 |
|
|
None of the following versions of GCC has an ABI that is compatible
|
3210 |
|
|
with any of the other versions in this list, with the exception that
|
3211 |
|
|
Red Hat 2.96 and Trillian 000171 are compatible with each other:
|
3212 |
|
|
3.1, 3.0.2, 3.0.1, 3.0, Red Hat 2.96, and Trillian 000717.
|
3213 |
|
|
This primarily affects C++ programs and programs that create shared libraries.
|
3214 |
|
|
GCC 3.1 or later is recommended for compiling linux, the kernel.
|
3215 |
|
|
As of version 3.1 GCC is believed to be fully ABI compliant, and hence no
|
3216 |
|
|
more major ABI changes are expected.
|
3217 |
|
|
|
3218 |
|
|
@html
|
3219 |
|
|
<hr />
|
3220 |
|
|
@end html
|
3221 |
|
|
@heading @anchor{ia64-x-hpux}ia64-*-hpux*
|
3222 |
|
|
Building GCC on this target requires the GNU Assembler. The bundled HP
|
3223 |
|
|
assembler will not work. To prevent GCC from using the wrong assembler,
|
3224 |
|
|
the option @option{--with-gnu-as} may be necessary.
|
3225 |
|
|
|
3226 |
|
|
The GCC libunwind library has not been ported to HPUX@. This means that for
|
3227 |
|
|
GCC versions 3.2.3 and earlier, @option{--enable-libunwind-exceptions}
|
3228 |
|
|
is required to build GCC@. For GCC 3.3 and later, this is the default.
|
3229 |
|
|
For gcc 3.4.3 and later, @option{--enable-libunwind-exceptions} is
|
3230 |
|
|
removed and the system libunwind library will always be used.
|
3231 |
|
|
|
3232 |
|
|
@html
|
3233 |
|
|
<hr />
|
3234 |
|
|
<!-- rs6000-ibm-aix*, powerpc-ibm-aix* -->
|
3235 |
|
|
@end html
|
3236 |
|
|
@heading @anchor{x-ibm-aix}*-ibm-aix*
|
3237 |
|
|
Support for AIX version 3 and older was discontinued in GCC 3.4.
|
3238 |
|
|
|
3239 |
|
|
``out of memory'' bootstrap failures may indicate a problem with
|
3240 |
|
|
process resource limits (ulimit). Hard limits are configured in the
|
3241 |
|
|
@file{/etc/security/limits} system configuration file.
|
3242 |
|
|
|
3243 |
|
|
To speed up the configuration phases of bootstrapping and installing GCC,
|
3244 |
|
|
one may use GNU Bash instead of AIX @command{/bin/sh}, e.g.,
|
3245 |
|
|
|
3246 |
|
|
@smallexample
|
3247 |
|
|
% CONFIG_SHELL=/opt/freeware/bin/bash
|
3248 |
|
|
% export CONFIG_SHELL
|
3249 |
|
|
@end smallexample
|
3250 |
|
|
|
3251 |
|
|
and then proceed as described in @uref{build.html,,the build
|
3252 |
|
|
instructions}, where we strongly recommend specifying an absolute path
|
3253 |
|
|
to invoke @var{srcdir}/configure.
|
3254 |
|
|
|
3255 |
|
|
Because GCC on AIX is built as a 32-bit executable by default,
|
3256 |
|
|
(although it can generate 64-bit programs) the GMP and MPFR libraries
|
3257 |
|
|
required by gfortran must be 32-bit libraries. Building GMP and MPFR
|
3258 |
|
|
as static archive libraries works better than shared libraries.
|
3259 |
|
|
|
3260 |
|
|
Errors involving @code{alloca} when building GCC generally are due
|
3261 |
|
|
to an incorrect definition of @code{CC} in the Makefile or mixing files
|
3262 |
|
|
compiled with the native C compiler and GCC@. During the stage1 phase of
|
3263 |
|
|
the build, the native AIX compiler @strong{must} be invoked as @command{cc}
|
3264 |
|
|
(not @command{xlc}). Once @command{configure} has been informed of
|
3265 |
|
|
@command{xlc}, one needs to use @samp{make distclean} to remove the
|
3266 |
|
|
configure cache files and ensure that @env{CC} environment variable
|
3267 |
|
|
does not provide a definition that will confuse @command{configure}.
|
3268 |
|
|
If this error occurs during stage2 or later, then the problem most likely
|
3269 |
|
|
is the version of Make (see above).
|
3270 |
|
|
|
3271 |
|
|
The native @command{as} and @command{ld} are recommended for bootstrapping
|
3272 |
|
|
on AIX 4 and required for bootstrapping on AIX 5L@. The GNU Assembler
|
3273 |
|
|
reports that it supports WEAK symbols on AIX 4, which causes GCC to try to
|
3274 |
|
|
utilize weak symbol functionality although it is not supported. The GNU
|
3275 |
|
|
Assembler and Linker do not support AIX 5L sufficiently to bootstrap GCC@.
|
3276 |
|
|
The native AIX tools do interoperate with GCC@.
|
3277 |
|
|
|
3278 |
|
|
Building @file{libstdc++.a} requires a fix for an AIX Assembler bug
|
3279 |
|
|
APAR IY26685 (AIX 4.3) or APAR IY25528 (AIX 5.1). It also requires a
|
3280 |
|
|
fix for another AIX Assembler bug and a co-dependent AIX Archiver fix
|
3281 |
|
|
referenced as APAR IY53606 (AIX 5.2) or a APAR IY54774 (AIX 5.1)
|
3282 |
|
|
|
3283 |
|
|
@samp{libstdc++} in GCC 3.4 increments the major version number of the
|
3284 |
|
|
shared object and GCC installation places the @file{libstdc++.a}
|
3285 |
|
|
shared library in a common location which will overwrite the and GCC
|
3286 |
|
|
3.3 version of the shared library. Applications either need to be
|
3287 |
|
|
re-linked against the new shared library or the GCC 3.1 and GCC 3.3
|
3288 |
|
|
versions of the @samp{libstdc++} shared object needs to be available
|
3289 |
|
|
to the AIX runtime loader. The GCC 3.1 @samp{libstdc++.so.4}, if
|
3290 |
|
|
present, and GCC 3.3 @samp{libstdc++.so.5} shared objects can be
|
3291 |
|
|
installed for runtime dynamic loading using the following steps to set
|
3292 |
|
|
the @samp{F_LOADONLY} flag in the shared object for @emph{each}
|
3293 |
|
|
multilib @file{libstdc++.a} installed:
|
3294 |
|
|
|
3295 |
|
|
Extract the shared objects from the currently installed
|
3296 |
|
|
@file{libstdc++.a} archive:
|
3297 |
|
|
@smallexample
|
3298 |
|
|
% ar -x libstdc++.a libstdc++.so.4 libstdc++.so.5
|
3299 |
|
|
@end smallexample
|
3300 |
|
|
|
3301 |
|
|
Enable the @samp{F_LOADONLY} flag so that the shared object will be
|
3302 |
|
|
available for runtime dynamic loading, but not linking:
|
3303 |
|
|
@smallexample
|
3304 |
|
|
% strip -e libstdc++.so.4 libstdc++.so.5
|
3305 |
|
|
@end smallexample
|
3306 |
|
|
|
3307 |
|
|
Archive the runtime-only shared object in the GCC 3.4
|
3308 |
|
|
@file{libstdc++.a} archive:
|
3309 |
|
|
@smallexample
|
3310 |
|
|
% ar -q libstdc++.a libstdc++.so.4 libstdc++.so.5
|
3311 |
|
|
@end smallexample
|
3312 |
|
|
|
3313 |
|
|
Linking executables and shared libraries may produce warnings of
|
3314 |
|
|
duplicate symbols. The assembly files generated by GCC for AIX always
|
3315 |
|
|
have included multiple symbol definitions for certain global variable
|
3316 |
|
|
and function declarations in the original program. The warnings should
|
3317 |
|
|
not prevent the linker from producing a correct library or runnable
|
3318 |
|
|
executable.
|
3319 |
|
|
|
3320 |
|
|
AIX 4.3 utilizes a ``large format'' archive to support both 32-bit and
|
3321 |
|
|
64-bit object modules. The routines provided in AIX 4.3.0 and AIX 4.3.1
|
3322 |
|
|
to parse archive libraries did not handle the new format correctly.
|
3323 |
|
|
These routines are used by GCC and result in error messages during
|
3324 |
|
|
linking such as ``not a COFF file''. The version of the routines shipped
|
3325 |
|
|
with AIX 4.3.1 should work for a 32-bit environment. The @option{-g}
|
3326 |
|
|
option of the archive command may be used to create archives of 32-bit
|
3327 |
|
|
objects using the original ``small format''. A correct version of the
|
3328 |
|
|
routines is shipped with AIX 4.3.2 and above.
|
3329 |
|
|
|
3330 |
|
|
Some versions of the AIX binder (linker) can fail with a relocation
|
3331 |
|
|
overflow severe error when the @option{-bbigtoc} option is used to link
|
3332 |
|
|
GCC-produced object files into an executable that overflows the TOC@. A fix
|
3333 |
|
|
for APAR IX75823 (OVERFLOW DURING LINK WHEN USING GCC AND -BBIGTOC) is
|
3334 |
|
|
available from IBM Customer Support and from its
|
3335 |
|
|
@uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com}
|
3336 |
|
|
website as PTF U455193.
|
3337 |
|
|
|
3338 |
|
|
The AIX 4.3.2.1 linker (bos.rte.bind_cmds Level 4.3.2.1) will dump core
|
3339 |
|
|
with a segmentation fault when invoked by any version of GCC@. A fix for
|
3340 |
|
|
APAR IX87327 is available from IBM Customer Support and from its
|
3341 |
|
|
@uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com}
|
3342 |
|
|
website as PTF U461879. This fix is incorporated in AIX 4.3.3 and above.
|
3343 |
|
|
|
3344 |
|
|
The initial assembler shipped with AIX 4.3.0 generates incorrect object
|
3345 |
|
|
files. A fix for APAR IX74254 (64BIT DISASSEMBLED OUTPUT FROM COMPILER FAILS
|
3346 |
|
|
TO ASSEMBLE/BIND) is available from IBM Customer Support and from its
|
3347 |
|
|
@uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com}
|
3348 |
|
|
website as PTF U453956. This fix is incorporated in AIX 4.3.1 and above.
|
3349 |
|
|
|
3350 |
|
|
AIX provides National Language Support (NLS)@. Compilers and assemblers
|
3351 |
|
|
use NLS to support locale-specific representations of various data
|
3352 |
|
|
formats including floating-point numbers (e.g., @samp{.} vs @samp{,} for
|
3353 |
|
|
separating decimal fractions). There have been problems reported where
|
3354 |
|
|
GCC does not produce the same floating-point formats that the assembler
|
3355 |
|
|
expects. If one encounters this problem, set the @env{LANG}
|
3356 |
|
|
environment variable to @samp{C} or @samp{En_US}.
|
3357 |
|
|
|
3358 |
|
|
By default, GCC for AIX 4.1 and above produces code that can be used on
|
3359 |
|
|
both Power or PowerPC processors.
|
3360 |
|
|
|
3361 |
|
|
A default can be specified with the @option{-mcpu=@var{cpu_type}}
|
3362 |
|
|
switch and using the configure option @option{--with-cpu-@var{cpu_type}}.
|
3363 |
|
|
|
3364 |
|
|
@html
|
3365 |
|
|
<hr />
|
3366 |
|
|
@end html
|
3367 |
|
|
@heading @anchor{iq2000-x-elf}iq2000-*-elf
|
3368 |
|
|
Vitesse IQ2000 processors. These are used in embedded
|
3369 |
|
|
applications. There are no standard Unix configurations.
|
3370 |
|
|
|
3371 |
|
|
@html
|
3372 |
|
|
<hr />
|
3373 |
|
|
@end html
|
3374 |
|
|
@heading @anchor{m32c-x-elf}m32c-*-elf
|
3375 |
|
|
Renesas M32C processor.
|
3376 |
|
|
This configuration is intended for embedded systems.
|
3377 |
|
|
|
3378 |
|
|
@html
|
3379 |
|
|
<hr />
|
3380 |
|
|
@end html
|
3381 |
|
|
@heading @anchor{m32r-x-elf}m32r-*-elf
|
3382 |
|
|
Renesas M32R processor.
|
3383 |
|
|
This configuration is intended for embedded systems.
|
3384 |
|
|
|
3385 |
|
|
@html
|
3386 |
|
|
<hr />
|
3387 |
|
|
@end html
|
3388 |
|
|
@heading @anchor{m6811-elf}m6811-elf
|
3389 |
|
|
Motorola 68HC11 family micro controllers. These are used in embedded
|
3390 |
|
|
applications. There are no standard Unix configurations.
|
3391 |
|
|
|
3392 |
|
|
@html
|
3393 |
|
|
<hr />
|
3394 |
|
|
@end html
|
3395 |
|
|
@heading @anchor{m6812-elf}m6812-elf
|
3396 |
|
|
Motorola 68HC12 family micro controllers. These are used in embedded
|
3397 |
|
|
applications. There are no standard Unix configurations.
|
3398 |
|
|
|
3399 |
|
|
@html
|
3400 |
|
|
<hr />
|
3401 |
|
|
@end html
|
3402 |
|
|
@heading @anchor{m68k-hp-hpux}m68k-hp-hpux
|
3403 |
|
|
HP 9000 series 300 or 400 running HP-UX@. HP-UX version 8.0 has a bug in
|
3404 |
|
|
the assembler that prevents compilation of GCC@. This
|
3405 |
|
|
bug manifests itself during the first stage of compilation, while
|
3406 |
|
|
building @file{libgcc2.a}:
|
3407 |
|
|
|
3408 |
|
|
@smallexample
|
3409 |
|
|
_floatdisf
|
3410 |
|
|
cc1: warning: `-g' option not supported on this version of GCC
|
3411 |
|
|
cc1: warning: `-g1' option not supported on this version of GCC
|
3412 |
|
|
./xgcc: Internal compiler error: program as got fatal signal 11
|
3413 |
|
|
@end smallexample
|
3414 |
|
|
|
3415 |
|
|
A patched version of the assembler is available as the file
|
3416 |
|
|
@uref{ftp://altdorf.ai.mit.edu/archive/cph/hpux-8.0-assembler}. If you
|
3417 |
|
|
have HP software support, the patch can also be obtained directly from
|
3418 |
|
|
HP, as described in the following note:
|
3419 |
|
|
|
3420 |
|
|
@quotation
|
3421 |
|
|
This is the patched assembler, to patch SR#1653-010439, where the
|
3422 |
|
|
assembler aborts on floating point constants.
|
3423 |
|
|
|
3424 |
|
|
The bug is not really in the assembler, but in the shared library
|
3425 |
|
|
version of the function ``cvtnum(3c)''. The bug on ``cvtnum(3c)'' is
|
3426 |
|
|
SR#4701-078451. Anyway, the attached assembler uses the archive
|
3427 |
|
|
library version of ``cvtnum(3c)'' and thus does not exhibit the bug.
|
3428 |
|
|
@end quotation
|
3429 |
|
|
|
3430 |
|
|
This patch is also known as PHCO_4484.
|
3431 |
|
|
|
3432 |
|
|
In addition gdb does not understand that native HP-UX format, so
|
3433 |
|
|
you must use gas if you wish to use gdb.
|
3434 |
|
|
|
3435 |
|
|
On HP-UX version 8.05, but not on 8.07 or more recent versions, the
|
3436 |
|
|
@command{fixproto} shell script triggers a bug in the system shell. If you
|
3437 |
|
|
encounter this problem, upgrade your operating system or use BASH (the
|
3438 |
|
|
GNU shell) to run @command{fixproto}. This bug will cause the fixproto
|
3439 |
|
|
program to report an error of the form:
|
3440 |
|
|
|
3441 |
|
|
@smallexample
|
3442 |
|
|
./fixproto: sh internal 1K buffer overflow
|
3443 |
|
|
@end smallexample
|
3444 |
|
|
|
3445 |
|
|
To fix this, you can also change the first line of the fixproto script
|
3446 |
|
|
to look like:
|
3447 |
|
|
|
3448 |
|
|
@smallexample
|
3449 |
|
|
#!/bin/ksh
|
3450 |
|
|
@end smallexample
|
3451 |
|
|
|
3452 |
|
|
@html
|
3453 |
|
|
<hr />
|
3454 |
|
|
@end html
|
3455 |
|
|
@heading @anchor{mips-x-x}mips-*-*
|
3456 |
|
|
If on a MIPS system you get an error message saying ``does not have gp
|
3457 |
|
|
sections for all it's [sic] sectons [sic]'', don't worry about it. This
|
3458 |
|
|
happens whenever you use GAS with the MIPS linker, but there is not
|
3459 |
|
|
really anything wrong, and it is okay to use the output file. You can
|
3460 |
|
|
stop such warnings by installing the GNU linker.
|
3461 |
|
|
|
3462 |
|
|
It would be nice to extend GAS to produce the gp tables, but they are
|
3463 |
|
|
optional, and there should not be a warning about their absence.
|
3464 |
|
|
|
3465 |
|
|
The libstdc++ atomic locking routines for MIPS targets requires MIPS II
|
3466 |
|
|
and later. A patch went in just after the GCC 3.3 release to
|
3467 |
|
|
make @samp{mips*-*-*} use the generic implementation instead. You can also
|
3468 |
|
|
configure for @samp{mipsel-elf} as a workaround. The
|
3469 |
|
|
@samp{mips*-*-linux*} target continues to use the MIPS II routines. More
|
3470 |
|
|
work on this is expected in future releases.
|
3471 |
|
|
|
3472 |
|
|
MIPS systems check for division by zero (unless
|
3473 |
|
|
@option{-mno-check-zero-division} is passed to the compiler) by
|
3474 |
|
|
generating either a conditional trap or a break instruction. Using
|
3475 |
|
|
trap results in smaller code, but is only supported on MIPS II and
|
3476 |
|
|
later. Also, some versions of the Linux kernel have a bug that
|
3477 |
|
|
prevents trap from generating the proper signal (@code{SIGFPE}). To enable
|
3478 |
|
|
the use of break, use the @option{--with-divide=breaks}
|
3479 |
|
|
@command{configure} option when configuring GCC@. The default is to
|
3480 |
|
|
use traps on systems that support them.
|
3481 |
|
|
|
3482 |
|
|
Cross-compilers for the MIPS as target using the MIPS assembler
|
3483 |
|
|
currently do not work, because the auxiliary programs
|
3484 |
|
|
@file{mips-tdump.c} and @file{mips-tfile.c} can't be compiled on
|
3485 |
|
|
anything but a MIPS. It does work to cross compile for a MIPS
|
3486 |
|
|
if you use the GNU assembler and linker.
|
3487 |
|
|
|
3488 |
|
|
The assembler from GNU binutils 2.17 and earlier has a bug in the way
|
3489 |
|
|
it sorts relocations for REL targets (o32, o64, EABI). This can cause
|
3490 |
|
|
bad code to be generated for simple C++ programs. Also the linker
|
3491 |
|
|
from GNU binutils versions prior to 2.17 has a bug which causes the
|
3492 |
|
|
runtime linker stubs in very large programs, like @file{libgcj.so}, to
|
3493 |
|
|
be incorrectly generated. Binutils CVS snapshots and releases made
|
3494 |
|
|
after Nov. 9, 2006 are thought to be free from both of these problems.
|
3495 |
|
|
|
3496 |
|
|
@html
|
3497 |
|
|
<hr />
|
3498 |
|
|
@end html
|
3499 |
|
|
@heading @anchor{mips-sgi-irix5}mips-sgi-irix5
|
3500 |
|
|
|
3501 |
|
|
In order to compile GCC on an SGI running IRIX 5, the @samp{compiler_dev.hdr}
|
3502 |
|
|
subsystem must be installed from the IDO CD-ROM supplied by SGI@.
|
3503 |
|
|
It is also available for download from
|
3504 |
|
|
@uref{ftp://ftp.sgi.com/sgi/IRIX5.3/iris-development-option-5.3.tardist}.
|
3505 |
|
|
|
3506 |
|
|
If you use the MIPS C compiler to bootstrap, it may be necessary
|
3507 |
|
|
to increase its table size for switch statements with the
|
3508 |
|
|
@option{-Wf,-XNg1500} option. If you use the @option{-O2}
|
3509 |
|
|
optimization option, you also need to use @option{-Olimit 3000}.
|
3510 |
|
|
|
3511 |
|
|
To enable debugging under IRIX 5, you must use GNU binutils 2.15 or
|
3512 |
|
|
later, and use the @option{--with-gnu-ld} @command{configure} option
|
3513 |
|
|
when configuring GCC@. You need to use GNU @command{ar} and @command{nm},
|
3514 |
|
|
also distributed with GNU binutils.
|
3515 |
|
|
|
3516 |
|
|
Some users have reported that @command{/bin/sh} will hang during bootstrap.
|
3517 |
|
|
This problem can be avoided by running the commands:
|
3518 |
|
|
|
3519 |
|
|
@smallexample
|
3520 |
|
|
% CONFIG_SHELL=/bin/ksh
|
3521 |
|
|
% export CONFIG_SHELL
|
3522 |
|
|
@end smallexample
|
3523 |
|
|
|
3524 |
|
|
before starting the build.
|
3525 |
|
|
|
3526 |
|
|
@html
|
3527 |
|
|
<hr />
|
3528 |
|
|
@end html
|
3529 |
|
|
@heading @anchor{mips-sgi-irix6}mips-sgi-irix6
|
3530 |
|
|
|
3531 |
|
|
If you are using SGI's MIPSpro @command{cc} as your bootstrap compiler, you must
|
3532 |
|
|
ensure that the N32 ABI is in use. To test this, compile a simple C
|
3533 |
|
|
file with @command{cc} and then run @command{file} on the
|
3534 |
|
|
resulting object file. The output should look like:
|
3535 |
|
|
|
3536 |
|
|
@smallexample
|
3537 |
|
|
test.o: ELF N32 MSB @dots{}
|
3538 |
|
|
@end smallexample
|
3539 |
|
|
|
3540 |
|
|
If you see:
|
3541 |
|
|
|
3542 |
|
|
@smallexample
|
3543 |
|
|
test.o: ELF 32-bit MSB @dots{}
|
3544 |
|
|
@end smallexample
|
3545 |
|
|
|
3546 |
|
|
or
|
3547 |
|
|
|
3548 |
|
|
@smallexample
|
3549 |
|
|
test.o: ELF 64-bit MSB @dots{}
|
3550 |
|
|
@end smallexample
|
3551 |
|
|
|
3552 |
|
|
then your version of @command{cc} uses the O32 or N64 ABI by default. You
|
3553 |
|
|
should set the environment variable @env{CC} to @samp{cc -n32}
|
3554 |
|
|
before configuring GCC@.
|
3555 |
|
|
|
3556 |
|
|
If you want the resulting @command{gcc} to run on old 32-bit systems
|
3557 |
|
|
with the MIPS R4400 CPU, you need to ensure that only code for the @samp{mips3}
|
3558 |
|
|
instruction set architecture (ISA) is generated. While GCC 3.x does
|
3559 |
|
|
this correctly, both GCC 2.95 and SGI's MIPSpro @command{cc} may change
|
3560 |
|
|
the ISA depending on the machine where GCC is built. Using one of them
|
3561 |
|
|
as the bootstrap compiler may result in @samp{mips4} code, which won't run at
|
3562 |
|
|
all on @samp{mips3}-only systems. For the test program above, you should see:
|
3563 |
|
|
|
3564 |
|
|
@smallexample
|
3565 |
|
|
test.o: ELF N32 MSB mips-3 @dots{}
|
3566 |
|
|
@end smallexample
|
3567 |
|
|
|
3568 |
|
|
If you get:
|
3569 |
|
|
|
3570 |
|
|
@smallexample
|
3571 |
|
|
test.o: ELF N32 MSB mips-4 @dots{}
|
3572 |
|
|
@end smallexample
|
3573 |
|
|
|
3574 |
|
|
instead, you should set the environment variable @env{CC} to @samp{cc
|
3575 |
|
|
-n32 -mips3} or @samp{gcc -mips3} respectively before configuring GCC@.
|
3576 |
|
|
|
3577 |
|
|
MIPSpro C 7.4 may cause bootstrap failures, due to a bug when inlining
|
3578 |
|
|
@code{memcmp}. Either add @code{-U__INLINE_INTRINSICS} to the @env{CC}
|
3579 |
|
|
environment variable as a workaround or upgrade to MIPSpro C 7.4.1m.
|
3580 |
|
|
|
3581 |
|
|
GCC on IRIX 6 is usually built to support the N32, O32 and N64 ABIs. If
|
3582 |
|
|
you build GCC on a system that doesn't have the N64 libraries installed
|
3583 |
|
|
or cannot run 64-bit binaries,
|
3584 |
|
|
you need to configure with @option{--disable-multilib} so GCC doesn't
|
3585 |
|
|
try to use them. This will disable building the O32 libraries, too.
|
3586 |
|
|
Look for @file{/usr/lib64/libc.so.1} to see if you
|
3587 |
|
|
have the 64-bit libraries installed.
|
3588 |
|
|
|
3589 |
|
|
To enable debugging for the O32 ABI, you must use GNU @command{as} from
|
3590 |
|
|
GNU binutils 2.15 or later. You may also use GNU @command{ld}, but
|
3591 |
|
|
this is not required and currently causes some problems with Ada.
|
3592 |
|
|
|
3593 |
|
|
The @option{--enable-threads} option doesn't currently work, a patch is
|
3594 |
|
|
in preparation for a future release. The @option{--enable-libgcj}
|
3595 |
|
|
option is disabled by default: IRIX 6 uses a very low default limit
|
3596 |
|
|
(20480) for the command line length. Although @command{libtool} contains a
|
3597 |
|
|
workaround for this problem, at least the N64 @samp{libgcj} is known not
|
3598 |
|
|
to build despite this, running into an internal error of the native
|
3599 |
|
|
@command{ld}. A sure fix is to increase this limit (@samp{ncargs}) to
|
3600 |
|
|
its maximum of 262144 bytes. If you have root access, you can use the
|
3601 |
|
|
@command{systune} command to do this.
|
3602 |
|
|
|
3603 |
|
|
@code{wchar_t} support in @samp{libstdc++} is not available for old
|
3604 |
|
|
IRIX 6.5.x releases, @math{x < 19}. The problem cannot be autodetected
|
3605 |
|
|
and in order to build GCC for such targets you need to configure with
|
3606 |
|
|
@option{--disable-wchar_t}.
|
3607 |
|
|
|
3608 |
|
|
See @uref{http://freeware.sgi.com/} for more
|
3609 |
|
|
information about using GCC on IRIX platforms.
|
3610 |
|
|
|
3611 |
|
|
@html
|
3612 |
|
|
<hr />
|
3613 |
|
|
@end html
|
3614 |
|
|
@heading @anchor{powerpc-x-x}powerpc-*-*
|
3615 |
|
|
|
3616 |
|
|
You can specify a default version for the @option{-mcpu=@var{cpu_type}}
|
3617 |
|
|
switch by using the configure option @option{--with-cpu-@var{cpu_type}}.
|
3618 |
|
|
|
3619 |
|
|
@html
|
3620 |
|
|
<hr />
|
3621 |
|
|
@end html
|
3622 |
|
|
@heading @anchor{powerpc-x-darwin}powerpc-*-darwin*
|
3623 |
|
|
PowerPC running Darwin (Mac OS X kernel).
|
3624 |
|
|
|
3625 |
|
|
Pre-installed versions of Mac OS X may not include any developer tools,
|
3626 |
|
|
meaning that you will not be able to build GCC from source. Tool
|
3627 |
|
|
binaries are available at
|
3628 |
|
|
@uref{http://developer.apple.com/darwin/projects/compiler/} (free
|
3629 |
|
|
registration required).
|
3630 |
|
|
|
3631 |
|
|
This version of GCC requires at least cctools-590.7.
|
3632 |
|
|
|
3633 |
|
|
The version of GCC shipped by Apple typically includes a number of
|
3634 |
|
|
extensions not available in a standard GCC release. These extensions
|
3635 |
|
|
are generally for backwards compatibility and best avoided.
|
3636 |
|
|
|
3637 |
|
|
@html
|
3638 |
|
|
<hr />
|
3639 |
|
|
@end html
|
3640 |
|
|
@heading @anchor{powerpc-x-elf}powerpc-*-elf, powerpc-*-sysv4
|
3641 |
|
|
PowerPC system in big endian mode, running System V.4.
|
3642 |
|
|
|
3643 |
|
|
@html
|
3644 |
|
|
<hr />
|
3645 |
|
|
@end html
|
3646 |
|
|
@heading @anchor{powerpc-x-linux-gnu}powerpc*-*-linux-gnu*
|
3647 |
|
|
|
3648 |
|
|
You will need
|
3649 |
|
|
@uref{ftp://ftp.kernel.org/pub/linux/devel/binutils,,binutils 2.15}
|
3650 |
|
|
or newer for a working GCC@.
|
3651 |
|
|
|
3652 |
|
|
@html
|
3653 |
|
|
<hr />
|
3654 |
|
|
@end html
|
3655 |
|
|
@heading @anchor{powerpc-x-netbsd}powerpc-*-netbsd*
|
3656 |
|
|
PowerPC system in big endian mode running NetBSD@. To build the
|
3657 |
|
|
documentation you will need Texinfo version 4.4 (NetBSD 1.5.1 included
|
3658 |
|
|
Texinfo version 3.12).
|
3659 |
|
|
|
3660 |
|
|
@html
|
3661 |
|
|
<hr />
|
3662 |
|
|
@end html
|
3663 |
|
|
@heading @anchor{powerpc-x-eabisim}powerpc-*-eabisim
|
3664 |
|
|
Embedded PowerPC system in big endian mode for use in running under the
|
3665 |
|
|
PSIM simulator.
|
3666 |
|
|
|
3667 |
|
|
@html
|
3668 |
|
|
<hr />
|
3669 |
|
|
@end html
|
3670 |
|
|
@heading @anchor{powerpc-x-eabi}powerpc-*-eabi
|
3671 |
|
|
Embedded PowerPC system in big endian mode.
|
3672 |
|
|
|
3673 |
|
|
@html
|
3674 |
|
|
<hr />
|
3675 |
|
|
@end html
|
3676 |
|
|
@heading @anchor{powerpcle-x-elf}powerpcle-*-elf, powerpcle-*-sysv4
|
3677 |
|
|
PowerPC system in little endian mode, running System V.4.
|
3678 |
|
|
|
3679 |
|
|
@html
|
3680 |
|
|
<hr />
|
3681 |
|
|
@end html
|
3682 |
|
|
@heading @anchor{powerpcle-x-eabisim}powerpcle-*-eabisim
|
3683 |
|
|
Embedded PowerPC system in little endian mode for use in running under
|
3684 |
|
|
the PSIM simulator.
|
3685 |
|
|
|
3686 |
|
|
@html
|
3687 |
|
|
<hr />
|
3688 |
|
|
@end html
|
3689 |
|
|
@heading @anchor{powerpcle-x-eabi}powerpcle-*-eabi
|
3690 |
|
|
Embedded PowerPC system in little endian mode.
|
3691 |
|
|
|
3692 |
|
|
@html
|
3693 |
|
|
<hr />
|
3694 |
|
|
@end html
|
3695 |
|
|
@heading @anchor{s390-x-linux}s390-*-linux*
|
3696 |
|
|
S/390 system running GNU/Linux for S/390@.
|
3697 |
|
|
|
3698 |
|
|
@html
|
3699 |
|
|
<hr />
|
3700 |
|
|
@end html
|
3701 |
|
|
@heading @anchor{s390x-x-linux}s390x-*-linux*
|
3702 |
|
|
zSeries system (64-bit) running GNU/Linux for zSeries@.
|
3703 |
|
|
|
3704 |
|
|
@html
|
3705 |
|
|
<hr />
|
3706 |
|
|
@end html
|
3707 |
|
|
@heading @anchor{s390x-ibm-tpf}s390x-ibm-tpf*
|
3708 |
|
|
zSeries system (64-bit) running TPF@. This platform is
|
3709 |
|
|
supported as cross-compilation target only.
|
3710 |
|
|
|
3711 |
|
|
@html
|
3712 |
|
|
<hr />
|
3713 |
|
|
@end html
|
3714 |
|
|
@c Please use Solaris 2 to refer to all release of Solaris, starting
|
3715 |
|
|
@c with 2.0 until 2.6, 7, 8, etc. Solaris 1 was a marketing name for
|
3716 |
|
|
@c SunOS 4 releases which we don't use to avoid confusion. Solaris
|
3717 |
|
|
@c alone is too unspecific and must be avoided.
|
3718 |
|
|
@heading @anchor{x-x-solaris2}*-*-solaris2*
|
3719 |
|
|
|
3720 |
|
|
Sun does not ship a C compiler with Solaris 2. To bootstrap and install
|
3721 |
|
|
GCC you first have to install a pre-built compiler, see the
|
3722 |
|
|
@uref{binaries.html,,binaries page} for details.
|
3723 |
|
|
|
3724 |
|
|
The Solaris 2 @command{/bin/sh} will often fail to configure
|
3725 |
|
|
@file{libstdc++-v3}, @file{boehm-gc} or @file{libjava}. We therefore
|
3726 |
|
|
recommend using the following initial sequence of commands
|
3727 |
|
|
|
3728 |
|
|
@smallexample
|
3729 |
|
|
% CONFIG_SHELL=/bin/ksh
|
3730 |
|
|
% export CONFIG_SHELL
|
3731 |
|
|
@end smallexample
|
3732 |
|
|
|
3733 |
|
|
and proceed as described in @uref{configure.html,,the configure instructions}.
|
3734 |
|
|
In addition we strongly recommend specifying an absolute path to invoke
|
3735 |
|
|
@var{srcdir}/configure.
|
3736 |
|
|
|
3737 |
|
|
Solaris 2 comes with a number of optional OS packages. Some of these
|
3738 |
|
|
are needed to use GCC fully, namely @code{SUNWarc},
|
3739 |
|
|
@code{SUNWbtool}, @code{SUNWesu}, @code{SUNWhea}, @code{SUNWlibm},
|
3740 |
|
|
@code{SUNWsprot}, and @code{SUNWtoo}. If you did not install all
|
3741 |
|
|
optional packages when installing Solaris 2, you will need to verify that
|
3742 |
|
|
the packages that GCC needs are installed.
|
3743 |
|
|
|
3744 |
|
|
To check whether an optional package is installed, use
|
3745 |
|
|
the @command{pkginfo} command. To add an optional package, use the
|
3746 |
|
|
@command{pkgadd} command. For further details, see the Solaris 2
|
3747 |
|
|
documentation.
|
3748 |
|
|
|
3749 |
|
|
Trying to use the linker and other tools in
|
3750 |
|
|
@file{/usr/ucb} to install GCC has been observed to cause trouble.
|
3751 |
|
|
For example, the linker may hang indefinitely. The fix is to remove
|
3752 |
|
|
@file{/usr/ucb} from your @env{PATH}.
|
3753 |
|
|
|
3754 |
|
|
The build process works more smoothly with the legacy Sun tools so, if you
|
3755 |
|
|
have @file{/usr/xpg4/bin} in your @env{PATH}, we recommend that you place
|
3756 |
|
|
@file{/usr/bin} before @file{/usr/xpg4/bin} for the duration of the build.
|
3757 |
|
|
|
3758 |
|
|
All releases of GNU binutils prior to 2.11.2 have known bugs on this
|
3759 |
|
|
platform. We recommend the use of GNU binutils 2.11.2 or later, or the
|
3760 |
|
|
vendor tools (Sun @command{as}, Sun @command{ld}). Note that your mileage
|
3761 |
|
|
may vary if you use a combination of the GNU tools and the Sun tools: while
|
3762 |
|
|
the combination GNU @command{as} + Sun @command{ld} should reasonably work,
|
3763 |
|
|
the reverse combination Sun @command{as} + GNU @command{ld} is known to
|
3764 |
|
|
cause memory corruption at runtime in some cases for C++ programs.
|
3765 |
|
|
|
3766 |
|
|
The stock GNU binutils 2.15 release is broken on this platform because of a
|
3767 |
|
|
single bug. It has been fixed on the 2.15 branch in the CVS repository.
|
3768 |
|
|
You can obtain a working version by checking out the binutils-2_15-branch
|
3769 |
|
|
from the CVS repository or applying the patch
|
3770 |
|
|
@uref{http://sourceware.org/ml/binutils-cvs/2004-09/msg00036.html} to the
|
3771 |
|
|
release.
|
3772 |
|
|
|
3773 |
|
|
We recommend using GNU binutils 2.16 or later in conjunction with GCC 4.x,
|
3774 |
|
|
or the vendor tools (Sun @command{as}, Sun @command{ld}). However, for
|
3775 |
|
|
Solaris 10 and above, an additional patch is required in order for the GNU
|
3776 |
|
|
linker to be able to cope with a new flavor of shared libraries. You
|
3777 |
|
|
can obtain a working version by checking out the binutils-2_16-branch from
|
3778 |
|
|
the CVS repository or applying the patch
|
3779 |
|
|
@uref{http://sourceware.org/ml/binutils-cvs/2005-07/msg00122.html} to the
|
3780 |
|
|
release.
|
3781 |
|
|
|
3782 |
|
|
Sun bug 4296832 turns up when compiling X11 headers with GCC 2.95 or
|
3783 |
|
|
newer: @command{g++} will complain that types are missing. These headers assume
|
3784 |
|
|
that omitting the type means @code{int}; this assumption worked for C89 but
|
3785 |
|
|
is wrong for C++, and is now wrong for C99 also.
|
3786 |
|
|
|
3787 |
|
|
@command{g++} accepts such (invalid) constructs with the option
|
3788 |
|
|
@option{-fpermissive}; it
|
3789 |
|
|
will assume that any missing type is @code{int} (as defined by C89).
|
3790 |
|
|
|
3791 |
|
|
There are patches for Solaris 2.6 (105633-56 or newer for SPARC,
|
3792 |
|
|
106248-42 or newer for Intel), Solaris 7 (108376-21 or newer for SPARC,
|
3793 |
|
|
108377-20 for Intel), and Solaris 8 (108652-24 or newer for SPARC,
|
3794 |
|
|
108653-22 for Intel) that fix this bug.
|
3795 |
|
|
|
3796 |
|
|
Sun bug 4927647 sometimes causes random spurious testsuite failures
|
3797 |
|
|
related to missing diagnostic output. This bug doesn't affect GCC
|
3798 |
|
|
itself, rather it is a kernel bug triggered by the @command{expect}
|
3799 |
|
|
program which is used only by the GCC testsuite driver. When the bug
|
3800 |
|
|
causes the @command{expect} program to miss anticipated output, extra
|
3801 |
|
|
testsuite failures appear.
|
3802 |
|
|
|
3803 |
|
|
There are patches for Solaris 8 (117350-12 or newer for SPARC,
|
3804 |
|
|
117351-12 or newer for Intel) and Solaris 9 (117171-11 or newer for
|
3805 |
|
|
SPARC, 117172-11 or newer for Intel) that address this problem.
|
3806 |
|
|
|
3807 |
|
|
@html
|
3808 |
|
|
<hr />
|
3809 |
|
|
@end html
|
3810 |
|
|
@heading @anchor{sparc-sun-solaris2}sparc-sun-solaris2*
|
3811 |
|
|
|
3812 |
|
|
When GCC is configured to use binutils 2.11.2 or later the binaries
|
3813 |
|
|
produced are smaller than the ones produced using Sun's native tools;
|
3814 |
|
|
this difference is quite significant for binaries containing debugging
|
3815 |
|
|
information.
|
3816 |
|
|
|
3817 |
|
|
Sun @command{as} 4.x is broken in that it cannot cope with long symbol names.
|
3818 |
|
|
A typical error message might look similar to the following:
|
3819 |
|
|
|
3820 |
|
|
@smallexample
|
3821 |
|
|
/usr/ccs/bin/as: "/var/tmp/ccMsw135.s", line 11041: error:
|
3822 |
|
|
can't compute value of an expression involving an external symbol.
|
3823 |
|
|
@end smallexample
|
3824 |
|
|
|
3825 |
|
|
This is Sun bug 4237974. This is fixed with patch 108908-02 for Solaris
|
3826 |
|
|
2.6 and has been fixed in later (5.x) versions of the assembler,
|
3827 |
|
|
starting with Solaris 7.
|
3828 |
|
|
|
3829 |
|
|
Starting with Solaris 7, the operating system is capable of executing
|
3830 |
|
|
64-bit SPARC V9 binaries. GCC 3.1 and later properly supports
|
3831 |
|
|
this; the @option{-m64} option enables 64-bit code generation.
|
3832 |
|
|
However, if all you want is code tuned for the UltraSPARC CPU, you
|
3833 |
|
|
should try the @option{-mtune=ultrasparc} option instead, which produces
|
3834 |
|
|
code that, unlike full 64-bit code, can still run on non-UltraSPARC
|
3835 |
|
|
machines.
|
3836 |
|
|
|
3837 |
|
|
When configuring on a Solaris 7 or later system that is running a kernel
|
3838 |
|
|
that supports only 32-bit binaries, one must configure with
|
3839 |
|
|
@option{--disable-multilib}, since we will not be able to build the
|
3840 |
|
|
64-bit target libraries.
|
3841 |
|
|
|
3842 |
|
|
GCC 3.3 and GCC 3.4 trigger code generation bugs in earlier versions of
|
3843 |
|
|
the GNU compiler (especially GCC 3.0.x versions), which lead to the
|
3844 |
|
|
miscompilation of the stage1 compiler and the subsequent failure of the
|
3845 |
|
|
bootstrap process. A workaround is to use GCC 3.2.3 as an intermediary
|
3846 |
|
|
stage, i.e.@: to bootstrap that compiler with the base compiler and then
|
3847 |
|
|
use it to bootstrap the final compiler.
|
3848 |
|
|
|
3849 |
|
|
GCC 3.4 triggers a code generation bug in versions 5.4 (Sun ONE Studio 7)
|
3850 |
|
|
and 5.5 (Sun ONE Studio 8) of the Sun compiler, which causes a bootstrap
|
3851 |
|
|
failure in form of a miscompilation of the stage1 compiler by the Sun
|
3852 |
|
|
compiler. This is Sun bug 4974440. This is fixed with patch 112760-07.
|
3853 |
|
|
|
3854 |
|
|
GCC 3.4 changed the default debugging format from STABS to DWARF-2 for
|
3855 |
|
|
32-bit code on Solaris 7 and later. If you use the Sun assembler, this
|
3856 |
|
|
change apparently runs afoul of Sun bug 4910101 (which is referenced as
|
3857 |
|
|
a x86-only problem by Sun, probably because they do not use DWARF-2).
|
3858 |
|
|
A symptom of the problem is that you cannot compile C++ programs like
|
3859 |
|
|
@command{groff} 1.19.1 without getting messages similar to the following:
|
3860 |
|
|
|
3861 |
|
|
@smallexample
|
3862 |
|
|
ld: warning: relocation error: R_SPARC_UA32: @dots{}
|
3863 |
|
|
external symbolic relocation against non-allocatable section
|
3864 |
|
|
.debug_info cannot be processed at runtime: relocation ignored.
|
3865 |
|
|
@end smallexample
|
3866 |
|
|
|
3867 |
|
|
To work around this problem, compile with @option{-gstabs+} instead of
|
3868 |
|
|
plain @option{-g}.
|
3869 |
|
|
|
3870 |
|
|
When configuring the GNU Multiple Precision Library (GMP) or the MPFR
|
3871 |
|
|
library on a Solaris 7 or later system, the canonical target triplet
|
3872 |
|
|
must be specified as the @command{build} parameter on the configure
|
3873 |
|
|
line. This triplet can be obtained by invoking ./config.guess in
|
3874 |
|
|
the toplevel source directory of GCC (and not that of GMP or MPFR).
|
3875 |
|
|
For example on a Solaris 7 system:
|
3876 |
|
|
|
3877 |
|
|
@smallexample
|
3878 |
|
|
% ./configure --build=sparc-sun-solaris2.7 --prefix=xxx
|
3879 |
|
|
@end smallexample
|
3880 |
|
|
|
3881 |
|
|
@html
|
3882 |
|
|
<hr />
|
3883 |
|
|
@end html
|
3884 |
|
|
@heading @anchor{sparc-sun-solaris27}sparc-sun-solaris2.7
|
3885 |
|
|
|
3886 |
|
|
Sun patch 107058-01 (1999-01-13) for Solaris 7/SPARC triggers a bug in
|
3887 |
|
|
the dynamic linker. This problem (Sun bug 4210064) affects GCC 2.8
|
3888 |
|
|
and later, including all EGCS releases. Sun formerly recommended
|
3889 |
|
|
107058-01 for all Solaris 7 users, but around 1999-09-01 it started to
|
3890 |
|
|
recommend it only for people who use Sun's compilers.
|
3891 |
|
|
|
3892 |
|
|
Here are some workarounds to this problem:
|
3893 |
|
|
@itemize @bullet
|
3894 |
|
|
@item
|
3895 |
|
|
Do not install Sun patch 107058-01 until after Sun releases a
|
3896 |
|
|
complete patch for bug 4210064. This is the simplest course to take,
|
3897 |
|
|
unless you must also use Sun's C compiler. Unfortunately 107058-01
|
3898 |
|
|
is preinstalled on some new Solaris 7-based hosts, so you may have to
|
3899 |
|
|
back it out.
|
3900 |
|
|
|
3901 |
|
|
@item
|
3902 |
|
|
Copy the original, unpatched Solaris 7
|
3903 |
|
|
@command{/usr/ccs/bin/as} into
|
3904 |
|
|
@command{/usr/local/libexec/gcc/sparc-sun-solaris2.7/3.4/as},
|
3905 |
|
|
adjusting the latter name to fit your local conventions and software
|
3906 |
|
|
version numbers.
|
3907 |
|
|
|
3908 |
|
|
@item
|
3909 |
|
|
Install Sun patch 106950-03 (1999-05-25) or later. Nobody with
|
3910 |
|
|
both 107058-01 and 106950-03 installed has reported the bug with GCC
|
3911 |
|
|
and Sun's dynamic linker. This last course of action is riskiest,
|
3912 |
|
|
for two reasons. First, you must install 106950 on all hosts that
|
3913 |
|
|
run code generated by GCC; it doesn't suffice to install it only on
|
3914 |
|
|
the hosts that run GCC itself. Second, Sun says that 106950-03 is
|
3915 |
|
|
only a partial fix for bug 4210064, but Sun doesn't know whether the
|
3916 |
|
|
partial fix is adequate for GCC@. Revision -08 or later should fix
|
3917 |
|
|
the bug. The current (as of 2004-05-23) revision is -24, and is included in
|
3918 |
|
|
the Solaris 7 Recommended Patch Cluster.
|
3919 |
|
|
@end itemize
|
3920 |
|
|
|
3921 |
|
|
GCC 3.3 triggers a bug in version 5.0 Alpha 03/27/98 of the Sun assembler,
|
3922 |
|
|
which causes a bootstrap failure when linking the 64-bit shared version of
|
3923 |
|
|
libgcc. A typical error message is:
|
3924 |
|
|
|
3925 |
|
|
@smallexample
|
3926 |
|
|
ld: fatal: relocation error: R_SPARC_32: file libgcc/sparcv9/_muldi3.o:
|
3927 |
|
|
symbol <unknown>: offset 0xffffffff7ec133e7 is non-aligned.
|
3928 |
|
|
@end smallexample
|
3929 |
|
|
|
3930 |
|
|
This bug has been fixed in the final 5.0 version of the assembler.
|
3931 |
|
|
|
3932 |
|
|
A similar problem was reported for version Sun WorkShop 6 99/08/18 of the
|
3933 |
|
|
Sun assembler, which causes a bootstrap failure with GCC 4.0.0:
|
3934 |
|
|
|
3935 |
|
|
@smallexample
|
3936 |
|
|
ld: fatal: relocation error: R_SPARC_DISP32:
|
3937 |
|
|
file .libs/libstdc++.lax/libsupc++convenience.a/vterminate.o:
|
3938 |
|
|
symbol <unknown>: offset 0xfccd33ad is non-aligned
|
3939 |
|
|
@end smallexample
|
3940 |
|
|
|
3941 |
|
|
This bug has been fixed in more recent revisions of the assembler.
|
3942 |
|
|
|
3943 |
|
|
@html
|
3944 |
|
|
<hr />
|
3945 |
|
|
@end html
|
3946 |
|
|
@heading @anchor{sparc-x-linux}sparc-*-linux*
|
3947 |
|
|
|
3948 |
|
|
GCC versions 3.0 and higher require binutils 2.11.2 and glibc 2.2.4
|
3949 |
|
|
or newer on this platform. All earlier binutils and glibc
|
3950 |
|
|
releases mishandled unaligned relocations on @code{sparc-*-*} targets.
|
3951 |
|
|
|
3952 |
|
|
|
3953 |
|
|
@html
|
3954 |
|
|
<hr />
|
3955 |
|
|
@end html
|
3956 |
|
|
@heading @anchor{sparc64-x-solaris2}sparc64-*-solaris2*
|
3957 |
|
|
|
3958 |
|
|
When configuring the GNU Multiple Precision Library (GMP) or the
|
3959 |
|
|
MPFR library, the canonical target triplet must be specified as
|
3960 |
|
|
the @command{build} parameter on the configure line. For example
|
3961 |
|
|
on a Solaris 7 system:
|
3962 |
|
|
|
3963 |
|
|
@smallexample
|
3964 |
|
|
% ./configure --build=sparc64-sun-solaris2.7 --prefix=xxx
|
3965 |
|
|
@end smallexample
|
3966 |
|
|
|
3967 |
|
|
The following compiler flags must be specified in the configure
|
3968 |
|
|
step in order to bootstrap this target with the Sun compiler:
|
3969 |
|
|
|
3970 |
|
|
@smallexample
|
3971 |
|
|
% CC="cc -xarch=v9 -xildoff" @var{srcdir}/configure [@var{options}] [@var{target}]
|
3972 |
|
|
@end smallexample
|
3973 |
|
|
|
3974 |
|
|
@option{-xarch=v9} specifies the SPARC-V9 architecture to the Sun toolchain
|
3975 |
|
|
and @option{-xildoff} turns off the incremental linker.
|
3976 |
|
|
|
3977 |
|
|
@html
|
3978 |
|
|
<hr />
|
3979 |
|
|
@end html
|
3980 |
|
|
@heading @anchor{sparcv9-x-solaris2}sparcv9-*-solaris2*
|
3981 |
|
|
|
3982 |
|
|
This is a synonym for sparc64-*-solaris2*.
|
3983 |
|
|
|
3984 |
|
|
@html
|
3985 |
|
|
<hr />
|
3986 |
|
|
@end html
|
3987 |
|
|
@heading @anchor{x-x-sysv}*-*-sysv*
|
3988 |
|
|
On System V release 3, you may get this error message
|
3989 |
|
|
while linking:
|
3990 |
|
|
|
3991 |
|
|
@smallexample
|
3992 |
|
|
ld fatal: failed to write symbol name @var{something}
|
3993 |
|
|
in strings table for file @var{whatever}
|
3994 |
|
|
@end smallexample
|
3995 |
|
|
|
3996 |
|
|
This probably indicates that the disk is full or your ulimit won't allow
|
3997 |
|
|
the file to be as large as it needs to be.
|
3998 |
|
|
|
3999 |
|
|
This problem can also result because the kernel parameter @code{MAXUMEM}
|
4000 |
|
|
is too small. If so, you must regenerate the kernel and make the value
|
4001 |
|
|
much larger. The default value is reported to be 1024; a value of 32768
|
4002 |
|
|
is said to work. Smaller values may also work.
|
4003 |
|
|
|
4004 |
|
|
On System V, if you get an error like this,
|
4005 |
|
|
|
4006 |
|
|
@smallexample
|
4007 |
|
|
/usr/local/lib/bison.simple: In function `yyparse':
|
4008 |
|
|
/usr/local/lib/bison.simple:625: virtual memory exhausted
|
4009 |
|
|
@end smallexample
|
4010 |
|
|
|
4011 |
|
|
@noindent
|
4012 |
|
|
that too indicates a problem with disk space, ulimit, or @code{MAXUMEM}.
|
4013 |
|
|
|
4014 |
|
|
On a System V release 4 system, make sure @file{/usr/bin} precedes
|
4015 |
|
|
@file{/usr/ucb} in @code{PATH}. The @command{cc} command in
|
4016 |
|
|
@file{/usr/ucb} uses libraries which have bugs.
|
4017 |
|
|
|
4018 |
|
|
@html
|
4019 |
|
|
<hr />
|
4020 |
|
|
@end html
|
4021 |
|
|
@heading @anchor{vax-dec-ultrix}vax-dec-ultrix
|
4022 |
|
|
Don't try compiling with VAX C (@command{vcc}). It produces incorrect code
|
4023 |
|
|
in some cases (for example, when @code{alloca} is used).
|
4024 |
|
|
|
4025 |
|
|
@html
|
4026 |
|
|
<hr />
|
4027 |
|
|
@end html
|
4028 |
|
|
@heading @anchor{x-x-vxworks}*-*-vxworks*
|
4029 |
|
|
Support for VxWorks is in flux. At present GCC supports @emph{only} the
|
4030 |
|
|
very recent VxWorks 5.5 (aka Tornado 2.2) release, and only on PowerPC@.
|
4031 |
|
|
We welcome patches for other architectures supported by VxWorks 5.5.
|
4032 |
|
|
Support for VxWorks AE would also be welcome; we believe this is merely
|
4033 |
|
|
a matter of writing an appropriate ``configlette'' (see below). We are
|
4034 |
|
|
not interested in supporting older, a.out or COFF-based, versions of
|
4035 |
|
|
VxWorks in GCC 3.
|
4036 |
|
|
|
4037 |
|
|
VxWorks comes with an older version of GCC installed in
|
4038 |
|
|
@file{@var{$WIND_BASE}/host}; we recommend you do not overwrite it.
|
4039 |
|
|
Choose an installation @var{prefix} entirely outside @var{$WIND_BASE}.
|
4040 |
|
|
Before running @command{configure}, create the directories @file{@var{prefix}}
|
4041 |
|
|
and @file{@var{prefix}/bin}. Link or copy the appropriate assembler,
|
4042 |
|
|
linker, etc.@: into @file{@var{prefix}/bin}, and set your @var{PATH} to
|
4043 |
|
|
include that directory while running both @command{configure} and
|
4044 |
|
|
@command{make}.
|
4045 |
|
|
|
4046 |
|
|
You must give @command{configure} the
|
4047 |
|
|
@option{--with-headers=@var{$WIND_BASE}/target/h} switch so that it can
|
4048 |
|
|
find the VxWorks system headers. Since VxWorks is a cross compilation
|
4049 |
|
|
target only, you must also specify @option{--target=@var{target}}.
|
4050 |
|
|
@command{configure} will attempt to create the directory
|
4051 |
|
|
@file{@var{prefix}/@var{target}/sys-include} and copy files into it;
|
4052 |
|
|
make sure the user running @command{configure} has sufficient privilege
|
4053 |
|
|
to do so.
|
4054 |
|
|
|
4055 |
|
|
GCC's exception handling runtime requires a special ``configlette''
|
4056 |
|
|
module, @file{contrib/gthr_supp_vxw_5x.c}. Follow the instructions in
|
4057 |
|
|
that file to add the module to your kernel build. (Future versions of
|
4058 |
|
|
VxWorks will incorporate this module.)
|
4059 |
|
|
|
4060 |
|
|
@html
|
4061 |
|
|
<hr />
|
4062 |
|
|
@end html
|
4063 |
|
|
@heading @anchor{x86-64-x-x}x86_64-*-*, amd64-*-*
|
4064 |
|
|
|
4065 |
|
|
GCC supports the x86-64 architecture implemented by the AMD64 processor
|
4066 |
|
|
(amd64-*-* is an alias for x86_64-*-*) on GNU/Linux, FreeBSD and NetBSD@.
|
4067 |
|
|
On GNU/Linux the default is a bi-arch compiler which is able to generate
|
4068 |
|
|
both 64-bit x86-64 and 32-bit x86 code (via the @option{-m32} switch).
|
4069 |
|
|
|
4070 |
|
|
@html
|
4071 |
|
|
<hr />
|
4072 |
|
|
@end html
|
4073 |
|
|
@heading @anchor{xtensa-x-elf}xtensa-*-elf
|
4074 |
|
|
|
4075 |
|
|
This target is intended for embedded Xtensa systems using the
|
4076 |
|
|
@samp{newlib} C library. It uses ELF but does not support shared
|
4077 |
|
|
objects. Designed-defined instructions specified via the
|
4078 |
|
|
Tensilica Instruction Extension (TIE) language are only supported
|
4079 |
|
|
through inline assembly.
|
4080 |
|
|
|
4081 |
|
|
The Xtensa configuration information must be specified prior to
|
4082 |
|
|
building GCC@. The @file{include/xtensa-config.h} header
|
4083 |
|
|
file contains the configuration information. If you created your
|
4084 |
|
|
own Xtensa configuration with the Xtensa Processor Generator, the
|
4085 |
|
|
downloaded files include a customized copy of this header file,
|
4086 |
|
|
which you can use to replace the default header file.
|
4087 |
|
|
|
4088 |
|
|
@html
|
4089 |
|
|
<hr />
|
4090 |
|
|
@end html
|
4091 |
|
|
@heading @anchor{xtensa-x-linux}xtensa-*-linux*
|
4092 |
|
|
|
4093 |
|
|
This target is for Xtensa systems running GNU/Linux. It supports ELF
|
4094 |
|
|
shared objects and the GNU C library (glibc). It also generates
|
4095 |
|
|
position-independent code (PIC) regardless of whether the
|
4096 |
|
|
@option{-fpic} or @option{-fPIC} options are used. In other
|
4097 |
|
|
respects, this target is the same as the
|
4098 |
|
|
@uref{#xtensa-*-elf,,@samp{xtensa-*-elf}} target.
|
4099 |
|
|
|
4100 |
|
|
@html
|
4101 |
|
|
<hr />
|
4102 |
|
|
@end html
|
4103 |
|
|
@heading @anchor{windows}Microsoft Windows (32-bit)
|
4104 |
|
|
|
4105 |
|
|
Ports of GCC are included with the
|
4106 |
|
|
@uref{http://www.cygwin.com/,,Cygwin environment}.
|
4107 |
|
|
|
4108 |
|
|
GCC will build under Cygwin without modification; it does not build
|
4109 |
|
|
with Microsoft's C++ compiler and there are no plans to make it do so.
|
4110 |
|
|
|
4111 |
|
|
@html
|
4112 |
|
|
<hr />
|
4113 |
|
|
@end html
|
4114 |
|
|
@heading @anchor{os2}OS/2
|
4115 |
|
|
|
4116 |
|
|
GCC does not currently support OS/2. However, Andrew Zabolotny has been
|
4117 |
|
|
working on a generic OS/2 port with pgcc. The current code can be found
|
4118 |
|
|
at @uref{http://www.goof.com/pcg/os2/,,http://www.goof.com/pcg/os2/}.
|
4119 |
|
|
|
4120 |
|
|
@html
|
4121 |
|
|
<hr />
|
4122 |
|
|
@end html
|
4123 |
|
|
@heading @anchor{older}Older systems
|
4124 |
|
|
|
4125 |
|
|
GCC contains support files for many older (1980s and early
|
4126 |
|
|
1990s) Unix variants. For the most part, support for these systems
|
4127 |
|
|
has not been deliberately removed, but it has not been maintained for
|
4128 |
|
|
several years and may suffer from bitrot.
|
4129 |
|
|
|
4130 |
|
|
Starting with GCC 3.1, each release has a list of ``obsoleted'' systems.
|
4131 |
|
|
Support for these systems is still present in that release, but
|
4132 |
|
|
@command{configure} will fail unless the @option{--enable-obsolete}
|
4133 |
|
|
option is given. Unless a maintainer steps forward, support for these
|
4134 |
|
|
systems will be removed from the next release of GCC@.
|
4135 |
|
|
|
4136 |
|
|
Support for old systems as hosts for GCC can cause problems if the
|
4137 |
|
|
workarounds for compiler, library and operating system bugs affect the
|
4138 |
|
|
cleanliness or maintainability of the rest of GCC@. In some cases, to
|
4139 |
|
|
bring GCC up on such a system, if still possible with current GCC, may
|
4140 |
|
|
require first installing an old version of GCC which did work on that
|
4141 |
|
|
system, and using it to compile a more recent GCC, to avoid bugs in the
|
4142 |
|
|
vendor compiler. Old releases of GCC 1 and GCC 2 are available in the
|
4143 |
|
|
@file{old-releases} directory on the @uref{../mirrors.html,,GCC mirror
|
4144 |
|
|
sites}. Header bugs may generally be avoided using
|
4145 |
|
|
@command{fixincludes}, but bugs or deficiencies in libraries and the
|
4146 |
|
|
operating system may still cause problems.
|
4147 |
|
|
|
4148 |
|
|
Support for older systems as targets for cross-compilation is less
|
4149 |
|
|
problematic than support for them as hosts for GCC; if an enthusiast
|
4150 |
|
|
wishes to make such a target work again (including resurrecting any of
|
4151 |
|
|
the targets that never worked with GCC 2, starting from the last
|
4152 |
|
|
version before they were removed), patches
|
4153 |
|
|
@uref{../contribute.html,,following the usual requirements} would be
|
4154 |
|
|
likely to be accepted, since they should not affect the support for more
|
4155 |
|
|
modern targets.
|
4156 |
|
|
|
4157 |
|
|
For some systems, old versions of GNU binutils may also be useful,
|
4158 |
|
|
and are available from @file{pub/binutils/old-releases} on
|
4159 |
|
|
@uref{http://sourceware.org/mirrors.html,,sourceware.org mirror sites}.
|
4160 |
|
|
|
4161 |
|
|
Some of the information on specific systems above relates to
|
4162 |
|
|
such older systems, but much of the information
|
4163 |
|
|
about GCC on such systems (which may no longer be applicable to
|
4164 |
|
|
current GCC) is to be found in the GCC texinfo manual.
|
4165 |
|
|
|
4166 |
|
|
@html
|
4167 |
|
|
<hr />
|
4168 |
|
|
@end html
|
4169 |
|
|
@heading @anchor{elf}all ELF targets (SVR4, Solaris 2, etc.)
|
4170 |
|
|
|
4171 |
|
|
C++ support is significantly better on ELF targets if you use the
|
4172 |
|
|
@uref{./configure.html#with-gnu-ld,,GNU linker}; duplicate copies of
|
4173 |
|
|
inlines, vtables and template instantiations will be discarded
|
4174 |
|
|
automatically.
|
4175 |
|
|
|
4176 |
|
|
|
4177 |
|
|
@html
|
4178 |
|
|
<hr />
|
4179 |
|
|
<p>
|
4180 |
|
|
@end html
|
4181 |
|
|
@ifhtml
|
4182 |
|
|
@uref{./index.html,,Return to the GCC Installation page}
|
4183 |
|
|
@end ifhtml
|
4184 |
|
|
@end ifset
|
4185 |
|
|
|
4186 |
|
|
@c ***Old documentation******************************************************
|
4187 |
|
|
@ifset oldhtml
|
4188 |
|
|
@include install-old.texi
|
4189 |
|
|
@html
|
4190 |
|
|
<hr />
|
4191 |
|
|
<p>
|
4192 |
|
|
@end html
|
4193 |
|
|
@ifhtml
|
4194 |
|
|
@uref{./index.html,,Return to the GCC Installation page}
|
4195 |
|
|
@end ifhtml
|
4196 |
|
|
@end ifset
|
4197 |
|
|
|
4198 |
|
|
@c ***GFDL********************************************************************
|
4199 |
|
|
@ifset gfdlhtml
|
4200 |
|
|
@include fdl.texi
|
4201 |
|
|
@html
|
4202 |
|
|
<hr />
|
4203 |
|
|
<p>
|
4204 |
|
|
@end html
|
4205 |
|
|
@ifhtml
|
4206 |
|
|
@uref{./index.html,,Return to the GCC Installation page}
|
4207 |
|
|
@end ifhtml
|
4208 |
|
|
@end ifset
|
4209 |
|
|
|
4210 |
|
|
@c ***************************************************************************
|
4211 |
|
|
@c Part 6 The End of the Document
|
4212 |
|
|
@ifinfo
|
4213 |
|
|
@comment node-name, next, previous, up
|
4214 |
|
|
@node Concept Index, , GNU Free Documentation License, Top
|
4215 |
|
|
@end ifinfo
|
4216 |
|
|
|
4217 |
|
|
@ifinfo
|
4218 |
|
|
@unnumbered Concept Index
|
4219 |
|
|
|
4220 |
|
|
@printindex cp
|
4221 |
|
|
|
4222 |
|
|
@contents
|
4223 |
|
|
@end ifinfo
|
4224 |
|
|
@bye
|