]>
Commit | Line | Data |
---|---|---|
f42974dc DW |
1 | \input texinfo.tex @c -*-texinfo-*- |
2 | @c @ifnothtml | |
3 | @c %**start of header | |
d7f8491b | 4 | @setfilename gccinstall.info |
f42974dc DW |
5 | @settitle Installing GCC |
6 | @setchapternewpage odd | |
7 | @c %**end of header | |
8 | @c @end ifnothtml | |
9 | ||
10 | @c Specify title for specific html page | |
11 | @ifset indexhtml | |
12 | @settitle Installing GCC | |
13 | @end ifset | |
14 | @ifset specifichtml | |
15 | @settitle Host/Target specific installation notes for GCC | |
16 | @end ifset | |
17 | @ifset downloadhtml | |
18 | @settitle Downloading GCC | |
19 | @end ifset | |
20 | @ifset configurehtml | |
21 | @settitle Installing GCC: Configuration | |
22 | @end ifset | |
23 | @ifset buildhtml | |
24 | @settitle Installing GCC: Building | |
25 | @end ifset | |
26 | @ifset testhtml | |
27 | @settitle Installing GCC: Testing | |
28 | @end ifset | |
29 | @ifset finalinstallhtml | |
30 | @settitle Installing GCC: Final installation | |
31 | @end ifset | |
32 | @ifset binarieshtml | |
33 | @settitle Installing GCC: Binaries | |
34 | @end ifset | |
73e2155a JM |
35 | @ifset oldhtml |
36 | @settitle Installing GCC: Old documentation | |
37 | @end ifset | |
aed5964b JM |
38 | @ifset gfdlhtml |
39 | @settitle Installing GCC: GNU Free Documentation License | |
40 | @end ifset | |
f42974dc | 41 | |
aed5964b | 42 | @c Copyright (C) 1988, 1989, 1992, 1993, 1994, 1995, 1996, 1997, 1998, |
4dd57c18 | 43 | @c 1999, 2000, 2001, 2002 Free Software Foundation, Inc. |
f42974dc DW |
44 | @c *** Converted to texinfo by Dean Wakerley, dean@wakerley.com |
45 | ||
46 | @c Include everything if we're not making html | |
47 | @ifnothtml | |
48 | @set indexhtml | |
49 | @set specifichtml | |
50 | @set downloadhtml | |
51 | @set configurehtml | |
52 | @set buildhtml | |
53 | @set testhtml | |
54 | @set finalinstallhtml | |
55 | @set binarieshtml | |
73e2155a | 56 | @set oldhtml |
aed5964b | 57 | @set gfdlhtml |
f42974dc DW |
58 | @end ifnothtml |
59 | ||
60 | @c Part 2 Summary Description and Copyright | |
bdefb2ab | 61 | @copying |
aed5964b | 62 | Copyright @copyright{} 1988, 1989, 1992, 1993, 1994, 1995, 1996, 1997, 1998, |
4dd57c18 | 63 | 1999, 2000, 2001, 2002 Free Software Foundation, Inc. |
aed5964b JM |
64 | @sp 1 |
65 | Permission is granted to copy, distribute and/or modify this document | |
66 | under the terms of the GNU Free Documentation License, Version 1.1 or | |
67 | any later version published by the Free Software Foundation; with no | |
68 | Invariant Sections, the Front-Cover texts being (a) (see below), and | |
69 | with the Back-Cover Texts being (b) (see below). A copy of the | |
70 | license is included in the section entitled ``@uref{./gfdl.html,,GNU | |
71 | Free Documentation License}''. | |
72 | ||
73 | (a) The FSF's Front-Cover Text is: | |
74 | ||
75 | A GNU Manual | |
76 | ||
77 | (b) The FSF's Back-Cover Text is: | |
78 | ||
79 | You have freedom to copy and modify this GNU Manual, like GNU | |
80 | software. Copies published by the Free Software Foundation raise | |
81 | funds for GNU development. | |
bdefb2ab | 82 | @end copying |
f42974dc | 83 | @ifinfo |
bdefb2ab | 84 | @insertcopying |
f42974dc DW |
85 | @end ifinfo |
86 | ||
87 | @c Part 3 Titlepage and Copyright | |
88 | @titlepage | |
89 | @sp 10 | |
90 | @comment The title is printed in a large font. | |
ef88b07d | 91 | @center @titlefont{Installing GCC} |
f42974dc DW |
92 | |
93 | @c The following two commands start the copyright page. | |
94 | @page | |
ef88b07d | 95 | @vskip 0pt plus 1filll |
bdefb2ab | 96 | @insertcopying |
f42974dc DW |
97 | @end titlepage |
98 | ||
99 | @c Part 4 Top node and Master Menu | |
100 | @ifinfo | |
101 | @node Top, , , (dir) | |
102 | @comment node-name, next, Previous, up | |
103 | ||
104 | @menu | |
105 | * Installing GCC:: This document describes the generic installation | |
106 | procedure for GCC as well as detailing some target | |
f9047ed3 | 107 | specific installation instructions. |
f42974dc DW |
108 | |
109 | * Specific:: Host/target specific installation notes for GCC. | |
110 | * Binaries:: Where to get pre-compiled binaries. | |
111 | ||
73e2155a JM |
112 | * Old:: Old installation documentation. |
113 | ||
aed5964b | 114 | * GNU Free Documentation License:: How you can copy and share this manual. |
f42974dc DW |
115 | * Concept Index:: This index has two entries. |
116 | @end menu | |
117 | @end ifinfo | |
118 | ||
119 | @c Part 5 The Body of the Document | |
120 | @c ***Installing GCC********************************************************** | |
6cfb3f16 | 121 | @ifnothtml |
f42974dc DW |
122 | @comment node-name, next, previous, up |
123 | @node Installing GCC, Binaries, , Top | |
6cfb3f16 | 124 | @end ifnothtml |
f42974dc | 125 | @ifset indexhtml |
f42974dc DW |
126 | @ifnothtml |
127 | @chapter Installing GCC | |
128 | @end ifnothtml | |
129 | ||
130 | The latest version of this document is always available at | |
f9047ed3 | 131 | @uref{http://gcc.gnu.org/install/,,http://gcc.gnu.org/install/}. |
f42974dc DW |
132 | |
133 | This document describes the generic installation procedure for GCC as well | |
f9047ed3 | 134 | as detailing some target specific installation instructions. |
f42974dc | 135 | |
f9047ed3 JM |
136 | GCC includes several components that previously were separate distributions |
137 | with their own installation instructions. This document supersedes all | |
eea81d3e | 138 | package specific installation instructions. |
f42974dc | 139 | |
f9047ed3 | 140 | @emph{Before} starting the build/install procedure please check the |
f42974dc | 141 | @ifnothtml |
eea81d3e | 142 | @ref{Specific, host/target specific installation notes}. |
f42974dc | 143 | @end ifnothtml |
c009f01f | 144 | @ifhtml |
f9047ed3 | 145 | @uref{specific.html,,host/target specific installation notes}. |
c009f01f | 146 | @end ifhtml |
f9047ed3 | 147 | We recommend you browse the entire generic installation instructions before |
f42974dc DW |
148 | you proceed. |
149 | ||
c009f01f | 150 | Lists of successful builds for released versions of GCC are |
b58bbfbb | 151 | available at @uref{http://gcc.gnu.org/buildstat.html}. |
c009f01f JJ |
152 | These lists are updated as new information becomes available. |
153 | ||
f9047ed3 | 154 | The installation procedure itself is broken into five steps. |
f42974dc DW |
155 | |
156 | @ifinfo | |
157 | @menu | |
158 | * Downloading the source:: | |
159 | * Configuration:: | |
160 | * Building:: | |
161 | * Testing:: (optional) | |
162 | * Final install:: | |
163 | @end menu | |
164 | @end ifinfo | |
c009f01f | 165 | @ifhtml |
f42974dc | 166 | @enumerate |
f9047ed3 | 167 | @item |
f42974dc | 168 | @uref{download.html,,Downloading the source} |
f42974dc | 169 | @item |
f9047ed3 JM |
170 | @uref{configure.html,,Configuration} |
171 | @item | |
172 | @uref{build.html,,Building} | |
173 | @item | |
174 | @uref{test.html,,Testing} (optional) | |
f42974dc DW |
175 | @item |
176 | @uref{finalinstall.html,,Final install} | |
177 | @end enumerate | |
c009f01f | 178 | @end ifhtml |
f42974dc | 179 | |
38209993 | 180 | Please note that GCC does not support @samp{make uninstall} and probably |
f9047ed3 | 181 | won't do so in the near future as this would open a can of worms. Instead, |
f42974dc | 182 | we suggest that you install GCC into a directory of its own and simply |
38209993 | 183 | remove that directory when you do not need that specific version of GCC |
eea81d3e RO |
184 | any longer, and, if shared libraries are installed there as well, no |
185 | more binaries exist that use them. | |
f42974dc | 186 | |
73e2155a JM |
187 | @ifhtml |
188 | There are also some @uref{old.html,,old installation instructions}, | |
189 | which are mostly obsolete but still contain some information which has | |
190 | not yet been merged into the main part of this manual. | |
191 | @end ifhtml | |
192 | ||
f42974dc | 193 | @html |
b8db17af | 194 | <hr /> |
f42974dc DW |
195 | <p> |
196 | @end html | |
197 | @ifhtml | |
198 | @uref{./index.html,,Return to the GCC Installation page} | |
aed5964b | 199 | |
bdefb2ab | 200 | @insertcopying |
f42974dc DW |
201 | @end ifhtml |
202 | @end ifset | |
203 | ||
204 | @c ***Downloading the source************************************************** | |
6cfb3f16 | 205 | @ifnothtml |
f42974dc DW |
206 | @comment node-name, next, previous, up |
207 | @node Downloading the source, Configuration, , Installing GCC | |
6cfb3f16 | 208 | @end ifnothtml |
f42974dc | 209 | @ifset downloadhtml |
f42974dc DW |
210 | @ifnothtml |
211 | @chapter Downloading GCC | |
212 | @end ifnothtml | |
213 | @cindex Downloading GCC | |
214 | @cindex Downloading the Source | |
215 | ||
eea81d3e RO |
216 | GCC is distributed via @uref{http://gcc.gnu.org/cvs.html,,CVS} and FTP |
217 | tarballs compressed with @command{gzip} or | |
6cfb3f16 | 218 | @command{bzip2}. It is possible to download a full distribution or specific |
f42974dc DW |
219 | components. |
220 | ||
221 | Please refer to our @uref{http://gcc.gnu.org/releases.html,,releases web page} | |
161d7b59 | 222 | for information on how to obtain GCC@. |
f42974dc DW |
223 | |
224 | The full distribution includes the C, C++, Objective-C, Fortran, Java, | |
8ecab453 GP |
225 | and Ada (in case of GCC 3.1 and later) compilers. The full distribution |
226 | also includes runtime libraries for C++, Objective-C, Fortran, and Java. | |
227 | In GCC 3.0 and later versions, GNU compiler testsuites are also included | |
228 | in the full distribution. | |
f42974dc DW |
229 | |
230 | If you choose to download specific components, you must download the core | |
eea81d3e | 231 | GCC distribution plus any language specific distributions you wish to |
6c0a4eab | 232 | use. The core distribution includes the C language front end as well as the |
767094dd | 233 | shared components. Each language has a tarball which includes the language |
6c0a4eab | 234 | front end as well as the language runtime (when appropriate). |
f42974dc DW |
235 | |
236 | Unpack the core distribution as well as any language specific | |
237 | distributions in the same directory. | |
238 | ||
239 | If you also intend to build binutils (either to upgrade an existing | |
240 | installation or for use in place of the corresponding tools of your | |
241 | OS), unpack the binutils distribution either in the same directory or | |
242 | a separate one. In the latter case, add symbolic links to any | |
243 | components of the binutils you intend to build alongside the compiler | |
6cfb3f16 JM |
244 | (@file{bfd}, @file{binutils}, @file{gas}, @file{gprof}, @file{ld}, |
245 | @file{opcodes}, @dots{}) to the directory containing the GCC sources. | |
f42974dc DW |
246 | |
247 | @html | |
b8db17af | 248 | <hr /> |
f42974dc DW |
249 | <p> |
250 | @end html | |
251 | @ifhtml | |
252 | @uref{./index.html,,Return to the GCC Installation page} | |
253 | @end ifhtml | |
254 | @end ifset | |
255 | ||
256 | @c ***Configuration*********************************************************** | |
6cfb3f16 | 257 | @ifnothtml |
f42974dc DW |
258 | @comment node-name, next, previous, up |
259 | @node Configuration, Building, Downloading the source, Installing GCC | |
6cfb3f16 | 260 | @end ifnothtml |
f42974dc | 261 | @ifset configurehtml |
f42974dc DW |
262 | @ifnothtml |
263 | @chapter Installing GCC: Configuration | |
264 | @end ifnothtml | |
265 | @cindex Configuration | |
266 | @cindex Installing GCC: Configuration | |
267 | ||
268 | Like most GNU software, GCC must be configured before it can be built. | |
269 | This document describes the recommended configuration procedure | |
270 | for both native and cross targets. | |
271 | ||
38209993 LG |
272 | We use @var{srcdir} to refer to the toplevel source directory for |
273 | GCC; we use @var{objdir} to refer to the toplevel build/object directory. | |
274 | ||
275 | If you obtained the sources via CVS, @var{srcdir} must refer to the top | |
276 | @file{gcc} directory, the one where the @file{MAINTAINERS} can be found, | |
277 | and not its @file{gcc} subdirectory, otherwise the build will fail. | |
f42974dc | 278 | |
b4b0fb02 RO |
279 | If either @var{srcdir} or @var{objdir} is located on an automounted NFS |
280 | file system, the shell's built-in @command{pwd} command will return | |
281 | temporary pathnames. Using these can lead to various sorts of build | |
282 | problems. To avoid this issue, set the @env{PWDCMD} environment | |
283 | variable to an automounter-aware @command{pwd} command, e.g., | |
284 | @command{pawd} or @command{amq -w}, during the configuration and build | |
285 | phases. | |
286 | ||
102b60d1 GP |
287 | First, we @strong{highly} recommend that GCC be built into a |
288 | separate directory than the sources which does @strong{not} reside | |
289 | within the source tree. This is how we generally build GCC; building | |
290 | where @var{srcdir} == @var{objdir} should still work, but doesn't | |
291 | get extensive testing; building where @var{objdir} is a subdirectory | |
292 | of @var{srcdir} is unsupported. | |
f42974dc | 293 | |
eea81d3e | 294 | If you have previously built GCC in the same directory for a |
f85b8d1a | 295 | different target machine, do @samp{make distclean} to delete all files |
377dfc82 GP |
296 | that might be invalid. One of the files this deletes is @file{Makefile}; |
297 | if @samp{make distclean} complains that @file{Makefile} does not exist | |
298 | or issues a message like ``don't know how to make distclean'' it probably | |
299 | means that the directory is already suitably clean. However, with the | |
300 | recommended method of building in a separate @var{objdir}, you should | |
301 | simply use a different @var{objdir} for each target. | |
f85b8d1a | 302 | |
38209993 LG |
303 | Second, when configuring a native system, either @command{cc} or |
304 | @command{gcc} must be in your path or you must set @env{CC} in | |
df002c7d DE |
305 | your environment before running configure. Otherwise the configuration |
306 | scripts may fail. | |
f42974dc | 307 | |
eea81d3e RO |
308 | Note that the bootstrap compiler and the resulting GCC must be link |
309 | compatible, else the bootstrap will fail with linker errors about | |
310 | incompatible object file formats. Several multilibed targets are | |
e69aa433 GP |
311 | affected by this requirement, see |
312 | @ifnothtml | |
313 | @ref{Specific, host/target specific installation notes}. | |
314 | @end ifnothtml | |
c009f01f | 315 | @ifhtml |
e69aa433 | 316 | @uref{specific.html,,host/target specific installation notes}. |
c009f01f | 317 | @end ifhtml |
eea81d3e | 318 | |
f42974dc DW |
319 | To configure GCC: |
320 | ||
321 | @example | |
38209993 LG |
322 | % mkdir @var{objdir} |
323 | % cd @var{objdir} | |
eea81d3e | 324 | % @var{srcdir}/configure [@var{options}] [@var{target}] |
f42974dc DW |
325 | @end example |
326 | ||
327 | ||
ef88b07d | 328 | @heading Target specification |
f42974dc DW |
329 | @itemize @bullet |
330 | @item | |
38209993 | 331 | GCC has code to correctly determine the correct value for @var{target} |
f9047ed3 | 332 | for nearly all native systems. Therefore, we highly recommend you not |
f42974dc DW |
333 | provide a configure target when configuring a native compiler. |
334 | ||
335 | @item | |
6cfb3f16 | 336 | @var{target} must be specified as @option{--target=@var{target}} |
f9047ed3 | 337 | when configuring a cross compiler; examples of valid targets would be |
f42974dc DW |
338 | i960-rtems, m68k-coff, sh-elf, etc. |
339 | ||
340 | @item | |
6cfb3f16 | 341 | Specifying just @var{target} instead of @option{--target=@var{target}} |
38209993 | 342 | implies that the host defaults to @var{target}. |
f42974dc DW |
343 | @end itemize |
344 | ||
345 | ||
ef88b07d | 346 | @heading Options specification |
f42974dc | 347 | |
ef88b07d | 348 | Use @var{options} to override several configure time options for |
80f9249a JM |
349 | GCC@. A list of supported @var{options} follows; @command{configure |
350 | --help} may list other options, but those not listed below may not | |
351 | work and should not normally be used. | |
f42974dc | 352 | |
ef88b07d JM |
353 | @table @code |
354 | @item --prefix=@var{dirname} | |
355 | Specify the toplevel installation | |
f42974dc DW |
356 | directory. This is the recommended way to install the tools into a directory |
357 | other than the default. The toplevel installation directory defaults to | |
6cfb3f16 | 358 | @file{/usr/local}. |
f42974dc | 359 | |
38209993 LG |
360 | We @strong{highly} recommend against @var{dirname} being the same or a |
361 | subdirectory of @var{objdir} or vice versa. | |
f42974dc DW |
362 | |
363 | These additional options control where certain parts of the distribution | |
364 | are installed. Normally you should not need to use these options. | |
ef88b07d | 365 | @table @code |
ab130aa5 JM |
366 | @item --exec-prefix=@var{dirname} |
367 | Specify the toplevel installation directory for architecture-dependent | |
368 | files. The default is @file{@var{prefix}}. | |
369 | ||
370 | @item --bindir=@var{dirname} | |
371 | Specify the installation directory for the executables called by users | |
372 | (such as @command{gcc} and @command{g++}). The default is | |
373 | @file{@var{exec-prefix}/bin}. | |
374 | ||
375 | @item --libdir=@var{dirname} | |
376 | Specify the installation directory for object code libraries and | |
161d7b59 | 377 | internal parts of GCC@. The default is @file{@var{exec-prefix}/lib}. |
ab130aa5 JM |
378 | |
379 | @item --with-slibdir=@var{dirname} | |
380 | Specify the installation directory for the shared libgcc library. The | |
381 | default is @file{@var{libdir}}. | |
382 | ||
383 | @item --infodir=@var{dirname} | |
384 | Specify the installation directory for documentation in info format. | |
385 | The default is @file{@var{prefix}/info}. | |
386 | ||
8567c70f TT |
387 | @item --datadir=@var{dirname} |
388 | Specify the installation directory for some architecture-independent | |
389 | data files referenced by GCC@. The default is @file{@var{prefix}/share}. | |
390 | ||
ab130aa5 JM |
391 | @item --mandir=@var{dirname} |
392 | Specify the installation directory for manual pages. The default is | |
393 | @file{@var{prefix}/man}. (Note that the manual pages are only extracts from | |
394 | the full GCC manuals, which are provided in Texinfo format. The | |
395 | @command{g77} manpage is unmaintained and may be out of date; the others | |
396 | are derived by an automatic conversion process from parts of the full | |
397 | manual.) | |
398 | ||
ef88b07d JM |
399 | @item --with-gxx-include-dir=@var{dirname} |
400 | Specify | |
eea81d3e | 401 | the installation directory for G++ header files. The default is |
ecb7d6b3 JM |
402 | @file{@var{prefix}/include/g++-v3}. |
403 | ||
ef88b07d | 404 | @end table |
f42974dc | 405 | |
b21d216c AF |
406 | @item --program-prefix=@var{prefix} |
407 | GCC supports some transformations of the names of its programs when | |
408 | installing them. This option prepends @var{prefix} to the names of | |
409 | programs to install in @var{bindir} (see above). For example, specifying | |
410 | @option{--program-prefix=foo-} would result in @samp{gcc} | |
411 | being installed as @file{/usr/local/bin/foo-gcc}. | |
412 | ||
413 | @item --program-suffix=@var{suffix} | |
414 | Appends @var{suffix} to the names of programs to install in @var{bindir} | |
415 | (see above). For example, specifying @option{--program-suffix=-3.1} | |
416 | would result in @samp{gcc} being installed as | |
417 | @file{/usr/local/bin/gcc-3.1}. | |
418 | ||
419 | @item --program-transform-name=@var{pattern} | |
420 | Applies the @samp{sed} script @var{pattern} to be applied to the names | |
421 | of programs to install in @var{bindir} (see above). @var{pattern} has to | |
422 | consist of one or more basic @samp{sed} editing commands, separated by | |
423 | semicolons. For example, if you want the @samp{gcc} program name to be | |
424 | transformed to the installed program @file{/usr/local/bin/myowngcc} and | |
425 | the @samp{g++} program name to be transformed to | |
426 | @file{/usr/local/bin/gspecial++} without changing other program names, | |
427 | you could use the pattern | |
428 | @option{--program-transform-name='s/^gcc$/myowngcc/; s/^g++$/gspecial++/'} | |
429 | to achieve this effect. | |
430 | ||
431 | All three options can be combined and used together, resulting in more | |
432 | complex conversion patterns. As a basic rule, @var{prefix} (and | |
433 | @var{suffix}) are prepended (appended) before further transformations | |
434 | can happen with a special transformation script @var{pattern}. | |
435 | ||
8c085f6f | 436 | As currently implemented, this option only takes effect for native |
b21d216c | 437 | builds; cross compiler binaries' names are not transformed even when a |
8c085f6f | 438 | transformation is explicitly asked for by one of these options. |
b21d216c AF |
439 | |
440 | For native builds, some of the installed programs are also installed | |
441 | with the target alias in front of their name, as in | |
442 | @samp{i686-pc-linux-gnu-gcc}. All of the above transformations happen | |
443 | before the target alias is prepended to the name - so, specifying | |
444 | @option{--program-prefix=foo-} and @option{program-suffix=-3.1}, the | |
445 | resulting binary would be installed as | |
446 | @file{/usr/local/bin/i686-pc-linux-gnu-foo-gcc-3.1}. | |
447 | ||
8ecab453 | 448 | As a last shortcoming, none of the installed Ada programs are |
b21d216c AF |
449 | transformed yet, which will be fixed in some time. |
450 | ||
ef88b07d JM |
451 | @item --with-local-prefix=@var{dirname} |
452 | Specify the | |
6ac48571 JM |
453 | installation directory for local include files. The default is |
454 | @file{/usr/local}. Specify this option if you want the compiler to | |
455 | search directory @file{@var{dirname}/include} for locally installed | |
456 | header files @emph{instead} of @file{/usr/local/include}. | |
457 | ||
458 | You should specify @option{--with-local-prefix} @strong{only} if your | |
459 | site has a different convention (not @file{/usr/local}) for where to put | |
460 | site-specific files. | |
461 | ||
462 | The default value for @option{--with-local-prefix} is @file{/usr/local} | |
463 | regardless of the value of @option{--prefix}. Specifying | |
464 | @option{--prefix} has no effect on which directory GCC searches for | |
465 | local header files. This may seem counterintuitive, but actually it is | |
466 | logical. | |
467 | ||
468 | The purpose of @option{--prefix} is to specify where to @emph{install | |
469 | GCC}. The local header files in @file{/usr/local/include}---if you put | |
161d7b59 | 470 | any in that directory---are not part of GCC@. They are part of other |
6ac48571 JM |
471 | programs---perhaps many others. (GCC installs its own header files in |
472 | another directory which is based on the @option{--prefix} value.) | |
473 | ||
48209ce5 JDA |
474 | Both the local-prefix include directory and the GCC-prefix include |
475 | directory are part of GCC's "system include" directories. Although these | |
476 | two directories are not fixed, they need to be searched in the proper | |
477 | order for the correct processing of the include_next directive. The | |
478 | local-prefix include directory is searched before the GCC-prefix | |
479 | include directory. Another characteristic of system include directories | |
480 | is that pedantic warnings are turned off for headers in these directories. | |
481 | ||
482 | Some autoconf macros add @option{-I @var{directory}} options to the | |
483 | compiler command line, to ensure that directories containing installed | |
484 | packages' headers are searched. When @var{directory} is one of GCC's | |
485 | system include directories, GCC will ignore the option so that system | |
486 | directories continue to be processed in the correct order. This | |
487 | may result in a search order different from what was specified but the | |
488 | directory will still be searched. | |
489 | ||
490 | GCC automatically searches for ordinary libraries using | |
491 | @env{GCC_EXEC_PREFIX}. Thus, when the same installation prefix is | |
492 | used for both GCC and packages, GCC will automatically search for | |
493 | both headers and libraries. This provides a configuration that is | |
494 | easy to use. GCC behaves in a manner similar to that when it is | |
495 | installed as a system compiler in @file{/usr}. | |
496 | ||
497 | Sites that need to install multiple versions of GCC may not want to | |
498 | use the above simple configuration. It is possible to use the | |
499 | @option{--program-prefix}, @option{--program-suffix} and | |
500 | @option{--program-transform-name} options to install multiple versions | |
501 | into a single directory, but it may be simpler to use different prefixes | |
502 | and the @option{--with-local-prefix} option to specify the location of the | |
503 | site-specific files for each version. It will then be necessary for | |
504 | users to specify explicitly the location of local site libraries | |
505 | (e.g., with @env{LIBRARY_PATH}). | |
506 | ||
507 | The same value can be used for both @option{--with-local-prefix} and | |
508 | @option{--prefix} provided it is not @file{/usr}. This can be used | |
509 | to avoid the default search of @file{/usr/local/include}. | |
510 | ||
6ac48571 JM |
511 | @strong{Do not} specify @file{/usr} as the @option{--with-local-prefix}! |
512 | The directory you use for @option{--with-local-prefix} @strong{must not} | |
513 | contain any of the system's standard header files. If it did contain | |
514 | them, certain programs would be miscompiled (including GNU Emacs, on | |
515 | certain targets), because this would override and nullify the header | |
4c64396e | 516 | file corrections made by the @command{fixincludes} script. |
6ac48571 JM |
517 | |
518 | Indications are that people who use this option use it based on mistaken | |
519 | ideas of what it is for. People use it as if it specified where to | |
161d7b59 | 520 | install part of GCC@. Perhaps they make this assumption because |
6ac48571 JM |
521 | installing GCC creates the directory. |
522 | ||
6cfb3f16 | 523 | @item --enable-shared[=@var{package}[,@dots{}]] |
0cb98517 AO |
524 | Build shared versions of libraries, if shared libraries are supported on |
525 | the target platform. Unlike GCC 2.95.x and earlier, shared libraries | |
eea81d3e RO |
526 | are enabled by default on all platforms that support shared libraries, |
527 | except for @samp{libobjc} which is built as a static library only by | |
528 | default. | |
0cb98517 AO |
529 | |
530 | If a list of packages is given as an argument, build shared libraries | |
531 | only for the listed packages. For other packages, only static libraries | |
532 | will be built. Package names currently recognized in the GCC tree are | |
533 | @samp{libgcc} (also known as @samp{gcc}), @samp{libstdc++} (not | |
534 | @samp{libstdc++-v3}), @samp{libffi}, @samp{zlib}, @samp{boehm-gc} and | |
535 | @samp{libjava}. Note that @samp{libobjc} does not recognize itself by | |
536 | any name, so, if you list package names in @option{--enable-shared}, | |
eea81d3e | 537 | you will only get static Objective-C libraries. @samp{libf2c} and |
0cb98517 AO |
538 | @samp{libiberty} do not support shared libraries at all. |
539 | ||
540 | Use @option{--disable-shared} to build only static libraries. Note that | |
541 | @option{--disable-shared} does not accept a list of package names as | |
542 | argument, only @option{--enable-shared} does. | |
f42974dc | 543 | |
ef88b07d JM |
544 | @item @anchor{with-gnu-as}--with-gnu-as |
545 | Specify that the compiler should assume that the | |
767094dd | 546 | assembler it finds is the GNU assembler. However, this does not modify |
377dfc82 GP |
547 | the rules to find an assembler and will result in confusion if the |
548 | assembler found is not actually the GNU assembler. (Confusion may also | |
8c26c999 JM |
549 | result if the compiler finds the GNU assembler but has not been |
550 | configured with @option{--with-gnu-as}.) If you have more than one | |
38209993 | 551 | assembler installed on your system, you may want to use this option in |
eea81d3e | 552 | connection with @option{--with-as=@var{pathname}}. |
38209993 | 553 | |
8c085f6f JJ |
554 | The following systems are the only ones where it makes a difference |
555 | whether you use the GNU assembler. On any other system, | |
556 | @option{--with-gnu-as} has no effect. | |
557 | ||
558 | @itemize bullet | |
559 | @item @samp{hppa1.0-@var{any}-@var{any}} | |
560 | @item @samp{hppa1.1-@var{any}-@var{any}} | |
561 | @item @samp{i386-@var{any}-sysv} | |
8c085f6f JJ |
562 | @item @samp{m68k-bull-sysv} |
563 | @item @samp{m68k-hp-hpux} | |
8c085f6f JJ |
564 | @item @samp{m68000-hp-hpux} |
565 | @item @samp{m68000-att-sysv} | |
566 | @item @samp{@var{any}-lynx-lynxos} | |
567 | @item @samp{mips-@var{any}} | |
568 | @end itemize | |
8c26c999 JM |
569 | |
570 | On the systems listed above (except for the HP-PA, for ISC on the | |
571 | 386, and for @samp{mips-sgi-irix5.*}), if you use the GNU assembler, | |
572 | you should also use the GNU linker (and specify @option{--with-gnu-ld}). | |
573 | ||
eea81d3e | 574 | @item --with-as=@var{pathname} |
ef88b07d | 575 | Specify that the |
38209993 LG |
576 | compiler should use the assembler pointed to by @var{pathname}, rather |
577 | than the one found by the standard rules to find an assembler, which | |
578 | are: | |
f42974dc DW |
579 | @itemize @bullet |
580 | @item | |
38209993 LG |
581 | Check the |
582 | @file{@var{exec_prefix}/lib/gcc-lib/@var{target}/@var{version}} | |
583 | directory, where @var{exec_prefix} defaults to @var{prefix} which | |
584 | defaults to @file{/usr/local} unless overridden by the | |
eea81d3e | 585 | @option{--prefix=@var{pathname}} switch described above. @var{target} is the |
b953cc4b | 586 | target system triple, such as @samp{sparc-sun-solaris2.7}, and |
eea81d3e | 587 | @var{version} denotes the GCC version, such as 3.0. |
f42974dc | 588 | @item |
e979f9e8 | 589 | Check operating system specific directories (e.g.@: @file{/usr/ccs/bin} on |
250d5688 | 590 | Sun Solaris 2). |
f42974dc | 591 | @end itemize |
767094dd | 592 | Note that these rules do not check for the value of @env{PATH}. You may |
38209993 LG |
593 | want to use @option{--with-as} if no assembler is installed in the |
594 | directories listed above, or if you have multiple assemblers installed | |
595 | and want to choose one that is not found by the above rules. | |
f42974dc | 596 | |
ef88b07d JM |
597 | @item @anchor{with-gnu-ld}--with-gnu-ld |
598 | Same as @uref{#with-gnu-as,,@option{--with-gnu-as}} | |
38209993 | 599 | but for linker. |
20293b4c | 600 | |
f42974dc | 601 | |
eea81d3e | 602 | @item --with-ld=@var{pathname} |
ef88b07d | 603 | Same as |
38209993 | 604 | @option{--with-as}, but for the linker. |
f42974dc | 605 | |
ef88b07d JM |
606 | @item --with-stabs |
607 | Specify that stabs debugging | |
38209993 LG |
608 | information should be used instead of whatever format the host normally |
609 | uses. Normally GCC uses the same debug format as the host system. | |
f42974dc | 610 | |
8c26c999 JM |
611 | On MIPS based systems and on Alphas, you must specify whether you want |
612 | GCC to create the normal ECOFF debugging format, or to use BSD-style | |
613 | stabs passed through the ECOFF symbol table. The normal ECOFF debug | |
161d7b59 JM |
614 | format cannot fully handle languages other than C@. BSD stabs format can |
615 | handle other languages, but it only works with the GNU debugger GDB@. | |
8c26c999 JM |
616 | |
617 | Normally, GCC uses the ECOFF debugging format by default; if you | |
161d7b59 | 618 | prefer BSD stabs, specify @option{--with-stabs} when you configure GCC@. |
8c26c999 JM |
619 | |
620 | No matter which default you choose when you configure GCC, the user | |
621 | can use the @option{-gcoff} and @option{-gstabs+} options to specify explicitly | |
622 | the debug format for a particular compilation. | |
623 | ||
624 | @option{--with-stabs} is meaningful on the ISC system on the 386, also, if | |
625 | @option{--with-gas} is used. It selects use of stabs debugging | |
626 | information embedded in COFF output. This kind of debugging information | |
627 | supports C++ well; ordinary COFF debugging information does not. | |
628 | ||
629 | @option{--with-stabs} is also meaningful on 386 systems running SVR4. It | |
630 | selects use of stabs debugging information embedded in ELF output. The | |
631 | C++ compiler currently (2.6.0) does not support the DWARF debugging | |
632 | information normally used on 386 SVR4 platforms; stabs provide a | |
633 | workable alternative. This requires gas and gdb, as the normal SVR4 | |
634 | tools can not generate or interpret stabs. | |
635 | ||
eea81d3e | 636 | @item --disable-multilib |
ef88b07d | 637 | Specify that multiple target |
eea81d3e RO |
638 | libraries to support different target variants, calling |
639 | conventions, etc should not be built. The default is to build a | |
640 | predefined set of them. | |
f42974dc | 641 | |
e8515283 DE |
642 | Some targets provide finer-grained control over which multilibs are built |
643 | (e.g., @option{--disable-softfloat}): | |
644 | @table @code | |
e8515283 DE |
645 | @item arc-*-elf* |
646 | biendian. | |
647 | ||
648 | @item arm-*-* | |
649 | fpu, 26bit, underscore, interwork, biendian, nofmult. | |
650 | ||
651 | @item m68*-*-* | |
652 | softfloat, m68881, m68000, m68020. | |
653 | ||
654 | @item mips*-*-* | |
655 | single-float, biendian, softfloat. | |
656 | ||
657 | @item powerpc*-*-*, rs6000*-*-* | |
658 | aix64, pthread, softfloat, powercpu, powerpccpu, powerpcos, biendian, | |
f282ffb3 | 659 | sysv, aix. |
e8515283 DE |
660 | |
661 | @end table | |
662 | ||
ef88b07d JM |
663 | @item --enable-threads |
664 | Specify that the target | |
38209993 LG |
665 | supports threads. This affects the Objective-C compiler and runtime |
666 | library, and exception handling for other languages like C++ and Java. | |
6ac48571 | 667 | On some systems, this is the default. |
f42974dc | 668 | |
f6160ed5 LR |
669 | In general, the best (and, in many cases, the only known) threading |
670 | model available will be configured for use. Beware that on some | |
671 | systems, gcc has not been taught what threading models are generally | |
3c6bb1db LR |
672 | available for the system. In this case, @option{--enable-threads} is an |
673 | alias for @option{--enable-threads=single}. | |
f6160ed5 LR |
674 | |
675 | @item --disable-threads | |
676 | Specify that threading support should be disabled for the system. | |
3c6bb1db | 677 | This is an alias for @option{--enable-threads=single}. |
f6160ed5 | 678 | |
ef88b07d JM |
679 | @item --enable-threads=@var{lib} |
680 | Specify that | |
38209993 LG |
681 | @var{lib} is the thread support library. This affects the Objective-C |
682 | compiler and runtime library, and exception handling for other languages | |
f85b8d1a JM |
683 | like C++ and Java. The possibilities for @var{lib} are: |
684 | ||
685 | @table @code | |
686 | @item aix | |
687 | AIX thread support. | |
688 | @item dce | |
689 | DCE thread support. | |
f85b8d1a | 690 | @item mach |
eea81d3e | 691 | Generic MACH thread support, known to work on NeXTSTEP@. (Please note |
3c6bb1db | 692 | that the file needed to support this configuration, @file{gthr-mach.h}, is |
f6160ed5 LR |
693 | missing and thus this setting will cause a known bootstrap failure.) |
694 | @item no | |
695 | This is an alias for @samp{single}. | |
f85b8d1a | 696 | @item posix |
c771326b | 697 | Generic POSIX thread support. |
f85b8d1a | 698 | @item pthreads |
f6160ed5 LR |
699 | Same as @samp{posix} on arm*-*-linux*, *-*-chorusos* and *-*-freebsd* |
700 | only. A future release of gcc might remove this alias or extend it | |
701 | to all platforms. | |
702 | @item rtems | |
703 | RTEMS thread support. | |
f85b8d1a JM |
704 | @item single |
705 | Disable thread support, should work for all platforms. | |
706 | @item solaris | |
eea81d3e | 707 | Sun Solaris 2 thread support. |
f85b8d1a JM |
708 | @item vxworks |
709 | VxWorks thread support. | |
710 | @item win32 | |
711 | Microsoft Win32 API thread support. | |
712 | @end table | |
f42974dc | 713 | |
ef88b07d JM |
714 | @item --with-cpu=@var{cpu} |
715 | Specify which cpu variant the | |
f42974dc | 716 | compiler should generate code for by default. This is currently |
377dfc82 | 717 | only supported on some ports, specifically arm, powerpc, and |
161d7b59 | 718 | SPARC@. If configure does not recognize the model name (e.g.@: arm700, |
ec5b5ef1 GP |
719 | 603e, or ultrasparc) you provide, please check the |
720 | @command{gcc/config.gcc} script for a complete list of supported models. | |
f42974dc | 721 | |
a004eb82 AH |
722 | @item --enable-altivec |
723 | Specify that the target supports AltiVec vector enhancements. This | |
724 | option will adjust the ABI for AltiVec enhancements, as well as generate | |
725 | AltiVec code when appropriate. This option is only available for | |
726 | PowerPC systems. | |
727 | ||
ef88b07d JM |
728 | @item --enable-target-optspace |
729 | Specify that target | |
38209993 LG |
730 | libraries should be optimized for code space instead of code speed. |
731 | This is the default for the m32r platform. | |
f42974dc | 732 | |
ab130aa5 JM |
733 | @item --disable-cpp |
734 | Specify that a user visible @command{cpp} program should not be installed. | |
735 | ||
736 | @item --with-cpp-install-dir=@var{dirname} | |
737 | Specify that the user visible @command{cpp} program should be installed | |
738 | in @file{@var{prefix}/@var{dirname}/cpp}, in addition to @var{bindir}. | |
f42974dc | 739 | |
07cf4226 DM |
740 | @item --enable-initfini-array |
741 | Force the use of sections @code{.init_array} and @code{.fini_array} | |
742 | (instead of @code{.init} and @code{.fini}) for constructors and | |
743 | destructors. Option @option{--disable-initfini-array} has the | |
744 | opposite effect. If neither option is specified, the configure script | |
745 | will try to guess whether the @code{.init_array} and | |
746 | @code{.fini_array} sections are supported and, if they are, use them. | |
747 | ||
ef88b07d JM |
748 | @item --enable-maintainer-mode |
749 | The build rules that | |
6cfb3f16 | 750 | regenerate the GCC master message catalog @file{gcc.pot} are normally |
767094dd JM |
751 | disabled. This is because it can only be rebuilt if the complete source |
752 | tree is present. If you have changed the sources and want to rebuild the | |
6ac48571 | 753 | catalog, configuring with @option{--enable-maintainer-mode} will enable |
767094dd | 754 | this. Note that you need a recent version of the @code{gettext} tools |
6ac48571 JM |
755 | to do so. |
756 | ||
ef88b07d JM |
757 | @item --enable-version-specific-runtime-libs |
758 | Specify | |
38209993 LG |
759 | that runtime libraries should be installed in the compiler specific |
760 | subdirectory (@file{@var{libsubdir}}) rather than the usual places. In | |
eea81d3e | 761 | addition, @samp{libstdc++}'s include files will be installed in |
38209993 | 762 | @file{@var{libsubdir}/include/g++} unless you overruled it by using |
6cfb3f16 | 763 | @option{--with-gxx-include-dir=@var{dirname}}. Using this option is |
38209993 | 764 | particularly useful if you intend to use several versions of GCC in |
eea81d3e RO |
765 | parallel. This is currently supported by @samp{libf2c} and |
766 | @samp{libstdc++}, and is the default for @samp{libobjc} which cannot be | |
767 | changed in this case. | |
38209993 | 768 | |
ef88b07d JM |
769 | @item --enable-languages=@var{lang1},@var{lang2},@dots{} |
770 | Specify that only a particular subset of compilers and | |
767094dd | 771 | their runtime libraries should be built. For a list of valid values for |
6cfb3f16 | 772 | @var{langN} you can issue the following command in the |
eea81d3e RO |
773 | @file{gcc} directory of your GCC source tree:@* |
774 | @example | |
775 | grep language= */config-lang.in | |
776 | @end example | |
777 | Currently, you can use any of the following: | |
e23381df | 778 | @code{ada}, @code{c}, @code{c++}, @code{f77}, @code{java}, @code{objc}. |
8ecab453 | 779 | Building the Ada compiler has special requirements, see below.@* |
38209993 | 780 | If you do not pass this flag, all languages available in the @file{gcc} |
6cfb3f16 | 781 | sub-tree will be configured. Re-defining @code{LANGUAGES} when calling |
ef88b07d | 782 | @samp{make bootstrap} @strong{does not} work anymore, as those |
38209993 | 783 | language sub-directories might not have been configured! |
f42974dc | 784 | |
ef88b07d JM |
785 | @item --disable-libgcj |
786 | Specify that the run-time libraries | |
f42974dc DW |
787 | used by GCJ should not be built. This is useful in case you intend |
788 | to use GCJ with some other run-time, or you're going to install it | |
789 | separately, or it just happens not to build on your particular | |
6c0a4eab | 790 | machine. In general, if the Java front end is enabled, the GCJ |
f42974dc | 791 | libraries will be enabled too, unless they're known to not work on |
eea81d3e | 792 | the target platform. If GCJ is enabled but @samp{libgcj} isn't built, you |
f42974dc | 793 | may need to port it; in this case, before modifying the top-level |
eea81d3e | 794 | @file{configure.in} so that @samp{libgcj} is enabled by default on this platform, |
f42974dc DW |
795 | you may use @option{--enable-libgcj} to override the default. |
796 | ||
ef88b07d JM |
797 | @item --with-dwarf2 |
798 | Specify that the compiler should | |
eea81d3e | 799 | use DWARF 2 debugging information as the default. |
f85b8d1a JM |
800 | |
801 | @item --enable-win32-registry | |
eea81d3e | 802 | @itemx --enable-win32-registry=@var{key} |
f85b8d1a | 803 | @itemx --disable-win32-registry |
6cfb3f16 | 804 | The @option{--enable-win32-registry} option enables Windows-hosted GCC |
f85b8d1a JM |
805 | to look up installations paths in the registry using the following key: |
806 | ||
807 | @smallexample | |
eea81d3e | 808 | @code{HKEY_LOCAL_MACHINE\SOFTWARE\Free Software Foundation\@var{key}} |
f85b8d1a JM |
809 | @end smallexample |
810 | ||
eea81d3e RO |
811 | @var{key} defaults to GCC version number, and can be overridden by the |
812 | @option{--enable-win32-registry=@var{key}} option. Vendors and distributors | |
f85b8d1a JM |
813 | who use custom installers are encouraged to provide a different key, |
814 | perhaps one comprised of vendor name and GCC version number, to | |
767094dd | 815 | avoid conflict with existing installations. This feature is enabled |
6cfb3f16 | 816 | by default, and can be disabled by @option{--disable-win32-registry} |
f85b8d1a JM |
817 | option. This option has no effect on the other hosts. |
818 | ||
819 | @item --nfp | |
820 | Specify that the machine does not have a floating point unit. This | |
c9693e96 LH |
821 | option only applies to @samp{m68k-sun-sunos@var{n}}. On any other |
822 | system, @option{--nfp} has no effect. | |
f85b8d1a JM |
823 | |
824 | @item --enable-checking | |
825 | @itemx --enable-checking=@var{list} | |
826 | When you specify this option, the compiler is built to perform checking | |
827 | of tree node types when referencing fields of that node, and some other | |
828 | internal consistency checks. This does not change the generated code, | |
829 | but adds error checking within the compiler. This will slow down the | |
830 | compiler and may only work properly if you are building the compiler | |
161d7b59 | 831 | with GCC@. This is on by default when building from CVS or snapshots, |
f85b8d1a JM |
832 | but off for releases. More control over the checks may be had by |
833 | specifying @var{list}; the categories of checks available are | |
4c76f856 JJ |
834 | @samp{misc}, @samp{tree}, @samp{gc}, @samp{rtl}, @samp{rtlflag} and |
835 | @samp{gcac}. The | |
836 | default when @var{list} is not specified is @samp{misc,tree,gc,rtlflag}; the | |
f85b8d1a JM |
837 | checks @samp{rtl} and @samp{gcac} are very expensive. |
838 | ||
22aa533e NS |
839 | @item --enable-coverage |
840 | @item --enable-coverage=@var{level} | |
841 | With this option, the compiler is built to collect self coverage | |
842 | information, every time it is run. This is for internal development | |
843 | purposes, and only works when the compiler is being built with gcc. The | |
844 | @var{level} argument controls whether the compiler is built optimized or | |
845 | not, values are @samp{opt} and @samp{noopt}. For coverage analysis you | |
846 | want to disable optimization, for performance analysis you want to | |
847 | enable optimization. When coverage is enabled, the default level is | |
848 | without optimization. | |
849 | ||
f85b8d1a JM |
850 | @item --enable-nls |
851 | @itemx --disable-nls | |
6cfb3f16 | 852 | The @option{--enable-nls} option enables Native Language Support (NLS), |
f85b8d1a | 853 | which lets GCC output diagnostics in languages other than American |
767094dd | 854 | English. Native Language Support is enabled by default if not doing a |
161d7b59 | 855 | canadian cross build. The @option{--disable-nls} option disables NLS@. |
f85b8d1a JM |
856 | |
857 | @item --with-included-gettext | |
c771326b | 858 | If NLS is enabled, the @option{--with-included-gettext} option causes the build |
021c4bfd | 859 | procedure to prefer its copy of GNU @command{gettext}. |
f85b8d1a JM |
860 | |
861 | @item --with-catgets | |
862 | If NLS is enabled, and if the host lacks @code{gettext} but has the | |
863 | inferior @code{catgets} interface, the GCC build procedure normally | |
864 | ignores @code{catgets} and instead uses GCC's copy of the GNU | |
6cfb3f16 | 865 | @code{gettext} library. The @option{--with-catgets} option causes the |
f85b8d1a | 866 | build procedure to use the host's @code{catgets} in this situation. |
80f9249a | 867 | |
5304400d CR |
868 | @item --with-libiconv-prefix=@var{dir} |
869 | Search for libiconv header files in @file{@var{dir}/include} and | |
870 | libiconv library files in @file{@var{dir}/lib}. | |
871 | ||
80f9249a JM |
872 | @item --with-system-zlib |
873 | Use installed zlib rather than that included with GCC@. This option | |
874 | only applies if the Java front end is being built. | |
9340544b ZW |
875 | |
876 | @item --enable-obsolete | |
877 | Enable configuration for an obsoleted system. If you attempt to | |
878 | configure GCC for a system (build, host, or target) which has been | |
879 | obsoleted, and you do not specify this flag, configure will halt with an | |
880 | error message. | |
881 | ||
882 | All support for systems which have been obsoleted in one release of GCC | |
883 | is removed entirely in the next major release, unless someone steps | |
884 | forward to maintain the port. | |
ef88b07d | 885 | @end table |
f42974dc DW |
886 | |
887 | Some options which only apply to building cross compilers: | |
ef88b07d | 888 | @table @code |
4977bab6 ZW |
889 | @item --with-sysroot |
890 | @itemx --with-sysroot=@var{dir} | |
891 | Tells GCC to consider @var{dir} as the root of a tree that contains a | |
892 | (subset of) the root filesystem of the target operating system. | |
893 | Target system headers, libraries and run-time object files will be | |
894 | searched in there. The specified directory is not copied into the | |
895 | install tree, unlike the options @option{--with-headers} and | |
896 | @option{--with-libs} that this option obsoletes. The default value, | |
897 | in case @option{--with-sysroot} is not given an argument, is | |
898 | @option{$@{gcc_tooldir@}/sys-root}. Starting the pathname with | |
899 | @option{$@{gcc_tooldir@}/}, in such a way that this variable is not | |
900 | expanded by the shell, but rather by the gcc Makefile, enables the | |
901 | target sysroot directory to be relocatable along with the entire | |
902 | install tree, but it will require the gcc tool directory to be created | |
903 | @emph{before} the build is started. This directory, as well as its | |
904 | intermediate pathnames, can be easily created by running @command{make | |
905 | install-gcc-tooldir} the gcc build directory. This is not done | |
906 | automatically to avoid touching the install tree without explicit | |
907 | permission. | |
908 | ||
65a824f6 JT |
909 | @item --with-headers |
910 | @itemx --with-headers=@var{dir} | |
4977bab6 | 911 | Deprecated in favor of @option{--with-sysroot}. |
65a824f6 JT |
912 | Specifies that target headers are available when building a cross compiler. |
913 | The @var{dir} argument specifies a directory which has the target include | |
914 | files. These include files will be copied into the @file{gcc} install | |
915 | directory. @emph{This option with the @var{dir} argument is required} when | |
916 | building a cross compiler, if @file{@var{prefix}/@var{target}/sys-include} | |
917 | doesn't pre-exist. If @file{@var{prefix}/@var{target}/sys-include} does | |
918 | pre-exist, the @var{dir} argument may be omitted. @command{fixincludes} | |
919 | will be run on these files to make them compatible with GCC. | |
920 | @item --with-libs | |
921 | @itemx --with-libs=``@var{dir1} @var{dir2} @dots{} @var{dirN}'' | |
4977bab6 | 922 | Deprecated in favor of @option{--with-sysroot}. |
38209993 LG |
923 | Specifies a list of directories which contain the target runtime |
924 | libraries. These libraries will be copied into the @file{gcc} install | |
65a824f6 JT |
925 | directory. If the directory list is omitted, this option has no |
926 | effect. | |
ef88b07d | 927 | @item --with-newlib |
eea81d3e | 928 | Specifies that @samp{newlib} is |
38209993 | 929 | being used as the target C library. This causes @code{__eprintf} to be |
eea81d3e RO |
930 | omitted from @file{libgcc.a} on the assumption that it will be provided by |
931 | @samp{newlib}. | |
ef88b07d | 932 | @end table |
f9047ed3 | 933 | |
38209993 LG |
934 | Note that each @option{--enable} option has a corresponding |
935 | @option{--disable} option and that each @option{--with} option has a | |
936 | corresponding @option{--without} option. | |
f42974dc DW |
937 | |
938 | @html | |
b8db17af | 939 | <hr /> |
f42974dc DW |
940 | <p> |
941 | @end html | |
942 | @ifhtml | |
943 | @uref{./index.html,,Return to the GCC Installation page} | |
944 | @end ifhtml | |
945 | @end ifset | |
946 | ||
947 | @c ***Building**************************************************************** | |
6cfb3f16 | 948 | @ifnothtml |
f42974dc DW |
949 | @comment node-name, next, previous, up |
950 | @node Building, Testing, Configuration, Installing GCC | |
6cfb3f16 | 951 | @end ifnothtml |
f42974dc | 952 | @ifset buildhtml |
f42974dc DW |
953 | @ifnothtml |
954 | @chapter Building | |
955 | @end ifnothtml | |
956 | @cindex Installing GCC: Building | |
957 | ||
958 | Now that GCC is configured, you are ready to build the compiler and | |
959 | runtime libraries. | |
960 | ||
58db9d1a | 961 | We @strong{highly} recommend that GCC be built using GNU make; |
f282ffb3 | 962 | other versions may work, then again they might not. |
58db9d1a AH |
963 | GNU make is required for compiling GNAT (the Ada compiler) and the Java |
964 | runtime library. | |
f42974dc DW |
965 | |
966 | (For example, many broken versions of make will fail if you use the | |
b8df899a JM |
967 | recommended setup where @var{objdir} is different from @var{srcdir}. |
968 | Other broken versions may recompile parts of the compiler when | |
969 | installing the compiler.) | |
f42974dc | 970 | |
b8df899a | 971 | Some commands executed when making the compiler may fail (return a |
df2a54e9 | 972 | nonzero status) and be ignored by @code{make}. These failures, which |
b8df899a JM |
973 | are often due to files that were not found, are expected, and can safely |
974 | be ignored. | |
975 | ||
976 | It is normal to have compiler warnings when compiling certain files. | |
977 | Unless you are a GCC developer, you can generally ignore these warnings | |
978 | unless they cause compilation to fail. | |
979 | ||
980 | On certain old systems, defining certain environment variables such as | |
6cfb3f16 | 981 | @env{CC} can interfere with the functioning of @command{make}. |
b8df899a JM |
982 | |
983 | If you encounter seemingly strange errors when trying to build the | |
984 | compiler in a directory other than the source directory, it could be | |
985 | because you have previously configured the compiler in the source | |
986 | directory. Make sure you have done all the necessary preparations. | |
987 | ||
988 | If you build GCC on a BSD system using a directory stored in an old System | |
4c64396e | 989 | V file system, problems may occur in running @command{fixincludes} if the |
b8df899a JM |
990 | System V file system doesn't support symbolic links. These problems |
991 | result in a failure to fix the declaration of @code{size_t} in | |
992 | @file{sys/types.h}. If you find that @code{size_t} is a signed type and | |
993 | that type mismatches occur, this could be the cause. | |
994 | ||
161d7b59 | 995 | The solution is not to use such a directory for building GCC@. |
f42974dc | 996 | |
f85b8d1a JM |
997 | When building from CVS or snapshots, or if you modify parser sources, |
998 | you need the Bison parser generator installed. Any version 1.25 or | |
999 | later should work; older versions may also work. If you do not modify | |
1000 | parser sources, releases contain the Bison-generated files and you do | |
1001 | not need Bison installed to build them. | |
1002 | ||
1003 | When building from CVS or snapshots, or if you modify Texinfo | |
a38f87a9 | 1004 | documentation, you need version 4.2 or later of Texinfo installed if you |
f85b8d1a JM |
1005 | want Info documentation to be regenerated. Releases contain Info |
1006 | documentation pre-built for the unmodified documentation in the release. | |
1007 | ||
f42974dc DW |
1008 | @section Building a native compiler |
1009 | ||
f9047ed3 | 1010 | For a native build issue the command @samp{make bootstrap}. This |
f42974dc DW |
1011 | will build the entire GCC system, which includes the following steps: |
1012 | ||
1013 | @itemize @bullet | |
1014 | @item | |
1015 | Build host tools necessary to build the compiler such as texinfo, bison, | |
1016 | gperf. | |
1017 | ||
1018 | @item | |
1019 | Build target tools for use by the compiler such as binutils (bfd, | |
eea81d3e | 1020 | binutils, gas, gprof, ld, and opcodes) |
f282ffb3 | 1021 | if they have been individually linked |
f42974dc DW |
1022 | or moved into the top level GCC source tree before configuring. |
1023 | ||
1024 | @item | |
1025 | Perform a 3-stage bootstrap of the compiler. | |
1026 | ||
1027 | @item | |
1028 | Perform a comparison test of the stage2 and stage3 compilers. | |
1029 | ||
1030 | @item | |
1031 | Build runtime libraries using the stage3 compiler from the previous step. | |
f9047ed3 | 1032 | |
f42974dc DW |
1033 | @end itemize |
1034 | ||
38209993 LG |
1035 | If you are short on disk space you might consider @samp{make |
1036 | bootstrap-lean} instead. This is identical to @samp{make | |
1037 | bootstrap} except that object files from the stage1 and | |
f42974dc DW |
1038 | stage2 of the 3-stage bootstrap of the compiler are deleted as |
1039 | soon as they are no longer needed. | |
1040 | ||
f42974dc DW |
1041 | If you want to save additional space during the bootstrap and in |
1042 | the final installation as well, you can build the compiler binaries | |
8c085f6f | 1043 | without debugging information as in the following example. This will save |
f42974dc DW |
1044 | roughly 40% of disk space both for the bootstrap and the final installation. |
1045 | (Libraries will still contain debugging information.) | |
1046 | ||
8c085f6f JJ |
1047 | @example |
1048 | make CFLAGS='-O' LIBCFLAGS='-g -O2' \ | |
1049 | LIBCXXFLAGS='-g -O2 -fno-implicit-templates' bootstrap | |
1050 | @end example | |
1051 | ||
eea81d3e RO |
1052 | If you wish to use non-default GCC flags when compiling the stage2 and |
1053 | stage3 compilers, set @code{BOOT_CFLAGS} on the command line when doing | |
f85b8d1a JM |
1054 | @samp{make bootstrap}. Non-default optimization flags are less well |
1055 | tested here than the default of @samp{-g -O2}, but should still work. | |
1056 | In a few cases, you may find that you need to specify special flags such | |
1057 | as @option{-msoft-float} here to complete the bootstrap; or, if the | |
1058 | native compiler miscompiles the stage1 compiler, you may need to work | |
1059 | around this, by choosing @code{BOOT_CFLAGS} to avoid the parts of the | |
1060 | stage1 compiler that were miscompiled, or by using @samp{make | |
1061 | bootstrap4} to increase the number of stages of bootstrap. | |
1062 | ||
6cfb3f16 | 1063 | If you used the flag @option{--enable-languages=@dots{}} to restrict |
f42974dc | 1064 | the compilers to be built, only those you've actually enabled will be |
767094dd | 1065 | built. This will of course only build those runtime libraries, for |
f42974dc | 1066 | which the particular compiler has been built. Please note, |
eea81d3e | 1067 | that re-defining @env{LANGUAGES} when calling @samp{make bootstrap} |
ef88b07d | 1068 | @strong{does not} work anymore! |
f42974dc | 1069 | |
f85b8d1a | 1070 | If the comparison of stage2 and stage3 fails, this normally indicates |
eea81d3e | 1071 | that the stage2 compiler has compiled GCC incorrectly, and is therefore |
f85b8d1a JM |
1072 | a potentially serious bug which you should investigate and report. (On |
1073 | a few systems, meaningful comparison of object files is impossible; they | |
1074 | always appear ``different''. If you encounter this problem, you will | |
1075 | need to disable comparison in the @file{Makefile}.) | |
f42974dc DW |
1076 | |
1077 | @section Building a cross compiler | |
1078 | ||
1079 | We recommend reading the | |
1080 | @uref{http://www.objsw.com/CrossGCC/,,crossgcc FAQ} | |
1081 | for information about building cross compilers. | |
1082 | ||
1083 | When building a cross compiler, it is not generally possible to do a | |
1084 | 3-stage bootstrap of the compiler. This makes for an interesting problem | |
161d7b59 | 1085 | as parts of GCC can only be built with GCC@. |
f42974dc DW |
1086 | |
1087 | To build a cross compiler, we first recommend building and installing a | |
1088 | native compiler. You can then use the native GCC compiler to build the | |
635771af JM |
1089 | cross compiler. The installed native compiler needs to be GCC version |
1090 | 2.95 or later. | |
f42974dc DW |
1091 | |
1092 | Assuming you have already installed a native copy of GCC and configured | |
6cfb3f16 | 1093 | your cross compiler, issue the command @command{make}, which performs the |
f42974dc DW |
1094 | following steps: |
1095 | ||
1096 | @itemize @bullet | |
1097 | @item | |
1098 | Build host tools necessary to build the compiler such as texinfo, bison, | |
1099 | gperf. | |
1100 | ||
1101 | @item | |
1102 | Build target tools for use by the compiler such as binutils (bfd, | |
1103 | binutils, gas, gprof, ld, and opcodes) | |
1104 | if they have been individually linked or moved into the top level GCC source | |
1105 | tree before configuring. | |
1106 | ||
1107 | @item | |
1108 | Build the compiler (single stage only). | |
1109 | ||
1110 | @item | |
1111 | Build runtime libraries using the compiler from the previous step. | |
1112 | @end itemize | |
1113 | ||
1114 | Note that if an error occurs in any step the make process will exit. | |
1115 | ||
1116 | @section Building in parallel | |
1117 | ||
38209993 LG |
1118 | If you have a multiprocessor system you can use @samp{make bootstrap |
1119 | MAKE="make -j 2" -j 2} or just @samp{make -j 2 bootstrap} | |
1120 | for GNU Make 3.79 and above instead of just @samp{make bootstrap} | |
161d7b59 | 1121 | when building GCC@. You can use a bigger number instead of two if |
f42974dc DW |
1122 | you like. In most cases, it won't help to use a number bigger than |
1123 | the number of processors in your machine. | |
1124 | ||
e23381df GB |
1125 | @section Building the Ada compiler |
1126 | ||
1127 | In order to build GNAT, the Ada compiler, you need a working GNAT | |
38e23049 JM |
1128 | compiler (GNAT version 3.13 or later, or GCC version 3.1 or later), |
1129 | since the Ada front end is written in Ada (with some | |
e23381df GB |
1130 | GNAT-specific extensions), and GNU make. |
1131 | ||
1132 | However, you do not need a full installation of GNAT, just the GNAT | |
1133 | binary @file{gnat1}, a copy of @file{gnatbind}, and a compiler driver | |
1134 | which can deal with Ada input (by invoking the @file{gnat1} binary). | |
1135 | You can specify this compiler driver by setting the @env{ADAC} | |
1136 | environment variable at the configure step. @command{configure} can | |
1137 | detect the driver automatically if it has got a common name such as | |
1138 | @command{gcc} or @command{gnatgcc}. Of course, you still need a working | |
1139 | C compiler (the compiler driver can be different or not). | |
38e23049 JM |
1140 | @command{configure} does not test whether the GNAT installation works |
1141 | and has a sufficiently recent version; if too old a GNAT version is | |
1142 | installed, the build will fail unless @option{--enable-languages} is | |
1143 | used to disable building the Ada front end. | |
e23381df GB |
1144 | |
1145 | Additional build tools (such as @command{gnatmake}) or a working GNAT | |
1146 | run-time library installation are usually @emph{not} required. However, | |
9026a957 | 1147 | if you want to bootstrap the compiler using a minimal version of GNAT, |
e23381df | 1148 | you have to issue the following commands before invoking @samp{make |
9026a957 | 1149 | bootstrap} (this assumes that you start with an unmodified and consistent |
e23381df GB |
1150 | source distribution): |
1151 | ||
1152 | @example | |
1153 | cd @var{srcdir}/gcc/ada | |
1154 | touch treeprs.ads [es]info.h nmake.ad[bs] | |
1155 | @end example | |
1156 | ||
1157 | At the moment, the GNAT library and several tools for GNAT are not built | |
f282ffb3 | 1158 | by @samp{make bootstrap}. You have to invoke |
e23381df GB |
1159 | @samp{make gnatlib_and_tools} in the @file{@var{objdir}/gcc} |
1160 | subdirectory before proceeding with the next steps. | |
1161 | ||
1162 | For example, you can build a native Ada compiler by issuing the | |
1163 | following commands (assuming @command{make} is GNU make): | |
1164 | ||
1165 | @example | |
1166 | cd @var{objdir} | |
f282ffb3 | 1167 | @var{srcdir}/configure --enable-languages=c,ada |
e23381df GB |
1168 | cd @var{srcdir}/gcc/ada |
1169 | touch treeprs.ads [es]info.h nmake.ad[bs] | |
f282ffb3 | 1170 | cd @var{objdir} |
e23381df GB |
1171 | make bootstrap |
1172 | cd gcc | |
1173 | make gnatlib_and_tools | |
1174 | cd .. | |
1175 | @end example | |
1176 | ||
1177 | Currently, when compiling the Ada front end, you cannot use the parallel | |
1178 | build feature described in the previous section. | |
1179 | ||
f42974dc | 1180 | @html |
b8db17af | 1181 | <hr /> |
f42974dc DW |
1182 | <p> |
1183 | @end html | |
1184 | @ifhtml | |
1185 | @uref{./index.html,,Return to the GCC Installation page} | |
1186 | @end ifhtml | |
1187 | @end ifset | |
1188 | ||
1189 | @c ***Testing***************************************************************** | |
6cfb3f16 | 1190 | @ifnothtml |
f42974dc DW |
1191 | @comment node-name, next, previous, up |
1192 | @node Testing, Final install, Building, Installing GCC | |
6cfb3f16 | 1193 | @end ifnothtml |
f42974dc | 1194 | @ifset testhtml |
f42974dc DW |
1195 | @ifnothtml |
1196 | @chapter Installing GCC: Testing | |
1197 | @end ifnothtml | |
1198 | @cindex Testing | |
1199 | @cindex Installing GCC: Testing | |
1200 | @cindex Testsuite | |
1201 | ||
f97903cc JJ |
1202 | Before you install GCC, we encourage you to run the testsuites and to |
1203 | compare your results with results from a similar configuration that have | |
1204 | been submitted to the | |
1205 | @uref{http://gcc.gnu.org/ml/gcc-testresults/,,gcc-testresults mailing list}. | |
1206 | This step is optional and may require you to download additional software, | |
1207 | but it can give you confidence in your new GCC installation or point out | |
1208 | problems before you install and start using your new GCC. | |
f42974dc | 1209 | |
f9047ed3 | 1210 | First, you must have @uref{download.html,,downloaded the testsuites}. |
f97903cc JJ |
1211 | These are part of the full distribution, but if you downloaded the |
1212 | ``core'' compiler plus any front ends, you must download the testsuites | |
1213 | separately. | |
f42974dc | 1214 | |
f97903cc JJ |
1215 | Second, you must have the testing tools installed. This includes |
1216 | a @uref{http://www.gnu.org/software/dejagnu/,,current version of DejaGnu}; | |
f42974dc | 1217 | dejagnu 1.3 is not sufficient. |
f97903cc | 1218 | It also includes Tcl and Expect; the DejaGnu site has links to these. |
f42974dc DW |
1219 | |
1220 | Now you may need specific preparations: | |
1221 | ||
1222 | @itemize @bullet | |
f42974dc DW |
1223 | |
1224 | @item | |
ecb7d6b3 | 1225 | The following environment variables may need to be set appropriately, as in |
f42974dc | 1226 | the following example (which assumes that DejaGnu has been installed |
6cfb3f16 | 1227 | under @file{/usr/local}): |
f42974dc DW |
1228 | |
1229 | @example | |
1230 | TCL_LIBRARY = /usr/local/share/tcl8.0 | |
1231 | DEJAGNULIBS = /usr/local/share/dejagnu | |
1232 | @end example | |
1233 | ||
1234 | On systems such as Cygwin, these paths are required to be actual | |
1235 | paths, not mounts or links; presumably this is due to some lack of | |
1236 | portability in the DejaGnu code. | |
1237 | ||
ecb7d6b3 JM |
1238 | If the directories where @command{runtest} and @command{expect} were |
1239 | installed are in the @env{PATH}, it should not be necessary to set these | |
1240 | environment variables. | |
1241 | ||
f42974dc DW |
1242 | @end itemize |
1243 | ||
1244 | Finally, you can run the testsuite (which may take a long time): | |
1245 | @example | |
ef88b07d | 1246 | cd @var{objdir}; make -k check |
f42974dc DW |
1247 | @end example |
1248 | ||
1249 | The testing process will try to test as many components in the GCC | |
2147b154 | 1250 | distribution as possible, including the C, C++, Objective-C and Fortran |
ecb7d6b3 | 1251 | compilers as well as the C++ and Java runtime libraries. |
f42974dc | 1252 | |
06809951 GP |
1253 | While running the testsuite, DejaGnu might emit messages resembling |
1254 | @samp{WARNING: Couldn't find the global config file.} or | |
1255 | @samp{WARNING: Couldn't find tool init file}. | |
1256 | These messages are harmless and do not affect the validity of the tests. | |
1257 | ||
f42974dc DW |
1258 | @section How can I run the test suite on selected tests? |
1259 | ||
1260 | As a first possibility to cut down the number of tests that are run it is | |
38209993 | 1261 | possible to use @samp{make check-gcc} or @samp{make check-g++} |
eea81d3e | 1262 | in the @file{gcc} subdirectory of the object directory. To further cut down the |
f42974dc DW |
1263 | tests the following is possible: |
1264 | ||
1265 | @example | |
6cfb3f16 | 1266 | make check-gcc RUNTESTFLAGS="execute.exp @var{other-options}" |
f42974dc DW |
1267 | @end example |
1268 | ||
eea81d3e | 1269 | This will run all @command{gcc} execute tests in the testsuite. |
f42974dc DW |
1270 | |
1271 | @example | |
6cfb3f16 | 1272 | make check-g++ RUNTESTFLAGS="old-deja.exp=9805* @var{other-options}" |
f42974dc DW |
1273 | @end example |
1274 | ||
eea81d3e | 1275 | This will run the @command{g++} ``old-deja'' tests in the testsuite where the filename |
6cfb3f16 | 1276 | matches @samp{9805*}. |
f42974dc | 1277 | |
6cfb3f16 JM |
1278 | The @file{*.exp} files are located in the testsuite directories of the GCC |
1279 | source, the most important ones being @file{compile.exp}, | |
1280 | @file{execute.exp}, @file{dg.exp} and @file{old-deja.exp}. | |
1281 | To get a list of the possible @file{*.exp} files, pipe the | |
38209993 | 1282 | output of @samp{make check} into a file and look at the |
6cfb3f16 | 1283 | @samp{Running @dots{} .exp} lines. |
f42974dc | 1284 | |
f702e700 JJ |
1285 | To run only the tests for a library, run @samp{make check} from the |
1286 | the library's testsuite in a subdirectory of the object directory: | |
1287 | @file{libstdc++-v3/testsuite} or @file{libcgj/testsuite}. | |
1288 | ||
1289 | @section Additional testing for Java Class Libraries | |
1290 | ||
1291 | The @uref{http://sources.redhat.com/mauve/,,Mauve Project} provides | |
1292 | a suite of tests for the Java Class Libraries. This suite can be run | |
1293 | as part of libgcj testing by placing the Mauve tree within the libjava | |
1294 | testsuite at @file{libjava/testsuite/libjava.mauve/mauve}, or by | |
1295 | specifying the location of that tree when invoking @samp{make}, as in | |
1296 | @samp{make MAUVEDIR=~/mauve check}. | |
1297 | ||
582f6e6d TT |
1298 | @uref{http://www-124.ibm.com/developerworks/oss/cvs/jikes/~checkout~/jacks/jacks.html,,Jacks} |
1299 | is a free test suite that tests Java compiler front ends. This suite | |
3b41afd9 | 1300 | can be run as part of libgcj testing by placing the Jacks tree within |
582f6e6d TT |
1301 | the libjava testsuite at @file{libjava/testsuite/libjava.jacks/jacks}. |
1302 | ||
f42974dc DW |
1303 | @section How to interpret test results |
1304 | ||
6cfb3f16 | 1305 | After the testsuite has run you'll find various @file{*.sum} and @file{*.log} |
767094dd | 1306 | files in the testsuite subdirectories. The @file{*.log} files contain a |
f42974dc | 1307 | detailed log of the compiler invocations and the corresponding |
767094dd | 1308 | results, the @file{*.sum} files summarize the results. These summaries list |
f42974dc DW |
1309 | all the tests that have been run with a corresponding status code: |
1310 | ||
1311 | @itemize @bullet | |
1312 | @item | |
1313 | PASS: the test passed as expected | |
1314 | @item | |
1315 | XPASS: the test unexpectedly passed | |
1316 | @item | |
1317 | FAIL: the test unexpectedly failed | |
1318 | @item | |
1319 | XFAIL: the test failed as expected | |
1320 | @item | |
1321 | UNSUPPORTED: the test is not supported on this platform | |
1322 | @item | |
1323 | ERROR: the testsuite detected an error | |
1324 | @item | |
1325 | WARNING: the testsuite detected a possible problem | |
1326 | @end itemize | |
1327 | ||
38209993 LG |
1328 | It is normal for some tests to report unexpected failures. At the |
1329 | current time our testing harness does not allow fine grained control | |
1330 | over whether or not a test is expected to fail. We expect to fix this | |
1331 | problem in future releases. | |
f42974dc DW |
1332 | |
1333 | ||
1334 | @section Submitting test results | |
1335 | ||
1336 | If you want to report the results to the GCC project, use the | |
767094dd | 1337 | @file{contrib/test_summary} shell script. Start it in the @var{objdir} with |
f42974dc DW |
1338 | |
1339 | @example | |
6cfb3f16 JM |
1340 | @var{srcdir}/contrib/test_summary -p your_commentary.txt \ |
1341 | -m gcc-testresults@@gcc.gnu.org |sh | |
f42974dc DW |
1342 | @end example |
1343 | ||
6cfb3f16 | 1344 | This script uses the @command{Mail} program to send the results, so |
767094dd | 1345 | make sure it is in your @env{PATH}. The file @file{your_commentary.txt} is |
f42974dc | 1346 | prepended to the testsuite summary and should contain any special |
767094dd | 1347 | remarks you have on your results or your build environment. Please |
f42974dc | 1348 | do not edit the testsuite result block or the subject line, as these |
f9047ed3 | 1349 | messages are automatically parsed and presented at the |
f42974dc DW |
1350 | @uref{http://gcc.gnu.org/testresults/,,GCC testresults} web |
1351 | page. Here you can also gather information on how specific tests | |
767094dd | 1352 | behave on different platforms and compare them with your results. A |
f42974dc DW |
1353 | few failing testcases are possible even on released versions and you |
1354 | should look here first if you think your results are unreasonable. | |
1355 | ||
aed5964b | 1356 | @html |
b8db17af | 1357 | <hr /> |
aed5964b JM |
1358 | <p> |
1359 | @end html | |
1360 | @ifhtml | |
1361 | @uref{./index.html,,Return to the GCC Installation page} | |
1362 | @end ifhtml | |
f42974dc DW |
1363 | @end ifset |
1364 | ||
1365 | @c ***Final install*********************************************************** | |
6cfb3f16 | 1366 | @ifnothtml |
f42974dc DW |
1367 | @comment node-name, next, previous, up |
1368 | @node Final install, , Testing, Installing GCC | |
6cfb3f16 | 1369 | @end ifnothtml |
f42974dc | 1370 | @ifset finalinstallhtml |
f42974dc DW |
1371 | @ifnothtml |
1372 | @chapter Installing GCC: Final installation | |
1373 | @end ifnothtml | |
1374 | ||
eea81d3e RO |
1375 | Now that GCC has been built (and optionally tested), you can install it with |
1376 | @example | |
1377 | cd @var{objdir}; make install | |
1378 | @end example | |
f42974dc | 1379 | |
06809951 GP |
1380 | We strongly recommend to install into a target directory where there is |
1381 | no previous version of GCC present. | |
1382 | ||
f42974dc | 1383 | That step completes the installation of GCC; user level binaries can |
38209993 LG |
1384 | be found in @file{@var{prefix}/bin} where @var{prefix} is the value you |
1385 | specified with the @option{--prefix} to configure (or @file{/usr/local} | |
ab130aa5 JM |
1386 | by default). (If you specified @option{--bindir}, that directory will |
1387 | be used instead; otherwise, if you specified @option{--exec-prefix}, | |
1388 | @file{@var{exec-prefix}/bin} will be used.) Headers for the C++ and | |
1389 | Java libraries are installed in @file{@var{prefix}/include}; libraries | |
1390 | in @file{@var{libdir}} (normally @file{@var{prefix}/lib}); internal | |
1391 | parts of the compiler in @file{@var{libdir}/gcc-lib}; documentation in | |
1392 | info format in @file{@var{infodir}} (normally @file{@var{prefix}/info}). | |
f42974dc | 1393 | |
2b46bc67 | 1394 | If you built a released version of GCC using @samp{make bootstrap} then please |
f97a5bda JJ |
1395 | quickly review the build status page for your release, available from |
1396 | @uref{http://gcc.gnu.org/buildstat.html}. | |
c5997381 JJ |
1397 | If your system is not listed for the version of GCC that you built, |
1398 | send a note to | |
eea81d3e RO |
1399 | @email{gcc@@gcc.gnu.org} indicating |
1400 | that you successfully built and installed GCC. | |
c5997381 | 1401 | Include the following information: |
f42974dc | 1402 | |
c5997381 JJ |
1403 | @itemize @bullet |
1404 | @item | |
1405 | Output from running @file{@var{srcdir}/config.guess}. Do not send us | |
1406 | that file itself, just the one-line output from running it. | |
1407 | ||
1408 | @item | |
1409 | The output of @samp{gcc -v} for your newly installed gcc. | |
1410 | This tells us which version of GCC you built and the options you passed to | |
1411 | configure. | |
1412 | ||
2b46bc67 JJ |
1413 | @item |
1414 | Whether you enabled all languages or a subset of them. If you used a | |
1415 | full distribution then this information is part of the configure | |
1416 | options in the output of @samp{gcc -v}, but if you downloaded the | |
1417 | ``core'' compiler plus additional front ends then it isn't apparent | |
1418 | which ones you built unless you tell us about it. | |
1419 | ||
c5997381 JJ |
1420 | @item |
1421 | If the build was for GNU/Linux, also include: | |
1422 | @itemize @bullet | |
1423 | @item | |
1424 | The distribution name and version (e.g., Red Hat 7.1 or Debian 2.2.3); | |
1425 | this information should be available from @file{/etc/issue}. | |
1426 | ||
1427 | @item | |
1428 | The version of the Linux kernel, available from @samp{uname --version} | |
1429 | or @samp{uname -a}. | |
1430 | ||
1431 | @item | |
1432 | The version of glibc you used; for RPM-based systems like Red Hat, | |
b9da07da JJ |
1433 | Mandrake, and SuSE type @samp{rpm -q glibc} to get the glibc version, |
1434 | and on systems like Debian and Progeny use @samp{dpkg -l libc6}. | |
c5997381 JJ |
1435 | @end itemize |
1436 | For other systems, you can include similar information if you think it is | |
1437 | relevant. | |
1438 | ||
1439 | @item | |
1440 | Any other information that you think would be useful to people building | |
1441 | GCC on the same configuration. The new entry in the build status list | |
1442 | will include a link to the archived copy of your message. | |
1443 | @end itemize | |
c009f01f JJ |
1444 | |
1445 | We'd also like to know if the | |
1446 | @ifnothtml | |
1447 | @ref{Specific, host/target specific installation notes} | |
1448 | @end ifnothtml | |
1449 | @ifhtml | |
1450 | @uref{specific.html,,host/target specific installation notes} | |
1451 | @end ifhtml | |
1452 | didn't include your host/target information or if that information is | |
1453 | incomplete or out of date. Send a note to | |
1454 | @email{gcc@@gcc.gnu.org} telling us how the information should be changed. | |
f42974dc DW |
1455 | |
1456 | If you find a bug, please report it following our | |
1457 | @uref{../bugs.html,,bug reporting guidelines}. | |
1458 | ||
ab130aa5 | 1459 | If you want to print the GCC manuals, do @samp{cd @var{objdir}; make |
a38f87a9 | 1460 | dvi}. You will need to have @command{texi2dvi} (version at least 4.2) |
ab130aa5 JM |
1461 | and @TeX{} installed. This creates a number of @file{.dvi} files in |
1462 | subdirectories of @file{@var{objdir}}; these may be converted for | |
1463 | printing with programs such as @command{dvips}. You can also | |
1464 | @uref{http://www.gnu.org/order/order.html,,buy printed manuals from the | |
1465 | Free Software Foundation}, though such manuals may not be for the most | |
161d7b59 | 1466 | recent version of GCC@. |
ab130aa5 | 1467 | |
f42974dc | 1468 | @html |
b8db17af | 1469 | <hr /> |
f42974dc DW |
1470 | <p> |
1471 | @end html | |
1472 | @ifhtml | |
1473 | @uref{./index.html,,Return to the GCC Installation page} | |
1474 | @end ifhtml | |
1475 | @end ifset | |
1476 | ||
1477 | @c ***Binaries**************************************************************** | |
6cfb3f16 | 1478 | @ifnothtml |
f42974dc DW |
1479 | @comment node-name, next, previous, up |
1480 | @node Binaries, Specific, Installing GCC, Top | |
6cfb3f16 | 1481 | @end ifnothtml |
f42974dc | 1482 | @ifset binarieshtml |
f42974dc DW |
1483 | @ifnothtml |
1484 | @chapter Installing GCC: Binaries | |
1485 | @end ifnothtml | |
1486 | @cindex Binaries | |
1487 | @cindex Installing GCC: Binaries | |
1488 | ||
161d7b59 | 1489 | We are often asked about pre-compiled versions of GCC@. While we cannot |
f42974dc DW |
1490 | provide these for all platforms, below you'll find links to binaries for |
1491 | various platforms where creating them by yourself is not easy due to various | |
1492 | reasons. | |
1493 | ||
1494 | Please note that we did not create these binaries, nor do we | |
1495 | support them. If you have any problems installing them, please | |
1496 | contact their makers. | |
1497 | ||
1498 | @itemize | |
1499 | @item | |
df002c7d DE |
1500 | AIX: |
1501 | @itemize | |
1502 | @item | |
ff4c5e7b | 1503 | @uref{http://www.bullfreeware.com,,Bull's Freeware and Shareware Archive for AIX}; |
df002c7d DE |
1504 | |
1505 | @item | |
8d5362b7 | 1506 | @uref{http://aixpdslib.seas.ucla.edu,,UCLA Software Library for AIX}. |
df002c7d | 1507 | @end itemize |
f42974dc DW |
1508 | |
1509 | @item | |
8d5362b7 GP |
1510 | DOS---@uref{http://www.delorie.com/djgpp/,,DJGPP}. |
1511 | ||
1512 | @item | |
1513 | Hitachi H8/300[HS]---@uref{http://h8300-hms.sourceforge.net/,,GNU | |
1514 | Development Tools for the Hitachi H8/300[HS] Series}. | |
f42974dc | 1515 | |
f404402c MW |
1516 | @item |
1517 | HP-UX: | |
1518 | @itemize | |
f42974dc DW |
1519 | @item |
1520 | @uref{http://hpux.cae.wisc.edu/,,HP-UX Porting Center}; | |
1521 | ||
f404402c MW |
1522 | @item |
1523 | @uref{ftp://sunsite.informatik.rwth-aachen.de/pub/packages/gcc_hpux/,,Binaries for HP-UX 11.00 at Aachen University of Technology}. | |
1524 | @end itemize | |
1525 | ||
f42974dc | 1526 | @item |
38209993 | 1527 | @uref{http://www.sco.com/skunkware/devtools/index.html#gcc,,SCO |
8d5362b7 | 1528 | OpenServer/Unixware}. |
f42974dc DW |
1529 | |
1530 | @item | |
8d5362b7 GP |
1531 | Sinix/Reliant Unix---@uref{ftp://ftp.siemens.de/sni/mr/pd/gnu/gcc,,Siemens}. |
1532 | ||
1533 | @item | |
1534 | Solaris 2 (SPARC, Intel)---@uref{http://www.sunfreeware.com/,,Sunfreeware}. | |
f42974dc DW |
1535 | |
1536 | @item | |
8d5362b7 | 1537 | SGI---@uref{http://freeware.sgi.com/,,SGI Freeware}. |
f42974dc DW |
1538 | |
1539 | @item | |
1540 | Windows 95, 98, and NT: | |
1541 | @itemize | |
1542 | @item | |
1543 | The @uref{http://sources.redhat.com/cygwin/,,Cygwin} project; | |
1544 | @item | |
cc92b8ab | 1545 | The @uref{http://www.mingw.org/,,MinGW} project. |
f42974dc DW |
1546 | @end itemize |
1547 | ||
1548 | @item | |
61d47787 | 1549 | @uref{ftp://ftp.thewrittenword.com/packages/free/by-name/,,The |
f42974dc DW |
1550 | Written Word} offers binaries for Solaris 2.5.1, 2.6, 2.7/SPARC, 2.7/Intel, |
1551 | IRIX 6.2, 6.5, Digital UNIX 4.0D, HP-UX 10.20, and HP-UX 11.00. | |
1552 | ||
f42974dc DW |
1553 | @end itemize |
1554 | ||
1555 | In addition to those specific offerings, you can get a binary | |
1556 | distribution CD-ROM from the | |
f9047ed3 | 1557 | @uref{http://www.fsf.org/order/order.html,,Free Software Foundation}. |
f42974dc | 1558 | It contains binaries for a number of platforms, and |
767094dd | 1559 | includes not only GCC, but other stuff as well. The current CD does |
f42974dc | 1560 | not contain the latest version of GCC, but it should allow |
767094dd | 1561 | bootstrapping the compiler. An updated version of that disk is in the |
f42974dc DW |
1562 | works. |
1563 | ||
1564 | @html | |
b8db17af | 1565 | <hr /> |
f42974dc DW |
1566 | <p> |
1567 | @end html | |
1568 | @ifhtml | |
1569 | @uref{./index.html,,Return to the GCC Installation page} | |
1570 | @end ifhtml | |
1571 | @end ifset | |
1572 | ||
1573 | @c ***Specific**************************************************************** | |
6cfb3f16 | 1574 | @ifnothtml |
f42974dc | 1575 | @comment node-name, next, previous, up |
73e2155a | 1576 | @node Specific, Old, Binaries, Top |
6cfb3f16 | 1577 | @end ifnothtml |
f42974dc | 1578 | @ifset specifichtml |
f42974dc DW |
1579 | @ifnothtml |
1580 | @chapter Host/target specific installation notes for GCC | |
1581 | @end ifnothtml | |
1582 | @cindex Specific | |
1583 | @cindex Specific installation notes | |
1584 | @cindex Target specific installation | |
1585 | @cindex Host specific installation | |
1586 | @cindex Target specific installation notes | |
1587 | ||
1588 | Please read this document carefully @emph{before} installing the | |
1589 | GNU Compiler Collection on your machine. | |
1590 | ||
ef88b07d | 1591 | @ifhtml |
f42974dc DW |
1592 | @itemize |
1593 | @item | |
333e14b0 | 1594 | @uref{#alpha*-*-*,,alpha*-*-*} |
f42974dc DW |
1595 | @item |
1596 | @uref{#alpha*-dec-osf*,,alpha*-dec-osf*} | |
1597 | @item | |
71b96724 RL |
1598 | @uref{#alphaev5-cray-unicosmk*,,alphaev5-cray-unicosmk*} |
1599 | @item | |
b8df899a JM |
1600 | @uref{#arc-*-elf,,arc-*-elf} |
1601 | @item | |
1602 | @uref{#arm-*-aout,,arm-*-aout} | |
1603 | @item | |
1604 | @uref{#arm-*-elf,,arm-*-elf} | |
1605 | @item | |
476c334e PB |
1606 | @uref{#arm*-*-linux-gnu,,arm*-*-linux-gnu} |
1607 | @item | |
f42974dc DW |
1608 | @uref{#avr,,avr} |
1609 | @item | |
0132e321 MH |
1610 | @uref{#c4x,,c4x} |
1611 | @item | |
f42974dc DW |
1612 | @uref{#dos,,DOS} |
1613 | @item | |
b8df899a JM |
1614 | @uref{#dsp16xx,,dsp16xx} |
1615 | @item | |
021c4bfd RO |
1616 | @uref{#*-*-freebsd*,,*-*-freebsd*} |
1617 | @item | |
f42974dc DW |
1618 | @uref{#h8300-hms,,h8300-hms} |
1619 | @item | |
1620 | @uref{#hppa*-hp-hpux*,,hppa*-hp-hpux*} | |
1621 | @item | |
1622 | @uref{#hppa*-hp-hpux9,,hppa*-hp-hpux9} | |
1623 | @item | |
1624 | @uref{#hppa*-hp-hpux10,,hppa*-hp-hpux10} | |
1625 | @item | |
1626 | @uref{#hppa*-hp-hpux11,,hppa*-hp-hpux11} | |
1627 | @item | |
b8df899a JM |
1628 | @uref{#i370-*-*,,i370-*-*} |
1629 | @item | |
f42974dc DW |
1630 | @uref{#*-*-linux-gnu,,*-*-linux-gnu} |
1631 | @item | |
b8df899a JM |
1632 | @uref{#ix86-*-linux*aout,,i?86-*-linux*aout} |
1633 | @item | |
f42974dc DW |
1634 | @uref{#ix86-*-linux*,,i?86-*-linux*} |
1635 | @item | |
b8df899a JM |
1636 | @uref{#ix86-*-sco,,i?86-*-sco} |
1637 | @item | |
1638 | @uref{#ix86-*-sco3.2v4,,i?86-*-sco3.2v4} | |
1639 | @item | |
f42974dc DW |
1640 | @uref{#ix86-*-sco3.2v5*,,i?86-*-sco3.2v5*} |
1641 | @item | |
f42974dc DW |
1642 | @uref{#ix86-*-udk,,i?86-*-udk} |
1643 | @item | |
b8df899a JM |
1644 | @uref{#ix86-*-esix,,i?86-*-esix} |
1645 | @item | |
b499d9ab JJ |
1646 | @uref{#ia64-*-linux,,ia64-*-linux} |
1647 | @item | |
b8df899a JM |
1648 | @uref{#*-lynx-lynxos,,*-lynx-lynxos} |
1649 | @item | |
f42974dc DW |
1650 | @uref{#*-ibm-aix*,,*-ibm-aix*} |
1651 | @item | |
e3223ea2 DC |
1652 | @uref{#ip2k-*-elf,,ip2k-*-elf} |
1653 | @item | |
b8df899a JM |
1654 | @uref{#m32r-*-elf,,m32r-*-elf} |
1655 | @item | |
1656 | @uref{#m68000-hp-bsd,,m68000-hp-bsd} | |
1657 | @item | |
1658 | @uref{#m6811-elf,,m6811-elf} | |
1659 | @item | |
1660 | @uref{#m6812-elf,,m6812-elf} | |
1661 | @item | |
b8df899a JM |
1662 | @uref{#m68k-att-sysv,,m68k-att-sysv} |
1663 | @item | |
9340544b | 1664 | @uref{#m68k-crds-unos,,m68k-crds-unos} |
b8df899a JM |
1665 | @item |
1666 | @uref{#m68k-hp-hpux,,m68k-hp-hpux} | |
1667 | @item | |
b8df899a JM |
1668 | @uref{#m68k-ncr-*,,m68k-ncr-*} |
1669 | @item | |
1670 | @uref{#m68k-sun,,m68k-sun} | |
1671 | @item | |
f42974dc DW |
1672 | @uref{#m68k-sun-sunos4.1.1,,m68k-sun-sunos4.1.1} |
1673 | @item | |
b8df899a JM |
1674 | @uref{#mips-*-*,,mips-*-*} |
1675 | @item | |
b953cc4b | 1676 | @uref{#mips-sgi-irix5,,mips-sgi-irix5} |
f42974dc | 1677 | @item |
b953cc4b | 1678 | @uref{#mips-sgi-irix6,,mips-sgi-irix6} |
f42974dc | 1679 | @item |
021c4bfd RO |
1680 | @uref{#powerpc*-*-*,,powerpc*-*-*, powerpc-*-sysv4} |
1681 | @item | |
4f2b1139 SS |
1682 | @uref{#powerpc-*-darwin*,,powerpc-*-darwin*} |
1683 | @item | |
b8df899a JM |
1684 | @uref{#powerpc-*-elf,,powerpc-*-elf, powerpc-*-sysv4} |
1685 | @item | |
f42974dc DW |
1686 | @uref{#powerpc-*-linux-gnu*,,powerpc-*-linux-gnu*} |
1687 | @item | |
edf1b3f3 AC |
1688 | @uref{#powerpc-*-netbsd*,,powerpc-*-netbsd*} |
1689 | @item | |
b8df899a JM |
1690 | @uref{#powerpc-*-eabiaix,,powerpc-*-eabiaix} |
1691 | @item | |
1692 | @uref{#powerpc-*-eabisim,,powerpc-*-eabisim} | |
1693 | @item | |
1694 | @uref{#powerpc-*-eabi,,powerpc-*-eabi} | |
1695 | @item | |
1696 | @uref{#powerpcle-*-elf,,powerpcle-*-elf, powerpcle-*-sysv4} | |
1697 | @item | |
1698 | @uref{#powerpcle-*-eabisim,,powerpcle-*-eabisim} | |
1699 | @item | |
1700 | @uref{#powerpcle-*-eabi,,powerpcle-*-eabi} | |
1701 | @item | |
1702 | @uref{#powerpcle-*-winnt,,powerpcle-*-winnt, powerpcle-*-pe} | |
1703 | @item | |
225cee28 | 1704 | @uref{#s390-*-linux*,,s390-*-linux*} |
91abf72d | 1705 | @item |
225cee28 | 1706 | @uref{#s390x-*-linux*,,s390x-*-linux*} |
91abf72d | 1707 | @item |
250d5688 | 1708 | @uref{#*-*-solaris2*,,*-*-solaris2*} |
f42974dc | 1709 | @item |
250d5688 | 1710 | @uref{#sparc-sun-solaris2*,,sparc-sun-solaris2*} |
f42974dc DW |
1711 | @item |
1712 | @uref{#sparc-sun-solaris2.7,,sparc-sun-solaris2.7} | |
1713 | @item | |
250d5688 | 1714 | @uref{#sparc-sun-sunos4*,,sparc-sun-sunos4*} |
f42974dc DW |
1715 | @item |
1716 | @uref{#sparc-unknown-linux-gnulibc1,,sparc-unknown-linux-gnulibc1} | |
1717 | @item | |
c6fa9728 JS |
1718 | @uref{#sparc-*-linux*,,sparc-*-linux*} |
1719 | @item | |
f42974dc DW |
1720 | @uref{#sparc64-*-*,,sparc64-*-*} |
1721 | @item | |
e403b4bc CR |
1722 | @uref{#sparcv9-*-solaris2*,,sparcv9-*-solaris2*} |
1723 | @item | |
b8df899a JM |
1724 | @uref{#*-*-sysv*,,*-*-sysv*} |
1725 | @item | |
1726 | @uref{#vax-dec-ultrix,,vax-dec-ultrix} | |
1727 | @item | |
4977bab6 ZW |
1728 | @uref{#*-*-vxworks*,,*-*-vxworks*} |
1729 | @item | |
fd29f6ea BW |
1730 | @uref{#xtensa-*-elf,,xtensa-*-elf} |
1731 | @item | |
1732 | @uref{#xtensa-*-linux*,,xtensa-*-linux*} | |
1733 | @item | |
f42974dc DW |
1734 | @uref{#windows,,Microsoft Windows} |
1735 | @item | |
1736 | @uref{#os2,,OS/2} | |
1737 | @item | |
1738 | @uref{#older,,Older systems} | |
1739 | @end itemize | |
1740 | ||
1741 | @itemize | |
1742 | @item | |
250d5688 | 1743 | @uref{#elf_targets,,all ELF targets} (SVR4, Solaris 2, etc.) |
f42974dc | 1744 | @end itemize |
ef88b07d | 1745 | @end ifhtml |
f42974dc DW |
1746 | |
1747 | ||
1748 | @html | |
1749 | <!-- -------- host/target specific issues start here ---------------- --> | |
b8db17af | 1750 | <hr /> |
f42974dc | 1751 | @end html |
333e14b0 LR |
1752 | @heading @anchor{alpha*-*-*}alpha*-*-* |
1753 | ||
1754 | This section contains general configuration information for all | |
1755 | alpha-based platforms using ELF (in particular, ignore this section for | |
161d7b59 | 1756 | DEC OSF/1, Digital UNIX and Tru64 UNIX)@. In addition to reading this |
f2541106 | 1757 | section, please read all other sections that match your target. |
333e14b0 | 1758 | |
021c4bfd RO |
1759 | We require binutils 2.11.2 or newer. |
1760 | Previous binutils releases had a number of problems with DWARF 2 | |
333e14b0 LR |
1761 | debugging information, not the least of which is incorrect linking of |
1762 | shared libraries. | |
1763 | ||
b8df899a | 1764 | @html |
b8db17af | 1765 | <hr /> |
b8df899a | 1766 | @end html |
f2541106 | 1767 | @heading @anchor{alpha*-dec-osf*}alpha*-dec-osf* |
b8df899a | 1768 | Systems using processors that implement the DEC Alpha architecture and |
f2541106 RO |
1769 | are running the DEC/Compaq Unix (DEC OSF/1, Digital UNIX, or Compaq |
1770 | Tru64 UNIX) operating system, for example the DEC Alpha AXP systems. | |
1771 | ||
c7bdf0a6 ZW |
1772 | As of GCC 3.2, versions before @code{alpha*-dec-osf4} are no longer |
1773 | supported. (These are the versions which identify themselves as DEC | |
1774 | OSF/1.) | |
9340544b | 1775 | |
6e92b3a1 RB |
1776 | In Digital Unix V4.0, virtual memory exhausted bootstrap failures |
1777 | may be fixed by configuring with @option{--with-gc=simple}, | |
1778 | reconfiguring Kernel Virtual Memory and Swap parameters | |
1779 | per the @command{/usr/sbin/sys_check} Tuning Suggestions, | |
1780 | or applying the patch in | |
1781 | @uref{http://gcc.gnu.org/ml/gcc/2002-08/msg00822.html}. | |
1782 | ||
f2541106 RO |
1783 | In Tru64 UNIX V5.1, Compaq introduced a new assembler that does not |
1784 | currently (2001-06-13) work with @command{mips-tfile}. As a workaround, | |
1785 | we need to use the old assembler, invoked via the barely documented | |
1786 | @option{-oldas} option. To bootstrap GCC, you either need to use the | |
1787 | Compaq C Compiler: | |
1788 | ||
1789 | @example | |
eea81d3e | 1790 | % CC=cc @var{srcdir}/configure [@var{options}] [@var{target}] |
f2541106 RO |
1791 | @end example |
1792 | ||
1793 | or you can use a copy of GCC 2.95.3 or higher built on Tru64 UNIX V4.0: | |
1794 | ||
1795 | @example | |
eea81d3e | 1796 | % CC=gcc -Wa,-oldas @var{srcdir}/configure [@var{options}] [@var{target}] |
f2541106 | 1797 | @end example |
b8df899a | 1798 | |
b953cc4b RO |
1799 | As of GNU binutils 2.11.2, neither GNU @command{as} nor GNU @command{ld} |
1800 | are supported on Tru64 UNIX, so you must not configure GCC with | |
1801 | @option{--with-gnu-as} or @option{--with-gnu-ld}. | |
1802 | ||
1803 | The @option{--enable-threads} options isn't supported yet. A patch is | |
18b467f1 | 1804 | in preparation for a future release. |
b953cc4b | 1805 | |
f0523f02 | 1806 | GCC writes a @samp{.verstamp} directive to the assembler output file |
b8df899a JM |
1807 | unless it is built as a cross-compiler. It gets the version to use from |
1808 | the system header file @file{/usr/include/stamp.h}. If you install a | |
1809 | new version of DEC Unix, you should rebuild GCC to pick up the new version | |
1810 | stamp. | |
1811 | ||
1812 | Note that since the Alpha is a 64-bit architecture, cross-compilers from | |
1813 | 32-bit machines will not generate code as efficient as that generated | |
1814 | when the compiler is running on a 64-bit machine because many | |
1815 | optimizations that depend on being able to represent a word on the | |
1816 | target in an integral value on the host cannot be performed. Building | |
1817 | cross-compilers on the Alpha for 32-bit machines has only been tested in | |
1818 | a few cases and may not work properly. | |
1819 | ||
1820 | @code{make compare} may fail on old versions of DEC Unix unless you add | |
6cfb3f16 | 1821 | @option{-save-temps} to @code{CFLAGS}. On these systems, the name of the |
b8df899a JM |
1822 | assembler input file is stored in the object file, and that makes |
1823 | comparison fail if it differs between the @code{stage1} and | |
6cfb3f16 | 1824 | @code{stage2} compilations. The option @option{-save-temps} forces a |
b8df899a | 1825 | fixed name to be used for the assembler input file, instead of a |
6cfb3f16 | 1826 | randomly chosen name in @file{/tmp}. Do not add @option{-save-temps} |
b8df899a | 1827 | unless the comparisons fail without that option. If you add |
6cfb3f16 | 1828 | @option{-save-temps}, you will have to manually delete the @samp{.i} and |
b8df899a JM |
1829 | @samp{.s} files after each series of compilations. |
1830 | ||
f0523f02 | 1831 | GCC now supports both the native (ECOFF) debugging format used by DBX |
161d7b59 | 1832 | and GDB and an encapsulated STABS format for use only with GDB@. See the |
6cfb3f16 | 1833 | discussion of the @option{--with-stabs} option of @file{configure} above |
b8df899a JM |
1834 | for more information on these formats and how to select them. |
1835 | ||
1836 | There is a bug in DEC's assembler that produces incorrect line numbers | |
1837 | for ECOFF format when the @samp{.align} directive is used. To work | |
f0523f02 | 1838 | around this problem, GCC will not emit such alignment directives |
b8df899a JM |
1839 | while writing ECOFF format debugging information even if optimization is |
1840 | being performed. Unfortunately, this has the very undesirable | |
6cfb3f16 JM |
1841 | side-effect that code addresses when @option{-O} is specified are |
1842 | different depending on whether or not @option{-g} is also specified. | |
b8df899a | 1843 | |
6cfb3f16 | 1844 | To avoid this behavior, specify @option{-gstabs+} and use GDB instead of |
161d7b59 | 1845 | DBX@. DEC is now aware of this problem with the assembler and hopes to |
b8df899a JM |
1846 | provide a fix shortly. |
1847 | ||
71b96724 | 1848 | @html |
b8db17af | 1849 | <hr /> |
71b96724 RL |
1850 | @end html |
1851 | @heading @anchor{alphaev5-cray-unicosmk*}alphaev5-cray-unicosmk* | |
1852 | Cray T3E systems running Unicos/Mk. | |
1853 | ||
1854 | This port is incomplete and has many known bugs. We hope to improve the | |
1855 | support for this target soon. Currently, only the C front end is supported, | |
1856 | and it is not possible to build parallel applications. Cray modules are not | |
1857 | supported; in particular, Craylibs are assumed to be in | |
1858 | @file{/opt/ctl/craylibs/craylibs}. | |
1859 | ||
1860 | You absolutely @strong{must} use GNU make on this platform. Also, you | |
1861 | need to tell GCC where to find the assembler and the linker. The | |
1862 | simplest way to do so is by providing @option{--with-as} and | |
1863 | @option{--with-ld} to @file{configure}, e.g.@: | |
1864 | ||
8c085f6f JJ |
1865 | @example |
1866 | configure --with-as=/opt/ctl/bin/cam --with-ld=/opt/ctl/bin/cld \ | |
1867 | --enable-languages=c | |
1868 | @end example | |
71b96724 RL |
1869 | |
1870 | The comparison test during @samp{make bootstrap} fails on Unicos/Mk | |
1871 | because the assembler inserts timestamps into object files. You should | |
1872 | be able to work around this by doing @samp{make all} after getting this | |
1873 | failure. | |
1874 | ||
b8df899a | 1875 | @html |
b8db17af | 1876 | <hr /> |
b8df899a JM |
1877 | @end html |
1878 | @heading @anchor{arc-*-elf}arc-*-elf | |
1879 | Argonaut ARC processor. | |
1880 | This configuration is intended for embedded systems. | |
1881 | ||
1882 | @html | |
b8db17af | 1883 | <hr /> |
b8df899a JM |
1884 | @end html |
1885 | @heading @anchor{arm-*-aout}arm-*-aout | |
1886 | Advanced RISC Machines ARM-family processors. These are often used in | |
1887 | embedded applications. There are no standard Unix configurations. | |
1888 | This configuration corresponds to the basic instruction sequences and will | |
1889 | produce @file{a.out} format object modules. | |
1890 | ||
1891 | You may need to make a variant of the file @file{arm.h} for your particular | |
1892 | configuration. | |
1893 | ||
1894 | @html | |
b8db17af | 1895 | <hr /> |
b8df899a JM |
1896 | @end html |
1897 | @heading @anchor{arm-*-elf}arm-*-elf | |
1898 | This configuration is intended for embedded systems. | |
1899 | ||
476c334e | 1900 | @html |
b8db17af | 1901 | <hr /> |
476c334e | 1902 | @end html |
ef88b07d | 1903 | @heading @anchor{arm*-*-linux-gnu}arm*-*-linux-gnu |
476c334e PB |
1904 | |
1905 | We require GNU binutils 2.10 or newer. | |
1906 | ||
f42974dc | 1907 | @html |
b8db17af | 1908 | <hr /> |
f42974dc | 1909 | @end html |
ef88b07d | 1910 | @heading @anchor{avr}avr |
f42974dc | 1911 | |
b8df899a | 1912 | ATMEL AVR-family micro controllers. These are used in embedded |
ca52d046 GP |
1913 | applications. There are no standard Unix configurations. |
1914 | @ifnothtml | |
1915 | @xref{AVR Options,, AVR Options, gcc, Using and Porting the GNU Compiler | |
1916 | Collection (GCC)}, | |
1917 | @end ifnothtml | |
98999d8b | 1918 | @ifhtml |
ca52d046 | 1919 | See ``AVR Options'' in the main manual |
98999d8b | 1920 | @end ifhtml |
ca52d046 | 1921 | for the list of supported MCU types. |
b8df899a | 1922 | |
161d7b59 | 1923 | Use @samp{configure --target=avr --enable-languages="c"} to configure GCC@. |
f42974dc DW |
1924 | |
1925 | Further installation notes and other useful information about AVR tools | |
1926 | can also be obtained from: | |
1927 | ||
1928 | @itemize @bullet | |
1929 | @item | |
de7999ba MM |
1930 | @uref{http://www.openavr.org,,http://www.openavr.org} |
1931 | @item | |
f42974dc DW |
1932 | @uref{http://home.overta.ru/users/denisc,,http://home.overta.ru/users/denisc} |
1933 | @item | |
63708ffc | 1934 | @uref{http://www.amelek.gda.pl/avr,,http://www.amelek.gda.pl/avr} |
f42974dc DW |
1935 | @end itemize |
1936 | ||
de7999ba | 1937 | We @emph{strongly} recommend using binutils 2.13 or newer. |
f42974dc DW |
1938 | |
1939 | The following error: | |
1940 | @example | |
1941 | Error: register required | |
1942 | @end example | |
1943 | ||
1944 | indicates that you should upgrade to a newer version of the binutils. | |
1945 | ||
0132e321 | 1946 | @html |
b8db17af | 1947 | <hr /> |
0132e321 MH |
1948 | @end html |
1949 | @heading @anchor{c4x}c4x | |
1950 | ||
1951 | Texas Instruments TMS320C3x and TMS320C4x Floating Point Digital Signal | |
1952 | Processors. These are used in embedded applications. There are no | |
d8393f64 GP |
1953 | standard Unix configurations. |
1954 | @ifnothtml | |
1955 | @xref{TMS320C3x/C4x Options,, TMS320C3x/C4x Options, gcc, Using and | |
1956 | Porting the GNU Compiler Collection (GCC)}, | |
1957 | @end ifnothtml | |
98999d8b | 1958 | @ifhtml |
d8393f64 | 1959 | See ``TMS320C3x/C4x Options'' in the main manual |
98999d8b | 1960 | @end ifhtml |
d8393f64 | 1961 | for the list of supported MCU types. |
0132e321 MH |
1962 | |
1963 | GCC can be configured as a cross compiler for both the C3x and C4x | |
1964 | architectures on the same system. Use @samp{configure --target=c4x | |
1965 | --enable-languages="c,c++"} to configure. | |
1966 | ||
1967 | ||
1968 | Further installation notes and other useful information about C4x tools | |
1969 | can also be obtained from: | |
1970 | ||
1971 | @itemize @bullet | |
1972 | @item | |
d8393f64 | 1973 | @uref{http://www.elec.canterbury.ac.nz/c4x/,,http://www.elec.canterbury.ac.nz/c4x/} |
0132e321 MH |
1974 | @end itemize |
1975 | ||
0b85d816 | 1976 | @html |
b8db17af | 1977 | <hr /> |
0b85d816 HPN |
1978 | @end html |
1979 | @heading @anchor{cris}CRIS | |
1980 | ||
1981 | CRIS is the CPU architecture in Axis Communications ETRAX system-on-a-chip | |
1982 | series. These are used in embedded applications. | |
1983 | ||
1984 | @ifnothtml | |
1985 | @xref{CRIS Options,, CRIS Options, gcc, Using and Porting the GNU Compiler | |
1986 | Collection (GCC)}, | |
1987 | @end ifnothtml | |
1988 | @ifhtml | |
1989 | See ``CRIS Options'' in the main manual | |
1990 | @end ifhtml | |
1991 | for a list of CRIS-specific options. | |
1992 | ||
1993 | There are a few different CRIS targets: | |
1994 | @table @code | |
1995 | @item cris-axis-aout | |
1996 | Old target. Includes a multilib for the @samp{elinux} a.out-based | |
1997 | target. No multilibs for newer architecture variants. | |
1998 | @item cris-axis-elf | |
1999 | Mainly for monolithic embedded systems. Includes a multilib for the | |
2000 | @samp{v10} core used in @samp{ETRAX 100 LX}. | |
2001 | @item cris-axis-linux-gnu | |
2002 | A GNU/Linux port for the CRIS architecture, currently targeting | |
2003 | @samp{ETRAX 100 LX} by default. | |
2004 | @end table | |
2005 | ||
2006 | For @code{cris-axis-aout} and @code{cris-axis-elf} you need binutils 2.11 | |
2007 | or newer. For @code{cris-axis-linux-gnu} you need binutils 2.12 or newer. | |
2008 | ||
2009 | Pre-packaged tools can be obtained from | |
2010 | @uref{ftp://ftp.axis.com/pub/axis/tools/cris/compiler-kit/}. More | |
2011 | information about this platform is available at | |
2012 | @uref{http://developer.axis.com/}. | |
2013 | ||
f42974dc | 2014 | @html |
b8db17af | 2015 | <hr /> |
f42974dc | 2016 | @end html |
ef88b07d | 2017 | @heading @anchor{dos}DOS |
f42974dc DW |
2018 | |
2019 | Please have a look at our @uref{binaries.html,,binaries page}. | |
2020 | ||
f0523f02 | 2021 | You cannot install GCC by itself on MSDOS; it will not compile under |
f85b8d1a JM |
2022 | any MSDOS compiler except itself. You need to get the complete |
2023 | compilation package DJGPP, which includes binaries as well as sources, | |
2024 | and includes all the necessary compilation tools and libraries. | |
2025 | ||
b8df899a | 2026 | @html |
b8db17af | 2027 | <hr /> |
b8df899a JM |
2028 | @end html |
2029 | @heading @anchor{dsp16xx}dsp16xx | |
2030 | A port to the AT&T DSP1610 family of processors. | |
2031 | ||
021c4bfd | 2032 | @html |
b8db17af | 2033 | <hr /> |
021c4bfd RO |
2034 | @end html |
2035 | @heading @anchor{*-*-freebsd*}*-*-freebsd* | |
2036 | ||
2037 | The version of binutils installed in @file{/usr/bin} is known to work unless | |
2038 | otherwise specified in any per-architecture notes. However, binutils | |
6b976d99 | 2039 | 2.12.1 or greater is known to improve overall testsuite results. |
021c4bfd | 2040 | |
c7bdf0a6 ZW |
2041 | FreeBSD 1 is no longer supported in GCC 3.2. |
2042 | ||
2043 | For FreeBSD 2 or any mutant a.out versions of FreeBSD 3: All | |
021c4bfd RO |
2044 | configuration support and files as shipped with GCC 2.95 are still in |
2045 | place. FreeBSD 2.2.7 has been known to bootstrap completely; however, | |
2046 | it is unknown which version of binutils was used (it is assumed that it | |
2047 | was the system copy in @file{/usr/bin}) and C++ EH failures were noted. | |
2048 | ||
2049 | For FreeBSD using the ELF file format: DWARF 2 debugging is now the | |
2050 | default for all CPU architectures. It had been the default on | |
2051 | FreeBSD/alpha since its inception. You may use @option{-gstabs} instead | |
2052 | of @option{-g}, if you really want the old debugging format. There are | |
2053 | no known issues with mixing object files and libraries with different | |
2054 | debugging formats. Otherwise, this release of GCC should now match more | |
2055 | of the configuration used in the stock FreeBSD configuration of GCC. In | |
2056 | particular, @option{--enable-threads} is now configured by default. | |
2057 | However, as a general user, do not attempt to replace the system | |
2058 | compiler with this release. Known to bootstrap and check with good | |
46fc709d LR |
2059 | results on FreeBSD 3.0, 3.4, 4.0, 4.2, 4.3, 4.4, 4.5-STABLE and 5-CURRENT@. |
2060 | ||
2061 | In principle, @option{--enable-threads} is now compatible with | |
2062 | @option{--enable-libgcj} on FreeBSD@. However, it has only been built | |
8c085f6f JJ |
2063 | and tested on @samp{i386-*-freebsd4.5} and @samp{alpha-*-freebsd5.0}. |
2064 | The static | |
6b976d99 LR |
2065 | library may be incorrectly built (symbols are missing at link time). |
2066 | There is a rare timing-based startup hang (probably involves an | |
c0478a66 | 2067 | assumption about the thread library). Multi-threaded boehm-gc (required for |
46fc709d LR |
2068 | libjava) exposes severe threaded signal-handling bugs on FreeBSD before |
2069 | 4.5-RELEASE. The alpha port may not fully bootstrap without some manual | |
8c085f6f | 2070 | intervention: @command{gcjh} will crash with a floating-point exception while |
46fc709d | 2071 | generating @file{java/lang/Double.h} (just copy the version built on |
8c085f6f JJ |
2072 | @samp{i386-*-freebsd*} and rerun the top-level @command{gmake} with no |
2073 | arguments and it | |
46fc709d LR |
2074 | should properly complete the bootstrap). Other CPU architectures |
2075 | supported by FreeBSD will require additional configuration tuning in, at | |
2076 | the very least, both boehm-gc and libffi. | |
021c4bfd | 2077 | |
bc3a44db LR |
2078 | Shared @file{libgcc_s.so} is now built and installed by default. |
2079 | ||
f42974dc | 2080 | @html |
b8db17af | 2081 | <hr /> |
f42974dc | 2082 | @end html |
ef88b07d | 2083 | @heading @anchor{h8300-hms}h8300-hms |
b8df899a | 2084 | Hitachi H8/300 series of processors. |
f42974dc DW |
2085 | |
2086 | Please have a look at our @uref{binaries.html,,binaries page}. | |
2087 | ||
b8df899a JM |
2088 | The calling convention and structure layout has changed in release 2.6. |
2089 | All code must be recompiled. The calling convention now passes the | |
2090 | first three arguments in function calls in registers. Structures are no | |
2091 | longer a multiple of 2 bytes. | |
2092 | ||
f42974dc | 2093 | @html |
b8db17af | 2094 | <hr /> |
f42974dc | 2095 | @end html |
ef88b07d | 2096 | @heading @anchor{hppa*-hp-hpux*}hppa*-hp-hpux* |
f42974dc | 2097 | |
021c4bfd | 2098 | We @emph{highly} recommend using gas/binutils 2.8 or newer on all hppa |
f9047ed3 | 2099 | platforms; you may encounter a variety of problems when using the HP |
581d9404 | 2100 | assembler. |
f42974dc DW |
2101 | |
2102 | Specifically, @option{-g} does not work on HP-UX (since that system | |
2103 | uses a peculiar debugging format which GCC does not know about), unless you | |
38209993 LG |
2104 | use GAS and GDB and configure GCC with the |
2105 | @uref{./configure.html#with-gnu-as,,@option{--with-gnu-as}} and | |
6cfb3f16 | 2106 | @option{--with-as=@dots{}} options. |
f42974dc | 2107 | |
08b3d104 JDA |
2108 | If you wish to use the pa-risc 2.0 architecture support with a 32-bit |
2109 | runtime, you must use either the HP assembler, gas/binutils 2.11 or newer, | |
2110 | or a recent | |
f42974dc DW |
2111 | @uref{ftp://sources.redhat.com/pub/binutils/snapshots,,snapshot of gas}. |
2112 | ||
d5355cb2 JDA |
2113 | There are two default scheduling models for instructions. These are |
2114 | PROCESSOR_7100LC and PROCESSOR_8000. They are selected from the pa-risc | |
2115 | architecture specified for the target machine when configuring. | |
2116 | PROCESSOR_8000 is the default. PROCESSOR_7100LC is selected when | |
2117 | the target is a @samp{hppa1*} machine. | |
806bf413 JDA |
2118 | |
2119 | The PROCESSOR_8000 model is not well suited to older processors. Thus, | |
2120 | it is important to completely specify the machine architecture when | |
2121 | configuring if you want a model other than PROCESSOR_8000. The macro | |
2122 | TARGET_SCHED_DEFAULT can be defined in BOOT_CFLAGS if a different | |
2123 | default scheduling model is desired. | |
2124 | ||
021c4bfd | 2125 | More specific information to @samp{hppa*-hp-hpux*} targets follows. |
f42974dc DW |
2126 | |
2127 | @html | |
b8db17af | 2128 | <hr /> |
f42974dc | 2129 | @end html |
ef88b07d | 2130 | @heading @anchor{hppa*-hp-hpux9}hppa*-hp-hpux9 |
f42974dc DW |
2131 | |
2132 | The HP assembler has major problems on this platform. We've tried to work | |
2133 | around the worst of the problems. However, those workarounds may be causing | |
2134 | linker crashes in some circumstances; the workarounds also probably prevent | |
2135 | shared libraries from working. Use the GNU assembler to avoid these problems. | |
2136 | ||
2137 | ||
2138 | The configuration scripts for GCC will also trigger a bug in the hpux9 | |
38209993 LG |
2139 | shell. To avoid this problem set @env{CONFIG_SHELL} to @file{/bin/ksh} |
2140 | and @env{SHELL} to @file{/bin/ksh} in your environment. | |
f42974dc DW |
2141 | |
2142 | ||
2143 | @html | |
b8db17af | 2144 | <hr /> |
f42974dc | 2145 | @end html |
ef88b07d | 2146 | @heading @anchor{hppa*-hp-hpux10}hppa*-hp-hpux10 |
f42974dc | 2147 | |
f9047ed3 | 2148 | For hpux10.20, we @emph{highly} recommend you pick up the latest sed patch |
161d7b59 | 2149 | @code{PHCO_19798} from HP@. HP has two sites which provide patches free of |
f42974dc DW |
2150 | charge: |
2151 | ||
2152 | @itemize @bullet | |
2153 | @item | |
2154 | @html | |
2155 | <a href="http://us-support.external.hp.com">US, Canada, Asia-Pacific, and | |
2156 | Latin-America</a> | |
2157 | @end html | |
2158 | @ifnothtml | |
2159 | @uref{http://us-support.external.hp.com,,}US, Canada, Asia-Pacific, and | |
2160 | Latin-America | |
2161 | @end ifnothtml | |
2162 | @item | |
2163 | @uref{http://europe-support.external.hp.com,,Europe} | |
2164 | @end itemize | |
2165 | ||
2166 | The HP assembler on these systems is much better than the hpux9 assembler, | |
2167 | but still has some problems. Most notably the assembler inserts timestamps | |
2168 | into each object file it creates, causing the 3-stage comparison test to fail | |
f9047ed3 JM |
2169 | during a @samp{make bootstrap}. You should be able to continue by |
2170 | saying @samp{make all} after getting the failure from @samp{make | |
38209993 | 2171 | bootstrap}. |
f42974dc DW |
2172 | |
2173 | ||
2174 | @html | |
b8db17af | 2175 | <hr /> |
f42974dc | 2176 | @end html |
ef88b07d | 2177 | @heading @anchor{hppa*-hp-hpux11}hppa*-hp-hpux11 |
f42974dc | 2178 | |
08b3d104 JDA |
2179 | GCC 3.0 and up support HP-UX 11. On 64-bit capable systems, there |
2180 | are two distinct ports. The @samp{hppa2.0w-hp-hpux11*} port generates | |
2181 | code for the 32-bit pa-risc runtime architecture. It uses the HP | |
2182 | linker and is currently the default selected by config.guess. The | |
2183 | optional @samp{hppa64-hp-hpux11*} port generates 64-bit code for the | |
2184 | pa-risc 2.0 architecture. It must be explicitly selected using the | |
2185 | @samp{--host=hppa64-hp-hpux11*} configure option. Different prefixes | |
2186 | must be used if both ports are to be installed on the same system. | |
2187 | ||
2188 | You must use GNU binutils 2.11 or above with the 32-bit port. Thread | |
2189 | support is not currently implemented, so @option{--enable-threads} does | |
8c085f6f JJ |
2190 | not work. See: |
2191 | ||
2192 | @itemize | |
2193 | @item @uref{http://gcc.gnu.org/ml/gcc-prs/2002-01/msg00551.html} | |
2194 | @item @uref{http://gcc.gnu.org/ml/gcc-bugs/2002-01/msg00663.html} | |
2195 | @end itemize | |
2196 | ||
6f2a28d7 | 2197 | GCC 2.95.x is not supported under HP-UX 11 and cannot be used to |
975c6e4e RO |
2198 | compile GCC 3.0 and up. Refer to @uref{binaries.html,,binaries} for |
2199 | information about obtaining precompiled GCC binaries for HP-UX. | |
f42974dc | 2200 | |
08b3d104 | 2201 | GNU binutils 2.13 or later is recommended with the 64-bit port. |
581d9404 JDA |
2202 | The HP assembler has many limitations and is not recommended. For |
2203 | example, it does not support weak symbols or alias definitions. | |
2204 | As a result, explicit template instantiations are required when | |
2205 | using C++. Either the HP or GNU linker can be used but it may be | |
2206 | necessary to use the GNU linker when dwarf2 exception support is | |
2207 | implemented. | |
2208 | ||
2209 | There are several possible approaches to building the distribution. | |
2210 | Binutils can be built first using the HP tools. Then, the GCC | |
2211 | distribution can be built. The second approach is to build GCC | |
2212 | first using the HP tools, then build binutils, then rebuild GCC. | |
2213 | There have been problems with various binary distributions, so | |
2214 | it is best not to start from a binary distribution. | |
2215 | ||
2216 | When starting with a HP compiler, it is preferable to use the ANSI | |
2217 | compiler as the bundled compiler only supports traditional C. | |
2218 | Bootstrapping with the bundled compiler is tested infrequently and | |
2219 | problems often arise because of the subtle differences in semantics | |
2220 | between traditional and ISO C. | |
2221 | ||
2222 | This port still is undergoing significant development. | |
08b3d104 | 2223 | |
b8df899a | 2224 | @html |
b8db17af | 2225 | <hr /> |
b8df899a JM |
2226 | @end html |
2227 | @heading @anchor{i370-*-*}i370-*-* | |
2228 | This port is very preliminary and has many known bugs. We hope to | |
2229 | have a higher-quality port for this machine soon. | |
2230 | ||
f42974dc | 2231 | @html |
b8db17af | 2232 | <hr /> |
f42974dc | 2233 | @end html |
ef88b07d | 2234 | @heading @anchor{*-*-linux-gnu}*-*-linux-gnu |
f42974dc DW |
2235 | |
2236 | If you use glibc 2.2 (or 2.1.9x), GCC 2.95.2 won't install | |
021c4bfd | 2237 | out-of-the-box. You'll get compile errors while building @samp{libstdc++}. |
f42974dc DW |
2238 | The patch @uref{glibc-2.2.patch,,glibc-2.2.patch}, that is to be |
2239 | applied in the GCC source tree, fixes the compatibility problems. | |
2240 | ||
e15ed790 | 2241 | @html |
e15ed790 AJ |
2242 | @end html |
2243 | ||
2244 | @html | |
2245 | <p> | |
2246 | @end html | |
2247 | ||
2248 | Currently Glibc 2.2.3 (and older releases) and GCC 3.0 are out of sync | |
161d7b59 | 2249 | since the latest exception handling changes for GCC@. Compiling glibc |
e15ed790 AJ |
2250 | with GCC 3.0 will give a binary incompatible glibc and therefore cause |
2251 | lots of problems and might make your system completly unusable. This | |
161d7b59 | 2252 | will definitly need fixes in glibc but might also need fixes in GCC@. We |
e15ed790 AJ |
2253 | strongly advise to wait for glibc 2.2.4 and to read the release notes of |
2254 | glibc 2.2.4 whether patches for GCC 3.0 are needed. You can use glibc | |
2255 | 2.2.3 with GCC 3.0, just do not try to recompile it. | |
2256 | ||
b8df899a | 2257 | @html |
b8db17af | 2258 | <hr /> |
b8df899a JM |
2259 | @end html |
2260 | @heading @anchor{ix86-*-linux*aout}i?86-*-linux*aout | |
2261 | Use this configuration to generate @file{a.out} binaries on Linux-based | |
767094dd | 2262 | GNU systems. This configuration is being superseded. You must use |
b8df899a JM |
2263 | gas/binutils version 2.5.2 or later. |
2264 | ||
f42974dc | 2265 | @html |
b8db17af | 2266 | <hr /> |
f42974dc | 2267 | @end html |
ef88b07d | 2268 | @heading @anchor{ix86-*-linux*}i?86-*-linux* |
f42974dc | 2269 | |
021c4bfd | 2270 | You will need binutils 2.9.1.0.15 or newer for exception handling to work. |
f42974dc DW |
2271 | |
2272 | If you receive Signal 11 errors when building on GNU/Linux, then it is | |
2273 | possible you have a hardware problem. Further information on this can be | |
2274 | found on @uref{http://www.bitwizard.nl/sig11/,,www.bitwizard.nl}. | |
2275 | ||
b8df899a | 2276 | @html |
b8db17af | 2277 | <hr /> |
b8df899a JM |
2278 | @end html |
2279 | @heading @anchor{ix86-*-sco}i?86-*-sco | |
2280 | Compilation with RCC is recommended. Also, it may be a good idea to | |
2281 | link with GNU malloc instead of the malloc that comes with the system. | |
2282 | ||
2283 | @html | |
b8db17af | 2284 | <hr /> |
b8df899a JM |
2285 | @end html |
2286 | @heading @anchor{ix86-*-sco3.2v4}i?86-*-sco3.2v4 | |
2287 | Use this configuration for SCO release 3.2 version 4. | |
2288 | ||
f42974dc | 2289 | @html |
b8db17af | 2290 | <hr /> |
f42974dc | 2291 | @end html |
ef88b07d | 2292 | @heading @anchor{ix86-*-sco3.2v5*}i?86-*-sco3.2v5* |
b8df899a | 2293 | Use this for the SCO OpenServer Release 5 family of operating systems. |
f42974dc DW |
2294 | |
2295 | Unlike earlier versions of GCC, the ability to generate COFF with this | |
2296 | target is no longer provided. | |
2297 | ||
021c4bfd | 2298 | Earlier versions of GCC emitted DWARF 1 when generating ELF to allow |
f42974dc | 2299 | the system debugger to be used. That support was too burdensome to |
021c4bfd | 2300 | maintain. GCC now emits only DWARF 2 for this target. This means you |
f42974dc | 2301 | may use either the UDK debugger or GDB to debug programs built by this |
161d7b59 | 2302 | version of GCC@. |
f42974dc | 2303 | |
021c4bfd | 2304 | Use of the @option{-march=pentiumpro} flag can result in |
f42974dc | 2305 | unrecognized opcodes when using the native assembler on OS versions before |
767094dd | 2306 | 5.0.6. (Support for P6 opcodes was added to the native ELF assembler in |
f9047ed3 | 2307 | that version.) While it's rather rare to see these emitted by GCC yet, |
f42974dc DW |
2308 | errors of the basic form: |
2309 | ||
2310 | @example | |
2311 | /usr/tmp/ccaNlqBc.s:22:unknown instruction: fcomip | |
2312 | /usr/tmp/ccaNlqBc.s:50:unknown instruction: fucomip | |
2313 | @end example | |
2314 | ||
2315 | are symptoms of this problem. You may work around this by not | |
2316 | building affected files with that flag, by using the GNU assembler, or | |
161d7b59 | 2317 | by using the assembler provided with the current version of the OS@. |
f42974dc DW |
2318 | Users of GNU assembler should see the note below for hazards on doing |
2319 | so. | |
2320 | ||
2321 | The native SCO assembler that is provided with the OS at no | |
2322 | charge is normally required. If, however, you must be able to use | |
2323 | the GNU assembler (perhaps you're compiling code with asms that | |
2324 | require GAS syntax) you may configure this package using the flags | |
38209993 LG |
2325 | @uref{./configure.html#with-gnu-as,,@option{--with-gnu-as}}. You must |
2326 | use a recent version of GNU binutils; versions past 2.9.1 seem to work | |
2327 | well. | |
f42974dc DW |
2328 | |
2329 | In general, the @option{--with-gnu-as} option isn't as well tested | |
2330 | as the native assembler. | |
2331 | ||
6cfb3f16 | 2332 | Look in @file{gcc/config/i386/sco5.h} (search for ``messy'') for |
f42974dc DW |
2333 | additional OpenServer-specific flags. |
2334 | ||
38209993 | 2335 | Systems based on OpenServer before 5.0.4 (@samp{uname -X} |
021c4bfd RO |
2336 | will tell you what you're running) require TLS597 from |
2337 | @uref{ftp://ftp.sco.com/TLS/,,ftp://ftp.sco.com/TLS/} | |
f42974dc DW |
2338 | for C++ constructors and destructors to work right. |
2339 | ||
2340 | The system linker in (at least) 5.0.4 and 5.0.5 will sometimes | |
2341 | do the wrong thing for a construct that GCC will emit for PIC | |
2342 | code. This can be seen as execution testsuite failures when using | |
6cfb3f16 | 2343 | @option{-fPIC} on @file{921215-1.c}, @file{931002-1.c}, @file{nestfunc-1.c}, and @file{gcov-1.c}. |
f42974dc | 2344 | For 5.0.5, an updated linker that will cure this problem is |
f9047ed3 | 2345 | available. You must install both |
38209993 | 2346 | @uref{ftp://ftp.sco.com/Supplements/rs505a/,,ftp://ftp.sco.com/Supplements/rs505a/} |
f42974dc DW |
2347 | and @uref{ftp://ftp.sco.com/SLS/,,OSS499A}. |
2348 | ||
2349 | The dynamic linker in OpenServer 5.0.5 (earlier versions may show | |
021c4bfd | 2350 | the same problem) aborts on certain G77-compiled programs. It's particularly |
f42974dc DW |
2351 | likely to be triggered by building Fortran code with the @option{-fPIC} flag. |
2352 | Although it's conceivable that the error could be triggered by other | |
2353 | code, only G77-compiled code has been observed to cause this abort. | |
2354 | If you are getting core dumps immediately upon execution of your | |
021c4bfd RO |
2355 | G77 program---and especially if it's compiled with @option{-fPIC}---try applying |
2356 | @uref{sco_osr5_g77.patch,,@file{sco_osr5_g77.patch}} to your @samp{libf2c} and | |
161d7b59 | 2357 | rebuilding GCC@. |
f42974dc DW |
2358 | Affected faults, when analyzed in a debugger, will show a stack |
2359 | backtrace with a fault occurring in @code{rtld()} and the program | |
f9047ed3 | 2360 | running as @file{/usr/lib/ld.so.1}. This problem has been reported to SCO |
f42974dc DW |
2361 | engineering and will hopefully be addressed in later releases. |
2362 | ||
2363 | ||
f42974dc | 2364 | @html |
b8db17af | 2365 | <hr /> |
f42974dc | 2366 | @end html |
ef88b07d | 2367 | @heading @anchor{ix86-*-udk}i?86-*-udk |
f42974dc DW |
2368 | |
2369 | This target emulates the SCO Universal Development Kit and requires that | |
f9047ed3 JM |
2370 | package be installed. (If it is installed, you will have a |
2371 | @file{/udk/usr/ccs/bin/cc} file present.) It's very much like the | |
b953cc4b | 2372 | @samp{i?86-*-unixware7*} target |
f42974dc DW |
2373 | but is meant to be used when hosting on a system where UDK isn't the |
2374 | default compiler such as OpenServer 5 or Unixware 2. This target will | |
f9047ed3 | 2375 | generate binaries that will run on OpenServer, Unixware 2, or Unixware 7, |
161d7b59 | 2376 | with the same warnings and caveats as the SCO UDK@. |
f42974dc | 2377 | |
f42974dc DW |
2378 | This target is a little tricky to build because we have to distinguish |
2379 | it from the native tools (so it gets headers, startups, and libraries | |
f9047ed3 | 2380 | from the right place) while making the tools not think we're actually |
f42974dc DW |
2381 | building a cross compiler. The easiest way to do this is with a configure |
2382 | command like this: | |
2383 | ||
8c085f6f JJ |
2384 | @example |
2385 | CC=/udk/usr/ccs/bin/cc @var{/your/path/to}/gcc/configure \ | |
2386 | --host=i686-pc-udk --target=i686-pc-udk --program-prefix=udk- | |
2387 | @end example | |
f42974dc | 2388 | |
6cfb3f16 | 2389 | @emph{You should substitute @samp{i686} in the above command with the appropriate |
f42974dc DW |
2390 | processor for your host.} |
2391 | ||
021c4bfd RO |
2392 | After the usual @samp{make bootstrap} and |
2393 | @samp{make install}, you can then access the UDK-targeted GCC | |
38209993 LG |
2394 | tools by adding @command{udk-} before the commonly known name. For |
2395 | example, to invoke the C compiler, you would use @command{udk-gcc}. | |
2396 | They will coexist peacefully with any native-target GCC tools you may | |
2397 | have installed. | |
f42974dc DW |
2398 | |
2399 | ||
b499d9ab | 2400 | @html |
b8db17af | 2401 | <hr /> |
b499d9ab JJ |
2402 | @end html |
2403 | @heading @anchor{ia64-*-linux}ia64-*-linux | |
2404 | IA-64 processor (also known as IPF, or Itanium Processor Family) | |
2405 | running GNU/Linux. | |
2406 | ||
2407 | The toolchain is not completely finished, so requirements will continue | |
2408 | to change. | |
2409 | GCC 3.0.1 and later require glibc 2.2.4. | |
2410 | GCC 3.0.2 requires binutils from 2001-09-05 or later. | |
2411 | GCC 3.0.1 requires binutils 2.11.1 or later. | |
2412 | ||
2413 | None of the following versions of GCC has an ABI that is compatible | |
2414 | with any of the other versions in this list, with the exception that | |
2415 | Red Hat 2.96 and Trillian 000171 are compatible with each other: | |
2416 | 3.0.2, 3.0.1, 3.0, Red Hat 2.96, and Trillian 000717. | |
2417 | This primarily affects C++ programs and programs that create shared libraries. | |
2418 | Because of these ABI incompatibilities, GCC 3.0.2 is not recommended for | |
2419 | user programs on GNU/Linux systems built using earlier compiler releases. | |
2420 | GCC 3.0.2 is recommended for compiling linux, the kernel. | |
2421 | GCC 3.0.2 is believed to be fully ABI compliant, and hence no more major | |
2422 | ABI changes are expected. | |
2423 | ||
b8df899a | 2424 | @html |
b8db17af | 2425 | <hr /> |
b8df899a JM |
2426 | @end html |
2427 | @heading @anchor{*-lynx-lynxos}*-lynx-lynxos | |
f0523f02 | 2428 | LynxOS 2.2 and earlier comes with GCC 1.x already installed as |
b8df899a | 2429 | @file{/bin/gcc}. You should compile with this instead of @file{/bin/cc}. |
f0523f02 | 2430 | You can tell GCC to use the GNU assembler and linker, by specifying |
b8df899a | 2431 | @samp{--with-gnu-as --with-gnu-ld} when configuring. These will produce |
f0523f02 | 2432 | COFF format object files and executables; otherwise GCC will use the |
b8df899a JM |
2433 | installed tools, which produce @file{a.out} format executables. |
2434 | ||
f42974dc | 2435 | @html |
b8db17af | 2436 | <hr /> |
f42974dc DW |
2437 | <!-- rs6000-ibm-aix*, powerpc-ibm-aix* --> |
2438 | @end html | |
ef88b07d | 2439 | @heading @anchor{*-ibm-aix*}*-ibm-aix* |
f42974dc DW |
2440 | |
2441 | AIX Make frequently has problems with GCC makefiles. GNU Make 3.76 or | |
2442 | newer is recommended to build on this platform. | |
2443 | ||
6cfb3f16 | 2444 | Errors involving @code{alloca} when building GCC generally are due |
021c4bfd | 2445 | to an incorrect definition of @code{CC} in the Makefile or mixing files |
161d7b59 | 2446 | compiled with the native C compiler and GCC@. During the stage1 phase of |
6cfb3f16 JM |
2447 | the build, the native AIX compiler @strong{must} be invoked as @command{cc} |
2448 | (not @command{xlc}). Once @command{configure} has been informed of | |
2449 | @command{xlc}, one needs to use @samp{make distclean} to remove the | |
38209993 | 2450 | configure cache files and ensure that @env{CC} environment variable |
f42974dc DW |
2451 | does not provide a definition that will confuse @command{configure}. |
2452 | If this error occurs during stage2 or later, then the problem most likely | |
2453 | is the version of Make (see above). | |
2454 | ||
2705baf5 | 2455 | The GNU Assembler incorrectly reports that it supports WEAK symbols on |
225cee28 DE |
2456 | AIX which causes GCC to try to utilize weak symbol functionality although |
2457 | it is not supported on the platform. The native @command{as} and | |
2705baf5 | 2458 | @command{ld} still are recommended. The native AIX tools do |
161d7b59 | 2459 | interoperate with GCC@. |
df002c7d | 2460 | |
04d2be8e | 2461 | Building @file{libstdc++.a} requires a fix for an AIX Assembler bug |
2705baf5 DE |
2462 | APAR IY26685 (AIX 4.3) or APAR IY25528 (AIX 5.1). |
2463 | ||
fdf68669 DE |
2464 | @samp{libstdc++} in GCC 3.2 increments the major version number of the |
2465 | shared object and GCC installation places the @file{libstdc++.a} | |
2466 | shared library in a common location which will overwrite the GCC 3.1 | |
2467 | version of the shared library. Applications either need to be | |
2468 | re-linked against the new shared library or the GCC 3.1 version of the | |
2469 | @samp{libstdc++} shared object needs to be available to the AIX | |
2470 | runtime loader. The GCC 3.1 @samp{libstdc++.so.4} shared object can | |
2471 | be installed for runtime dynamic loading using the following steps to | |
2472 | set the @samp{F_LOADONLY} flag in the shared object for @emph{each} | |
2473 | multilib @file{libstdc++.a} installed: | |
2474 | ||
2475 | Extract the shared object from each the GCC 3.1 @file{libstdc++.a} | |
2476 | archive: | |
2477 | @example | |
35fb4cf6 | 2478 | % ar -x libstdc++.a libstdc++.so.4 |
fdf68669 DE |
2479 | @end example |
2480 | ||
2481 | Enable the @samp{F_LOADONLY} flag so that the shared object will be | |
2482 | available for runtime dynamic loading, but not linking: | |
2483 | @example | |
2484 | % strip -e libstdc++.so.4 | |
2485 | @end example | |
2486 | ||
2487 | Archive the runtime-only shared object in the GCC 3.2 | |
2488 | @file{libstdc++.a} archive: | |
2489 | @example | |
35fb4cf6 | 2490 | % ar -q libstdc++.a libstdc++.so.4 |
fdf68669 DE |
2491 | @end example |
2492 | ||
df002c7d DE |
2493 | Linking executables and shared libraries may produce warnings of |
2494 | duplicate symbols. The assembly files generated by GCC for AIX always | |
2495 | have included multiple symbol definitions for certain global variable | |
2496 | and function declarations in the original program. The warnings should | |
2497 | not prevent the linker from producing a correct library or runnable | |
2498 | executable. | |
2499 | ||
6cfb3f16 | 2500 | AIX 4.3 utilizes a ``large format'' archive to support both 32-bit and |
df002c7d DE |
2501 | 64-bit object modules. The routines provided in AIX 4.3.0 and AIX 4.3.1 |
2502 | to parse archive libraries did not handle the new format correctly. | |
2503 | These routines are used by GCC and result in error messages during | |
6cfb3f16 | 2504 | linking such as ``not a COFF file''. The version of the routines shipped |
df002c7d DE |
2505 | with AIX 4.3.1 should work for a 32-bit environment. The @option{-g} |
2506 | option of the archive command may be used to create archives of 32-bit | |
6cfb3f16 | 2507 | objects using the original ``small format''. A correct version of the |
d5d8d540 | 2508 | routines is shipped with AIX 4.3.2 and above. |
df002c7d | 2509 | |
f42974dc DW |
2510 | Some versions of the AIX binder (linker) can fail with a relocation |
2511 | overflow severe error when the @option{-bbigtoc} option is used to link | |
161d7b59 | 2512 | GCC-produced object files into an executable that overflows the TOC@. A fix |
f42974dc DW |
2513 | for APAR IX75823 (OVERFLOW DURING LINK WHEN USING GCC AND -BBIGTOC) is |
2514 | available from IBM Customer Support and from its | |
d5d8d540 | 2515 | @uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com} |
f42974dc DW |
2516 | website as PTF U455193. |
2517 | ||
df002c7d | 2518 | The AIX 4.3.2.1 linker (bos.rte.bind_cmds Level 4.3.2.1) will dump core |
161d7b59 | 2519 | with a segmentation fault when invoked by any version of GCC@. A fix for |
df002c7d | 2520 | APAR IX87327 is available from IBM Customer Support and from its |
d5d8d540 | 2521 | @uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com} |
df002c7d | 2522 | website as PTF U461879. This fix is incorporated in AIX 4.3.3 and above. |
f42974dc DW |
2523 | |
2524 | The initial assembler shipped with AIX 4.3.0 generates incorrect object | |
2525 | files. A fix for APAR IX74254 (64BIT DISASSEMBLED OUTPUT FROM COMPILER FAILS | |
2526 | TO ASSEMBLE/BIND) is available from IBM Customer Support and from its | |
d5d8d540 | 2527 | @uref{http://techsupport.services.ibm.com/,,techsupport.services.ibm.com} |
f42974dc DW |
2528 | website as PTF U453956. This fix is incorporated in AIX 4.3.1 and above. |
2529 | ||
161d7b59 | 2530 | AIX provides National Language Support (NLS)@. Compilers and assemblers |
df002c7d | 2531 | use NLS to support locale-specific representations of various data |
6cfb3f16 | 2532 | formats including floating-point numbers (e.g., @samp{.} vs @samp{,} for |
df002c7d DE |
2533 | separating decimal fractions). There have been problems reported where |
2534 | GCC does not produce the same floating-point formats that the assembler | |
c771326b | 2535 | expects. If one encounters this problem, set the @env{LANG} |
6cfb3f16 | 2536 | environment variable to @samp{C} or @samp{En_US}. |
f42974dc | 2537 | |
5791e6da DE |
2538 | By default, GCC for AIX 4.1 and above produces code that can be used on |
2539 | both Power or PowerPC processors. | |
2540 | ||
d5d8d540 DE |
2541 | A default can be specified with the @option{-mcpu=@var{cpu_type}} |
2542 | switch and using the configure option @option{--with-cpu-@var{cpu_type}}. | |
f42974dc | 2543 | |
e3223ea2 | 2544 | @html |
b8db17af | 2545 | <hr /> |
e3223ea2 DC |
2546 | @end html |
2547 | @heading @anchor{ip2k-*-elf}ip2k-*-elf | |
2548 | Ubicom IP2022 micro controller. | |
2549 | This configuration is intended for embedded systems. | |
2550 | There are no standard Unix configurations. | |
2551 | ||
2552 | Use @samp{configure --target=ip2k-elf --enable-languages=c} to configure GCC@. | |
2553 | ||
b8df899a | 2554 | @html |
b8db17af | 2555 | <hr /> |
b8df899a JM |
2556 | @end html |
2557 | @heading @anchor{m32r-*-elf}m32r-*-elf | |
2558 | Mitsubishi M32R processor. | |
2559 | This configuration is intended for embedded systems. | |
2560 | ||
2561 | @html | |
b8db17af | 2562 | <hr /> |
b8df899a JM |
2563 | @end html |
2564 | @heading @anchor{m68000-hp-bsd}m68000-hp-bsd | |
161d7b59 | 2565 | HP 9000 series 200 running BSD@. Note that the C compiler that comes |
f0523f02 JM |
2566 | with this system cannot compile GCC; contact @email{law@@cygnus.com} |
2567 | to get binaries of GCC for bootstrapping. | |
b8df899a JM |
2568 | |
2569 | @html | |
b8db17af | 2570 | <hr /> |
b8df899a JM |
2571 | @end html |
2572 | @heading @anchor{m6811-elf}m6811-elf | |
2573 | Motorola 68HC11 family micro controllers. These are used in embedded | |
2574 | applications. There are no standard Unix configurations. | |
2575 | ||
2576 | @html | |
b8db17af | 2577 | <hr /> |
b8df899a JM |
2578 | @end html |
2579 | @heading @anchor{m6812-elf}m6812-elf | |
2580 | Motorola 68HC12 family micro controllers. These are used in embedded | |
2581 | applications. There are no standard Unix configurations. | |
2582 | ||
b8df899a | 2583 | @html |
b8db17af | 2584 | <hr /> |
b8df899a JM |
2585 | @end html |
2586 | @heading @anchor{m68k-att-sysv}m68k-att-sysv | |
161d7b59 | 2587 | AT&T 3b1, a.k.a.@: 7300 PC@. This version of GCC cannot |
b8df899a JM |
2588 | be compiled with the system C compiler, which is too buggy. |
2589 | You will need to get a previous version of GCC and use it to | |
2590 | bootstrap. Binaries are available from the OSU-CIS archive, at | |
2591 | @uref{ftp://archive.cis.ohio-state.edu/pub/att7300/}. | |
2592 | ||
b8df899a | 2593 | @html |
b8db17af | 2594 | <hr /> |
b8df899a | 2595 | @end html |
9340544b | 2596 | @heading @anchor{m68k-crds-unos}m68k-crds-unos |
b8df899a JM |
2597 | Use @samp{configure unos} for building on Unos. |
2598 | ||
2599 | The Unos assembler is named @code{casm} instead of @code{as}. For some | |
2600 | strange reason linking @file{/bin/as} to @file{/bin/casm} changes the | |
f0523f02 | 2601 | behavior, and does not work. So, when installing GCC, you should |
b8df899a JM |
2602 | install the following script as @file{as} in the subdirectory where |
2603 | the passes of GCC are installed: | |
2604 | ||
2605 | @example | |
2606 | #!/bin/sh | |
2607 | casm $* | |
2608 | @end example | |
2609 | ||
2610 | The default Unos library is named @file{libunos.a} instead of | |
f0523f02 | 2611 | @file{libc.a}. To allow GCC to function, either change all |
6cfb3f16 | 2612 | references to @option{-lc} in @file{gcc.c} to @option{-lunos} or link |
b8df899a JM |
2613 | @file{/lib/libc.a} to @file{/lib/libunos.a}. |
2614 | ||
2615 | @cindex @code{alloca}, for Unos | |
f0523f02 | 2616 | When compiling GCC with the standard compiler, to overcome bugs in |
6cfb3f16 JM |
2617 | the support of @code{alloca}, do not use @option{-O} when making stage 2. |
2618 | Then use the stage 2 compiler with @option{-O} to make the stage 3 | |
b8df899a JM |
2619 | compiler. This compiler will have the same characteristics as the usual |
2620 | stage 2 compiler on other systems. Use it to make a stage 4 compiler | |
2621 | and compare that with stage 3 to verify proper compilation. | |
2622 | ||
2623 | (Perhaps simply defining @code{ALLOCA} in @file{x-crds} as described in | |
2624 | the comments there will make the above paragraph superfluous. Please | |
2625 | inform us of whether this works.) | |
2626 | ||
2627 | Unos uses memory segmentation instead of demand paging, so you will need | |
2628 | a lot of memory. 5 Mb is barely enough if no other tasks are running. | |
2629 | If linking @file{cc1} fails, try putting the object files into a library | |
2630 | and linking from that library. | |
2631 | ||
2632 | @html | |
b8db17af | 2633 | <hr /> |
b8df899a JM |
2634 | @end html |
2635 | @heading @anchor{m68k-hp-hpux}m68k-hp-hpux | |
161d7b59 JM |
2636 | HP 9000 series 300 or 400 running HP-UX@. HP-UX version 8.0 has a bug in |
2637 | the assembler that prevents compilation of GCC@. This | |
b8df899a JM |
2638 | bug manifests itself during the first stage of compilation, while |
2639 | building @file{libgcc2.a}: | |
2640 | ||
2641 | @smallexample | |
2642 | _floatdisf | |
2643 | cc1: warning: `-g' option not supported on this version of GCC | |
2644 | cc1: warning: `-g1' option not supported on this version of GCC | |
2645 | ./xgcc: Internal compiler error: program as got fatal signal 11 | |
2646 | @end smallexample | |
2647 | ||
2648 | A patched version of the assembler is available as the file | |
2649 | @uref{ftp://altdorf.ai.mit.edu/archive/cph/hpux-8.0-assembler}. If you | |
2650 | have HP software support, the patch can also be obtained directly from | |
2651 | HP, as described in the following note: | |
2652 | ||
2653 | @quotation | |
2654 | This is the patched assembler, to patch SR#1653-010439, where the | |
2655 | assembler aborts on floating point constants. | |
2656 | ||
2657 | The bug is not really in the assembler, but in the shared library | |
2658 | version of the function ``cvtnum(3c)''. The bug on ``cvtnum(3c)'' is | |
2659 | SR#4701-078451. Anyway, the attached assembler uses the archive | |
2660 | library version of ``cvtnum(3c)'' and thus does not exhibit the bug. | |
2661 | @end quotation | |
2662 | ||
2663 | This patch is also known as PHCO_4484. | |
2664 | ||
021c4bfd | 2665 | In addition, if you wish to use gas, you must use |
b8df899a JM |
2666 | gas version 2.1 or later, and you must use the GNU linker version 2.1 or |
2667 | later. Earlier versions of gas relied upon a program which converted the | |
2668 | gas output into the native HP-UX format, but that program has not been | |
2669 | kept up to date. gdb does not understand that native HP-UX format, so | |
2670 | you must use gas if you wish to use gdb. | |
2671 | ||
2672 | On HP-UX version 8.05, but not on 8.07 or more recent versions, the | |
2673 | @code{fixproto} shell script triggers a bug in the system shell. If you | |
2674 | encounter this problem, upgrade your operating system or use BASH (the | |
2675 | GNU shell) to run @code{fixproto}. This bug will cause the fixproto | |
2676 | program to report an error of the form: | |
2677 | ||
2678 | @example | |
2679 | ./fixproto: sh internal 1K buffer overflow | |
2680 | @end example | |
2681 | ||
2682 | To fix this, you can also change the first line of the fixproto script | |
2683 | to look like: | |
2684 | ||
2685 | @example | |
2686 | #!/bin/ksh | |
2687 | @end example | |
2688 | ||
2689 | ||
b8df899a | 2690 | @html |
b8db17af | 2691 | <hr /> |
b8df899a JM |
2692 | @end html |
2693 | @heading @anchor{m68k-ncr-*}m68k-ncr-* | |
2694 | On the Tower models 4@var{n}0 and 6@var{n}0, by default a process is not | |
2695 | allowed to have more than one megabyte of memory. GCC cannot compile | |
6cfb3f16 | 2696 | itself (or many other programs) with @option{-O} in that much memory. |
b8df899a JM |
2697 | |
2698 | To solve this problem, reconfigure the kernel adding the following line | |
2699 | to the configuration file: | |
2700 | ||
2701 | @smallexample | |
2702 | MAXUMEM = 4096 | |
2703 | @end smallexample | |
2704 | ||
2705 | ||
2706 | @html | |
b8db17af | 2707 | <hr /> |
b8df899a JM |
2708 | @end html |
2709 | @heading @anchor{m68k-sun}m68k-sun | |
2710 | Sun 3. We do not provide a configuration file to use the Sun FPA by | |
2711 | default, because programs that establish signal handlers for floating | |
161d7b59 | 2712 | point traps inherently cannot work with the FPA@. |
f42974dc DW |
2713 | |
2714 | @html | |
b8db17af | 2715 | <hr /> |
f42974dc | 2716 | @end html |
ef88b07d | 2717 | @heading @anchor{m68k-sun-sunos4.1.1}m68k-sun-sunos4.1.1 |
f42974dc DW |
2718 | |
2719 | It is reported that you may need the GNU assembler on this platform. | |
2720 | ||
2721 | ||
b8df899a | 2722 | @html |
b8db17af | 2723 | <hr /> |
b8df899a JM |
2724 | @end html |
2725 | @heading @anchor{mips-*-*}mips-*-* | |
b8df899a JM |
2726 | If on a MIPS system you get an error message saying ``does not have gp |
2727 | sections for all it's [sic] sectons [sic]'', don't worry about it. This | |
2728 | happens whenever you use GAS with the MIPS linker, but there is not | |
2729 | really anything wrong, and it is okay to use the output file. You can | |
2730 | stop such warnings by installing the GNU linker. | |
2731 | ||
2732 | It would be nice to extend GAS to produce the gp tables, but they are | |
2733 | optional, and there should not be a warning about their absence. | |
2734 | ||
b953cc4b RO |
2735 | @heading @anchor{mips-sgi-irix5}mips-sgi-irix5 |
2736 | ||
2737 | This configuration has considerable problems, which will be fixed in a | |
2738 | future release. | |
f42974dc | 2739 | |
213ba345 RO |
2740 | In order to compile GCC on an SGI running IRIX 5, the ``compiler_dev.hdr'' |
2741 | subsystem must be installed from the IDO CD-ROM supplied by Silicon | |
2742 | Graphics. It is also available for download from | |
2743 | @uref{http://www.sgi.com/developers/devtools/apis/ido.html,,http://www.sgi.com/developers/devtools/apis/ido.html}. | |
f42974dc | 2744 | |
213ba345 RO |
2745 | @code{make compare} may fail on version 5 of IRIX unless you add |
2746 | @option{-save-temps} to @code{CFLAGS}. On these systems, the name of the | |
2747 | assembler input file is stored in the object file, and that makes | |
2748 | comparison fail if it differs between the @code{stage1} and | |
2749 | @code{stage2} compilations. The option @option{-save-temps} forces a | |
2750 | fixed name to be used for the assembler input file, instead of a | |
2751 | randomly chosen name in @file{/tmp}. Do not add @option{-save-temps} | |
2752 | unless the comparisons fail without that option. If you do you | |
2753 | @option{-save-temps}, you will have to manually delete the @samp{.i} and | |
2754 | @samp{.s} files after each series of compilations. | |
f42974dc | 2755 | |
213ba345 RO |
2756 | If you use the MIPS C compiler to bootstrap, it may be necessary |
2757 | to increase its table size for switch statements with the | |
2758 | @option{-Wf,-XNg1500} option. If you use the @option{-O2} | |
2759 | optimization option, you also need to use @option{-Olimit 3000}. | |
f42974dc | 2760 | |
b953cc4b | 2761 | To enable debugging under IRIX 5, you must use GNU @command{as} 2.11.2 |
f282ffb3 | 2762 | or later, |
213ba345 RO |
2763 | and use the @option{--with-gnu-as} configure option when configuring GCC. |
2764 | GNU @command{as} is distributed as part of the binutils package. | |
f282ffb3 | 2765 | When using release 2.11.2, you need to apply a patch |
b953cc4b RO |
2766 | @uref{http://sources.redhat.com/ml/binutils/2001-07/msg00352.html,,http://sources.redhat.com/ml/binutils/2001-07/msg00352.html} |
2767 | which will be included in the next release of binutils. | |
f42974dc | 2768 | |
213ba345 RO |
2769 | When building GCC, the build process loops rebuilding @command{cc1} over |
2770 | and over again. This happens on @samp{mips-sgi-irix5.2}, and possibly | |
2771 | other platforms. It has been reported that this is a known bug in the | |
2772 | @command{make} shipped with IRIX 5.2. We recommend you use GNU | |
2773 | @command{make} instead of the vendor supplied @command{make} program; | |
2774 | however, you may have success with @command{smake} on IRIX 5.2 if you do | |
2775 | not have GNU @command{make} available. | |
f42974dc DW |
2776 | |
2777 | @html | |
b8db17af | 2778 | <hr /> |
f42974dc | 2779 | @end html |
b953cc4b | 2780 | @heading @anchor{mips-sgi-irix6}mips-sgi-irix6 |
f42974dc | 2781 | |
213ba345 | 2782 | If you are using IRIX @command{cc} as your bootstrap compiler, you must |
f42974dc DW |
2783 | ensure that the N32 ABI is in use. To test this, compile a simple C |
2784 | file with @command{cc} and then run @command{file} on the | |
2785 | resulting object file. The output should look like: | |
2786 | ||
2787 | @example | |
213ba345 | 2788 | test.o: ELF N32 MSB @dots{} |
f42974dc DW |
2789 | @end example |
2790 | ||
2791 | If you see: | |
213ba345 RO |
2792 | |
2793 | @example | |
2794 | test.o: ELF 32-bit MSB @dots{} | |
2795 | @end example | |
2796 | ||
2797 | or | |
2798 | ||
f42974dc | 2799 | @example |
213ba345 | 2800 | test.o: ELF 64-bit MSB @dots{} |
f42974dc DW |
2801 | @end example |
2802 | ||
213ba345 | 2803 | then your version of @command{cc} uses the O32 or N64 ABI by default. You |
38209993 | 2804 | should set the environment variable @env{CC} to @samp{cc -n32} |
161d7b59 | 2805 | before configuring GCC@. |
f42974dc | 2806 | |
0fca60ab RO |
2807 | If you want the resulting @command{gcc} to run on old 32-bit systems |
2808 | with the MIPS R4400 CPU, you need to ensure that only code for the mips3 | |
2809 | instruction set architecture (ISA) is generated. While GCC 3.x does | |
2810 | this correctly, both GCC 2.95 and SGI's MIPSpro @command{cc} may change | |
2811 | the ISA depending on the machine where GCC is built. Using one of them | |
2812 | as the bootstrap compiler may result in mips4 code, which won't run at | |
2813 | all on mips3-only systems. For the test program above, you should see: | |
2814 | ||
2815 | @example | |
2816 | test.o: ELF N32 MSB mips-3 @dots{} | |
2817 | @end example | |
2818 | ||
2819 | If you get: | |
2820 | ||
2821 | @example | |
2822 | test.o: ELF N32 MSB mips-4 @dots{} | |
2823 | @end example | |
2824 | ||
2825 | instead, you should set the environment variable @env{CC} to @samp{cc | |
2826 | -n32 -mips3} or @samp{gcc -mips3} respectively before configuring GCC@. | |
2827 | ||
213ba345 RO |
2828 | GCC on IRIX 6 is usually built to support both the N32 and N64 ABIs. If |
2829 | you build GCC on a system that doesn't have the N64 libraries installed, | |
2830 | you need to configure with @option{--disable-multilib} so GCC doesn't | |
2831 | try to use them. Look for @file{/usr/lib64/libc.so.1} to see if you | |
2832 | have the 64-bit libraries installed. | |
2833 | ||
2834 | You must @emph{not} use GNU @command{as} (which isn't built anyway as of | |
2835 | binutils 2.11.2) on IRIX 6 platforms; doing so will only cause problems. | |
2836 | ||
f42974dc | 2837 | GCC does not currently support generating O32 ABI binaries in the |
b953cc4b | 2838 | @samp{mips-sgi-irix6} configurations. It is possible to create a GCC |
213ba345 | 2839 | with O32 ABI only support by configuring it for the @samp{mips-sgi-irix5} |
b953cc4b RO |
2840 | target and using a patched GNU @command{as} 2.11.2 as documented in the |
2841 | @uref{#mips-sgi-irix5,,@samp{mips-sgi-irix5}} section above. Using the | |
2842 | native assembler requires patches to GCC which will be included in a | |
2843 | future release. It is | |
213ba345 | 2844 | expected that O32 ABI support will be available again in a future release. |
f42974dc | 2845 | |
b953cc4b RO |
2846 | The @option{--enable-threads} option doesn't currently work, a patch is |
2847 | in preparation for a future release. The @option{--enable-libgcj} | |
2848 | option is disabled by default: IRIX 6 uses a very low default limit | |
2849 | (20480) for the command line length. Although libtool contains a | |
2850 | workaround for this problem, at least the N64 @samp{libgcj} is known not | |
2851 | to build despite this, running into an internal error of the native | |
2852 | @command{ld}. A sure fix is to increase this limit (@samp{ncargs}) to | |
2853 | its maximum of 262144 bytes. If you have root access, you can use the | |
2854 | @command{systune} command to do this. | |
2855 | ||
f42974dc | 2856 | GCC does not correctly pass/return structures which are |
767094dd JM |
2857 | smaller than 16 bytes and which are not 8 bytes. The problem is very |
2858 | involved and difficult to fix. It affects a number of other targets also, | |
57694e40 | 2859 | but IRIX 6 is affected the most, because it is a 64-bit target, and 4 byte |
767094dd | 2860 | structures are common. The exact problem is that structures are being padded |
e979f9e8 | 2861 | at the wrong end, e.g.@: a 4 byte structure is loaded into the lower 4 bytes |
f42974dc DW |
2862 | of the register when it should be loaded into the upper 4 bytes of the |
2863 | register. | |
2864 | ||
2865 | GCC is consistent with itself, but not consistent with the SGI C compiler | |
2866 | (and the SGI supplied runtime libraries), so the only failures that can | |
2867 | happen are when there are library functions that take/return such | |
213ba345 RO |
2868 | structures. There are very few such library functions. Currently this |
2869 | is known to affect @code{inet_ntoa}, @code{inet_lnaof}, | |
46d2e8d7 RO |
2870 | @code{inet_netof}, @code{inet_makeaddr}, and @code{semctl}. Until the |
2871 | bug is fixed, GCC contains workarounds for the known affected functions. | |
f42974dc | 2872 | |
3aa8219e GP |
2873 | See @uref{http://freeware.sgi.com/,,http://freeware.sgi.com/} for more |
2874 | information about using GCC on IRIX platforms. | |
f42974dc | 2875 | |
b8df899a | 2876 | @html |
b8db17af | 2877 | <hr /> |
b8df899a | 2878 | @end html |
021c4bfd | 2879 | @heading @anchor{powerpc*-*-*}powerpc-*-* |
b8df899a | 2880 | |
6cfb3f16 JM |
2881 | You can specify a default version for the @option{-mcpu=@var{cpu_type}} |
2882 | switch by using the configure option @option{--with-cpu-@var{cpu_type}}. | |
b8df899a | 2883 | |
4f2b1139 | 2884 | @html |
b8db17af | 2885 | <hr /> |
4f2b1139 SS |
2886 | @end html |
2887 | @heading @anchor{powerpc-*-darwin*}powerpc-*-darwin* | |
2888 | PowerPC running Darwin (Mac OS X kernel). | |
2889 | ||
2890 | GCC 3.0 does not support Darwin, but 3.1 and later releases will work. | |
2891 | ||
2892 | Pre-installed versions of Mac OS X may not include any developer tools, | |
2893 | meaning that you will not be able to build GCC from source. Tool | |
2894 | binaries are available at | |
2895 | @uref{http://www.opensource.apple.com/projects/darwin} (free | |
2896 | registration required). | |
2897 | ||
2898 | Versions of the assembler prior to ``cctools-364'' cannot handle the | |
8c085f6f | 2899 | 4-argument form of @code{rlwinm} and related mask-using instructions. Darwin |
4f2b1139 SS |
2900 | 1.3 (Mac OS X 10.0) uses cctools-353 for instance. To get cctools-364, |
2901 | check out @file{cctools} with tag @samp{Apple-364}, build it, and | |
2902 | install the assembler as @file{usr/bin/as}. See | |
2903 | @uref{http://www.opensource.apple.com/tools/cvs/docs.html} for details. | |
2904 | ||
2905 | Also, the default stack limit of 512K is too small, and a bootstrap will | |
2906 | typically fail when self-compiling @file{expr.c}. Set the stack to 800K | |
2907 | or more, for instance by doing @samp{limit stack 800}. It's also | |
2908 | convenient to use the GNU preprocessor instead of Apple's during the | |
2909 | first stage of bootstrapping; this is automatic when doing @samp{make | |
2910 | bootstrap}, but to do it from the toplevel objdir you will need to say | |
2911 | @samp{make CC='cc -no-cpp-precomp' bootstrap}. | |
2912 | ||
2913 | Note that the version of GCC shipped by Apple typically includes a | |
2914 | number of extensions not available in a standard GCC release. These | |
2915 | extensions are generally specific to Mac programming. | |
2916 | ||
021c4bfd | 2917 | @html |
b8db17af | 2918 | <hr /> |
021c4bfd RO |
2919 | @end html |
2920 | @heading @anchor{powerpc-*-elf}powerpc-*-elf, powerpc-*-sysv4 | |
2921 | PowerPC system in big endian mode, running System V.4. | |
2922 | ||
f42974dc | 2923 | @html |
b8db17af | 2924 | <hr /> |
f42974dc | 2925 | @end html |
ef88b07d | 2926 | @heading @anchor{powerpc-*-linux-gnu*}powerpc-*-linux-gnu* |
f42974dc | 2927 | |
f9047ed3 | 2928 | You will need |
e8a7b0c2 | 2929 | @uref{ftp://ftp.kernel.org/pub/linux/devel/binutils,,binutils 2.13.90.0.10} |
791a949f | 2930 | or newer for a working GCC@. |
f42974dc | 2931 | |
edf1b3f3 | 2932 | @html |
b8db17af | 2933 | <hr /> |
edf1b3f3 AC |
2934 | @end html |
2935 | @heading @anchor{powerpc-*-netbsd*}powerpc-*-netbsd* | |
2936 | PowerPC system in big endian mode running NetBSD@. To build the | |
a38f87a9 | 2937 | documentation you will need Texinfo version 4.2 (NetBSD 1.5.1 included |
edf1b3f3 AC |
2938 | Texinfo version 3.12). |
2939 | ||
b8df899a | 2940 | @html |
b8db17af | 2941 | <hr /> |
b8df899a JM |
2942 | @end html |
2943 | @heading @anchor{powerpc-*-eabiaix}powerpc-*-eabiaix | |
6cfb3f16 | 2944 | Embedded PowerPC system in big endian mode with @option{-mcall-aix} selected as |
b8df899a JM |
2945 | the default. |
2946 | ||
b8df899a | 2947 | @html |
b8db17af | 2948 | <hr /> |
b8df899a JM |
2949 | @end html |
2950 | @heading @anchor{powerpc-*-eabisim}powerpc-*-eabisim | |
2951 | Embedded PowerPC system in big endian mode for use in running under the | |
2952 | PSIM simulator. | |
2953 | ||
b8df899a | 2954 | @html |
b8db17af | 2955 | <hr /> |
b8df899a JM |
2956 | @end html |
2957 | @heading @anchor{powerpc-*-eabi}powerpc-*-eabi | |
2958 | Embedded PowerPC system in big endian mode. | |
2959 | ||
b8df899a | 2960 | @html |
b8db17af | 2961 | <hr /> |
b8df899a JM |
2962 | @end html |
2963 | @heading @anchor{powerpcle-*-elf}powerpcle-*-elf, powerpcle-*-sysv4 | |
2964 | PowerPC system in little endian mode, running System V.4. | |
2965 | ||
b8df899a | 2966 | @html |
b8db17af | 2967 | <hr /> |
b8df899a JM |
2968 | @end html |
2969 | @heading @anchor{powerpcle-*-eabisim}powerpcle-*-eabisim | |
2970 | Embedded PowerPC system in little endian mode for use in running under | |
2971 | the PSIM simulator. | |
2972 | ||
2973 | @html | |
b8db17af | 2974 | <hr /> |
b8df899a JM |
2975 | @end html |
2976 | @heading @anchor{powerpcle-*-eabi}powerpcle-*-eabi | |
2977 | Embedded PowerPC system in little endian mode. | |
2978 | ||
b8df899a | 2979 | @html |
b8db17af | 2980 | <hr /> |
b8df899a JM |
2981 | @end html |
2982 | @heading @anchor{powerpcle-*-winnt}powerpcle-*-winnt, powerpcle-*-pe | |
161d7b59 | 2983 | PowerPC system in little endian mode running Windows NT@. |
b8df899a | 2984 | |
91abf72d | 2985 | @html |
b8db17af | 2986 | <hr /> |
91abf72d HP |
2987 | @end html |
2988 | @heading @anchor{s390-*-linux*}s390-*-linux* | |
f282ffb3 | 2989 | S/390 system running Linux for S/390@. |
91abf72d HP |
2990 | |
2991 | @html | |
b8db17af | 2992 | <hr /> |
91abf72d HP |
2993 | @end html |
2994 | @heading @anchor{s390x-*-linux*}s390x-*-linux* | |
57694e40 | 2995 | zSeries system (64-bit) running Linux for zSeries@. |
91abf72d | 2996 | |
f42974dc | 2997 | @html |
b8db17af | 2998 | <hr /> |
f42974dc | 2999 | @end html |
250d5688 RO |
3000 | @c Please use Solaris 2 to refer to all release of Solaris, starting |
3001 | @c with 2.0 until 2.6, 7, and 8. Solaris 1 was a marketing name for | |
3002 | @c SunOS 4 releases which we don't use to avoid confusion. Solaris | |
3003 | @c alone is too unspecific and must be avoided. | |
3004 | @heading @anchor{*-*-solaris2*}*-*-solaris2* | |
f42974dc | 3005 | |
250d5688 | 3006 | Sun does not ship a C compiler with Solaris 2. To bootstrap and install |
dbd210ef KC |
3007 | GCC you first have to install a pre-built compiler, see our |
3008 | @uref{binaries.html,,binaries page} for details. | |
f42974dc | 3009 | |
250d5688 RO |
3010 | The Solaris 2 @command{/bin/sh} will often fail to configure |
3011 | @file{libstdc++-v3}, @file{boehm-gc} or | |
021c4bfd | 3012 | @file{libjava}. If you encounter this problem, set @env{CONFIG_SHELL} to |
29f21e4f | 3013 | @command{/bin/ksh} in your environment before running @command{configure}. |
e6855a2d | 3014 | |
b8df899a | 3015 | Solaris 2 comes with a number of optional OS packages. Some of these |
dbd210ef KC |
3016 | packages are needed to use GCC fully, namely @code{SUNWarc}, |
3017 | @code{SUNWbtool}, @code{SUNWesu}, @code{SUNWhea}, @code{SUNWlibm}, | |
3018 | @code{SUNWsprot}, and @code{SUNWtoo}. If you did not install all | |
250d5688 | 3019 | optional packages when installing Solaris 2, you will need to verify that |
b8df899a JM |
3020 | the packages that GCC needs are installed. |
3021 | ||
3022 | To check whether an optional package is installed, use | |
dbd210ef | 3023 | the @command{pkginfo} command. To add an optional package, use the |
250d5688 | 3024 | @command{pkgadd} command. For further details, see the Solaris 2 |
b8df899a JM |
3025 | documentation. |
3026 | ||
250d5688 | 3027 | Trying to use the linker and other tools in |
b8df899a JM |
3028 | @file{/usr/ucb} to install GCC has been observed to cause trouble. |
3029 | For example, the linker may hang indefinitely. The fix is to remove | |
250d5688 | 3030 | @file{/usr/ucb} from your @env{PATH}. |
f42974dc | 3031 | |
021c4bfd RO |
3032 | All releases of GNU binutils prior to 2.11.2 have known bugs on this |
3033 | platform. We recommend the use of GNU binutils 2.11.2 or the vendor | |
3034 | tools (Sun @command{as}, Sun @command{ld}). | |
f42974dc | 3035 | |
250d5688 RO |
3036 | Sun bug 4296832 turns up when compiling X11 headers with GCC 2.95 or |
3037 | newer: @command{g++} will complain that types are missing. These headers assume | |
3038 | that omitting the type means @code{int}; this assumption worked for C89 but | |
3039 | is wrong for C++, and is now wrong for C99 also. | |
3040 | ||
13ba36b4 | 3041 | @command{g++} accepts such (invalid) constructs with the option |
250d5688 RO |
3042 | @option{-fpermissive}; it |
3043 | will assume that any missing type is @code{int} (as defined by C89). | |
3044 | ||
3045 | There are patches for Solaris 2.6 (105633-56 or newer for SPARC, | |
3046 | 106248-42 or newer for Intel), Solaris 7 (108376-21 or newer for SPARC, | |
3047 | 108377-20 for Intel), and Solaris 8 (108652-24 or newer for SPARC, | |
3048 | 108653-22 for Intel) that fix this bug. | |
f42974dc | 3049 | |
dbd210ef | 3050 | @html |
b8db17af | 3051 | <hr /> |
dbd210ef | 3052 | @end html |
250d5688 | 3053 | @heading @anchor{sparc-sun-solaris2*}sparc-sun-solaris2* |
dbd210ef | 3054 | |
1405141b DN |
3055 | When GCC is configured to use binutils 2.11.2 or later the binaries |
3056 | produced are smaller than the ones produced using Sun's native tools; | |
3057 | this difference is quite significant for binaries containing debugging | |
3058 | information. | |
3059 | ||
250d5688 | 3060 | Sun @command{as} 4.x is broken in that it cannot cope with long symbol names. |
dbd210ef KC |
3061 | A typical error message might look similar to the following: |
3062 | ||
8c085f6f JJ |
3063 | @smallexample |
3064 | /usr/ccs/bin/as: "/var/tmp/ccMsw135.s", line 11041: error: | |
3065 | can't compute value of an expression involving an external symbol. | |
3066 | @end smallexample | |
dbd210ef | 3067 | |
250d5688 RO |
3068 | This is Sun bug 4237974. This is fixed with patch 108908-02 for Solaris |
3069 | 2.6 and has been fixed in later (5.x) versions of the assembler, | |
3070 | starting with Solaris 7. | |
dbd210ef | 3071 | |
03b272d2 | 3072 | Starting with Solaris 7, the operating system is capable of executing |
975c6e4e RO |
3073 | 64-bit SPARC V9 binaries. GCC 3.1 and later properly supports |
3074 | this; the @option{-m64} option enables 64-bit code generation. | |
3075 | However, if all you want is code tuned for the UltraSPARC CPU, you | |
3076 | should try the @option{-mtune=ultrasparc} option instead, which produces | |
3077 | code that, unlike full 64-bit code, can still run on non-UltraSPARC | |
edf1c8df | 3078 | machines. |
03b272d2 | 3079 | |
975c6e4e | 3080 | When configuring on a Solaris 7 or later system that is running a kernel |
8947df0c RH |
3081 | that supports only 32-bit binaries, one must configure with |
3082 | @option{--disable-multilib}, since we will not be able to build the | |
3083 | 64-bit target libraries. | |
3fc602a0 | 3084 | |
f42974dc | 3085 | @html |
b8db17af | 3086 | <hr /> |
f42974dc | 3087 | @end html |
ef88b07d | 3088 | @heading @anchor{sparc-sun-solaris2.7}sparc-sun-solaris2.7 |
f42974dc | 3089 | |
250d5688 | 3090 | Sun patch 107058-01 (1999-01-13) for Solaris 7/SPARC triggers a bug in |
f42974dc DW |
3091 | the dynamic linker. This problem (Sun bug 4210064) affects GCC 2.8 |
3092 | and later, including all EGCS releases. Sun formerly recommended | |
3093 | 107058-01 for all Solaris 7 users, but around 1999-09-01 it started to | |
3094 | recommend it only for people who use Sun's compilers. | |
f9047ed3 | 3095 | |
f42974dc DW |
3096 | Here are some workarounds to this problem: |
3097 | @itemize @bullet | |
3098 | @item | |
3099 | Do not install Sun patch 107058-01 until after Sun releases a | |
3100 | complete patch for bug 4210064. This is the simplest course to take, | |
3101 | unless you must also use Sun's C compiler. Unfortunately 107058-01 | |
250d5688 | 3102 | is preinstalled on some new Solaris 7-based hosts, so you may have to |
f42974dc | 3103 | back it out. |
f9047ed3 | 3104 | |
f42974dc DW |
3105 | @item |
3106 | Copy the original, unpatched Solaris 7 | |
3107 | @command{/usr/ccs/bin/as} into | |
975c6e4e | 3108 | @command{/usr/local/lib/gcc-lib/sparc-sun-solaris2.7/3.1/as}, |
f42974dc DW |
3109 | adjusting the latter name to fit your local conventions and software |
3110 | version numbers. | |
3111 | ||
3112 | @item | |
3113 | Install Sun patch 106950-03 (1999-05-25) or later. Nobody with | |
3114 | both 107058-01 and 106950-03 installed has reported the bug with GCC | |
3115 | and Sun's dynamic linker. This last course of action is riskiest, | |
3116 | for two reasons. First, you must install 106950 on all hosts that | |
3117 | run code generated by GCC; it doesn't suffice to install it only on | |
3118 | the hosts that run GCC itself. Second, Sun says that 106950-03 is | |
3119 | only a partial fix for bug 4210064, but Sun doesn't know whether the | |
161d7b59 | 3120 | partial fix is adequate for GCC@. Revision -08 or later should fix |
250d5688 | 3121 | the bug. The current (as of 2001-09-24) revision is -14, and is included in |
f282ffb3 | 3122 | the Solaris 7 Recommended Patch Cluster. |
f9047ed3 | 3123 | @end itemize |
f42974dc DW |
3124 | |
3125 | ||
f42974dc DW |
3126 | @html |
3127 | <p> | |
b8db17af | 3128 | <hr /> |
f42974dc | 3129 | @end html |
250d5688 | 3130 | @heading @anchor{sparc-sun-sunos4*}sparc-sun-sunos4* |
f42974dc | 3131 | |
021c4bfd | 3132 | A bug in the SunOS 4 linker will cause it to crash when linking |
6cfb3f16 | 3133 | @option{-fPIC} compiled objects (and will therefore not allow you to build |
f42974dc DW |
3134 | shared libraries). |
3135 | ||
3136 | To fix this problem you can either use the most recent version of | |
021c4bfd | 3137 | binutils or get the latest SunOS 4 linker patch (patch ID 100170-10) |
f42974dc DW |
3138 | from Sun's patch site. |
3139 | ||
dbd210ef KC |
3140 | Sometimes on a Sun 4 you may observe a crash in the program |
3141 | @command{genflags} or @command{genoutput} while building GCC. This is said to | |
3142 | be due to a bug in @command{sh}. You can probably get around it by running | |
3143 | @command{genflags} or @command{genoutput} manually and then retrying the | |
3144 | @command{make}. | |
f42974dc DW |
3145 | |
3146 | @html | |
b8db17af | 3147 | <hr /> |
f42974dc | 3148 | @end html |
ef88b07d | 3149 | @heading @anchor{sparc-unknown-linux-gnulibc1}sparc-unknown-linux-gnulibc1 |
f42974dc | 3150 | |
f9047ed3 | 3151 | It has been reported that you might need |
021c4bfd | 3152 | @uref{ftp://ftp.yggdrasil.com/private/hjl,,binutils 2.8.1.0.23} |
f42974dc DW |
3153 | for this platform, too. |
3154 | ||
3155 | ||
c6fa9728 | 3156 | @html |
b8db17af | 3157 | <hr /> |
c6fa9728 JS |
3158 | @end html |
3159 | @heading @anchor{sparc-*-linux*}sparc-*-linux* | |
3160 | ||
3161 | GCC versions 3.0 and higher require binutils 2.11.2 and glibc 2.2.4 | |
3162 | or newer on this platform. All earlier binutils and glibc | |
3163 | releases mishandled unaligned relocations on @code{sparc-*-*} targets. | |
3164 | ||
3165 | ||
f42974dc | 3166 | @html |
b8db17af | 3167 | <hr /> |
f42974dc | 3168 | @end html |
ef88b07d | 3169 | @heading @anchor{sparc64-*-*}sparc64-*-* |
f42974dc DW |
3170 | |
3171 | GCC version 2.95 is not able to compile code correctly for | |
3172 | @code{sparc64} targets. Users of the Linux kernel, at least, | |
e9d21442 | 3173 | can use the @code{sparc32} program to start up a new shell |
f42974dc | 3174 | invocation with an environment that causes @command{configure} to |
021c4bfd | 3175 | recognize (via @samp{uname -a}) the system as @samp{sparc-*-*} instead. |
f42974dc | 3176 | |
e403b4bc | 3177 | @html |
b8db17af | 3178 | <hr /> |
e403b4bc CR |
3179 | @end html |
3180 | @heading @anchor{sparcv9-*-solaris2*}sparcv9-*-solaris2* | |
3181 | ||
3182 | The following compiler flags must be specified in the configure | |
3183 | step in order to bootstrap this target with the Sun compiler: | |
3184 | ||
3185 | @example | |
3186 | % CC="cc -xildoff -xarch=v9" @var{srcdir}/configure [@var{options}] [@var{target}] | |
3187 | @end example | |
3188 | ||
3189 | @option{-xildoff} turns off the incremental linker, and @option{-xarch=v9} | |
3190 | specifies the v9 architecture to the Sun linker and assembler. | |
f42974dc | 3191 | |
b8df899a | 3192 | @html |
b8db17af | 3193 | <hr /> |
b8df899a JM |
3194 | @end html |
3195 | @heading @anchor{#*-*-sysv*}*-*-sysv* | |
3196 | On System V release 3, you may get this error message | |
3197 | while linking: | |
3198 | ||
3199 | @smallexample | |
3200 | ld fatal: failed to write symbol name @var{something} | |
3201 | in strings table for file @var{whatever} | |
3202 | @end smallexample | |
3203 | ||
021c4bfd | 3204 | This probably indicates that the disk is full or your ulimit won't allow |
b8df899a JM |
3205 | the file to be as large as it needs to be. |
3206 | ||
3207 | This problem can also result because the kernel parameter @code{MAXUMEM} | |
3208 | is too small. If so, you must regenerate the kernel and make the value | |
3209 | much larger. The default value is reported to be 1024; a value of 32768 | |
3210 | is said to work. Smaller values may also work. | |
3211 | ||
3212 | On System V, if you get an error like this, | |
3213 | ||
3214 | @example | |
3215 | /usr/local/lib/bison.simple: In function `yyparse': | |
3216 | /usr/local/lib/bison.simple:625: virtual memory exhausted | |
3217 | @end example | |
3218 | ||
3219 | @noindent | |
021c4bfd | 3220 | that too indicates a problem with disk space, ulimit, or @code{MAXUMEM}. |
b8df899a | 3221 | |
f85b8d1a JM |
3222 | On a System V release 4 system, make sure @file{/usr/bin} precedes |
3223 | @file{/usr/ucb} in @code{PATH}. The @code{cc} command in | |
3224 | @file{/usr/ucb} uses libraries which have bugs. | |
b8df899a JM |
3225 | |
3226 | @html | |
b8db17af | 3227 | <hr /> |
b8df899a JM |
3228 | @end html |
3229 | @heading @anchor{vax-dec-ultrix}vax-dec-ultrix | |
8aeea6e6 | 3230 | Don't try compiling with VAX C (@code{vcc}). It produces incorrect code |
b8df899a JM |
3231 | in some cases (for example, when @code{alloca} is used). |
3232 | ||
4977bab6 ZW |
3233 | @html |
3234 | <hr /> | |
3235 | @end html | |
3236 | @heading @anchor{*-*-vxworks*}*-*-vxworks* | |
3237 | Support for VxWorks is in flux. At present GCC supports @emph{only} the | |
3238 | very recent VxWorks 5.5 (aka Tornado 2.2) release, and only on PowerPC. | |
3239 | We welcome patches for other architectures supported by VxWorks 5.5. | |
3240 | Support for VxWorks AE would also be welcome; we believe this is merely | |
3241 | a matter of writing an appropriate ``configlette'' (see below). We are | |
3242 | not interested in supporting older, a.out or COFF-based, versions of | |
3243 | VxWorks in GCC 3. | |
3244 | ||
3245 | VxWorks comes with an older version of GCC installed in | |
3246 | @file{@var{$WIND_BASE}/host}; we recommend you do not overwrite it. | |
3247 | Choose an installation @var{prefix} entirely outside @var{$WIND_BASE}. | |
3248 | Before running @command{configure}, create the directories @file{@var{prefix}} | |
3249 | and @file{@var{prefix}/bin}. Link or copy the appropriate assembler, | |
3250 | linker, etc. into @file{@var{prefix}/bin}, and set your @var{PATH} to | |
3251 | include that directory while running both @command{configure} and | |
3252 | @command{make}. | |
3253 | ||
3254 | You must give @command{configure} the | |
3255 | @option{--with-headers=@var{$WIND_BASE}/target/h} switch so that it can | |
3256 | find the VxWorks system headers. Since VxWorks is a cross compilation | |
3257 | target only, you must also specify @option{--target=@var{target}}. | |
3258 | @command{configure} will attempt to create the directory | |
3259 | @file{@var{prefix}/@var{target}/sys-include} and copy files into it; | |
3260 | make sure the user running @command{configure} has sufficient privilege | |
3261 | to do so. | |
3262 | ||
3263 | GCC's exception handling runtime requires a special ``configlette'' | |
3264 | module, @file{contrib/gthr_supp_vxw_5x.c}. Follow the instructions in | |
3265 | that file to add the module to your kernel build. (Future versions of | |
3266 | VxWorks will incorporate this module.) | |
3267 | ||
fd29f6ea | 3268 | @html |
b8db17af | 3269 | <hr /> |
fd29f6ea BW |
3270 | @end html |
3271 | @heading @anchor{xtensa-*-elf}xtensa-*-elf | |
3272 | ||
3273 | This target is intended for embedded Xtensa systems using the | |
3274 | @samp{newlib} C library. It uses ELF but does not support shared | |
3275 | objects. Designed-defined instructions specified via the | |
3276 | Tensilica Instruction Extension (TIE) language are only supported | |
3277 | through inline assembly. | |
3278 | ||
3279 | The Xtensa configuration information must be specified prior to | |
3280 | building GCC@. The @file{gcc/config/xtensa/xtensa-config.h} header | |
3281 | file contains the configuration information. If you created your | |
3282 | own Xtensa configuration with the Xtensa Processor Generator, the | |
3283 | downloaded files include a customized copy of this header file, | |
3284 | which you can use to replace the default header file. | |
3285 | ||
3286 | @html | |
b8db17af | 3287 | <hr /> |
fd29f6ea BW |
3288 | @end html |
3289 | @heading @anchor{xtensa-*-linux*}xtensa-*-linux* | |
3290 | ||
3291 | This target is for Xtensa systems running GNU/Linux. It supports ELF | |
3292 | shared objects and the GNU C library (glibc). It also generates | |
3293 | position-independent code (PIC) regardless of whether the | |
3294 | @option{-fpic} or @option{-fPIC} options are used. In other | |
f282ffb3 | 3295 | respects, this target is the same as the |
fd29f6ea BW |
3296 | @uref{#xtensa-*-elf,,@samp{xtensa-*-elf}} target. |
3297 | ||
f42974dc | 3298 | @html |
b8db17af | 3299 | <hr /> |
f42974dc | 3300 | @end html |
57694e40 | 3301 | @heading @anchor{windows}Microsoft Windows (32-bit) |
f42974dc | 3302 | |
f9047ed3 | 3303 | A port of GCC 2.95.x is included with the |
f42974dc DW |
3304 | @uref{http://www.cygwin.com/,,Cygwin environment}. |
3305 | ||
3306 | Current (as of early 2001) snapshots of GCC will build under Cygwin | |
3307 | without modification. | |
3308 | ||
3309 | @html | |
b8db17af | 3310 | <hr /> |
f42974dc | 3311 | @end html |
ef88b07d | 3312 | @heading @anchor{os2}OS/2 |
f42974dc DW |
3313 | |
3314 | GCC does not currently support OS/2. However, Andrew Zabolotny has been | |
14976c58 | 3315 | working on a generic OS/2 port with pgcc. The current code can be found |
f42974dc DW |
3316 | at @uref{http://www.goof.com/pcg/os2/,,http://www.goof.com/pcg/os2/}. |
3317 | ||
f9047ed3 | 3318 | An older copy of GCC 2.8.1 is included with the EMX tools available at |
f42974dc DW |
3319 | @uref{ftp://ftp.leo.org/pub/comp/os/os2/leo/devtools/emx+gcc/,, |
3320 | ftp://ftp.leo.org/pub/comp/os/os2/leo/devtools/emx+gcc/}. | |
3321 | ||
3322 | @html | |
b8db17af | 3323 | <hr /> |
f42974dc | 3324 | @end html |
ef88b07d | 3325 | @heading @anchor{older}Older systems |
f9047ed3 JM |
3326 | |
3327 | GCC contains support files for many older (1980s and early | |
3328 | 1990s) Unix variants. For the most part, support for these systems | |
3329 | has not been deliberately removed, but it has not been maintained for | |
c7bdf0a6 | 3330 | several years and may suffer from bitrot. |
f9047ed3 | 3331 | |
c7bdf0a6 | 3332 | Starting with GCC 3.1, each release has a list of ``obsoleted'' systems. |
9340544b ZW |
3333 | Support for these systems is still present in that release, but |
3334 | @command{configure} will fail unless the @option{--enable-obsolete} | |
c7bdf0a6 ZW |
3335 | option is given. Unless a maintainer steps forward, support for these |
3336 | systems will be removed from the next release of GCC@. | |
f9047ed3 JM |
3337 | |
3338 | Support for old systems as hosts for GCC can cause problems if the | |
3339 | workarounds for compiler, library and operating system bugs affect the | |
161d7b59 | 3340 | cleanliness or maintainability of the rest of GCC@. In some cases, to |
f9047ed3 JM |
3341 | bring GCC up on such a system, if still possible with current GCC, may |
3342 | require first installing an old version of GCC which did work on that | |
c7bdf0a6 ZW |
3343 | system, and using it to compile a more recent GCC, to avoid bugs in the |
3344 | vendor compiler. Old releases of GCC 1 and GCC 2 are available in the | |
3345 | @file{old-releases} directory on the @uref{../mirrors.html,,GCC mirror | |
3346 | sites}. Header bugs may generally be avoided using | |
3347 | @command{fixincludes}, but bugs or deficiencies in libraries and the | |
3348 | operating system may still cause problems. | |
3349 | ||
3350 | Support for older systems as targets for cross-compilation is less | |
3351 | problematic than support for them as hosts for GCC; if an enthusiast | |
3352 | wishes to make such a target work again (including resurrecting any of | |
3353 | the targets that never worked with GCC 2, starting from the last CVS | |
3354 | version before they were removed), patches | |
3355 | @uref{../contribute.html,,following the usual requirements} would be | |
3356 | likely to be accepted, since they should not affect the support for more | |
3357 | modern targets. | |
f9047ed3 JM |
3358 | |
3359 | For some systems, old versions of GNU binutils may also be useful, | |
021c4bfd | 3360 | and are available from @file{pub/binutils/old-releases} on |
f42974dc | 3361 | @uref{http://sources.redhat.com/mirrors.html,,sources.redhat.com mirror sites}. |
f9047ed3 JM |
3362 | |
3363 | Some of the information on specific systems above relates to | |
3364 | such older systems, but much of the information | |
3365 | about GCC on such systems (which may no longer be applicable to | |
f42974dc | 3366 | current GCC) is to be found in the GCC texinfo manual. |
f9047ed3 | 3367 | |
f42974dc | 3368 | @html |
b8db17af | 3369 | <hr /> |
f42974dc | 3370 | @end html |
250d5688 | 3371 | @heading @anchor{elf_targets}all ELF targets (SVR4, Solaris 2, etc.) |
f42974dc | 3372 | |
38209993 LG |
3373 | C++ support is significantly better on ELF targets if you use the |
3374 | @uref{./configure.html#with-gnu-ld,,GNU linker}; duplicate copies of | |
3375 | inlines, vtables and template instantiations will be discarded | |
3376 | automatically. | |
f42974dc DW |
3377 | |
3378 | ||
3379 | @html | |
b8db17af | 3380 | <hr /> |
f42974dc DW |
3381 | <p> |
3382 | @end html | |
3383 | @ifhtml | |
3384 | @uref{./index.html,,Return to the GCC Installation page} | |
3385 | @end ifhtml | |
3386 | @end ifset | |
3387 | ||
73e2155a JM |
3388 | @c ***Old documentation****************************************************** |
3389 | @ifset oldhtml | |
3390 | @include install-old.texi | |
3391 | @html | |
b8db17af | 3392 | <hr /> |
73e2155a JM |
3393 | <p> |
3394 | @end html | |
3395 | @ifhtml | |
3396 | @uref{./index.html,,Return to the GCC Installation page} | |
3397 | @end ifhtml | |
3398 | @end ifset | |
3399 | ||
aed5964b JM |
3400 | @c ***GFDL******************************************************************** |
3401 | @ifset gfdlhtml | |
3402 | @include fdl.texi | |
3403 | @html | |
b8db17af | 3404 | <hr /> |
aed5964b JM |
3405 | <p> |
3406 | @end html | |
3407 | @ifhtml | |
3408 | @uref{./index.html,,Return to the GCC Installation page} | |
3409 | @end ifhtml | |
3410 | @end ifset | |
3411 | ||
f42974dc DW |
3412 | @c *************************************************************************** |
3413 | @c Part 6 The End of the Document | |
3414 | @ifinfo | |
3415 | @comment node-name, next, previous, up | |
aed5964b | 3416 | @node Concept Index, , GNU Free Documentation License, Top |
f42974dc DW |
3417 | @end ifinfo |
3418 | ||
3419 | @ifinfo | |
3420 | @unnumbered Concept Index | |
3421 | ||
3422 | @printindex cp | |
3423 | ||
3424 | @contents | |
3425 | @end ifinfo | |
3426 | @bye |