5 @node Installation, Maintenance, Library Summary, Top
8 @c %MENU% How to install the GNU C Library
9 @appendix Installing @theglibc{}
11 Before you do anything else, you should read the FAQ at
12 @url{https://sourceware.org/glibc/wiki/FAQ}. It answers common
13 questions and describes problems you may experience with compilation
16 You will need recent versions of several GNU tools: definitely GCC and
17 GNU Make, and possibly others. @xref{Tools for Compilation}, below.
21 * Configuring and compiling:: How to compile and test GNU libc.
22 * Running make install:: How to install it once you've got it
24 * Tools for Compilation:: You'll need these first.
25 * Linux:: Specific advice for GNU/Linux systems.
26 * Reporting Bugs:: So they'll get fixed.
30 @node Configuring and compiling
31 @appendixsec Configuring and compiling @theglibc{}
35 @Theglibc{} cannot be compiled in the source directory. You must build
36 it in a separate build directory. For example, if you have unpacked
37 the @glibcadj{} sources in @file{/src/gnu/glibc-@var{version}},
39 @file{/src/gnu/glibc-build} to put the object files in. This allows
40 removing the whole build directory in case an error occurs, which is
41 the safest way to get a fresh start and should always be done.
43 From your object directory, run the shell script @file{configure} located
44 at the top level of the source tree. In the scenario above, you'd type
47 $ ../glibc-@var{version}/configure @var{args@dots{}}
50 Please note that even though you're building in a separate build
51 directory, the compilation may need to create or modify files and
52 directories in the source directory.
55 @code{configure} takes many options, but the only one that is usually
56 mandatory is @samp{--prefix}. This option tells @code{configure}
57 where you want @theglibc{} installed. This defaults to @file{/usr/local},
58 but the normal setting to install as the standard system library is
59 @samp{--prefix=/usr} for @gnulinuxsystems{} and @samp{--prefix=} (an
60 empty prefix) for @gnuhurdsystems{}.
62 It may also be useful to pass @samp{CC=@var{compiler}} and
63 @code{CFLAGS=@var{flags}} arguments to @code{configure}. @code{CC}
64 selects the C compiler that will be used, and @code{CFLAGS} sets
65 optimization options for the compiler. Any compiler options required
66 for all compilations, such as options selecting an ABI or a processor
67 for which to generate code, should be included in @code{CC}. Options
68 that may be overridden by the @glibcadj{} build system for particular
69 files, such as for optimization and debugging, should go in
70 @code{CFLAGS}. The default value of @code{CFLAGS} is @samp{-g -O2},
71 and @theglibc{} cannot be compiled without optimization, so if
72 @code{CFLAGS} is specified it must enable optimization. For example:
75 $ ../glibc-@var{version}/configure CC="gcc -m32" CFLAGS="-O3"
78 The following list describes all of the available options for
82 @item --prefix=@var{directory}
83 Install machine-independent data files in subdirectories of
84 @file{@var{directory}}. The default is to install in @file{/usr/local}.
86 @item --exec-prefix=@var{directory}
87 Install the library and other machine-dependent files in subdirectories
88 of @file{@var{directory}}. The default is to the @samp{--prefix}
89 directory if that option is specified, or @file{/usr/local} otherwise.
91 @item --with-headers=@var{directory}
92 Look for kernel header files in @var{directory}, not
93 @file{/usr/include}. @Theglibc{} needs information from the kernel's header
94 files describing the interface to the kernel. @Theglibc{} will normally
95 look in @file{/usr/include} for them,
96 but if you specify this option, it will look in @var{DIRECTORY} instead.
98 This option is primarily of use on a system where the headers in
99 @file{/usr/include} come from an older version of @theglibc{}. Conflicts can
100 occasionally happen in this case. You can also use this option if you want to
101 compile @theglibc{} with a newer set of kernel headers than the ones found in
104 @item --enable-kernel=@var{version}
105 This option is currently only useful on @gnulinuxsystems{}. The
106 @var{version} parameter should have the form X.Y.Z and describes the
107 smallest version of the Linux kernel the generated library is expected
108 to support. The higher the @var{version} number is, the less
109 compatibility code is added, and the faster the code gets.
111 @item --with-binutils=@var{directory}
112 Use the binutils (assembler and linker) in @file{@var{directory}}, not
113 the ones the C compiler would default to. You can use this option if
114 the default binutils on your system cannot deal with all the constructs
115 in @theglibc{}. In that case, @code{configure} will detect the
116 problem and suppress these constructs, so that the library will still be
117 usable, but functionality may be lost---for example, you can't build a
118 shared libc with old binutils.
120 @item --with-nonshared-cflags=@var{cflags}
121 Use additional compiler flags @var{cflags} to build the parts of the
122 library which are always statically linked into applications and
123 libraries even with shared linking (that is, the object files contained
124 in @file{lib*_nonshared.a} libraries). The build process will
125 automatically use the appropriate flags, but this option can be used to
126 set additional flags required for building applications and libraries,
127 to match local policy. For example, if such a policy requires that all
128 code linked into applications must be built with source fortification,
129 @samp{--with-nonshared-cflags=-Wp,-D_FORTIFY_SOURCE=2} will make sure
130 that the objects in @file{libc_nonshared.a} are compiled with this flag
131 (although this will not affect the generated code in this particular
132 case and potentially change debugging information and metadata only).
134 @c disable static doesn't work currently
135 @c @item --disable-static
136 @c Don't build static libraries. Static libraries aren't that useful these
137 @c days, but we recommend you build them in case you need them.
139 @item --disable-shared
140 Don't build shared libraries even if it is possible. Not all systems
141 support shared libraries; you need ELF support and (currently) the GNU
144 @item --enable-static-pie
145 Enable static position independent executable (static PIE) support.
146 Static PIE is similar to static executable, but can be loaded at any
147 address without help from a dynamic linker. All static programs as
148 well as static tests are built as static PIE, except for those marked
149 with no-pie. The resulting glibc can be used with the GCC option,
150 -static-pie, which is available with GCC 8 or above, to create static
151 PIE. This option also implies that glibc programs and tests are created
152 as dynamic position independent executables (PIE) by default.
155 @itemx --enable-cet=permissive
156 Enable Intel Control-flow Enforcement Technology (CET) support. When
157 @theglibc{} is built with @option{--enable-cet} or
158 @option{--enable-cet=permissive}, the resulting library
159 is protected with indirect branch tracking (IBT) and shadow stack
160 (SHSTK)@. When CET is enabled, @theglibc{} is compatible with all
161 existing executables and shared libraries. This feature is currently
162 supported on i386, x86_64 and x32 with GCC 8 and binutils 2.29 or later.
163 Note that when CET is enabled, @theglibc{} requires CPUs capable of
164 multi-byte NOPs, like x86-64 processors as well as Intel Pentium Pro or
165 newer. With @option{--enable-cet}, it is an error to dlopen a non CET
166 enabled shared library in CET enabled application. With
167 @option{--enable-cet=permissive}, CET is disabled when dlopening a
168 non CET enabled shared library in CET enabled application.
170 NOTE: @option{--enable-cet} has been tested for i686, x86_64 and x32
171 on non-CET processors. @option{--enable-cet} has been tested for
172 i686, x86_64 and x32 on CET processors.
174 @item --disable-profile
175 Don't build libraries with profiling information. You may want to use
176 this option if you don't plan to do profiling.
178 @item --enable-static-nss
179 Compile static versions of the NSS (Name Service Switch) libraries.
180 This is not recommended because it defeats the purpose of NSS; a program
181 linked statically with the NSS libraries cannot be dynamically
182 reconfigured to use a different name database.
184 @item --enable-hardcoded-path-in-tests
185 By default, dynamic tests are linked to run with the installed C library.
186 This option hardcodes the newly built C library path in dynamic tests
187 so that they can be invoked directly.
189 @item --disable-timezone-tools
190 By default, timezone related utilities (@command{zic}, @command{zdump},
191 and @command{tzselect}) are installed with @theglibc{}. If you are building
192 these independently (e.g. by using the @samp{tzcode} package), then this
193 option will allow disabling the install of these.
195 Note that you need to make sure the external tools are kept in sync with
196 the versions that @theglibc{} expects as the data formats may change over
197 time. Consult the @file{timezone} subdirectory for more details.
199 @item --enable-stack-protector
200 @itemx --enable-stack-protector=strong
201 @itemx --enable-stack-protector=all
202 Compile the C library and all other parts of the glibc package
203 (including the threading and math libraries, NSS modules, and
204 transliteration modules) using the GCC @option{-fstack-protector},
205 @option{-fstack-protector-strong} or @option{-fstack-protector-all}
206 options to detect stack overruns. Only the dynamic linker and a small
207 number of routines called directly from assembler are excluded from this
210 @item --enable-bind-now
211 Disable lazy binding for installed shared objects and programs. This
212 provides additional security hardening because it enables full RELRO
213 and a read-only global offset table (GOT), at the cost of slightly
214 increased program load times.
218 @item --enable-pt_chown
219 The file @file{pt_chown} is a helper binary for @code{grantpt}
220 (@pxref{Allocation, Pseudo-Terminals}) that is installed setuid root to
221 fix up pseudo-terminal ownership. It is not built by default because
222 systems using the Linux kernel are commonly built with the @code{devpts}
223 filesystem enabled and mounted at @file{/dev/pts}, which manages
224 pseudo-terminal ownership automatically. By using
225 @samp{--enable-pt_chown}, you may build @file{pt_chown} and install it
226 setuid and owned by @code{root}. The use of @file{pt_chown} introduces
227 additional security risks to the system and you should enable it only if
228 you understand and accept those risks.
230 @item --disable-werror
231 By default, @theglibc{} is built with @option{-Werror}. If you wish
232 to build without this option (for example, if building with a newer
233 version of GCC than this version of @theglibc{} was tested with, so
234 new warnings cause the build with @option{-Werror} to fail), you can
235 configure with @option{--disable-werror}.
237 @item --disable-mathvec
238 By default for x86_64, @theglibc{} is built with the vector math library.
239 Use this option to disable the vector math library.
241 @item --enable-tunables
242 Tunables support allows additional library parameters to be customized at
243 runtime. This feature is enabled by default. This option can take the
248 This is the default if no option is passed to configure. This enables tunables
249 and selects the default frontend (currently @samp{valstring}).
252 This option disables tunables.
255 This enables tunables and selects the @samp{valstring} frontend for tunables.
256 This frontend allows users to specify tunables as a colon-separated list in a
257 single environment variable @env{GLIBC_TUNABLES}.
260 @item --disable-crypt
261 Do not install the passphrase-hashing library @file{libcrypt} or the
262 header file @file{crypt.h}. @file{unistd.h} will still declare the
263 function @code{crypt}. Using this option does not change the set of
264 programs that may need to be linked with @option{-lcrypt}; it only
265 means that @theglibc{} will not provide that library.
267 This option is for hackers and distributions experimenting with
268 independently-maintained implementations of libcrypt. It may become
269 the default in a future release.
271 @item --disable-experimental-malloc
272 By default, a per-thread cache is enabled in @code{malloc}. While
273 this cache can be disabled on a per-application basis using tunables
274 (set glibc.malloc.tcache_count to zero), this option can be used to
275 remove it from the build completely.
277 @item --build=@var{build-system}
278 @itemx --host=@var{host-system}
279 These options are for cross-compiling. If you specify both options and
280 @var{build-system} is different from @var{host-system}, @code{configure}
281 will prepare to cross-compile @theglibc{} from @var{build-system} to be used
282 on @var{host-system}. You'll probably need the @samp{--with-headers}
283 option too, and you may have to override @var{configure}'s selection of
284 the compiler and/or binutils.
286 If you only specify @samp{--host}, @code{configure} will prepare for a
287 native compile but use what you specify instead of guessing what your
288 system is. This is most useful to change the CPU submodel. For example,
289 if @code{configure} guesses your machine as @code{i686-pc-linux-gnu} but
290 you want to compile a library for 586es, give
291 @samp{--host=i586-pc-linux-gnu} or just @samp{--host=i586-linux} and add
292 the appropriate compiler flags (@samp{-mcpu=i586} will do the trick) to
295 If you specify just @samp{--build}, @code{configure} will get confused.
297 @item --with-pkgversion=@var{version}
298 Specify a description, possibly including a build number or build
299 date, of the binaries being built, to be included in
300 @option{--version} output from programs installed with @theglibc{}.
301 For example, @option{--with-pkgversion='FooBar GNU/Linux glibc build
302 123'}. The default value is @samp{GNU libc}.
304 @item --with-bugurl=@var{url}
305 Specify the URL that users should visit if they wish to report a bug,
306 to be included in @option{--help} output from programs installed with
307 @theglibc{}. The default value refers to the main bug-reporting
308 information for @theglibc{}.
311 To build the library and related programs, type @code{make}. This will
312 produce a lot of output, some of which may look like errors from
313 @code{make} but aren't. Look for error messages from @code{make}
314 containing @samp{***}. Those indicate that something is seriously wrong.
316 The compilation process can take a long time, depending on the
317 configuration and the speed of your machine. Some complex modules may
318 take a very long time to compile, as much as several minutes on slower
319 machines. Do not panic if the compiler appears to hang.
321 If you want to run a parallel make, simply pass the @samp{-j} option
322 with an appropriate numeric parameter to @code{make}. You need a recent
323 GNU @code{make} version, though.
325 To build and run test programs which exercise some of the library
326 facilities, type @code{make check}. If it does not complete
327 successfully, do not use the built library, and report a bug after
328 verifying that the problem is not already known. @xref{Reporting Bugs},
329 for instructions on reporting bugs. Note that some of the tests assume
330 they are not being run by @code{root}. We recommend you compile and
331 test @theglibc{} as an unprivileged user.
333 Before reporting bugs make sure there is no problem with your system.
334 The tests (and later installation) use some pre-existing files of the
335 system such as @file{/etc/passwd}, @file{/etc/nsswitch.conf} and others.
336 These files must all contain correct and sensible content.
338 Normally, @code{make check} will run all the tests before reporting
339 all problems found and exiting with error status if any problems
340 occurred. You can specify @samp{stop-on-test-failure=y} when running
341 @code{make check} to make the test run stop and exit with an error
342 status immediately when a failure occurs.
344 To format the @cite{GNU C Library Reference Manual} for printing, type
345 @w{@code{make dvi}}. You need a working @TeX{} installation to do
346 this. The distribution builds the on-line formatted version of the
347 manual, as Info files, as part of the build process. You can build
348 them manually with @w{@code{make info}}.
350 The library has a number of special-purpose configuration parameters
351 which you can find in @file{Makeconfig}. These can be overwritten with
352 the file @file{configparms}. To change them, create a
353 @file{configparms} in your build directory and add values as appropriate
354 for your system. The file is included and parsed by @code{make} and has
355 to follow the conventions for makefiles.
357 It is easy to configure @theglibc{} for cross-compilation by
358 setting a few variables in @file{configparms}. Set @code{CC} to the
359 cross-compiler for the target you configured the library for; it is
360 important to use this same @code{CC} value when running
361 @code{configure}, like this: @samp{configure @var{target}
362 CC=@var{target}-gcc}. Set @code{BUILD_CC} to the compiler to use for programs
363 run on the build system as part of compiling the library. You may need to
364 set @code{AR} to cross-compiling versions of @code{ar}
365 if the native tools are not configured to work with
366 object files for the target you configured for. When cross-compiling
367 @theglibc{}, it may be tested using @samp{make check
368 test-wrapper="@var{srcdir}/scripts/cross-test-ssh.sh @var{hostname}"},
369 where @var{srcdir} is the absolute directory name for the main source
370 directory and @var{hostname} is the host name of a system that can run
371 the newly built binaries of @theglibc{}. The source and build
372 directories must be visible at the same locations on both the build
373 system and @var{hostname}.
375 In general, when testing @theglibc{}, @samp{test-wrapper} may be set
376 to the name and arguments of any program to run newly built binaries.
377 This program must preserve the arguments to the binary being run, its
378 working directory and the standard input, output and error file
379 descriptors. If @samp{@var{test-wrapper} env} will not work to run a
380 program with environment variables set, then @samp{test-wrapper-env}
381 must be set to a program that runs a newly built program with
382 environment variable assignments in effect, those assignments being
383 specified as @samp{@var{var}=@var{value}} before the name of the
384 program to be run. If multiple assignments to the same variable are
385 specified, the last assignment specified must take precedence.
386 Similarly, if @samp{@var{test-wrapper} env -i} will not work to run a
387 program with an environment completely empty of variables except those
388 directly assigned, then @samp{test-wrapper-env-only} must be set; its
389 use has the same syntax as @samp{test-wrapper-env}, the only
390 difference in its semantics being starting with an empty set of
391 environment variables rather than the ambient set.
394 @node Running make install
395 @appendixsec Installing the C Library
398 To install the library and its header files, and the Info files of the
399 manual, type @code{make install}. This will
400 build things, if necessary, before installing them; however, you should
401 still compile everything first. If you are installing @theglibc{} as your
402 primary C library, we recommend that you shut the system down to
403 single-user mode first, and reboot afterward. This minimizes the risk
404 of breaking things when the library changes out from underneath.
406 @samp{make install} will do the entire job of upgrading from a
407 previous installation of @theglibc{} version 2.x. There may sometimes
409 left behind from the previous installation, but those are generally
410 harmless. If you want to avoid leaving headers behind you can do
411 things in the following order.
413 You must first build the library (@samp{make}), optionally check it
414 (@samp{make check}), switch the include directories and then install
415 (@samp{make install}). The steps must be done in this order. Not moving
416 the directory before install will result in an unusable mixture of header
417 files from both libraries, but configuring, building, and checking the
418 library requires the ability to compile and run programs against the old
419 library. The new @file{/usr/include}, after switching the include
420 directories and before installing the library should contain the Linux
421 headers, but nothing else. If you do this, you will need to restore
422 any headers from libraries other than @theglibc{} yourself after installing the
425 You can install @theglibc{} somewhere other than where you configured
426 it to go by setting the @code{DESTDIR} GNU standard make variable on
427 the command line for @samp{make install}. The value of this variable
428 is prepended to all the paths for installation. This is useful when
429 setting up a chroot environment or preparing a binary distribution.
430 The directory should be specified with an absolute file name. Installing
431 with the @code{prefix} and @code{exec_prefix} GNU standard make variables
432 set is not supported.
434 @Theglibc{} includes a daemon called @code{nscd}, which you
435 may or may not want to run. @code{nscd} caches name service lookups; it
436 can dramatically improve performance with NIS+, and may help with DNS as
439 One auxiliary program, @file{/usr/libexec/pt_chown}, is installed setuid
440 @code{root} if the @samp{--enable-pt_chown} configuration option is used.
441 This program is invoked by the @code{grantpt} function; it sets the
442 permissions on a pseudoterminal so it can be used by the calling process.
443 If you are using a Linux kernel with the @code{devpts} filesystem enabled
444 and mounted at @file{/dev/pts}, you don't need this program.
446 After installation you should configure the timezone and install locales
447 for your system. The time zone configuration ensures that your system
448 time matches the time for your current timezone. The locales ensure that
449 the display of information on your system matches the expectations of
450 your language and geographic region.
452 @Theglibc{} is able to use two kinds of localization information sources, the
453 first is a locale database named @file{locale-archive} which is generally
454 installed as @file{/usr/lib/locale/locale-archive}. The locale archive has the
455 benefit of taking up less space and being very fast to load, but only if you
456 plan to install sixty or more locales. If you plan to install one or two
457 locales you can instead install individual locales into their self-named
458 directories e.g.@: @file{/usr/lib/locale/en_US.utf8}. For example to install
459 the German locale using the character set for UTF-8 with name @code{de_DE} into
460 the locale archive issue the command @samp{localedef -i de_DE -f UTF-8 de_DE},
461 and to install just the one locale issue the command @samp{localedef
462 --no-archive -i de_DE -f UTF-8 de_DE}. To configure all locales that are
463 supported by @theglibc{}, you can issue from your build directory the command
464 @samp{make localedata/install-locales} to install all locales into the locale
465 archive or @samp{make localedata/install-locale-files} to install all locales
466 as files in the default configured locale installation directory (derived from
467 @samp{--prefix} or @code{--localedir}). To install into an alternative system
468 root use @samp{DESTDIR} e.g.@: @samp{make localedata/install-locale-files
469 DESTDIR=/opt/glibc}, but note that this does not change the configured prefix.
471 To configure the locally used timezone, set the @code{TZ} environment
472 variable. The script @code{tzselect} helps you to select the right value.
473 As an example, for Germany, @code{tzselect} would tell you to use
474 @samp{TZ='Europe/Berlin'}. For a system wide installation (the given
475 paths are for an installation with @samp{--prefix=/usr}), link the
476 timezone file which is in @file{/usr/share/zoneinfo} to the file
477 @file{/etc/localtime}. For Germany, you might execute @samp{ln -s
478 /usr/share/zoneinfo/Europe/Berlin /etc/localtime}.
480 @node Tools for Compilation
481 @appendixsec Recommended Tools for Compilation
482 @cindex installation tools
483 @cindex tools, for installing library
485 We recommend installing the following GNU tools before attempting to
490 GNU @code{make} 4.0 or newer
492 As of relase time, GNU @code{make} 4.2.1 is the newest verified to work
493 to build @theglibc{}.
498 GCC 6.2 or higher is required. In general it is recommended to use
499 the newest version of the compiler that is known to work for building
500 @theglibc{}, as newer compilers usually produce better code. As of
501 release time, GCC 9.2.1 is the newest compiler verified to work to build
504 For PowerPC 64-bits little-endian (powerpc64le), a GCC version with support
505 for @option{-mno-gnu-attribute}, @option{-mabi=ieeelongdouble}, and
506 @option{-mabi=ibmlondouble} is required. Likewise, the compiler must also
507 support passing @option{-mlong-double-128} with the preceding options. As
508 of release, this implies GCC 7.4 and newer (excepting GCC 7.5.0, see GCC
509 PR94200). These additional features are required for building the GNU C
510 Library with support for IEEE long double.
512 @c powerpc64le performs an autoconf test to verify the compiler compiles with
513 @c commands like "$CC -c foo.c -mabi=ibmlongdouble -mlong-double-128".
515 For multi-arch support it is recommended to use a GCC which has been built with
516 support for GNU indirect functions. This ensures that correct debugging
517 information is generated for functions selected by IFUNC resolvers. This
518 support can either be enabled by configuring GCC with
519 @samp{--enable-gnu-indirect-function}, or by enabling it by default by setting
520 @samp{default_gnu_indirect_function} variable for a particular architecture in
521 the GCC source file @file{gcc/config.gcc}.
523 You can use whatever compiler you like to compile programs that use
526 Check the FAQ for any special compiler issues on particular platforms.
529 GNU @code{binutils} 2.25 or later
531 You must use GNU @code{binutils} (as and ld) to build @theglibc{}.
532 No other assembler or linker has the necessary functionality at the
533 moment. As of release time, GNU @code{binutils} 2.32 is the newest
534 verified to work to build @theglibc{}.
536 For PowerPC 64-bits little-endian (powerpc64le), @command{objcopy} is required
537 to support @option{--update-section}. This option requires binutils 2.26 or
541 GNU @code{texinfo} 4.7 or later
543 To correctly translate and install the Texinfo documentation you need
544 this version of the @code{texinfo} package. Earlier versions do not
545 understand all the tags used in the document, and the installation
546 mechanism for the info files is not present or works differently.
547 As of release time, @code{texinfo} 6.6 is the newest verified to work
548 to build @theglibc{}.
551 GNU @code{awk} 3.1.2, or higher
553 @code{awk} is used in several places to generate files.
554 Some @code{gawk} extensions are used, including the @code{asorti}
555 function, which was introduced in version 3.1.2 of @code{gawk}.
556 As of release time, @code{gawk} version 5.0.1 is the newest verified
557 to work to build @theglibc{}.
560 GNU @code{bison} 2.7 or later
562 @code{bison} is used to generate the @code{yacc} parser code in the @file{intl}
563 subdirectory. As of release time, @code{bison} version 3.4.1 is the newest
564 verified to work to build @theglibc{}.
569 Perl is not required, but if present it is used in some tests and the
570 @code{mtrace} program, to build the @glibcadj{} manual. As of release
571 time @code{perl} version 5.30.1 is the newest verified to work to
575 GNU @code{sed} 3.02 or newer
577 @code{Sed} is used in several places to generate files. Most scripts work
578 with any version of @code{sed}. As of release time, @code{sed} version
579 4.5 is the newest verified to work to build @theglibc{}.
584 Python is required to build @theglibc{}. As of release time, Python
585 3.7.4 is the newest verified to work for building and testing
590 The pretty printer tests drive GDB through test programs and compare
591 its output to the printers'. PExpect is used to capture the output of
592 GDB, and should be compatible with the Python version in your system.
593 As of release time PExpect 4.3 is the newest verified to work to test
597 GDB 7.8 or later with support for Python 2.7/3.4 or later
599 GDB itself needs to be configured with Python support in order to use
600 the pretty printers. Notice that your system having Python available
601 doesn't imply that GDB supports it, nor that your system's Python and
602 GDB's have the same version. As of release time GNU @code{debugger}
603 8.3 is the newest verified to work to test the pretty printers.
605 Unless Python, PExpect and GDB with Python support are present, the
606 printer tests will report themselves as @code{UNSUPPORTED}. Notice
607 that some of the printer tests require @theglibc{} to be compiled with
612 If you change any of the @file{configure.ac} files you will also need
616 GNU @code{autoconf} 2.69 (exactly)
620 and if you change any of the message translation files you will need
624 GNU @code{gettext} 0.10.36 or later
626 As of release time, GNU @code{gettext} version 0.19.8.1 is the newest
627 version verified to work to build @theglibc{}.
632 You may also need these packages if you upgrade your source tree using
633 patches, although we try to avoid this.
636 @appendixsec Specific advice for @gnulinuxsystems{}
637 @cindex kernel header files
639 If you are installing @theglibc{} on @gnulinuxsystems{}, you need to have
640 the header files from a 3.2 or newer kernel around for reference.
641 (For the ia64 architecture, you need version 3.2.18 or newer because this
642 is the first version with support for the @code{accept4} system call.)
643 These headers must be installed using @samp{make headers_install}; the
644 headers present in the kernel source directory are not suitable for
645 direct use by @theglibc{}. You do not need to use that kernel, just have
646 its headers installed where @theglibc{} can access them, referred to here as
647 @var{install-directory}. The easiest way to do this is to unpack it
648 in a directory such as @file{/usr/src/linux-@var{version}}. In that
649 directory, run @samp{make headers_install
650 INSTALL_HDR_PATH=@var{install-directory}}. Finally, configure @theglibc{}
651 with the option @samp{--with-headers=@var{install-directory}/include}.
652 Use the most recent kernel you can get your hands on. (If you are
653 cross-compiling @theglibc{}, you need to specify
654 @samp{ARCH=@var{architecture}} in the @samp{make headers_install}
655 command, where @var{architecture} is the architecture name used by the
656 Linux kernel, such as @samp{x86} or @samp{powerpc}.)
658 After installing @theglibc{}, you may need to remove or rename
659 directories such as @file{/usr/include/linux} and
660 @file{/usr/include/asm}, and replace them with copies of directories
661 such as @file{linux} and @file{asm} from
662 @file{@var{install-directory}/include}. All directories present in
663 @file{@var{install-directory}/include} should be copied, except that
664 @theglibc{} provides its own version of @file{/usr/include/scsi}; the
665 files provided by the kernel should be copied without replacing those
666 provided by @theglibc{}. The @file{linux}, @file{asm} and
667 @file{asm-generic} directories are required to compile programs using
668 @theglibc{}; the other directories describe interfaces to the kernel but
669 are not required if not compiling programs using those interfaces.
670 You do not need to copy kernel headers if you did not specify an
671 alternate kernel header source using @samp{--with-headers}.
673 The Filesystem Hierarchy Standard for @gnulinuxsystems{} expects some
674 components of the @glibcadj{} installation to be in
675 @file{/lib} and some in @file{/usr/lib}. This is handled automatically
676 if you configure @theglibc{} with @samp{--prefix=/usr}. If you set some other
677 prefix or allow it to default to @file{/usr/local}, then all the
678 components are installed there.
681 @appendixsec Reporting Bugs
682 @cindex reporting bugs
683 @cindex bugs, reporting
685 There are probably bugs in @theglibc{}. There are certainly
686 errors and omissions in this manual. If you report them, they will get
687 fixed. If you don't, no one will ever know about them and they will
688 remain unfixed for all eternity, if not longer.
690 It is a good idea to verify that the problem has not already been
691 reported. Bugs are documented in two places: The file @file{BUGS}
692 describes a number of well known bugs and the central @glibcadj{}
693 bug tracking system has a
695 @url{https://sourceware.org/bugzilla/}. The WWW
696 interface gives you access to open and closed reports. A closed report
697 normally includes a patch or a hint on solving the problem.
699 To report a bug, first you must find it. With any luck, this will be the
700 hard part. Once you've found a bug, make sure it's really a bug. A
701 good way to do this is to see if @theglibc{} behaves the same way
702 some other C library does. If so, probably you are wrong and the
703 libraries are right (but not necessarily). If not, one of the libraries
704 is probably wrong. It might not be @theglibc{}. Many historical
705 Unix C libraries permit things that we don't, such as closing a file
708 If you think you have found some way in which @theglibc{} does not
709 conform to the ISO and POSIX standards (@pxref{Standards and
710 Portability}), that is definitely a bug. Report it!
712 Once you're sure you've found a bug, try to narrow it down to the
713 smallest test case that reproduces the problem. In the case of a C
714 library, you really only need to narrow it down to one library
715 function call, if possible. This should not be too difficult.
717 The final step when you have a simple test case is to report the bug.
718 Do this at @value{REPORT_BUGS_TO}.
720 If you are not sure how a function should behave, and this manual
721 doesn't tell you, that's a bug in the manual. Report that too! If the
722 function's behavior disagrees with the manual, then either the library
723 or the manual has a bug, so report the disagreement. If you find any
724 errors or omissions in this manual, please report them to the
725 bug database. If you refer to specific
726 sections of the manual, please include the section names for easier