The new <pkg>-show-depends simply outputs the list of dependencies for
the given package.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The recent change to use config.guess to figure out GNU_HOST_NAME
broken the x86x86fix detection for 32bit hosts, as config.guess
returns something like i686-pc-linux-gnu rather than something with
*-unknown-linux-* like we used to use.
Fix it by simply looking for "linux" instead.
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
There is a build problem with netsnmp on external toolchain that isn't
easy to fix: libtool strips the --sysroot that we are passing it,
which make the link step fail.
In preparation for the release, just make sure this package isn't
visible to external toolchain users. Of course, on the long run, we
should fix this.
We also disable the build of the Quagga SNMP support because it
selects netsnmp.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
External toolchain builds absolutely require the --sysroot option to
be passed at link time.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Without CFLAGS, the build breaks because --sysroot is missing at link
time, and this option is absolutely mandatory for external toolchain
builds.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Since lvm2 uses autoconf, there's no point in overriding CC and other
variables at build time. This causes build breakage as CFLAGS aren't
used at link time, therefore --sysroot lacks, which breaks the build
with external toolchains.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
mutt has been upgraded, but the mutt-1.5.16-makedoc-hostcc.patch
wasn't upgraded accordingly, causing build failures in the
documentation. Therefore, we add a new patch
mutt-1.5.17+20080114-nodoc.patch that disables the construction of the
documentation.
As this patch modifies Makefile.am, we need to autoreconf the
package. The simplest and cleanest way to do this is to convert the
package to the autotools infrastructure.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Since we no longer need to build pango on the host, cairo on the host
is no longer needed.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
We needed to build pango for the host in order to compile
pango-querymodules for the host, which was used to produce
/etc/pango/pango.modules. Unfortunately:
* This produces an incorrect /etc/pango/pango.modules (no modules
detected in my case), probably because the host pango-querymodules
was looking at host pango modules
* This requires to build pango for the host, which requires to build
cairo for the host, which requires to build X11 for the host.
To make things work and remove the dependency between pango and
host-pango, we introduce a S25pango script that creates
/etc/pango/pango.modules on startup if it doesn't exist, just as we do
with libgtk2 for /etc/gtk-2.0/gdk-pixbuf.loaders.
Since host-pango is no longer needed, we remove all definitions
related to it.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This patch adds a patch to NTP to fix the following build failure,
using a workaround patch found on the Fedora bugzilla:
ntp_loopfilter.c: In function 'local_clock':
ntp_loopfilter.c:571: error: 'MOD_NANO' undeclared (first use in this function)
ntp_loopfilter.c:571: error: (Each undeclared identifier is reported only once
ntp_loopfilter.c:571: error: for each function it appears in.)
ntp_loopfilter.c: In function 'loop_config':
ntp_loopfilter.c:896: error: 'MOD_NANO' undeclared (first use in this function)
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This package uses several pmap_*() functions in the C library, that
are only available if the C library has RPC support.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
A C library compiled with locale support already has an iconv()
implementation. In that case, libiconv is useless. Therefore, disallow
the selection of libiconv when the configuration says that the
toolchain has locale support.
This fixes build issues similar to the one we had with gettext:
libiconv headers are installed, but -liconv is not added because the
./configure script detects that the C library supports iconv.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This allows to make sure that an user, or a randpackageconfig, will
not select the gettext or libintl package when using an external
toolchain based on glibc/eglibc.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
ltrace failed to build because of missing arguments to gcc to find the
header files. This is due to the fact that the existing ltrace.mk was
setting CC and LD at build time to incorrect values. Keeping the
values set at configure time is just the right thing to do.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The patch had an incorrect name, and the libtool patch was applied
while it shouldn't, and the package wasn't autoreconfed while the
patch changes some Makefile.am files.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
On a x86-64 host, libcap tried to install things in
$(STAGING_DIR)/lib64 and $(TARGET_DIR)/lib64. Therefore, pass lib= and
prefix=, as required by the strange build system used by libcap.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Fixes the ULONG_MAX definition problem:
In file included from cache.c:43:
../include/netlink-local.h: In function '__str2type':../include/netlink-local.h:218: error: 'ULONG_MAX' undeclared (first use in this function)
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
netplug uses nested functions, and gcc generates a warning that
is turned into an error:
cc1: warnings being treated as errors
if_info.c: In function 'ifsm_scriptdone':
if_info.c:289: error: generating trampoline in object (requires executable stack)
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
The less package already selects BR2_PACKAGE_CURSES, but it fails to
depend on it in the makefile. Therefore, there are cases where less
gets built before curses, and the build fails:
checking for working terminal libraries... Cannot find terminal libraries - configure failed
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
When OpenSSL is not enabled, hostapd relies on an internal
library. However, this library must be enabled, otherwise the
compilation of hostapd fails with:
/home/test/ctng-arm-eglibc/bin/../lib/gcc/arm-unknown-linux-gnueabi/4.3.4/../../../../arm-unknown-linux-gnueabi/bin/ld: cannot find -ltommath
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Version 4.1.5 fails to build due to the infamous getline() conflict
issue. 4.2.1 doesn't have this problem.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Before 2010.02, VLC didn't build, because it didn't pass the
appropriate --disable- options according to the libraries available in
Buildroot. Now, 2010.05 is going to be released, and no one bothered
to fix VLC. Therefore, let's mark VLC as broken.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
linux-fusion is composed of a kernel module, and linux-fusion.mk uses
several variable definitions that only exist when Buildroot compiles a
kernel.
We also add the same dependency to BR2_PACKAGE_DIRECTFB_MULTI, which
relies on linux-fusion.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Some python versions print x.y rather than x.y.z as version number, so
make regexp more robust.
Reported-by: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
libgcrypt depends on libgpg-error, but it needs to know where the
gpg-error-config utility is.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
hal integrates its own version of intltool tools, but they require
libxml-parser-perl to be installed on the host.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
For some unknown reason, gawk.mk was overriding CC to $(TARGET_CC) at
build time, while it had already been set at configure time to
"$(TARGET_CC) $(TARGET_CFLAGS)" which is already the good value.
Setting it to just $(TARGET_CC) breaks the compilation with external
toolchains because we must pass the --sysroot option. But anyway, this
was incorrect as we were loosing some CFLAGS set by Buildroot.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
When neither CLI nor CGI is selected, PHP's configure script fails
with:
checking whether to build CGI binary... configure: error: No SAPIs selected.
Of couse, the help text of the options says that at least one of them
should be selected, but when doing testing with randpackageconfig,
noone is reading these help texts.
Therefore, based on the suggestion of Yann E. Morin, modify the
Config.in organization so that at least one of the option is selected.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
We need to pass CFLAGS and LDFLAGS, and to rename the patch so that it
actually gets applied.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Pass CFLAGS and LDFLAGS so that when building squashfs, it finds the
zlib installed in $(HOST_DIR).
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
When xlib_libX11 was bumped to 1.3.2 in commit
1d956c9190, the keysymdef patch was
dropped. However, this patch is still needed in order to be able to
tell ./configure where the keysymdef header file is installed.
The patch has been updated for 1.3.2.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
When building libxcb, the variable XCBPROTO_XCBPYTHONDIR must point to
the location where the Python modules needed to run the c_client.py
program are installed. The path
$(STAGING_DIR)/usr/lib/python2.6/site-packages was hardcoded. However,
it doesn't work when the version of Python installed on the host is
Python 2.5.
Therefore, add a little bit of magic to compute the host Python
version.
We also verify that Python is available on the host, as we don't build
it in Buildroot.
Fixes bug #1531.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The MTD utils require MTD headers in the toolchain, and in several
external toolchains, they are not present. In order to make the build
work by default, let's disable the MTD Busybox applets in our default
configuration.
Fixes bug #1669
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Closes#849, #1135
customize.mk used to copy files with a wrong command, resulting in one
directory level in excess. For example, source/etc/myfile would have
produced /etc/etc/myfile in the target filesystem instead of /etc/myfile.
Signed-off-by: Luca Ceresoli <luca@lucaceresoli.net>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
When the selected C library is glibc, the C library shouldn't be
mentionned in REAL_GNU_TARGET_NAME. In other words:
arm-unknown-linux-uclibcgnueabi must be used for uClibc
arm-unknown-linux-gnueabi must be used for glibc
This fixes the build of GDB on the target, as reported by Quotient
Remainder <quotientvremainder@gmail.com>.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
HOST_SED_BINARY is used 2 times during startup, and is relatively
expensive to run, so make it a simple expanded variable (:=) rather
than a recursively expanded one (=).
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
As noticed by Thomas, we call the UPPERCASE macro a lot, and it slows down
startup quite a bit.
Optimize it by implementing it in make, rather than forking a shell + tr.
The implementation is heavily based on the 'up' macro from gmsl
(http://gmsl.sf.net)
With this in place, startup time is ~5 times lower.
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Instead of assuming that mkfontdir and mkfontscale are available on
the development system, use the one installed in $(HOST_DIR).
The FCCACHE variable is removed, because, for some reason, the
installation process of xfonts disables the call to fc-cache when a
DESTDIR is passed. See the definition of XORG_FONT_FCCACHE in
aclocal.m4 in font packages:
AC_DEFUN([XORG_FONT_FCCACHE],[
AC_PATH_PROG(FCCACHE, fc-cache)
FCCACHE_WARN='echo "** Warning: fonts.cache not built" ; echo "** Generate this file manually on host system using fc-cache"'
if test x"$FCCACHE" = x ; then
RUN_FCCACHE="${FCCACHE_WARN}"
else
RUN_FCCACHE='@(if test -z "$(DESTDIR)"; then echo $(FCCACHE) $(fontdir); $(FCCACHE) $(fontdir); else'
RUN_FCCACHE="${RUN_FCCACHE} ${FCCACHE_WARN} ; fi)"
fi
AC_SUBST([RUN_FCCACHE])
])
In addition to this, we update the font dependencies to include
font-utils for the host, mkfontdir for the host, mkfontscale for the
host and bdftopcf for the host.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Now that xapp_bdftopcf is enabled on the host, we also need its
dependencies to be available for the host. The dependency of
host-xapp_bdftopcf is host-xlib_libXfont, which itself depends on a
bunch of other packages. Some of them were already available for the
host, some not (xfont_encodings, xlib_xtrans, xproto_fontcacheproto,
xproto_fontsproto) and are therefore added by this patch.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Compiling xfonts require having bdftopcf installed on the
host. Therefore, enable host support for this package.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
When compiling xfonts, host utilities like bdftopcf or ucs2any, part
of xfont_font-util are needed. Until now, xfont_font-util was
installing ugly symbolic links, assuming that these tools were
available on the development machine.
This patch enables the host package for xfont_font-util, removes the
post-install.sh script, and keeps only the useful part of this script
in a post install hook.
host-xfont_font-util will be used later by xfont packages.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
In order to solve issues of libtool trying to link target components
against host libraries, it seems that specifying -L$(STAGING_DIR)/lib
and -L$(STAGING_DIR)/usr/lib works.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
In c1b6242fdc, we added
$(HOST_DIR)/usr/lib to LD_LIBRARY_PATH when building target packages,
because the build of target packages sometimes require host tools
installed in $(HOST_DIR)/usr/bin which themselves require host
libaries installed in $(HOST_DIR)/usr/lib.
Unfortunately, this solution didn't work, as libtool then tried to
link target binaries against host libraries. So $(HOST_DIR)/usr/lib
got removed from LD_LIBRARY_PATH in
0d1830b07d.
However, this meant that we went back to the previous situation, in
which host tools used during compilation of target components might
require host libraries. An example :
make[2]: Entering directory `/home/test/buildroot/output.ctng-arm-eglibc-2010-05-04-12-41-00/build/xfont_font-adobe-100dpi-1.0.1'
/home/test/buildroot/output.ctng-arm-eglibc-2010-05-04-12-41-00/host/usr/bin/mkfontdir /home/test/buildroot/output.ctng-arm-eglibc-2010-05-04-12-41-00/target/usr/share/fonts/X11/100dpi
/home/test/buildroot/output.ctng-arm-eglibc-2010-05-04-12-41-00/host/usr/bin/mkfontscale: error while loading shared libraries: libfontenc.so.1: cannot open shared object file: No such file or directory
Therefore, we try another solution: make sure that host binaries are
linked with an -rpath option, so that $(HOST_DIR)/usr/lib doesn't need
to be in LD_LIBRARY_PATH for them to find their libraries.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
On my target rootfs it is installed documentation for:
atk, cairo, gdk, gdk-pixbuf, gio, glib, gobject, gtk, pango
It ocupies 34M of space. Configure option --disable-gtk-doc doesn't
prevent installation of it. So we should remove them manually if
documentation is disabled.
[Peter: also remove manual rm's from hal/libxml2/libglade]
Signed-off-by: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Closes#1525
Kconfig selects are not recursive, so ensure we select the dependencies
of our dependencies as well.
Cleanup target installation:
- Remove tasks handled by the generic post-build cleanup
- Hal doesn't install etc/rc.d, so don't try to delete it
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Fix a few typos/mistakes introduced in commit ea448fe:
- bzip2 soname is not the same as package version
- It is PKG_INSTALL_STAGING, not PKG_STAGING_INSTALL
- Typo: BZIP2_INSTALL_TARGET_CMDS, not B2IP_INSTALL_TARGET_CMDS
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Some packages like icu requires to be compiled against the host system
first to be able to compile against the target. This is due to the
usage of self generated binaries by the package to build itself. When
the generated tools also depends on generated libraries it is required
to add the path to these libraries in the library path
(LD_LIBRARY_PATH) especially for the configure step.
Adding $(HOST_DIR)/usr/lib to LD_LIBRARY_PATH for target compilation
might break the link step by mixing host libraries and target
binaries.
Signed-off-by: Lionel Landwerlin <llandwerlin@gmail.com>
Acked-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Also convert to generic package infrastructure
Signed-off-by: Lionel Landwerlin <llandwerlin@gmail.com>
Acked-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
It got broken by the introduction of the new autotools infrastructure
based on the generic infrastructure.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Instead of asking the user about the GNU target suffix, just compute
it automatically from the other configuration options.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
In both internal and external toolchain cases, KERNEL_CROSS was
defined to *exactly* the same value as TARGET_CROSS. It isn't modified
anywhere, and is just used by kernel compilation and pcmcia
compilation.
Therefore, get rid of KERNEL_CROSS and use TARGET_CROSS instead.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
The definition of TOOLCHAIN_DIR is the same regardless of whether
external or internal toolchains are used. Moreover, move its
definition together with all the other *_DIR definitions.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
When FOO_LIBTOOL_PATCH is defined, assume that the same value should
be used for HOST_FOO_LIBTOOL_PATCH.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
OpenSSL is not using the autotools as its build system. Therefore, we
must use the generic infrastructure instead of the autotools one.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Do not build against libgcrypt on the host, since we don't build
libgcrypt ourself, and it might fail even when libgcrypt is installed
on the host:
...configure...
checking for libgcrypt-config... /usr/bin/libgcrypt-config
Crypto extensions will be available.
...build...
gcc: /usr/lib/libgcrypt.so: No such file or directory
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
The autoreconf check was incorrect, missing a $ sign to properly
reference a package-specific variable. There was no visible effect
until now since :
* The existing syntax allowed to access the value defined in the
package specific .mk file, so when AUTORECONF was set to YES by a
package, it was working.
* The default value in Makefile.autotools.in was NO. In fact, when a
package .mkf file wasn't defining the AUTORECONF variable, the
Makefile.autotools.in test was testing the empty string against
'YES', which was false, leading to the AUTORECONF not being done,
which was the desired effect.
However, in a later patch, we intend to change this default value.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
install -D needs the destination filename, not just it's directory,
otherwise we end up with a pkgconfig file instead.
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
autoheader is used by autoreconf, therefor we need to give the right
path to this tools for host/target autoreconfiguration.
Signed-off-by: Lionel Landwerlin <llandwerlin@gmail.com>
Acked-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
