The 'systemctl disable' operation should not be done in postrm() because
the .service file does not exist anymore and the script will fail hence.
After moving this command into prerm(), postrm() becomes empty and can
be removed completely.
Signed-off-by: Enrico Scholz <enrico.scholz@sigma-chemnitz.de>
Acked-by: Otavio Salvador <otavio@ossystems.com.br>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
* Fix packaging issues for the examples package.
* Disable build for machines without neon, since cmake detection is broken.
Signed-off-by: Philip Balister <philip@balister.org>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
This version drops the usrp E100 utilities. Also, the recipe depends on
python-cheetah-native now.
Signed-off-by: Philip Balister <philip@balister.org>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
* use INC_PR
* fix RDEPENDS
* use normal PV
* move checksums from .inc to .bb
* upgrade, because old version fails to build with a lot of errors like this:
compat-wireless-3.5-rc5-1/include/linux/compat-3.5.h:102:8: error: redefinition of 'struct tc_fq_codel_xstats'
* 3.5.4-1, 3.6.8-1 can failed when kernel option is missing:
| compat-wireless-all/3.5.4-r1.0/compat-wireless-3.5.4-1/drivers/net/ethernet/broadcom/b44.c:2342:2:
error: implicit declaration of function 'ssb_pcihost_register' [-Werror=implicit-function-declaration]
| cc1: some warnings being treated as errors
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Reference: http://www.spinics.net/lists/autofs/msg00139.html
Loading autofs module is #ifdef'ed in the source, so
there is no need to check for /proc (which is only used
to load module) or modprobe. Both modprobe and /proc
are always in the fixed location so there's no need to
check for these to start with.
Signed-off-by: yanjun.zhu <yanjun.zhu@windriver.com>
Signed-off-by: Joe MacDonald <joe.macdonald@windriver.com>
* fails without:
| temacs: error while loading shared libraries: libharfbuzz.so.0: cannot open shared object file: No such file or directory
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
* pcre_info was removed in newer pcre and abiword now fails to build
/usr/lib/libgoffice-0.8.so: undefined reference to pcre_info
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
An argument was missed in the manual setuptools configuration which meant that
the tool was installed into
/data/poky-master/tmp/sysroots/x86_64-linux/usr/bin/mako-render. Using the
setuptools class reduces duplication and puts the tool into /usr/bin where it
belongs.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Add obsolete_automake_macros.patch that replaces automake macros
no longer supported by automake-1.13 with modern constructs.
Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
* Removed systemd support from meta-networking recipe
and added bbappend to add its support.
Signed-off-by: Noor Ahsan <noor_ahsan@mentor.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
* Autofs was being inherited from systemd. This class is only
present in meta-systemd. So bitbake was giving parse error
about systemd class. So removed systemd from inherit and
removed SYSTEMD_PACKAGES and SYSTEMD_SERVICE.
Signed-off-by: Noor Ahsan <noor_ahsan@mentor.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
gst123 is useful for testing as it simply uses PlayBin, but can display the
pipeline being used for QA purposes.
Signed-off-by: Ross Burton <ross.burton@intel.com>
0.10.x is the stable series, 0.11.x is the development series that some packages
(specifically dLeyna) require.
Signed-off-by: Ross Burton <ross.burton@intel.com>
0.18.x is the stable series, 0.19.x is the development series that some packages
(specifically dLeyna) require.
Signed-off-by: Ross Burton <ross.burton@intel.com>
0.12.x is the stable series, 0.13.x is the development series that some packages
(specifically dLeyna) require.
Signed-off-by: Ross Burton <ross.burton@intel.com>
* Convert --enable-neon selection to use neon tune feature.
* Drop neon patch since fftw now has direct neon support.
* Remove SRC_URI from each bb file and use only SRC_URI from fftw.inc file.
The old recipe selected neon for all armv7 machines, which was not
entirely correct.
Signed-off-by: Philip Balister <philip@balister.org>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
This essentially fixes two problems
1. We were ignoring systemd for multilibs now we dont
2. We were replacing '-systemd' string in PN which is
a problem whenre -systemd appears more than once
e.g. lib32-systemd-units-system it was removing both
essentially we should only chop the end of PN if it
ends with '-systemd'
Signed-off-by: Khem Raj <raj.khem@gmail.com>
we were carrying a wrong patch to fix rootlibdir
to adapt systemd to OE's notion of base_libdir
but thats not right and the units dont get initialised
properly when we have base_libdir which is not /lib
This patch fixes the use of base_libdir where
it should have been catering to notions of systemd/udev
for putting unit files and init scripts in /lib
and not in base_libdir
It also ignores the lib32-* packages so that systemd
build works for multilib enabled systems.
Adapt to nativesdk renaming
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Enrico Scholz
Martin Jansa
Andrei Gherzan
Andrei Gherzan
Philip Balister
yanjun.zhu
Jack Mitchell
Ross Burton
Andreas Müller
Marcin Juszkiewicz
Khem Raj
Marko Lindqvist
Noor Ahsan
Roy.Li