* Use cmake as build system. Projects also using cmake and depending on fltk
would not find all required components. Note that FindFLTK.cmake is part of
cmake.
* Add a native recipe and split out common - we need fluid (Fast Light User
Interface Designer) as working binary
Signed-off-by: Andreas Müller <schnitzeltony@googlemail.com>
Subpackage python3-${PN} is not right for multilib and it will be
extended to such as lib32-python3-lib32-opencv.
Replace PN with BPN to fix it.
Signed-off-by: Kai Kang <kai.kang@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
The current setting of S is not right for multilib. Remove the setting
and use the default value.
And library install directory is not right for multilib either. It fails
with [installed-vs-shipped] QA error:
| ERROR: gflags-2.2.0-r0 do_package: QA Issue: gflags: Files/directories
| were installed but not shipped in any package:
| /usr/lib/libgflags.so
| /usr/lib/libgflags_nothreads.so.2.2
Set cmake variable LIB_INSTALL_DIR with ${baselib} to fix the issue.
Signed-off-by: Kai Kang <kai.kang@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Additional patches are upstream to fix build with gcc6 and clang
Add systemd unit file
Create missing symlink for dev shared object
Signed-off-by: Khem Raj <raj.khem@gmail.com>
* based on discussion in pndeprecated thread:
https://patchwork.openembedded.org/patch/137573/
update the messages to warn possible users that the
recipe will be removed before the end of the next development
cycle (before Yocto 2.4 is released).
* updated with:
sed -i 's/^\(PNBLACKLIST.*".*\)"/\1 - the recipe will be removed on 2017-09-01 unless the issue is fixed"/g' `git grep PNBLACKLIST | sed 's/:.*//g' | sort -u | xargs`
* then noticed couple recipes being blacklisted only based on
DISTRO_FEATURES, so removed those:
meta-networking/recipes-support/lksctp-tools/lksctp-tools_1.0.17.bb
meta-oe/recipes-connectivity/bluez/bluez-hcidump_2.5.bb
meta-oe/recipes-connectivity/bluez/bluez4_4.101.bb
meta-oe/recipes-connectivity/bluez/gst-plugin-bluetooth_4.101.bb
meta-oe/recipes-navigation/foxtrotgps/foxtrotgps_1.1.1.bb
meta-oe/recipes-navigation/gypsy/gypsy.inc
meta-oe/recipes-navigation/navit/navit.inc
meta-oe/recipes-support/opensync/libsyncml_0.5.4.bb
* if it isn't fixed by this date, it's fair game to be removed
whenever someone gets around to i
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
* fixes:
No GNU_HASH in the elf binary:
'/tmp/work/armv5te-oe-linux-gnueabi/pngcheck/2.3.0-r0/packages-split/pngcheck/usr/bin/pngcheck'
[ldflags]
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
The recipe opencv_3.2.bb inherits to either distutils-base or
distutils3-base classes or both which do add appropriate
build dependency on python or python3 respectively.
There's no need to have python in DEPENDS explicitly.
It can even make things needlessly complicated in case a user
wants to build python3-only images.
The redundant build dependency on python is removed from DEPENDS.
Signed-off-by: Dmitry Rozhkov <dmitry.rozhkov@linux.intel.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
it has a tclsh need for executing a script during build
Fixes
| ./gen-rules.tcl
| /usr/bin/env: ‘tclsh’: No such file or directory
| make: *** [Makefile:26: 40-usb_modeswitch.rules] Error 127
| WARNING: /mnt/b/build/tmp-musl/work/all-oe-linux/usb-modeswitch-data/20170205-r0/temp/run.do_compile.30695:1 exit 1 from 'exit 1'
Signed-off-by: Khem Raj <raj.khem@gmail.com>
* 5.1.14 fails with default linux-yocto version:
CC [M] qemux86-oe-linux/vboxguestdrivers/5.1.14-r0/vbox_module/vboxsf/lnkops.o
qemux86-oe-linux/vboxguestdrivers/5.1.14-r0/vbox_module/vboxsf/lnkops.c:93:23: error: 'generic_readlink' undeclared here (not in a function)
.readlink = generic_readlink,
^~~~~~~~~~~~~~~~
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
* otherwise we get the failure from do_install with useless log:
DEBUG: Executing shell function do_install
install: cannot stat 'vboxsf.ko': No such file or directory
WARNING: exit code 1 from a shell command.
* instead of the real issue like 5.1.14 currently has with default linux-yocto
version:
CC [M] qemux86-oe-linux/vboxguestdrivers/5.1.14-r0/vbox_module/vboxsf/lnkops.o
qemux86-oe-linux/vboxguestdrivers/5.1.14-r0/vbox_module/vboxsf/lnkops.c:93:23: error: 'generic_readlink' undeclared here (not in a function)
.readlink = generic_readlink,
^~~~~~~~~~~~~~~~
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
lvm2-native doesn't have packages, so the RRECOMMENDS must be limited
to the target case. This fixes:
ERROR: Nothing RPROVIDES 'lvm2-native-scripts-native' (but virtual:native:.../meta-openembedded/meta-oe/recipes-support/lvm2/lvm2_2.02.166.bb RDEPENDS on or otherwise requires it)
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Applications like kpartx and cryptsetup were broken by moving only
libdevicemapper itself into a separate package: as a result of that
change, lvm2 was not getting pulled into images anymore although
libdevicemapper depends on dmsetup and udev rules to be fully
functional.
For example, "kpartx -as" started to hang while waiting for the udev
rules to trigger, which is what creates the /dev/mapper/ entries for
the new partitions (see also
https://github.com/docker/docker/issues/22025#issuecomment-243943728).
Putting udev rules and dmsetup also into libdevicemapper is perhaps
counter-intuitive, but necessary to keep the package functioning. A
full lvm2 installation is guaranteed to pull them in, too, both
because of implicit library dependencies and (just to be sure) an
explicit RDEPENDS.
lvm2-native doesn't have packages, so this RDEPENDS must be limited to
the target case.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Rather than having the lvm2 package depend on bash, modify the init
scripts it may contain to use /bin/sh instead of /bin/bash.
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
With musl charset.alias is generated due to a gnu-config bug
Fixes
QA Issue: zile: Files/directories were installed but not shipped in any package:
/usr/lib
/usr/lib/charset.alias
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Fix the following error for qemuppc.
error: 'siginfo_t' does not name a type
Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
To fix:
file /usr/share/man/man1/eject.1 conflicts between attempted installs
of util-linux-doc-2.29.1-r0.core2_64 and eject-doc-2.1.5-r1.core2_64
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
The newly resurrected devmem2.c file besides having a new location, also
updates its header, producing different checksums, which would conflict
with any previously cached or mirrored instances. To avoid such conflicts,
use a different filename in fetch(). Rename it back to original name at
unpack() for devmem2-fixups-2.patch to succeed w/o modifications.
WARNING: devmem2-1.0-r7 do_fetch: Fetcher failure for URL: 'http://www.free-electrons.com/pub/mirror/devmem2.c'. Checksum mismatch!
File: '/OE/master/downloads/devmem2.c' has md5 checksum be12c0132a1ae118cbf5e79d98427c1d when e23f236e94be4c429aa1ceac0f01544b was expected
File: '/OE/master/downloads/devmem2.c' has sha256 checksum ec382c90af3ef2f49695ff14a4d6521e58ac482c4e29d6c9ebca8768f699c191 when 3b15515693bae1ebd14d914e46d388edfec2175829ea1576a7a0c8606ebbe639 was expected
If this change is expected (e.g. you have upgraded to a new version without updating the checksums) then you can use these lines within the recipe:
SRC_URI[md5sum] = "be12c0132a1ae118cbf5e79d98427c1d"
SRC_URI[sha256sum] = "ec382c90af3ef2f49695ff14a4d6521e58ac482c4e29d6c9ebca8768f699c191"
Otherwise you should retry the download and/or check with upstream to determine if the file has become corrupted or otherwise unexpectedly modified.
WARNING: devmem2-1.0-r7 do_fetch: Renaming /OE/sources/devmem2.c to /OE/sources/devmem2.c_bad-checksum_be12c0132a1ae118cbf5e79d98427c1d
Signed-off-by: Denys Dmytriyenko <denys@ti.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
8.0.0427 is currently used in Arch Linux so we know it has at least had
some testing.
Upstream has fixed the following CVEs since 8.0.0022:
* CVE-2016-1248 in 8.0.0056
* CVE-2017-5953 in 8.0.0322
* CVE-2017-6349 in 8.0.0377
* CVE-2017-6350 in 8.0.0378
See http://www.cvedetails.com/product/14270/VIM-VIM.html?vendor_id=8218
Patches are updated as src/configure.in was renamed upstream.
Signed-off-by: Paul Barker <pbarker@toganlabs.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Replace some "+=/=+" with "=" when setting BBCLASSEXTEND, they are
redundant and inconsistent with the same setting in other recipes.
Signed-off-by: Ming Liu <peter.x.liu@external.atlascopco.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
WARNING: lvm2-2.02.166-r0 do_package_qa: QA Issue: /etc/init.d/lvm2-monitor contained in package lvm2 requires /bin/bash, but no providers found in RDEPENDS_lvm2? [file-rdeps]
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
* some recipes might use xxd
* with stricter host binaries available in PATH:
http://lists.openembedded.org/pipermail/openembedded-core/2017-March/133886.html
we cannot use xxd from host and it's not so common to add it into
HOSTTOOL_NONFATAL
* there is also smaller solution for xxd -i (which is what I need):
https://github.com/gregkh/xxdi
but after creating xxdi-native recipe for this I've discovered that
perl on my host as well as perlnative doesn't provide File::Slurp
used by this, so adding BBCLASSEXTEND here is easier solution
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
The backtick as a special character causes a syntax error by the parser
in comment line:
Error performing backtick substitution in configuration file;
error='missing closing backtick (`) character',
filename='/etc/syslog-ng/syslog-ng.conf', line='55:1'
Replace it with single quote and bump the conf version to 3.8.
Signed-off-by: Yi Zhao <yi.zhao@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Update license to reflect new Apache 2.0 license
Fix build with clang
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
