This patch adds the cross package for the crash recipe. I've built and
tested crash-cross for ARM on x86_64 host only.
The way to pass GDB_CONF_FLAGS and --target to GDB build without
removing -m32 from CFLAGS is awkward but that is what I could come up
with, without changing the patch files included with the recipe or
breaking target and native builds. Anyone got any ideas on how to
better do this?
Thanks,
Goran
Signed-off-by: Goran Cengic <cengic@gmail.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
checking for sys/acl.h... no
| configure: error: *** ACL headers not found.
| NOTE: The following config.log files may provide further information.
add acl to DEPENDS
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
LICENSE checksum has changed due to OpenSSL copyright year being
bumped from 1998-2011 to 1998-2016.
Signed-off-by: Jonathan Liu <net147@gmail.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
libmodbus has two release streams:
- stable: which currently is 3.0.x and does only receive fixes
- development: 3.1.x which receives new features and according to upstream is also very stable
Some users require access to the newer features.
So a recipe for the development branch but leave the stable branch the default.
This is similar to what is done for the nginx recipe which also has two release streams.
Signed-off-by: Pascal Bach <pascal.bach@siemens.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
The SRC_URI is up again, no need to blacklist the recipe if we disable parallel make
The following error can ocurr in a race condition:
scan.l:38:18: fatal error: gram.h: No such file or directory
http://errors.yoctoproject.org/Errors/Details/56158/
That is, the scan.l file is parsed before the gram.h file is generated.
To avoid this, we clear the PARALLEL_MAKE variable.
Signed-off-by: Adrian Dudau <adrian.dudau@enea.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
* If llvm.inc is used for native and nativesdk class packages, the
resulting llvm-config is broken since "${base_libdir}" is prefixed
with the full patch to sysroot
* Replace "/lib" with "/${baselib}" instead of "${base_libdir}" as
"${base_libdir}" is the full path and "${baselib}" is the name of
the multilib directory.
Signed-off-by: Jacob Stiffler <j-stiffler@ti.com>
Acked-by: Denys Dmytriyenko <denys@ti.com>
Tested-by: Denys Dmytriyenko <denys@ti.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
add native and nativesdk extend, curl-native/nativesdk need them.
replace the hardcode /etc with ${sysconfdir}, /var with ${localstatedir}
Signed-off-by: Roy Li <rongqing.li@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Moreover, the messages are misleading, saying "Building for arch blah",
even when this recipe is not being built. It merely checks if the recipe
is cmaptible with the currently used architecture.
Signed-off-by: Denys Dmytriyenko <denys@ti.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Also misc recipe cleanup, take license text from tinyxml2.cpp source
file instead of using the entire readme.md, and build using cmake.
Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Without this default theme install , openbox is unable to start, and shows the
following message:
dragonboard-410c:/home/linaro# openbox
ObRender-Message: Unable to load the theme 'Clearlooks'
ObRender-Message: Falling back to the default theme 'Clearlooks'
ObRender-Message: Unable to load the theme 'Clearlooks'
Openbox-Message: Unable to load a theme.
As discussed in [1] , the solution is to make sure that openbox-theme-clearlooks
is always included.
[1] https://www.mail-archive.com/openembedded-devel@lists.openembedded.org/msg47153.html
Signed-off-by: Nicolas Dechesne <nicolas.dechesne@linaro.org>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
OpenCV wrongly assumes that the X86 variable would be set on Intel 32
bit architectures.
This patch adds the missing definition on those cases.
Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
opencv-3.1+gitAUTOINC+92387b1ef8: libopencv-videoio rdepends on
libgphotoport, but it isn't a build dependency, missing libgphoto2 in
DEPENDS or PACKAGECONFIG? [build-deps]
opencv-3.1+gitAUTOINC+92387b1ef8: libopencv-videoio rdepends on
libgphoto2, but it isn't a build dependency, missing libgphoto2 in
DEPENDS or PACKAGECONFIG? [build-deps]
opencv-3.1+gitAUTOINC+92387b1ef8: libopencv-dnn rdepends on protobuf,
but it isn't a build dependency, missing protobuf in DEPENDS or
PACKAGECONFIG? [build-deps]
Credit-to: Martin Jansa <martin.jansa@gmail.com>
Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Libdir was pointing to the wrong location
also fix for QA Issue: opencv.pc failed sanity test (tmpdir) in path
Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
This bug was introducing a loop dependency of opencv->opencv that was
confusing dpkg.
Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
No further updates were required, aside from changing the
checksums and renaming the file.
Signed-off-by: Ruben De Smet <ruben.de.smet@rubdos.be>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
The recipes that are using waf-samba were all
moved to meta-networking, so move the bbclass
and related cross-answers files as well.
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Any PACKAGECONFIG default assigned with ??= is lost by subsequent
assignments with +=, therefore lvm2 is currently building with the
readline PACKAGECONFIG disabled. Make that explicitly clear in the
recipe. Also minor formatting updates, to align with style guide.
Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
* configure doesn't have config option ot disable it and it's autodetected from sysroot
causing:
WARN: jq: jq rdepends on onig, but it isn't a build dependency?
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
PT Sans is an open source licensed font released in 2010 by ParaType.
The ParaTypeFFL-1.3 license file has been submitted to OE-core.
Signed-off-by: Ioan-Adrian Ratiu <adrian.ratiu@ni.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
* fixes following issue found with test-dependencies
WARN: libftdi: libftdi rdepends on libgcc, but it isn't a build dependency?
WARN: libftdi: libftdi rdepends on libstdc++, but it isn't a build dependency?
* when boost is found it was still building libftdipp1.so.2.2.0
* as shown in log.do_package:
DEBUG: calculating shlib requirements for libftdi
DEBUG: libftdi: Dependency libftdi1.so.2 requires package libftdi (used by files: /home/jenkins/oe/world/shr-core/tmp-glibc/work/i586-oe-linux/libftdi/1.2-r0/packages-split/libftdi/usr/lib/libftdipp1.so.2.2.0)
DEBUG: libftdi: Dependency libusb-1.0.so.0 requires package libusb1 (used by files: /home/jenkins/oe/world/shr-core/tmp-glibc/work/i586-oe-linux/libftdi/1.2-r0/packages-split/libftdi/usr/lib/libftdipp1.so.2.2.0)
DEBUG: libftdi: Dependency libstdc++.so.6 requires package libstdc++ (used by files: /home/jenkins/oe/world/shr-core/tmp-glibc/work/i586-oe-linux/libftdi/1.2-r0/packages-split/libftdi/usr/lib/libftdipp1.so.2.2.0)
DEBUG: libftdi: Dependency libgcc_s.so.1 requires package libgcc (used by files: /home/jenkins/oe/world/shr-core/tmp-glibc/work/i586-oe-linux/libftdi/1.2-r0/packages-split/libftdi/usr/lib/libftdipp1.so.2.2.0)
DEBUG: libftdi: Dependency libusb-1.0.so.0 requires package libusb1 (used by files: /home/jenkins/oe/world/shr-core/tmp-glibc/work/i586-oe-linux/libftdi/1.2-r0/packages-split/libftdi/usr/lib/libftdi1.so.2.2.0)
DEBUG: libftdi: Dependency libc.so.6 requires package glibc (used by files: /home/jenkins/oe/world/shr-core/tmp-glibc/work/i586-oe-linux/libftdi/1.2-r0/packages-split/libftdi/usr/lib/libftdi1.so.2.2.0)
DEBUG: Python function package_do_shlibs finished
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Explicitly disable ipv6 support if the distro does not include it.
Signed-off-by: Richard Tollerton <rich.tollerton@ni.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Set the SH to the target path to avoid host contamination,
which causes the following issue on some hosts:
| WARNING: QA Issue: /usr/bin/smicache_libsmi contained in
package libsmi requires /usr/bin/sh, but no providers
found in its RDEPENDS [file-rdeps]
| Computing transaction...error: Can't install
libsmi-0.4.8-r0.0@corei7_64: no package provides /usr/bin/sh
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Derek Straka
Koen Kooi
Pascal Bach
Goran Cengic
Armin Kuster
Jonathan Liu
Adrian Dudau
Jacob Stiffler
Roy Li
Denys Dmytriyenko
Andre McCurdy
Nicolas Dechesne
Ricardo Ribalda Delgado
Ruben De Smet
Martin Jansa
Khem Raj
Yannick Gicquel
Catalin Enache
Qi.Chen@windriver.com
Jackie Huang
mike.looijmans@topic.nl
Ioan-Adrian Ratiu
José Bollo
Richard Tollerton