[oe] [meta-oe][PATCH] kconfig-frontends: drop unneeded target flex/bison deps
From: Christopher Larson
Signed-off-by: Christopher Larson
---
.../recipes-devtools/kconfig-frontends/kconfig-frontends_4.11.0.1.bb| 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
a/meta-oe/recipes-devtools/kconfig-frontends/kconfig-frontends_4.11.0.1.bb
b/meta-oe/recipes-devtools/kconfig-frontends/kconfig-frontends_4.11.0.1.bb
index fae80a79e..ec959d27d 100644
--- a/meta-oe/recipes-devtools/kconfig-frontends/kconfig-frontends_4.11.0.1.bb
+++ b/meta-oe/recipes-devtools/kconfig-frontends/kconfig-frontends_4.11.0.1.bb
@@ -11,7 +11,7 @@ HOMEPAGE = "https://gitlab.com/ymorin/kconfig-frontends";
LICENSE = "GPL-2.0"
LIC_FILES_CHKSUM = "file://COPYING;md5=9b8cf60ff39767ff04b671fca8302408"
SECTION = "devel"
-DEPENDS += "ncurses flex bison gperf-native bison-native"
+DEPENDS += "ncurses flex-native gperf-native bison-native"
RDEPENDS_${PN} += "python3 bash"
SRC_URI =
"git://gitlab.com/ymorin/kconfig-frontends.git;protocol=https;branch=4.11.x \
file://0001-Makefile-ensure-frontends-exits-before-writing-into-.patch \
--
2.11.1
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH] libmxml: Use autotools bbclass and cleanup the recipe a bit.
EXTRA_AUTORECONF = “--exclude=autopoint,autoheader"
On Aug 21, 2019, 12:12 PM -0700, Piotr Tworek , wrote:
> The package does use autotools based build systemd but it comes with a
> bit of a twist. It explicitly disallows running autoheader on the source
> tree. Unfortunately this is what autoreconf invoked by autotools bbclass
> does. Still, there is an easy fix for this. When AUTOHEADER env variable
> is defined this is what autoreconf will invoke instead of autoheader
> binary found in PATH. If we set it to /bin/true we'll esentially satisfy
> mxml requirements without having to patch autotools bbclass.
>
> This patch also adds PACKAGECONFIG option allowing the user to toggle
> threading support in the library. Not sure how useful it is, but upstream
> does offer such option, it works, and its not a lot of work to expose it
> via PACKAGECONFIG.
>
> Another improvement is related to package optimization handling.
> mxml has a nasty habit of disregarding optimization flags passed by the
> user. It always appends -Os -g to CFLAGS in its configure script. Its
> a bit suprising to find out DEBUG_BUILD=1 in local.conf has no impact on
> libraries produced by mxml recipe. This can be fixed with a simple sed
> script.
>
> Sed is also used to enable verbose compilation output. Having full
> compiler invocations visible in the logs is generally useful, but
> more importantly it allows some of the QA checks to their work.
>
> While doing this cleanup I've also renamed the recipe from libxml_git.bb
> to libxml_3.0.bb. The git keyword suggests this recipe is for some
> random git snapshot, while in reality its for latest stable upstream
> release of the package. Fortunately for us upstream also tags releases
> in git so we can use those tags to fetch correct version of the sources.
>
> Last but not least this patch removes the bin package. Current versions of
> mxml no longer provide any tools. The package was empty.
>
> Signed-off-by: Piotr Tworek
> ---
> .../recipes-support/libmxml/libmxml_3.0.bb | 35 +++
> .../recipes-support/libmxml/libmxml_git.bb | 44 ---
> 2 files changed, 35 insertions(+), 44 deletions(-)
> create mode 100644 meta-oe/recipes-support/libmxml/libmxml_3.0.bb
> delete mode 100644 meta-oe/recipes-support/libmxml/libmxml_git.bb
>
> diff --git a/meta-oe/recipes-support/libmxml/libmxml_3.0.bb
> b/meta-oe/recipes-support/libmxml/libmxml_3.0.bb
> new file mode 100644
> index 0..cc83e19c9
> --- /dev/null
> +++ b/meta-oe/recipes-support/libmxml/libmxml_3.0.bb
> @@ -0,0 +1,35 @@
> +DESCRIPTION = "Tiny XML Library"
> +LICENSE = "Apache-2.0"
> +LIC_FILES_CHKSUM = "file://LICENSE;md5=86d3f3a95c324c9479bd8986968f4327"
> +HOMEPAGE = "https://www.msweet.org/mxml/";
> +BUGTRACKER = "https://github.com/michaelrsweet/mxml/issues";
> +
> +SRC_URI = "git://github.com/michaelrsweet/mxml.git"
> +SRCREV = "v${PV}"
> +S = "${WORKDIR}/git"
> +
> +inherit autotools
> +
> +PACKAGECONFIG ??= "threads"
> +PACKAGECONFIG[threads] = "--enable-threads,--disable-threads"
> +
> +# Package does not support out of tree builds.
> +B = "${S}"
> +
> +do_configure_prepend() {
> + # Respect optimization CFLAGS specified by OE.
> + sed -e 's/-Os -g//' -i ${S}/configure.ac
> +
> + # Enable verbose compilation output. This is required for extra QA checks
> to work.
> + sed -e '/.SILENT:/d' -i ${S}/Makefile.in
> +
> + # The package uses autotools but it explicitly states it does not
> + # support running autoheader. The following export should make sure
> + # autoreconf used by autotools.bbclass invokes /bin/true instead
> + export AUTOHEADER=$(which true)
> +}
> +
> +do_install() {
> + # Package uses DSTROOT instread of standard DESTDIR to specify install
> location.
> + oe_runmake install DSTROOT=${D}
> +}
> diff --git a/meta-oe/recipes-support/libmxml/libmxml_git.bb
> b/meta-oe/recipes-support/libmxml/libmxml_git.bb
> deleted file mode 100644
> index b17fc67e0..0
> --- a/meta-oe/recipes-support/libmxml/libmxml_git.bb
> +++ /dev/null
> @@ -1,44 +0,0 @@
> -DESCRIPTION = "Tiny XML Library"
> -LICENSE = "Apache-2.0"
> -LIC_FILES_CHKSUM = "file://LICENSE;md5=86d3f3a95c324c9479bd8986968f4327"
> -HOMEPAGE = "https://www.msweet.org/mxml/";
> -BUGTRACKER = "https://github.com/michaelrsweet/mxml/issues";
> -
> -SRC_URI = "git://github.com/michaelrsweet/mxml.git"
> -SRCREV = "c7755b6992a2afdd34dde47fc9be97f1237cfded"
> -S = "${WORKDIR}/git"
> -# v3.0
> -PV = "3.0"
> -
> -CONFIGUREOPTS = " --prefix=${prefix} \
> - --bindir=${bindir} \
> - --sbindir=${sbindir} \
> - --libexecdir=${libexecdir} \
> - --datadir=${datadir} \
> - --sysconfdir=${sysconfdir} \
> - --sharedstatedir=${sharedstatedir} \
> - --localstatedir=${localstatedir} \
> - --libdir=${libdir} \
> - --includedir=${includedir} \
> - --oldincludedir=${oldincludedir} \
> - --infodir=${infodir} \
> - --mandir=${mandir} \
> - --host=${TARGET_SYS} \
> - --build=${BUILD_SYS} \
> - ${PACKAGECONFIG_CONFARGS} \
> - "
> -
> -do_configure() {
> - install
Re: [oe] [meta-oe][PATCH] minicoredumper: relocate admin-binaries to sbin
The commit message is generic whereas the commit is not. The commit changes the startup script to match the install path, but the commit message doesn't mention the startup scripts at all. On Tue, Apr 10, 2018 at 8:23 AM Christophe PRIOUZEAU < christophe.priouz...@st.com> wrote: > Sine the commit "relocate admin-binaries to sbin" on minicoredumper, > minicoredumper binaries are installed on /usr/sbin directory instead > of /usr/bin. > src/minicoredumper/Makefile.am: sbin_PROGRAMS = minicoredumper > > > Signed-off-by: Christophe Priouzeau > --- > meta-oe/recipes-kernel/minicoredumper/files/minicoredumper.init| 2 +- > meta-oe/recipes-kernel/minicoredumper/files/minicoredumper.service | 2 +- > 2 files changed, 2 insertions(+), 2 deletions(-) > > diff --git > a/meta-oe/recipes-kernel/minicoredumper/files/minicoredumper.init > b/meta-oe/recipes-kernel/minicoredumper/files/minicoredumper.init > index 90ef7fa..9c96e31 100644 > --- a/meta-oe/recipes-kernel/minicoredumper/files/minicoredumper.init > +++ b/meta-oe/recipes-kernel/minicoredumper/files/minicoredumper.init > @@ -22,7 +22,7 @@ > > case "$1" in > start) > -echo "|/usr/bin/minicoredumper %p %u %g %s %t %h %e" > > /proc/sys/kernel/core_pattern > +echo "|/usr/sbin/minicoredumper %p %u %g %s %t %h %e" > > /proc/sys/kernel/core_pattern > ;; > stop) > echo "core" > /proc/sys/kernel/core_pattern > diff --git > a/meta-oe/recipes-kernel/minicoredumper/files/minicoredumper.service > b/meta-oe/recipes-kernel/minicoredumper/files/minicoredumper.service > index 851b54c..45d098f 100644 > --- a/meta-oe/recipes-kernel/minicoredumper/files/minicoredumper.service > +++ b/meta-oe/recipes-kernel/minicoredumper/files/minicoredumper.service > @@ -4,7 +4,7 @@ Description=Enable minicoredumper. > [Service] > Type=oneshot > RemainAfterExit=yes > -ExecStart=/bin/sh -c '/bin/echo "|/usr/bin/minicoredumper %%p %%u %%g %%s > %%t %%h %%e" > /proc/sys/kernel/core_pattern' > +ExecStart=/bin/sh -c '/bin/echo "|/usr/sbin/minicoredumper %%p %%u %%g > %%s %%t %%h %%e" > /proc/sys/kernel/core_pattern' > ExecStop=/bin/sh -c '/bin/echo "core" > /proc/sys/kernel/core_pattern' > > [Install] > -- > 2.7.4 > -- > ___ > Openembedded-devel mailing list > Openembedded-devel@lists.openembedded.org > http://lists.openembedded.org/mailman/listinfo/openembedded-devel > -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] libtalloc failure due to waf
n: FileNotFoundError: [Errno 2] No such file or directory: > > '/OE/master/build/tmp-glibc/work/armv7ahf-neon-oe-linux- > gnueabi/libtalloc/2.1.10-r0/talloc-2.1.10/waf' > > > > ERROR: libtalloc-2.1.10-r0 do_configure: Function failed: > waf_preconfigure > > ERROR: Logfile of failure stored in: > > /OE/master/build/tmp-glibc/work/armv7ahf-neon-oe-linux- > gnueabi/libtalloc/2.1.10-r0/temp/log.do_configure.52699 > > ERROR: Task > > (/OE/master/sources/meta-openembedded/meta-networking/ > recipes-support/libtalloc/libtalloc_2.1.10.bb:do_configure) > > failed with exit code '1' > > -- > > ___ > > Openembedded-devel mailing list > > Openembedded-devel@lists.openembedded.org > > http://lists.openembedded.org/mailman/listinfo/openembedded-devel > > > -- > ___ > Openembedded-devel mailing list > Openembedded-devel@lists.openembedded.org > http://lists.openembedded.org/mailman/listinfo/openembedded-devel > -- Christopher Larson kergoth at gmail dot com Founder - BitBake, OpenEmbedded, OpenZaurus Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-python][PATCH] python-pyroute2: uprev to v0.4.21 (from 0.3.22)
; +RDEPENDS_${PN} += " \
> >> +${PYTHON_PN}-json \
> >> +"
> >> diff --git a/meta-python/recipes-devtools/python/python-
> >> pyroute2/import-simplejson-as-json.patch b/meta-python/recipes-
> >> devtools/python/python-pyroute2/import-simplejson-as-json.patch
> >> deleted file mode 100644
> >> index d73da07..000
> >> --- a/meta-python/recipes-devtools/python/python-
> >> pyroute2/import-simplejson-as-json.patch
> >> +++ /dev/null
> >> @@ -1,26 +0,0 @@
> >> -Upstream-Status: Inappropriate [wrong dependency]
> >> -
> >> -Subject: [PATCH] import simplejson as json
> >> -
> >> -At runtime python-pyroute2 tries to import json, but the module is not
> >> -available, import simplejson as json solve the issue.
> >> -
> >> -Signed-off-by: Fabio Berton
> >>
> >> - pyroute2/netlink/rtnl/ifinfmsg.py | 2 +-
> >> - 1 file changed, 1 insertion(+), 1 deletion(-)
> >> -
> >> -diff --git a/pyroute2/netlink/rtnl/ifinfmsg.py
> b/pyroute2/netlink/rtnl/
> >> ifinfmsg.py
> >> -index 4f14f25..55643eb 100644
> >> a/pyroute2/netlink/rtnl/ifinfmsg.py
> >> -+++ b/pyroute2/netlink/rtnl/ifinfmsg.py
> >> -@@ -1,5 +1,5 @@
> >> - import os
> >> --import json
> >> -+import simplejson as json
> >> - import errno
> >> - import select
> >> - import struct
> >> ---
> >> -2.1.4
> >> -
> >> diff --git a/meta-python/recipes-devtools/python/python-
> pyroute2_0.3.22.bb
> >> b/meta-python/recipes-devtools/python/python-pyroute2_0.4.21.bb
> >> similarity index 100%
> >> rename from meta-python/recipes-devtools/python/python-pyroute2_0.3.22.
> bb
> >> rename to meta-python/recipes-devtools/python/python-pyroute2_0.4.21.bb
> >> diff --git a/meta-python/recipes-devtools/python/python3-
> >> pyroute2_0.3.22.bb b/meta-python/recipes-devtools/python/python3-
> >> pyroute2_0.4.21.bb
> >> similarity index 100%
> >> rename from meta-python/recipes-devtools/python/python3-pyroute2_0.3.
> 22.bb
> >> rename to meta-python/recipes-devtools/python/python3-pyroute2_0.4.
> 21.bb
> >> --
> >> 2.7.4
> >>
> >> --
> >> ___
> >> Openembedded-devel mailing list
> >> Openembedded-devel@lists.openembedded.org
> >> http://lists.openembedded.org/mailman/listinfo/openembedded-devel
> >>
> > --
> > ___
> > Openembedded-devel mailing list
> > Openembedded-devel@lists.openembedded.org
> > http://lists.openembedded.org/mailman/listinfo/openembedded-devel
> --
> ___
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-devel
>
--
Christopher Larson
kergoth at gmail dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [PATCH] meson: don't pass localedir
On Mon, Jun 5, 2017 at 2:46 AM, Ross Burton wrote: > For historic reasons the localedir directory is where binary locales are > stored > (/usr/lib/locale) , not where application translations belong (typically, > /usr/share/locale). Don't pass localedir explicitly, and let Meson use the > default of $datadir/locale to match the behaviour of autotools.bbclass > and the > expectations of the system. > > Signed-off-by: Ross Burton > Maybe we should think about renaming that variable for clarity. -- Christopher Larson kergoth at gmail dot com Founder - BitBake, OpenEmbedded, OpenZaurus Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [PATCH V2] picocom: Fix build
On Thu, Jan 19, 2017 at 5:15 PM, Marek Vasut wrote:
> The picocom complains about missing GNU_HASH from the binary during
> QA stage, this is because the picocom Makefile overrides CPPFLAGS,
> CFLAGS and LDFLAGS. Fix this by passing those as an argument to make.
> Moreover, since picocom 1.7 now accepts VERSION variable and the
> UUCP_LOCK_DIR is set to /var/lock by default, drop the CPPFLAGS
> override altogether and replace it simply with passing VERSION
> argument to make to precisely retain the original intention.
>
> Signed-off-by: Marek Vasut
> ---
> V2: Use EXTRA_OEMAKE instead of custom do_configure
> ---
> meta-oe/recipes-support/picocom/picocom_1.7.bb | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/meta-oe/recipes-support/picocom/picocom_1.7.bb
> b/meta-oe/recipes-support/picocom/picocom_1.7.bb
> index dbee65609..ca0b8f419 100644
> --- a/meta-oe/recipes-support/picocom/picocom_1.7.bb
> +++ b/meta-oe/recipes-support/picocom/picocom_1.7.bb
> @@ -9,7 +9,8 @@ SRC_URI = "http://picocom.googlecode.
> com/files/picocom-${PV}.tar.gz"
> SRC_URI[md5sum] = "8eaba1d31407e8408674d6e57af447ef"
> SRC_URI[sha256sum] = "d0f31c8f7a215a76922d30c81a52b9
> a2348c89e02a84935517002b3bc2c1129e"
>
> -CPPFLAGS_append = '-DVERSION_STR=\\"${PV}\\"
> -DUUCP_LOCK_DIR=\\"/var/lock\\" -DHIGH_BAUD'
> +EXTRA_OEMAKE_append = "'CC=${CC}' 'LD=${LD}' 'VERSION=${PV}' \
I'd suggest ditching the pointless _append while you're at it (just use =
or +=). Besides, the use of _append was missing a space separator, so the
CC wouldn't be obeyed.
--
Christopher Larson
clarson at kergoth dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [PATCH] devtools: Add recipe for Chromium Embedded Controller utilities
On Wed, Dec 28, 2016 at 11:31 AM, Moritz Fischer wrote: > Add a recipe to build 'ectool', a utility to interact with the embedded > controller on Google's Chromebooks (and other platforms using it). > > Signed-off-by: Moritz Fischer > Thanks for your contribution, they’re always appreciated. In the future, try to keep in mind that the subject line needs a correct prefix for oe commit message standards (prefix with recipe name, not recipes subdir name), and the email subject should mention which layer you want it applied to (though one can determine it based on context, best practice is to include it). I’ll leave it to the layer maintainer to determine whether you need to re-send or just keep it in mind for future patches. Thanks again, -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [PATCH] hwloc: Add
On Thu, Dec 22, 2016 at 2:34 PM, Martin Jansa
wrote:
> SUMMARY is for short summary, put long one into DESCRIPTION and create
> short one for SUMMARY.
>
> Also missing sha265sum, so I doubt it was tested properly.
>
> On Thu, Dec 22, 2016 at 9:02 PM, Haris Okanovic
> wrote:
>
> > https://www.open-mpi.org/projects/hwloc/
> >
> > "The Portable Hardware Locality (hwloc) software package provides a
> > portable abstraction (across OS, versions, architectures, ...) of the
> > hierarchical topology of modern architectures, including NUMA memory
> > nodes, sockets, shared caches, cores and simultaneous multithreading.
> > It also gathers various system attributes such as cache and memory
> > information as well as the locality of I/O devices such as network
> > interfaces, InfiniBand HCAs or GPUs."
> >
> > This recipe adds libhwloc and lstopo utility as two packages.
> >
> > Testing: Built in OE/Krogoth. Installed to x86_64 machine. Successfully
> > ran `lstopo` and verified it printed correct CPU and cache info.
> >
> > Signed-off-by: Haris Okanovic
> > ---
> > meta-oe/recipes-extended/hwloc/hwloc_1.11.5.bb | 25
> > +
> > 1 file changed, 25 insertions(+)
> > create mode 100644 meta-oe/recipes-extended/hwloc/hwloc_1.11.5.bb
> >
> > diff --git a/meta-oe/recipes-extended/hwloc/hwloc_1.11.5.bb
> > b/meta-oe/recipes-extended/hwloc/hwloc_1.11.5.bb
> > new file mode 100644
> > index 000..3a88f5b
> > --- /dev/null
> > +++ b/meta-oe/recipes-extended/hwloc/hwloc_1.11.5.bb
> > @@ -0,0 +1,25 @@
> > +SUMMARY = "The Portable Hardware Locality (hwloc) software package \
> > + provides a portable abstraction of the hierarchical topology of modern
> \
> > + architectures."
> > +HOMEPAGE = "https://www.open-mpi.org/software/hwloc/";
> > +SECTION = "base"
> > +LICENSE = "BSD"
> > +LIC_FILES_CHKSUM = "file://COPYING;md5=3282e20dc3cec311deda3c6d4b1f99
> 0b"
> > +
> > +SRC_URI = "https://www.open-mpi.org/software/${PN}/v1.11/
> > downloads/${BP}.tar.bz2"
> > +SRC_URI[md5sum] = "96c34136ff416d2b13a7821c27477bed"
> > +
> > +inherit autotools
> > +
> > +# Split hwloc library into separate subpackage
> > +PACKAGES_prepend = "lib${PN}"
> > +FILES_lib${PN} += "${libdir}/lib${PN}.so*"
> > +${PN}_RDEPENDS += "lib${PN}"
> > +
> > +do_install_append() {
> > +# Let ldconfig do it's thing, no need for these symlinks
> > + test -L "${D}${libdir}/lib${PN}.so.5"
> > +test -L "${D}${libdir}/lib${PN}.so"
> > +rm "${D}${libdir}/lib${PN}.so.5"
> > +rm "${D}${libdir}/lib${PN}.so"
>
This link manipulation is also inconsistent with every other recipe. It’s
not appropriate. If your distro wants to rely on ldconfig to handle the
links, then your distro could do this for all recipes rather than just one.
This isn’t how things are done in our existing recipes.
--
Christopher Larson
clarson at kergoth dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Maintainer - Tslib
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-webserver][PATCH] apache2: the download has been moved to the 'archives'
On Wed, Dec 21, 2016 at 2:53 PM, Mark Asselstine < mark.asselst...@windriver.com> wrote: > This release of apache2 is no longer found in the main downloads area > but has been moved to the 'archives'. We should uprev apache2 but for > now this will at least get the builds working again. > > Signed-off-by: Mark Asselstine > Wouldn’t it make sense to do a recipe-local addition to MIRRORS to automatically fall back to the apache archive? -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java][PATCH] move from bb.data.getVar(x, d, 1) to d.getVar(x, True)
On Tue, Nov 29, 2016 at 9:48 AM, Cody P Schafer wrote:
> bb.data.getVar was removed, need to use the modern mechanism.
>
> Signed-off-by: Cody P Schafer
> ---
> classes/java-library.bbclass | 4 ++--
> recipes-core/icedtea/icedtea7-native.inc | 2 +-
> recipes-core/openjdk/openjdk-7-common.inc | 2 +-
> recipes-core/openjdk/openjdk-8-common.inc | 10 +-
> recipes-core/openjdk/openjdk-common.inc | 6 +++---
> 5 files changed, 12 insertions(+), 12 deletions(-)
>
> diff --git a/classes/java-library.bbclass b/classes/java-library.bbclass
> index 144cd2f..58d5a40 100644
> --- a/classes/java-library.bbclass
> +++ b/classes/java-library.bbclass
> @@ -23,8 +23,8 @@ def java_package_name(d):
>pre=""
>post=""
>
> - bpn = bb.data.getVar('BPN', d, 1)
> - ml = bb.data.getVar('MLPREFIX', d, 1)
> + bpn = d.getVar('BPN', True)
> + ml = d.getVar('MLPREFIX', True)
>
If you’re working on master, even the ‘, True’ is no longer needed, FYI.
--
Christopher Larson
clarson at kergoth dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Maintainer - Tslib
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH] xterm: use ${libdir} to avoid installed-vs-shipped error on ppc64
On Fri, Oct 28, 2016 at 3:07 AM, wrote:
> xterm installs files into PREFIX/lib/X11/app-defaults by default.
> This leads to QA issue on ppc64 which defines libdir as /usr/lib64:
> ERROR: QA Issue: xterm: Files/directories were installed but not shipped
> in any package:
> /usr/lib
> /usr/lib/X11
> /usr/lib/X11/app-defaults
> /usr/lib/X11/app-defaults/UXTerm
> /usr/lib/X11/app-defaults/KOI8RXTerm-color
> /usr/lib/X11/app-defaults/XTerm-color
> /usr/lib/X11/app-defaults/UXTerm-color
> /usr/lib/X11/app-defaults/KOI8RXTerm
> /usr/lib/X11/app-defaults/XTerm
>
> Fix it by passing --with-app-defaults=${libdir}/X11/app-defaults to
> configure.
>
> Signed-off-by: Ting Liu
> ---
> meta-oe/recipes-graphics/xorg-app/xterm_325.bb | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/meta-oe/recipes-graphics/xorg-app/xterm_325.bb
> b/meta-oe/recipes-graphics/xorg-app/xterm_325.bb
> index e0baf3c..48556c7 100644
> --- a/meta-oe/recipes-graphics/xorg-app/xterm_325.bb
> +++ b/meta-oe/recipes-graphics/xorg-app/xterm_325.bb
> @@ -14,6 +14,7 @@ PACKAGECONFIG[xft] =
> "--enable-freetype,--disable-freetype,libxft
> fontconfig fre
>
> EXTRA_OECONF = " --x-includes=${STAGING_INCDIR} \
> --x-libraries=${STAGING_LIBDIR} \
> + --with-app-defaults=${libdir}/X11/app-defaults \
> FREETYPE_CONFIG=${STAGING_BINDIR_CROSS}/freetype-config
> \
> --disable-imake \
> --disable-rpath-hack \
>
Are you sure this is correct? I highly doubt those files are actually
multilib specific. Would it not be better to just package these files in
their current location?
--
Christopher Larson
clarson at kergoth dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Maintainer - Tslib
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe] [PATCH] bash2dash conversion
The purpose of ‘command’ is to *run* a command, not check if it’s
available, and ‘type’ is mandated in SuSv3/SuSv4/POSIX. If dash doesn’t
provide it, it’s broken.
On Sat, Sep 17, 2016 at 7:52 PM, wrote:
> From: "Tim K. Chan"
>
> Change bash style to dash style
>
> Signed-off-by: Tim K. Chan
> [Adjust context]
> Signed-off-by: He Zhe
> ---
> .../iscsi-initiator-utils/iscsi-initiator-utils_2.0-873.bb | 2
> +-
> meta-oe/recipes-benchmark/lmbench/lmbench_3.0-a9.bb | 6
> --
> meta-oe/recipes-connectivity/krb5/krb5_1.13.2.bb| 2
> +-
> meta-oe/recipes-extended/sblim-sfcb/sblim-sfcb_1.4.9.bb | 2
> +-
> meta-oe/recipes-support/postgresql/postgresql.inc | 2
> +-
> 5 files changed, 8 insertions(+), 6 deletions(-)
>
> diff --git a/meta-networking/recipes-daemons/iscsi-initiator-utils/
> iscsi-initiator-utils_2.0-873.bb b/meta-networking/recipes-
> daemons/iscsi-initiator-utils/iscsi-initiator-utils_2.0-873.bb
> index 03f6459..17a5c1e 100644
> --- a/meta-networking/recipes-daemons/iscsi-initiator-utils/
> iscsi-initiator-utils_2.0-873.bb
> +++ b/meta-networking/recipes-daemons/iscsi-initiator-utils/
> iscsi-initiator-utils_2.0-873.bb
> @@ -103,7 +103,7 @@ pkg_postinst_${PN}() {
>
> if [ -e /etc/init.d/populate-volatile.sh ]; then
> /etc/init.d/populate-volatile.sh update
> -elif command -v systemd-tmpfiles >/dev/null; then
> +elif command -p systemd-tmpfiles >/dev/null; then
> systemd-tmpfiles --create ${sysconfdir}/tmpfiles.d/iscsi.conf
> fi
> }
> diff --git a/meta-oe/recipes-benchmark/lmbench/lmbench_3.0-a9.bb
> b/meta-oe/recipes-benchmark/lmbench/lmbench_3.0-a9.bb
> index 3c3b4b5..2961b6b 100644
> --- a/meta-oe/recipes-benchmark/lmbench/lmbench_3.0-a9.bb
> +++ b/meta-oe/recipes-benchmark/lmbench/lmbench_3.0-a9.bb
> @@ -33,7 +33,9 @@ do_configure() {
> }
>
> do_compile () {
> -. ${CONFIG_SITE}
> +for CONFIG_SITE_ITEM in $CONFIG_SITE; do
> + . $CONFIG_SITE_ITEM
> +done
> if [ X"$ac_cv_uint" = X"yes" ]; then
> CFLAGS="${CFLAGS} -DHAVE_uint"
> fi
> @@ -68,7 +70,7 @@ do_install () {
>
> pkg_postinst_${PN} () {
> if [ -z "$D" ]; then
> -if command -v systemd-tmpfiles >/dev/null; then
> +if command -p systemd-tmpfiles >/dev/null; then
> systemd-tmpfiles --create ${sysconfdir}/tmpfiles.d/
> lmbench.conf
> elif [ -e ${sysconfdir}/init.d/populate-volatile.sh ]; then
> ${sysconfdir}/init.d/populate-volatile.sh update
> diff --git a/meta-oe/recipes-connectivity/krb5/krb5_1.13.2.bb
> b/meta-oe/recipes-connectivity/krb5/krb5_1.13.2.bb
> index 3a3886b..ba0e2ac 100644
> --- a/meta-oe/recipes-connectivity/krb5/krb5_1.13.2.bb
> +++ b/meta-oe/recipes-connectivity/krb5/krb5_1.13.2.bb
> @@ -108,7 +108,7 @@ do_install_append() {
>
> pkg_postinst_${PN} () {
> if [ -z "$D" ]; then
> -if command -v systemd-tmpfiles >/dev/null; then
> +if command -p systemd-tmpfiles >/dev/null; then
> systemd-tmpfiles --create ${sysconfdir}/tmpfiles.d/krb5.conf
> elif [ -e ${sysconfdir}/init.d/populate-volatile.sh ]; then
> ${sysconfdir}/init.d/populate-volatile.sh update
> diff --git a/meta-oe/recipes-extended/sblim-sfcb/sblim-sfcb_1.4.9.bb
> b/meta-oe/recipes-extended/sblim-sfcb/sblim-sfcb_1.4.9.bb
> index 84e465d..b1346b3 100644
> --- a/meta-oe/recipes-extended/sblim-sfcb/sblim-sfcb_1.4.9.bb
> +++ b/meta-oe/recipes-extended/sblim-sfcb/sblim-sfcb_1.4.9.bb
> @@ -67,7 +67,7 @@ pkg_postinst_${PN} () {
>
> if [ x"$D" != "x" ]; then
> OPTS="--root=$D"
> -if type systemctl >/dev/null 2>/dev/null; then
> +if command -p systemctl >/dev/null 2>/dev/null; then
> systemctl $OPTS ${SYSTEMD_AUTO_ENABLE} ${SYSTEMD_SERVICE}
> fi
> exit 1
> diff --git a/meta-oe/recipes-support/postgresql/postgresql.inc
> b/meta-oe/recipes-support/postgresql/postgresql.inc
> index e473f58..9592c79 100644
> --- a/meta-oe/recipes-support/postgresql/postgresql.inc
> +++ b/meta-oe/recipes-support/postgresql/postgresql.inc
> @@ -179,7 +179,7 @@ do_install_append() {
>
> # install COPYRIGHT README HISTORY
> install -d -m 0755 ${D}${docdir}/${BPN}
> -for i in ${B}/{COPYRIGHT,README,HISTORY}
> ${B}/doc/{KNOWN_BUGS,MISSING_FEATURES,README*,bug.template}; do
> +for i in ${B}/COPYRIGHT ${B}/README ${B}/HISTORY ${B}/doc/KNOWN_BUGS
> ${B}/doc/MISSING_FEATURES ${B}/doc/README* ${B}/doc/bug.template; do
>
Re: [oe] [meta-oe][PATCH] geany-plugins: include GPLv3 in the main LICENSE
On Mon, Aug 29, 2016 at 1:52 PM, Martin Jansa wrote: > * since last changes in oe-core LICENSE variable needs to list all > licenses used > by individual packages otherwise bitbake warning like this is shown: > WARNING: meta-openembedded/meta-oe/recipes-devtools/geany/geany- > plugins_1.27.bb: LICENSE_geany-plugins-commander includes licenses > (GPLv3) that are not listed in LICENSE > as possible improvement you can set LICENSE to GPLv2 to all plugins > which are GPLv2-only, because now they will "inherit" both GPLv2 & > GPLv3 which sucks as described here: > https://www.mail-archive.com/openembedded-core@lists. > openembedded.org/msg81645.html > > Signed-off-by: Martin Jansa > Would be nice if we had a general way to change the inherited LICENSE as used by the binary packages. Adding a single indirection to package.bbclass to default to LICENSE_DEFAULT rather than LICENSE and defaulting LICENSE_DEFAULT to LICENSE would do (or similar), then this recipe could set LICENSE_DEFAULT to GPLv2. In the meantime.. We have a bbappend in meta-mentor for this on a branch which handles it for this recipe reasonably, I think: https://github.com/MentorEmbedded/meta-mentor/pull/843/commits/d852857 -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-multimedia][PATCH] libebml: don't rely on make -e, obey LDFLAGS
On Sun, Jul 24, 2016 at 5:01 AM, Martin Jansa wrote: > On Fri, Jun 17, 2016 at 01:34:49PM -0700, Christopher Larson wrote: > > From: Christopher Larson > > > > We need to obey LDFLAGS to get the correct hash style for external > toolchains, > > and passing vars explicitly is better than relying on implicitly > overriding > > everything blindly from the environment. > > > > Signed-off-by: Christopher Larson > > --- > > .../recipes-mkv/libebml/libebml/ldflags.patch | 57 > ++ > > .../libebml/libebml/override-uname.patch | 42 > > .../recipes-mkv/libebml/libebml_1.3.0.bb | 44 > +++-- > > 3 files changed, 138 insertions(+), 5 deletions(-) > > create mode 100644 > meta-multimedia/recipes-mkv/libebml/libebml/ldflags.patch > > create mode 100644 > meta-multimedia/recipes-mkv/libebml/libebml/override-uname.patch > > > > diff --git a/meta-multimedia/recipes-mkv/libebml/libebml/ldflags.patch > b/meta-multimedia/recipes-mkv/libebml/libebml/ldflags.patch > > new file mode 100644 > > index 000..a01c754 > > --- /dev/null > > +++ b/meta-multimedia/recipes-mkv/libebml/libebml/ldflags.patch > > @@ -0,0 +1,57 @@ > > +Use LD and obey LDFLAGS > > + > > +Signed-off-by: Christopher Larson > > +Upstream-Status: Pending > > + > > +diff --git a/make/linux/Makefile b/make/linux/Makefile > > +index 391b6e3..a2ac13e 100644 > > +--- a/make/linux/Makefile > > b/make/linux/Makefile > > +@@ -64,6 +64,7 @@ objects_so:=$(patsubst %$(EXTENSION),%.lo,$(sources)) > > + WARNINGFLAGS=-Wall -Wextra -Wno-unknown-pragmas -ansi > -fno-gnu-keywords -Wshadow > > + COMPILEFLAGS=$(WARNINGFLAGS) $(CXXFLAGS) $(CPPFLAGS) $(DEBUGFLAGS) > $(INCLUDE) > > + DEPENDFLAGS = $(CXXFLAGS) $(INCLUDE) > > ++LINKFLAGS=$(LDFLAGS) > > + > > + ifeq (Darwin,$(shell uname -s)) > > + all: staticlib > > +@@ -91,7 +92,7 @@ $(LIBRARY): $(objects) > > + $(RANLIB) $@ > > + > > + $(LIBRARY_SO): $(objects_so) > > +-$(CXX) -shared -Wl,-soname,$(LIBRARY_SO_VER) -o $(LIBRARY_SO_VER) > $(objects_so) > > ++$(LD) $(LINKFLAGS) -shared -Wl,-soname,$(LIBRARY_SO_VER) -o > $(LIBRARY_SO_VER) $(objects_so) > > + rm -f $(LIBRARY_SO) > > + ln -s $(LIBRARY_SO_VER) $(LIBRARY_SO) > > + > > +diff --git a/make/mingw32/Makefile b/make/mingw32/Makefile > > +index e5986ef..6ca16aa 100644 > > +--- a/make/mingw32/Makefile > > b/make/mingw32/Makefile > > +@@ -18,6 +18,7 @@ DEBUGFLAGS=-g -DDEBUG > > + endif > > + CROSS = > > + CXX = $(CROSS)g++ > > ++LD = $(CXX) > > + CC = $(CROSS)gcc > > + WINDRES = $(CROSS)windres > > + RANLIB = $(CROSS)ranlib > > +@@ -33,6 +34,7 @@ LIBS = libebml.a > > + endif > > + INCS = -I"$(shell pwd)/../.." > > + COMPILEFLAGS = $(DEBUGFLAGS) $(INCS) $(DLLFLAGS) $(CXXFLAGS) > > ++LINKFLAGS = $(LDFLAGS) > > + > > + .PHONY: all all-before all-after clean clean-custom > > + > > +@@ -54,7 +56,7 @@ libebml.a: $(OBJ) > > + $(RANLIB) $@ > > + > > + libebml.dll: $(OBJ) > > +-$(CXX) -shared -Wl,--export-all -Wl,--out-implib=$@.a -o $@ $(OBJ) > > ++$(LD) $(LINKFLAGS) -shared -Wl,--export-all -Wl,--out-implib=$@.a > -o $@ $(OBJ) > > + > > + depend: > > + @echo Calculating dependecies: > > +-- > > +2.8.0 > > diff --git > a/meta-multimedia/recipes-mkv/libebml/libebml/override-uname.patch > b/meta-multimedia/recipes-mkv/libebml/libebml/override-uname.patch > > new file mode 100644 > > index 000..57eb345 > > --- /dev/null > > +++ b/meta-multimedia/recipes-mkv/libebml/libebml/override-uname.patch > > @@ -0,0 +1,42 @@ > > +Allow override of the 'uname -s' for cross-compilation > > + > > +Signed-off-by: Christopher Larson > > +Upstream-Status: Pending > > + > > +diff --git a/make/linux/Makefile b/make/linux/Makefile > > +index a2ac13e..4188bee 100644 > > +--- a/make/linux/Makefile > > b/make/linux/Makefile > > +@@ -9,9 +9,11 @@ > > + # 'make DEBUG=yes'. > > + # > > + > > ++TARGET_OS ?= $(shell uname -s) > > ++ > > + # Paths > > + # BeOS wants the libs and headers in /boot/home/config > > +-ifeq (BeOS,$(shell uname -s)) > > ++ifeq (BeOS,$(TARGET_OS)) > > + prefix=/boot/home/config > > + else > > + prefix=/usr/local > > +@@ -66,7 +68,7 @@ COMPILEFLAGS=$(WARNINGFLAGS) $(CXXFLAGS) $(CPPFLAGS) > $(DEBUGFLAGS) $(INCLUDE) > > + DEPENDF
Re: [oe] [meta-networking][PATCH 2/2] meta-networking: override SECURITY_CFLAGS for c-ares
On Fri, Jul 15, 2016 at 5:54 AM, wrote: > From: André Draszik > > c-ares doesn't build if the distro has enabled usage of the > security_flags.inc file as it is picky about what is placed > into CPPFLAGS and CFLAGS. It complains and errors out if any > preprocessor options appear in CFLAGS. > > Fix this by providing an additional include file that is > require'd from conf/layer.conf. > > Signed-off-by: André Draszik > This commit message is a little misleading. The fact that you added a .inc isn't what fixed the problem, the move of the defines from CFLAGS to CPPFLAGS was. -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH] CANopenNode: recipe for building the CANopenSocket tools
On Wed, Jul 13, 2016 at 8:00 AM, Gary Thomas wrote:
> On 2016-07-13 16:57, cynthiachiu wrote:
>
>> Initial check-in for the CANopenSocket tools recipe.
>>
>> Signed-off-by: cynthiachiu
>> ---
>> .../canopennode/canopensocket_git.bb | 31
>> ++
>> 1 file changed, 31 insertions(+)
>> create mode 100644 meta-oe/recipes-extended/canopennode/
>> canopensocket_git.bb
>>
>> diff --git a/meta-oe/recipes-extended/canopennode/canopensocket_git.bb
>> b/meta-oe/recipes-extended/canopennode/canopensocket_git.bb
>> new file mode 100644
>> index 000..4506698
>> --- /dev/null
>> +++ b/meta-oe/recipes-extended/canopennode/canopensocket_git.bb
>> @@ -0,0 +1,31 @@
>> +SUMMARY = "Linux CANOpen tools"
>> +DESCRIPTION = "Linux CANOpen Protocol Stack Tools"
>> +LICENSE = "GPLv2"
>> +LIC_FILES_CHKSUM =
>> "file://gpl-2.0.txt;md5=b234ee4d69f5fce4486a80fdaf4a4263"
>> +
>> +SRC_URI = "git://github.com/CANopenNode/CANopenSocket.git"
>> +SRCREV = "${AUTOREV}"
>>
>
> Can't you pick a known, working ${REV}? Using ${AUTOREV} makes it
> impossible
> to use this recipe disconnected from the internet (yes, that's still a
> concern
> in 2016!)
>
>
> +
>> +S = "${WORKDIR}/git"
>> +
>> +do_compile_prepend() {
>> +cd ${S}
>> +git submodule init
>> +git submodule update
>>
>
This fetches things at do_compile time, which is not appropriate. Use
gitsm:// instead of git:// and let the fetcher handle the submodules.
--
Christopher Larson
clarson at kergoth dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Maintainer - Tslib
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-networking][PATCH 1/1] openconnect: PACKAGECONFIG certain dependencies
On Fri, Jul 1, 2016 at 1:25 PM, Slater, Joseph wrote: > > -Original Message- > > From: openembedded-devel-boun...@lists.openembedded.org [mailto: > openembedded-devel- > > boun...@lists.openembedded.org] On Behalf Of Christopher Larson > > Sent: Thursday, June 30, 2016 10:13 PM > > To: Openembedded Discussion > > Subject: Re: [oe] [meta-networking][PATCH 1/1] openconnect: > PACKAGECONFIG certain > > dependencies > > > > On Thu, Jun 30, 2016 at 12:46 PM, Joe Slater > wrote: > > > > > +PACKAGECONFIG ??= "gnutls lz4 libproxy" > > > + > > > +# config defaults > > > +PACKAGECONFIG[gnutls] = ",--without-gnutls,gnutls," > > > +PACKAGECONFIG[lz4] = ",--without-lz4,lz4," > > > +PACKAGECONFIG[libproxy] = ",--without-libproxy,libproxy," > > > + > > > +# not config defaults > > > +PACKAGECONFIG[pcsc-lite] = ",--without-libpcsclite,pcsc-lite," > > > > > > > Why are the options not being passed for the enabled case, here? We > prefer > > to be explicit, rather than relying on defaults. > > I can easily add them. I left them out because configure help only > mentions the --without > options to override the defaults. It'd be best to add them, then we're explicitly specifying what we want, and potentially avoid issues if the upstream default behavior changes during an upgrade in the future. -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-networking][PATCH 1/1] openconnect: PACKAGECONFIG certain dependencies
On Thu, Jun 30, 2016 at 12:46 PM, Joe Slater wrote: > +PACKAGECONFIG ??= "gnutls lz4 libproxy" > + > +# config defaults > +PACKAGECONFIG[gnutls] = ",--without-gnutls,gnutls," > +PACKAGECONFIG[lz4] = ",--without-lz4,lz4," > +PACKAGECONFIG[libproxy] = ",--without-libproxy,libproxy," > + > +# not config defaults > +PACKAGECONFIG[pcsc-lite] = ",--without-libpcsclite,pcsc-lite," > Why are the options not being passed for the enabled case, here? We prefer to be explicit, rather than relying on defaults. -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-multimedia][PATCH] libebml: don't rely on make -e, obey LDFLAGS
From: Christopher Larson
We need to obey LDFLAGS to get the correct hash style for external toolchains,
and passing vars explicitly is better than relying on implicitly overriding
everything blindly from the environment.
Signed-off-by: Christopher Larson
---
.../recipes-mkv/libebml/libebml/ldflags.patch | 57 ++
.../libebml/libebml/override-uname.patch | 42
.../recipes-mkv/libebml/libebml_1.3.0.bb | 44 +++--
3 files changed, 138 insertions(+), 5 deletions(-)
create mode 100644 meta-multimedia/recipes-mkv/libebml/libebml/ldflags.patch
create mode 100644
meta-multimedia/recipes-mkv/libebml/libebml/override-uname.patch
diff --git a/meta-multimedia/recipes-mkv/libebml/libebml/ldflags.patch
b/meta-multimedia/recipes-mkv/libebml/libebml/ldflags.patch
new file mode 100644
index 000..a01c754
--- /dev/null
+++ b/meta-multimedia/recipes-mkv/libebml/libebml/ldflags.patch
@@ -0,0 +1,57 @@
+Use LD and obey LDFLAGS
+
+Signed-off-by: Christopher Larson
+Upstream-Status: Pending
+
+diff --git a/make/linux/Makefile b/make/linux/Makefile
+index 391b6e3..a2ac13e 100644
+--- a/make/linux/Makefile
b/make/linux/Makefile
+@@ -64,6 +64,7 @@ objects_so:=$(patsubst %$(EXTENSION),%.lo,$(sources))
+ WARNINGFLAGS=-Wall -Wextra -Wno-unknown-pragmas -ansi -fno-gnu-keywords
-Wshadow
+ COMPILEFLAGS=$(WARNINGFLAGS) $(CXXFLAGS) $(CPPFLAGS) $(DEBUGFLAGS) $(INCLUDE)
+ DEPENDFLAGS = $(CXXFLAGS) $(INCLUDE)
++LINKFLAGS=$(LDFLAGS)
+
+ ifeq (Darwin,$(shell uname -s))
+ all: staticlib
+@@ -91,7 +92,7 @@ $(LIBRARY): $(objects)
+ $(RANLIB) $@
+
+ $(LIBRARY_SO): $(objects_so)
+- $(CXX) -shared -Wl,-soname,$(LIBRARY_SO_VER) -o $(LIBRARY_SO_VER)
$(objects_so)
++ $(LD) $(LINKFLAGS) -shared -Wl,-soname,$(LIBRARY_SO_VER) -o
$(LIBRARY_SO_VER) $(objects_so)
+ rm -f $(LIBRARY_SO)
+ ln -s $(LIBRARY_SO_VER) $(LIBRARY_SO)
+
+diff --git a/make/mingw32/Makefile b/make/mingw32/Makefile
+index e5986ef..6ca16aa 100644
+--- a/make/mingw32/Makefile
b/make/mingw32/Makefile
+@@ -18,6 +18,7 @@ DEBUGFLAGS=-g -DDEBUG
+ endif
+ CROSS =
+ CXX = $(CROSS)g++
++LD = $(CXX)
+ CC = $(CROSS)gcc
+ WINDRES = $(CROSS)windres
+ RANLIB = $(CROSS)ranlib
+@@ -33,6 +34,7 @@ LIBS = libebml.a
+ endif
+ INCS = -I"$(shell pwd)/../.."
+ COMPILEFLAGS = $(DEBUGFLAGS) $(INCS) $(DLLFLAGS) $(CXXFLAGS)
++LINKFLAGS = $(LDFLAGS)
+
+ .PHONY: all all-before all-after clean clean-custom
+
+@@ -54,7 +56,7 @@ libebml.a: $(OBJ)
+ $(RANLIB) $@
+
+ libebml.dll: $(OBJ)
+- $(CXX) -shared -Wl,--export-all -Wl,--out-implib=$@.a -o $@ $(OBJ)
++ $(LD) $(LINKFLAGS) -shared -Wl,--export-all -Wl,--out-implib=$@.a -o $@
$(OBJ)
+
+ depend:
+ @echo Calculating dependecies:
+--
+2.8.0
diff --git a/meta-multimedia/recipes-mkv/libebml/libebml/override-uname.patch
b/meta-multimedia/recipes-mkv/libebml/libebml/override-uname.patch
new file mode 100644
index 000..57eb345
--- /dev/null
+++ b/meta-multimedia/recipes-mkv/libebml/libebml/override-uname.patch
@@ -0,0 +1,42 @@
+Allow override of the 'uname -s' for cross-compilation
+
+Signed-off-by: Christopher Larson
+Upstream-Status: Pending
+
+diff --git a/make/linux/Makefile b/make/linux/Makefile
+index a2ac13e..4188bee 100644
+--- a/make/linux/Makefile
b/make/linux/Makefile
+@@ -9,9 +9,11 @@
+ # 'make DEBUG=yes'.
+ #
+
++TARGET_OS ?= $(shell uname -s)
++
+ # Paths
+ # BeOS wants the libs and headers in /boot/home/config
+-ifeq (BeOS,$(shell uname -s))
++ifeq (BeOS,$(TARGET_OS))
+ prefix=/boot/home/config
+ else
+ prefix=/usr/local
+@@ -66,7 +68,7 @@ COMPILEFLAGS=$(WARNINGFLAGS) $(CXXFLAGS) $(CPPFLAGS)
$(DEBUGFLAGS) $(INCLUDE)
+ DEPENDFLAGS = $(CXXFLAGS) $(INCLUDE)
+ LINKFLAGS=$(LDFLAGS)
+
+-ifeq (Darwin,$(shell uname -s))
++ifeq (Darwin,$(TARGET_OS))
+ all: staticlib
+ else
+ all: staticlib sharedlib
+@@ -116,7 +118,7 @@ depend:
+ $(CXX) $(DEPENDFLAGS) -MM -MT $$o $$i >> .depend ; \
+ done
+
+-ifeq (Darwin,$(shell uname -s))
++ifeq (Darwin,$(TARGET_OS))
+ install: install_staticlib install_headers
+ else
+ install: install_staticlib install_sharedlib install_headers
+--
+2.8.0
diff --git a/meta-multimedia/recipes-mkv/libebml/libebml_1.3.0.bb
b/meta-multimedia/recipes-mkv/libebml/libebml_1.3.0.bb
index 71e2bfc..b756a39 100644
--- a/meta-multimedia/recipes-mkv/libebml/libebml_1.3.0.bb
+++ b/meta-multimedia/recipes-mkv/libebml/libebml_1.3.0.bb
@@ -2,15 +2,49 @@ SUMMARY = "libebml is a C++ libary to parse EBML files"
LICENSE = "LGPLv2.1"
LIC_FILES_CHKSUM = "file://LICENSE.LGPL;md5=f14599a2f089f6ff8c97e2baa4e3d575"
-SRC_URI = "http://dl.matroska.org/downloads/libebml/libebml-${PV}.tar.bz2";
+SRC_URI = "\
+http://dl.matroska.org/downloads/libebml/libebml-${PV}.tar.bz2 \
+file://ldflags.patch \
+file://override-uname.pa
[oe] [meta-oe][PATCH] vboxguestdrivers: obey LDFLAGS
From: Christopher Larson
We need to obey LDFLAGS to get the correct hash style for external toolchains.
Signed-off-by: Christopher Larson
---
meta-oe/recipes-support/vboxguestdrivers/vboxguestdrivers_4.3.36.bb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
a/meta-oe/recipes-support/vboxguestdrivers/vboxguestdrivers_4.3.36.bb
b/meta-oe/recipes-support/vboxguestdrivers/vboxguestdrivers_4.3.36.bb
index 781e3d5..081e977 100644
--- a/meta-oe/recipes-support/vboxguestdrivers/vboxguestdrivers_4.3.36.bb
+++ b/meta-oe/recipes-support/vboxguestdrivers/vboxguestdrivers_4.3.36.bb
@@ -40,7 +40,7 @@ do_export_sources() {
# compile and install mount utility
do_compile_append() {
-oe_runmake -C ${S}/utils
+oe_runmake 'LD=${CC}' 'LDFLAGS=${LDFLAGS}' -C ${S}/utils
}
module_do_install() {
--
2.8.0
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 2/2] mpv: use waf.bbclass
From: Christopher Larson
Signed-off-by: Christopher Larson
---
meta-oe/recipes-multimedia/mplayer/mpv_0.15.0.bb | 23 ++-
1 file changed, 6 insertions(+), 17 deletions(-)
diff --git a/meta-oe/recipes-multimedia/mplayer/mpv_0.15.0.bb
b/meta-oe/recipes-multimedia/mplayer/mpv_0.15.0.bb
index 5fac298..768c87a 100644
--- a/meta-oe/recipes-multimedia/mplayer/mpv_0.15.0.bb
+++ b/meta-oe/recipes-multimedia/mplayer/mpv_0.15.0.bb
@@ -12,14 +12,14 @@ LICENSE = "GPLv2+"
LIC_FILES_CHKSUM = "file://LICENSE;md5=91f1cb870c1cc2d31351a4d2595441cb"
SRC_URI = "https://github.com/mpv-player/mpv/archive/v${PV}.tar.gz;name=mpv \
- http://www.freehackers.org/~tnagy/release/waf-1.8.12;name=waf \
+
http://www.freehackers.org/~tnagy/release/waf-1.8.12;name=waf;subdir=${BPN}-${PV}
\
"
SRC_URI[mpv.md5sum] = "9042bd3fbff2bc8ba0b7fadaa4a22101"
SRC_URI[mpv.sha256sum] =
"7d31217ba8572f364fcea2955733f821374ae6d8c6d8f22f8bc63c44c0400bdc"
SRC_URI[waf.md5sum] = "cef4ee82206b1843db082d0b0506bf71"
SRC_URI[waf.sha256sum] =
"01bf2beab2106d1558800c8709bc2c8e496d3da4a2ca343fe091f22fca60c98b"
-inherit pkgconfig pythonnative distro_features_check
+inherit waf pkgconfig pythonnative distro_features_check
# Note: both lua and libass are required to get on-screen-display (controls)
PACKAGECONFIG ??= "lua libass"
@@ -28,6 +28,7 @@ PACKAGECONFIG[libass] =
"--enable-libass,--disable-libass,libass"
PACKAGECONFIG[libarchive] =
"--enable-libarchive,--disable-libarchive,libarchive"
PACKAGECONFIG[jack] = "--enable-jack, --disable-jack, jack"
+SIMPLE_TARGET_SYS = "${@'${TARGET_SYS}'.replace('${TARGET_VENDOR}', '')}"
EXTRA_OECONF = " \
--prefix=${prefix} \
--target=${SIMPLE_TARGET_SYS} \
@@ -50,21 +51,9 @@ EXTRA_OECONF = " \
--disable-vapoursynth-lazy \
"
-do_configure() {
-if [ ! -L ../waf ]; then
-chmod a+x ../waf-1.8.12
- ln -s waf-1.8.12 ../waf
-fi
-export SIMPLE_TARGET_SYS="$(echo ${TARGET_SYS} | sed
s:${TARGET_VENDOR}::g)"
-../waf configure ${EXTRA_OECONF}
-}
-
-do_compile () {
-../waf build
-}
-
-do_install() {
-../waf install --destdir=${D}
+do_configure_prepend () {
+ln -sf waf-1.8.12 ${S}/waf
+chmod +x ${S}/waf
}
FILES_${PN} += "${datadir}/icons"
--
2.8.0
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 1/2] mpv: add DESCRIPTION
From: Christopher Larson Signed-off-by: Christopher Larson --- meta-oe/recipes-multimedia/mplayer/mpv_0.15.0.bb | 1 + 1 file changed, 1 insertion(+) diff --git a/meta-oe/recipes-multimedia/mplayer/mpv_0.15.0.bb b/meta-oe/recipes-multimedia/mplayer/mpv_0.15.0.bb index 6a258b5..5fac298 100644 --- a/meta-oe/recipes-multimedia/mplayer/mpv_0.15.0.bb +++ b/meta-oe/recipes-multimedia/mplayer/mpv_0.15.0.bb @@ -1,4 +1,5 @@ SUMMARY = "Open Source multimedia player" +DESCRIPTION = "mpv is a fork of mplayer2 and MPlayer. It shares some features with the former projects while introducing many more." SECTION = "multimedia" HOMEPAGE = "http://www.mpv.io/"; DEPENDS = "zlib ffmpeg jpeg virtual/libx11 xsp libxv \ -- 2.8.0 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 0/2] mpv: add desc & use waf.bbclass
From: Christopher Larson The following changes since commit 52213998eb4ead7d24cba012c937bcc25e89c2ca: mpv: Media Player (2016-06-08 14:55:30 +0200) are available in the git repository at: git://github.com/kergoth/meta-openembedded mpv-use-waf-bbclass https://github.com/kergoth/meta-openembedded/tree/mpv-use-waf-bbclass Christopher Larson (2): mpv: add DESCRIPTION mpv: use waf.bbclass meta-oe/recipes-multimedia/mplayer/mpv_0.15.0.bb | 24 +++- 1 file changed, 7 insertions(+), 17 deletions(-) -- 2.8.0 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [OE-core] State of bitbake world 2016-06-07
On Thu, Jun 9, 2016 at 12:03 AM, Martin Jansa
wrote:
> This report includes python3 changes, there are few more failures, but
> it's hard to see what is new, so I've sent couple PNBLACKLISTs for
> recipes reported in previous 2 reports which are still failing with
> glibc-2.24 and gcc-6. Hopefully next report will show only "new" issues
> caused by python3 switch.
>
> == Number of issues - stats ==
> {| class='wikitable'
> !|Date !!colspan='3'|Failed tasks
> !!colspan='6'|Failed depencencies!!|Signatures
> !!colspan='12'|QA !!Comment
> |-
> || ||qemuarm ||qemux86 ||qemux86_64
> ||qemuarm||max||min ||qemux86||max||min ||all ||already-stripped
> ||libdir||textrel ||build-deps||file-rdeps
> ||version-going-backwards ||host-user-contaminated
> ||installed-vs-shipped ||unknown-configure-option
> ||symlink-to-sysroot||invalid-pkgconfig ||pkgname ||
> |-
> ||2016-06-07||34||40||39||84||38||37||92
> ||42||41||0 ||0 ||0 ||1
> ||3 ||18||0 ||0 ||0
> ||0 ||0 ||0 ||0 ||
> |}
>
> http://www.openembedded.org/wiki/Bitbake_World_Status
chkconfig-alternatives-native is failing, but natives are excluded from
world by default. This recipe should only be built if it's the preferred
update-alternatives-native provider, otherwise you'll hit the conflict that
your build is hitting. If you're undoing the EXCLUDE_FROM_WORLD for all
natives, better keep it here. I am surprised the bits that skip
non-preferred recipes didn't happen here, though.
--
Christopher Larson
clarson at kergoth dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Maintainer - Tslib
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH] gitpkgv: ensure that we close our opened files
Ignore this, just noticed the fix is already on master-next.
On Fri, Jun 3, 2016 at 11:36 AM, Christopher Larson
wrote:
> From: Christopher Larson
>
> Avoids warnings like this with python3:
>
> WARNING:
> .../meta-openembedded/meta-filesystems/recipes-filesystems/smbnetfs/
> smbnetfs_git.bb: :94: ResourceWarning: unclosed file
> <_io.TextIOWrapper
> name='.../build/downloads/git2/smbnetfs.git.sourceforge.net.gitroot.smbnetfs.smbnetfs/oe-gitpkgv_ace1c519d4'
> mode='r' encoding='UTF-8'>
>
> Signed-off-by: Christopher Larson
> ---
> meta-oe/classes/gitpkgv.bbclass | 6 --
> 1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/meta-oe/classes/gitpkgv.bbclass
> b/meta-oe/classes/gitpkgv.bbclass
> index 1cba00c..4866fac 100644
> --- a/meta-oe/classes/gitpkgv.bbclass
> +++ b/meta-oe/classes/gitpkgv.bbclass
> @@ -87,11 +87,13 @@ def get_git_pkgv(d, use_tags):
>
> if commits != "":
> oe.path.remove(rev_file, recurse=False)
> -open(rev_file, "w").write("%d\n" % int(commits))
> +with open(rev_file, "w") as f:
> +f.write("%d\n" % int(commits))
> else:
> commits = "0"
> else:
> -commits = open(rev_file, "r").readline(128).strip()
> + with open(rev_file, "r") as f:
> +commits = f.readline(128).strip()
>
> if use_tags:
> try:
> --
> 2.8.0
>
>
--
Christopher Larson
kergoth at gmail dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Maintainer - Tslib
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH] gitpkgv: ensure that we close our opened files
From: Christopher Larson
Avoids warnings like this with python3:
WARNING:
.../meta-openembedded/meta-filesystems/recipes-filesystems/smbnetfs/smbnetfs_git.bb:
:94: ResourceWarning: unclosed file <_io.TextIOWrapper
name='.../build/downloads/git2/smbnetfs.git.sourceforge.net.gitroot.smbnetfs.smbnetfs/oe-gitpkgv_ace1c519d4'
mode='r' encoding='UTF-8'>
Signed-off-by: Christopher Larson
---
meta-oe/classes/gitpkgv.bbclass | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/meta-oe/classes/gitpkgv.bbclass b/meta-oe/classes/gitpkgv.bbclass
index 1cba00c..4866fac 100644
--- a/meta-oe/classes/gitpkgv.bbclass
+++ b/meta-oe/classes/gitpkgv.bbclass
@@ -87,11 +87,13 @@ def get_git_pkgv(d, use_tags):
if commits != "":
oe.path.remove(rev_file, recurse=False)
-open(rev_file, "w").write("%d\n" % int(commits))
+with open(rev_file, "w") as f:
+f.write("%d\n" % int(commits))
else:
commits = "0"
else:
-commits = open(rev_file, "r").readline(128).strip()
+with open(rev_file, "r") as f:
+commits = f.readline(128).strip()
if use_tags:
try:
--
2.8.0
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH v2] plymouth: allow disabling the dracut dep
From: Christopher Larson
Add a 'initrd' PACKAGECONFIG to disable emission of the plymouth-initrd
package which includes initramfs files for use by dracut. Disabling this will
avoid the runtime dependency on dracut, which is useful if one doesn't need or
want dracut and doesn't want to pull in meta-initramfs.
Signed-off-by: Christopher Larson
---
meta-oe/recipes-core/plymouth/plymouth_0.9.2.bb | 18 +++---
1 file changed, 11 insertions(+), 7 deletions(-)
diff --git a/meta-oe/recipes-core/plymouth/plymouth_0.9.2.bb
b/meta-oe/recipes-core/plymouth/plymouth_0.9.2.bb
index 9ca8502..de23dcd 100644
--- a/meta-oe/recipes-core/plymouth/plymouth_0.9.2.bb
+++ b/meta-oe/recipes-core/plymouth/plymouth_0.9.2.bb
@@ -24,29 +24,33 @@ EXTRA_OECONF += " --enable-shared --disable-static
--disable-gtk --disable-docum
${@bb.utils.contains('DISTRO_FEATURES', 'systemd',
'--enable-systemd-integration --with-system-root-install', '', d)} \
"
+PACKAGECONFIG ??= "pango initrd"
+PACKAGECONFIG_append_x86 = " drm"
+PACKAGECONFIG_append_x86-64 = " drm"
+
PACKAGECONFIG[drm] = "--enable-drm,--disable-drm,libdrm"
PACKAGECONFIG[pango] = "--enable-pango,--disable-pango,pango"
PACKAGECONFIG[gtk] = "--enable-gtk,--disable-gtk,gtk+"
-
-PACKAGECONFIG ??= "pango"
+PACKAGECONFIG[initrd] = ",,,"
LOGO ??= "${datadir}/plymouth/bizcom.png"
-PACKAGECONFIG_append_x86 = " drm"
-PACKAGECONFIG_append_x86-64 = " drm"
-
inherit autotools pkgconfig systemd
-
do_install_append() {
install -d ${D}${systemd_unitdir}/system
install -m 644 ${B}/systemd-units/*.service
${D}${systemd_unitdir}/system
install -m 644 ${B}/systemd-units/systemd-ask-password-plymouth.path
${D}${systemd_unitdir}/system
# Remove /var/run from package as plymouth will populate it on startup
rm -fr "${D}${localstatedir}/run"
+
+ if ! ${@bb.utils.contains('PACKAGECONFIG', 'initrd', 'true', 'false',
d)}; then
+ rm -rf "${D}${libexecdir}"
+ fi
}
-PACKAGES =+ "${PN}-initrd ${PN}-set-default-theme"
+PACKAGES =. "${@bb.utils.contains('PACKAGECONFIG', 'initrd', '${PN}-initrd ',
'', d)}"
+PACKAGES =+ "${PN}-set-default-theme"
FILES_${PN}-initrd = "${libexecdir}/plymouth/*"
FILES_${PN}-set-default-theme = "${sbindir}/plymouth-set-default-theme"
--
2.8.0
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH] plymouth: allow disabling the dracut dep
On Tue, May 31, 2016 at 12:15 PM, Christopher Larson wrote: > From: Christopher Larson > > Add a 'initrd' PACKAGECONFIG to disable emission of the plymouth-initrd > package which includes initramfs files for use by dracut. Disabling this > will > avoid the runtime dependency on dracut, which is useful if one doesn't > need or > want dracut and doesn't want to pull in meta-initramfs. > > Signed-off-by: Christopher Larson > Ignore this, needs a rebase. v2 forthcoming. -- Christopher Larson kergoth at gmail dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH] plymouth: allow disabling the dracut dep
From: Christopher Larson
Add a 'initrd' PACKAGECONFIG to disable emission of the plymouth-initrd
package which includes initramfs files for use by dracut. Disabling this will
avoid the runtime dependency on dracut, which is useful if one doesn't need or
want dracut and doesn't want to pull in meta-initramfs.
Signed-off-by: Christopher Larson
---
meta-oe/recipes-core/plymouth/plymouth_0.9.2.bb | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/meta-oe/recipes-core/plymouth/plymouth_0.9.2.bb
b/meta-oe/recipes-core/plymouth/plymouth_0.9.2.bb
index 9ca8502..922e96b 100644
--- a/meta-oe/recipes-core/plymouth/plymouth_0.9.2.bb
+++ b/meta-oe/recipes-core/plymouth/plymouth_0.9.2.bb
@@ -37,6 +37,8 @@ PACKAGECONFIG_append_x86-64 = " drm"
inherit autotools pkgconfig systemd
+PACKAGECONFIG = "initrd"
+PACKAGECONFIG[initrd] = ",,,"
do_install_append() {
install -d ${D}${systemd_unitdir}/system
@@ -44,9 +46,14 @@ do_install_append() {
install -m 644 ${B}/systemd-units/systemd-ask-password-plymouth.path
${D}${systemd_unitdir}/system
# Remove /var/run from package as plymouth will populate it on startup
rm -fr "${D}${localstatedir}/run"
+
+ if ! ${@bb.utils.contains('PACKAGECONFIG', 'initrd', 'true', 'false',
d)}; then
+ rm -rf "${D}${libexecdir}"
+ fi
}
-PACKAGES =+ "${PN}-initrd ${PN}-set-default-theme"
+PACKAGES =. "${@bb.utils.contains('PACKAGECONFIG', 'initrd', '${PN}-initrd ',
'', d)}"
+PACKAGES =+ "${PN}-set-default-theme"
FILES_${PN}-initrd = "${libexecdir}/plymouth/*"
FILES_${PN}-set-default-theme = "${sbindir}/plymouth-set-default-theme"
--
2.8.0
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCHv2] luajit: fix build issues, obey more vars
From: Christopher Larson
Rework how variables are passed in to ensure that more of our flag vars are
obeyed, make it obey more of our target path vars, and use fewer of its
hardcoded flags. Also add verbosity to the compile output for debugging.
Without this, errors were seen for us:
| In file included from
.../lib/gcc/i686-pc-linux-gnu/5.2.0/include-fixed/64/sgxx-glibc/syslimits.h:7:0,
| from
.../lib/gcc/i686-pc-linux-gnu/5.2.0/include-fixed/64/sgxx-glibc/limits.h:34,
| from luaconf.h:12,
| from lua.h:16,
| from lj_arch.h:9:
|
.../lib/gcc/i686-pc-linux-gnu/5.2.0/include-fixed/64/sgxx-glibc/limits.h:168:61:
fatal error: limits.h: No such file or directory
| compilation terminated.
| Makefile:233: *** Unsupported target architecture. Stop.
| make[1]: Leaving directory '.../luajit/2.0.4-r0/LuaJIT-2.0.4/src'
Signed-off-by: Christopher Larson
---
v2 changes: fixed http://errors.yoctoproject.org/Errors/Details/68382/ by
ensuring LUA_TARGET_OS is correct. The lua buildsystem didn't recognize the os
with the LIBCEXTENSION+ABIEXTENSION.
meta-oe/recipes-devtools/luajit/luajit_2.0.4.bb | 56 +++--
1 file changed, 43 insertions(+), 13 deletions(-)
diff --git a/meta-oe/recipes-devtools/luajit/luajit_2.0.4.bb
b/meta-oe/recipes-devtools/luajit/luajit_2.0.4.bb
index a0252aa..964dc1d 100644
--- a/meta-oe/recipes-devtools/luajit/luajit_2.0.4.bb
+++ b/meta-oe/recipes-devtools/luajit/luajit_2.0.4.bb
@@ -15,11 +15,6 @@ inherit pkgconfig binconfig
BBCLASSEXTEND = "native"
-do_configure_prepend() {
-sed -i 's:PREFIX= /usr/local:PREFIX= ${prefix}:g' ${S}/Makefile
-sed -i 's:MULTILIB= lib:MULTILIB= ${baselib}:g' ${S}/Makefile
-}
-
# http://luajit.org/install.html#cross
# Host luajit needs to be compiled with the same pointer size
# If you want to cross-compile to any 32 bit target on an x64 OS,
@@ -30,20 +25,52 @@ BUILD_CC_ARCH_append_powerpc = ' -m32'
BUILD_CC_ARCH_append_x86 = ' -m32'
BUILD_CC_ARCH_append_arm = ' -m32'
-EXTRA_OEMAKE_append_class-target = '\
-CROSS=${HOST_PREFIX} \
-HOST_CC="${BUILD_CC} ${BUILD_CC_ARCH}" \
-TARGET_CFLAGS="${TOOLCHAIN_OPTIONS} ${TARGET_CC_ARCH}" \
-TARGET_LDFLAGS="${TOOLCHAIN_OPTIONS}" \
-TARGET_SHLDFLAGS="${TOOLCHAIN_OPTIONS}" \
-'
+# The lua makefiles expect the TARGET_SYS to be from uname -s
+# Values: Windows, Linux, Darwin, iOS, SunOS, PS3, GNU/kFreeBSD
+LUA_TARGET_OS = "Unknown"
+LUA_TARGET_OS_darwin = "Darwin"
+LUA_TARGET_OS_linux = "Linux"
+LUA_TARGET_OS_linux-gnueabi = "Linux"
+LUA_TARGET_OS_mingw32 = "Windows"
+
+# We don't want the lua buildsystem's compiler optimizations, or its
+# stripping, and we don't want it to pick up CFLAGS or LDFLAGS, as those apply
+# to both host and target compiles
+EXTRA_OEMAKE = "\
+Q= E='@:' \
+\
+CCOPT= CCOPT_x86= CFLAGS= LDFLAGS= TARGET_STRIP='@:' \
+\
+'TARGET_SYS=${LUA_TARGET_OS}' \
+\
+'CC=${CC}' \
+'TARGET_AR=${AR} rcus' \
+'TARGET_CFLAGS=${CFLAGS}' \
+'TARGET_LDFLAGS=${LDFLAGS}' \
+'TARGET_SHLDFLAGS=${LDFLAGS}' \
+'HOST_CC=${BUILD_CC}' \
+'HOST_CFLAGS=${BUILD_CFLAGS}' \
+'HOST_LDFLAGS=${BUILD_LDFLAGS}' \
+\
+'PREFIX=${prefix}' \
+'MULTILIB=${baselib}' \
+"
do_compile () {
oe_runmake
}
+# There's INSTALL_LIB and INSTALL_SHARE also, but the lua binary hardcodes the
+# '/share' and '/' + LUA_MULTILIB paths, so we don't want to break those
+# expectations.
+EXTRA_OEMAKEINST = "\
+'DESTDIR=${D}' \
+'INSTALL_BIN=${D}${bindir}' \
+'INSTALL_INC=${D}${includedir}/luajit-$(MAJVER).$(MINVER)' \
+'INSTALL_MAN=${D}${mandir}/man1' \
+"
do_install () {
-oe_runmake 'DESTDIR=${D}' install
+oe_runmake ${EXTRA_OEMAKEINST} install
rmdir ${D}${datadir}/lua/5.* \
${D}${datadir}/lua \
${D}${libdir}/lua/5.* \
@@ -52,6 +79,9 @@ do_install () {
PACKAGES += 'luajit-common'
+# See the comment for EXTRA_OEMAKEINST. This is needed to ensure the hardcoded
+# paths are packaged regardless of what the libdir and datadir paths are.
+FILES_${PN} += "${prefix}/${baselib} ${prefix}/share"
FILES_${PN} += "${libdir}/libluajit-5.1.so.2 \
${libdir}/libluajit-5.1.so.${PV} \
"
--
2.8.0
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH 2/6] luajit: fix build issues, obey more vars
On Sat, May 28, 2016 at 8:41 PM, Christopher Larson wrote: > On Tue, May 17, 2016 at 9:10 AM, Christopher Larson > wrote: > >> From: Christopher Larson >> >> Rework how variables are passed in to ensure that more of our flag vars >> are >> obeyed, make it obey more of our target path vars, and use fewer of its >> hardcoded flags. Also add verbosity to the compile output for debugging. >> >> Without this, errors were seen for us: >> >> | In file included from >> .../lib/gcc/i686-pc-linux-gnu/5.2.0/include-fixed/64/sgxx-glibc/syslimits.h:7:0, >> | from >> .../lib/gcc/i686-pc-linux-gnu/5.2.0/include-fixed/64/sgxx-glibc/limits.h:34, >> | from luaconf.h:12, >> | from lua.h:16, >> | from lj_arch.h:9: >> | >> .../lib/gcc/i686-pc-linux-gnu/5.2.0/include-fixed/64/sgxx-glibc/limits.h:168:61: >> fatal error: limits.h: No such file or directory >> | compilation terminated. >> | Makefile:233: *** Unsupported target architecture. Stop. >> | make[1]: Leaving directory '.../luajit/2.0.4-r0/LuaJIT-2.0.4/src' >> >> Signed-off-by: Christopher Larson >> > > Were there issues with this one? I noticed the other patches in this > series were applied, but this one wasn't. > Ah, nevermind, I see it's on master-next. -- Christopher Larson kergoth at gmail dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH 2/6] luajit: fix build issues, obey more vars
On Tue, May 17, 2016 at 9:10 AM, Christopher Larson wrote: > From: Christopher Larson > > Rework how variables are passed in to ensure that more of our flag vars are > obeyed, make it obey more of our target path vars, and use fewer of its > hardcoded flags. Also add verbosity to the compile output for debugging. > > Without this, errors were seen for us: > > | In file included from > .../lib/gcc/i686-pc-linux-gnu/5.2.0/include-fixed/64/sgxx-glibc/syslimits.h:7:0, > | from > .../lib/gcc/i686-pc-linux-gnu/5.2.0/include-fixed/64/sgxx-glibc/limits.h:34, > | from luaconf.h:12, > | from lua.h:16, > | from lj_arch.h:9: > | > .../lib/gcc/i686-pc-linux-gnu/5.2.0/include-fixed/64/sgxx-glibc/limits.h:168:61: > fatal error: limits.h: No such file or directory > | compilation terminated. > | Makefile:233: *** Unsupported target architecture. Stop. > | make[1]: Leaving directory '.../luajit/2.0.4-r0/LuaJIT-2.0.4/src' > > Signed-off-by: Christopher Larson > Were there issues with this one? I noticed the other patches in this series were applied, but this one wasn't. -- Christopher Larson kergoth at gmail dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH v3 2/2] mpv: Media Player
On Thu, Apr 21, 2016 at 7:49 AM, Ahsan, Noor wrote: > Hello Gary, > > Are you planning to send updated PR for MPV? > > Noor > Any news on this series? -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 6/6] x11vnc: remove old libtool macros from acinclude.m4
From: Christopher Larson
These can cause build problems. We only want macros in acinclude.m4 which
aren't available for aclocal to get elsewhere.
Signed-off-by: Christopher Larson
---
meta-oe/recipes-graphics/x11vnc/x11vnc_0.9.13.bb | 6 ++
1 file changed, 6 insertions(+)
diff --git a/meta-oe/recipes-graphics/x11vnc/x11vnc_0.9.13.bb
b/meta-oe/recipes-graphics/x11vnc/x11vnc_0.9.13.bb
index 7e92965..64c8a5d 100644
--- a/meta-oe/recipes-graphics/x11vnc/x11vnc_0.9.13.bb
+++ b/meta-oe/recipes-graphics/x11vnc/x11vnc_0.9.13.bb
@@ -24,3 +24,9 @@ PACKAGECONFIG ??= "${@bb.utils.contains('DISTRO_FEATURES',
'zeroconf', 'avahi',
PACKAGECONFIG[avahi] = "--with-avahi,--without-avahi,avahi"
PACKAGECONFIG[xinerama] = "--with-xinerama,--without-xinerama,libxinerama"
PACKAGECONFIG[libvncserver] =
"--with-system-libvncserver,--without-system-libvncserver,libvncserver"
+
+do_prepare_sources () {
+# Remove old libtool macros from acinclude.m4
+sed -i -e '/^# libtool.m4/q' acinclude.m4
+}
+do_patch[postfuncs] += "do_prepare_sources"
--
2.8.0
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 5/6] networkmanager: add missing dep on dbus-glib-native
From: Christopher Larson The networkmanager build runs dbus-binding-tool, which only exists if dbus-glib-native has been built, and will fail otherwise. Signed-off-by: Christopher Larson --- meta-oe/recipes-connectivity/networkmanager/networkmanager_1.0.10.bb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/meta-oe/recipes-connectivity/networkmanager/networkmanager_1.0.10.bb b/meta-oe/recipes-connectivity/networkmanager/networkmanager_1.0.10.bb index c8447e6..c18afff 100644 --- a/meta-oe/recipes-connectivity/networkmanager/networkmanager_1.0.10.bb +++ b/meta-oe/recipes-connectivity/networkmanager/networkmanager_1.0.10.bb @@ -7,7 +7,7 @@ LIC_FILES_CHKSUM = "file://COPYING;md5=cbbffd568227ada506640fe950a4823b \ file://docs/api/html/license.html;md5=51d7fb67bde992e58533a8481cee070b \ " -DEPENDS = "intltool-native libnl dbus dbus-glib libgudev util-linux libndp libnewt polkit" +DEPENDS = "intltool-native libnl dbus dbus-glib dbus-glib-native libgudev util-linux libndp libnewt polkit" inherit gnomebase gettext systemd bluetooth bash-completion vala gobject-introspection -- 2.8.0 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH 1/6] mg: stop relying on make -e
On Tue, May 17, 2016 at 9:10 AM, Christopher Larson wrote: > From: Christopher Larson > > Relying on 'make -e' isn't ideal, as it's less implicit behavior, relies on > variable exports, and hides the variables being used from someone reading > the > recipe. > > Signed-off-by: Christopher Larson Erm, clearly meant 'explicit' here :) -- Christopher Larson kergoth at gmail dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 4/6] networkmanager: add missing dep on intltool-native
From: Christopher Larson Recipes using intltool need this dependency. Signed-off-by: Christopher Larson --- meta-oe/recipes-connectivity/networkmanager/networkmanager_1.0.10.bb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/meta-oe/recipes-connectivity/networkmanager/networkmanager_1.0.10.bb b/meta-oe/recipes-connectivity/networkmanager/networkmanager_1.0.10.bb index 831ddf0..c8447e6 100644 --- a/meta-oe/recipes-connectivity/networkmanager/networkmanager_1.0.10.bb +++ b/meta-oe/recipes-connectivity/networkmanager/networkmanager_1.0.10.bb @@ -7,7 +7,7 @@ LIC_FILES_CHKSUM = "file://COPYING;md5=cbbffd568227ada506640fe950a4823b \ file://docs/api/html/license.html;md5=51d7fb67bde992e58533a8481cee070b \ " -DEPENDS = "libnl dbus dbus-glib libgudev util-linux libndp libnewt polkit" +DEPENDS = "intltool-native libnl dbus dbus-glib libgudev util-linux libndp libnewt polkit" inherit gnomebase gettext systemd bluetooth bash-completion vala gobject-introspection -- 2.8.0 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 3/6] openlmi-networking: add missing dep on konkretcmpi-native
From: Christopher Larson
Without it, we can hit errors like:
| CMake Error at
.../build/tmp/sysroots/qemux86-64/usr/share/cmake/Modules/OpenLMIMacros.cmake:105
(message):
| KonkretCMPI failed: No such file or directory
Signed-off-by: Christopher Larson
---
meta-oe/recipes-extended/openlmi/openlmi-networking_0.3.1.bb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta-oe/recipes-extended/openlmi/openlmi-networking_0.3.1.bb
b/meta-oe/recipes-extended/openlmi/openlmi-networking_0.3.1.bb
index 5442910..7cfcbe2 100644
--- a/meta-oe/recipes-extended/openlmi/openlmi-networking_0.3.1.bb
+++ b/meta-oe/recipes-extended/openlmi/openlmi-networking_0.3.1.bb
@@ -6,7 +6,7 @@ HOMEPAGE = "http://www.openlmi.org/";
LICENSE = "LGPL-2.1+"
LIC_FILES_CHKSUM = "file://COPYING;md5=7c13b3376cea0ce68d2d2da0a1b3a72c"
SECTION = "System/Management"
-DEPENDS = "openlmi-providers konkretcmpi sblim-cmpi-devel cim-schema-exper
networkmanager dbus libcheck glib-2.0"
+DEPENDS = "openlmi-providers konkretcmpi konkretcmpi-native sblim-cmpi-devel
cim-schema-exper networkmanager dbus libcheck glib-2.0"
SRC_URI = "http://fedorahosted.org/released/${BPN}/${BP}.tar.gz \
file://0001-fix-lib64-can-not-be-shiped-in-64bit-target.patch \
--
2.8.0
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 2/6] luajit: fix build issues, obey more vars
From: Christopher Larson
Rework how variables are passed in to ensure that more of our flag vars are
obeyed, make it obey more of our target path vars, and use fewer of its
hardcoded flags. Also add verbosity to the compile output for debugging.
Without this, errors were seen for us:
| In file included from
.../lib/gcc/i686-pc-linux-gnu/5.2.0/include-fixed/64/sgxx-glibc/syslimits.h:7:0,
| from
.../lib/gcc/i686-pc-linux-gnu/5.2.0/include-fixed/64/sgxx-glibc/limits.h:34,
| from luaconf.h:12,
| from lua.h:16,
| from lj_arch.h:9:
|
.../lib/gcc/i686-pc-linux-gnu/5.2.0/include-fixed/64/sgxx-glibc/limits.h:168:61:
fatal error: limits.h: No such file or directory
| compilation terminated.
| Makefile:233: *** Unsupported target architecture. Stop.
| make[1]: Leaving directory '.../luajit/2.0.4-r0/LuaJIT-2.0.4/src'
Signed-off-by: Christopher Larson
---
meta-oe/recipes-devtools/luajit/luajit_2.0.4.bb | 53 +++--
1 file changed, 40 insertions(+), 13 deletions(-)
diff --git a/meta-oe/recipes-devtools/luajit/luajit_2.0.4.bb
b/meta-oe/recipes-devtools/luajit/luajit_2.0.4.bb
index a0252aa..66bf3e4 100644
--- a/meta-oe/recipes-devtools/luajit/luajit_2.0.4.bb
+++ b/meta-oe/recipes-devtools/luajit/luajit_2.0.4.bb
@@ -15,11 +15,6 @@ inherit pkgconfig binconfig
BBCLASSEXTEND = "native"
-do_configure_prepend() {
-sed -i 's:PREFIX= /usr/local:PREFIX= ${prefix}:g' ${S}/Makefile
-sed -i 's:MULTILIB= lib:MULTILIB= ${baselib}:g' ${S}/Makefile
-}
-
# http://luajit.org/install.html#cross
# Host luajit needs to be compiled with the same pointer size
# If you want to cross-compile to any 32 bit target on an x64 OS,
@@ -30,20 +25,49 @@ BUILD_CC_ARCH_append_powerpc = ' -m32'
BUILD_CC_ARCH_append_x86 = ' -m32'
BUILD_CC_ARCH_append_arm = ' -m32'
-EXTRA_OEMAKE_append_class-target = '\
-CROSS=${HOST_PREFIX} \
-HOST_CC="${BUILD_CC} ${BUILD_CC_ARCH}" \
-TARGET_CFLAGS="${TOOLCHAIN_OPTIONS} ${TARGET_CC_ARCH}" \
-TARGET_LDFLAGS="${TOOLCHAIN_OPTIONS}" \
-TARGET_SHLDFLAGS="${TOOLCHAIN_OPTIONS}" \
-'
+# The lua makefiles expect the TARGET_SYS to be from uname -s
+# Values: Windows, Linux, Darwin, iOS, SunOS, PS3, GNU/kFreeBSD
+LUA_TARGET_OS = "${@'${TARGET_OS}'[0].upper() + '${TARGET_OS}'[1:]}"
+LUA_TARGET_OS_mingw32 = "Windows"
+
+# We don't want the lua buildsystem's compiler optimizations, or its
+# stripping, and we don't want it to pick up CFLAGS or LDFLAGS, as those apply
+# to both host and target compiles
+EXTRA_OEMAKE = "\
+Q= E='@:' \
+\
+CCOPT= CCOPT_x86= CFLAGS= LDFLAGS= TARGET_STRIP='@:' \
+\
+'TARGET_SYS=${LUA_TARGET_OS}' \
+\
+'CC=${CC}' \
+'TARGET_AR=${AR} rcus' \
+'TARGET_CFLAGS=${CFLAGS}' \
+'TARGET_LDFLAGS=${LDFLAGS}' \
+'TARGET_SHLDFLAGS=${LDFLAGS}' \
+'HOST_CC=${BUILD_CC}' \
+'HOST_CFLAGS=${BUILD_CFLAGS}' \
+'HOST_LDFLAGS=${BUILD_LDFLAGS}' \
+\
+'PREFIX=${prefix}' \
+'MULTILIB=${baselib}' \
+"
do_compile () {
oe_runmake
}
+# There's INSTALL_LIB and INSTALL_SHARE also, but the lua binary hardcodes the
+# '/share' and '/' + LUA_MULTILIB paths, so we don't want to break those
+# expectations.
+EXTRA_OEMAKEINST = "\
+'DESTDIR=${D}' \
+'INSTALL_BIN=${D}${bindir}' \
+'INSTALL_INC=${D}${includedir}/luajit-$(MAJVER).$(MINVER)' \
+'INSTALL_MAN=${D}${mandir}/man1' \
+"
do_install () {
-oe_runmake 'DESTDIR=${D}' install
+oe_runmake ${EXTRA_OEMAKEINST} install
rmdir ${D}${datadir}/lua/5.* \
${D}${datadir}/lua \
${D}${libdir}/lua/5.* \
@@ -52,6 +76,9 @@ do_install () {
PACKAGES += 'luajit-common'
+# See the comment for EXTRA_OEMAKEINST. This is needed to ensure the hardcoded
+# paths are packaged regardless of what the libdir and datadir paths are.
+FILES_${PN} += "${prefix}/${baselib} ${prefix}/share"
FILES_${PN} += "${libdir}/libluajit-5.1.so.2 \
${libdir}/libluajit-5.1.so.${PV} \
"
--
2.8.0
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 0/6] Various fixes from meta-mentor
From: Christopher Larson Working through my upstream patch submission backlog. The following changes since commit 37297c3222a40b916d98ccd06b757b595aebc756: nbd: fix LIC_FILES_CHKSUM (2016-05-12 11:53:30 -0400) are available in the git repository at: git://github.com/kergoth/meta-openembedded various-mentor-fixes https://github.com/kergoth/meta-openembedded/tree/various-mentor-fixes Christopher Larson (6): mg: stop relying on make -e luajit: fix build issues, obey more vars openlmi-networking: add missing dep on konkretcmpi-native networkmanager: add missing dep on intltool-native networkmanager: add missing dep on dbus-glib-native x11vnc: remove old libtool macros from acinclude.m4 .../networkmanager/networkmanager_1.0.10.bb| 2 +- meta-oe/recipes-devtools/luajit/luajit_2.0.4.bb| 53 -- .../openlmi/openlmi-networking_0.3.1.bb| 2 +- meta-oe/recipes-graphics/x11vnc/x11vnc_0.9.13.bb | 6 +++ meta-oe/recipes-support/mg/mg_20110905.bb | 12 - 5 files changed, 59 insertions(+), 16 deletions(-) -- 2.8.0 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 1/6] mg: stop relying on make -e
From: Christopher Larson
Relying on 'make -e' isn't ideal, as it's less implicit behavior, relies on
variable exports, and hides the variables being used from someone reading the
recipe.
Signed-off-by: Christopher Larson
---
meta-oe/recipes-support/mg/mg_20110905.bb | 12 +++-
1 file changed, 11 insertions(+), 1 deletion(-)
diff --git a/meta-oe/recipes-support/mg/mg_20110905.bb
b/meta-oe/recipes-support/mg/mg_20110905.bb
index e924bbd..95aeb62 100644
--- a/meta-oe/recipes-support/mg/mg_20110905.bb
+++ b/meta-oe/recipes-support/mg/mg_20110905.bb
@@ -11,7 +11,17 @@ SRC_URI =
"http://homepage.boetes.org/software/mg/mg-${PV}.tar.gz \
SRC_URI[md5sum] = "2de35316fa8ebafe6003efaae70b723e"
SRC_URI[sha256sum] =
"1cd37d7e6a3eecc890a5718c38b8f38495057ba93856762a756ccee2f9618229"
-EXTRA_OEMAKE = "-e MAKEFLAGS="
+# CFLAGS isn't in EXTRA_OEMAKE, as the makefile picks it up via ?=
+EXTRA_OEMAKE = "\
+'CC=${CC}' \
+'LDFLAGS=${LDFLAGS}' \
+\
+'prefix=${prefix}' \
+'bindir=${bindir}' \
+'libdir=${libdir}' \
+'includedir=${includedir}' \
+'mandir=${mandir}' \
+"
do_configure () {
sed -i Makefile.in -e 's,^prefix=.*,prefix=${prefix},'
--
2.8.0
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [PATCH] Fix "failed sanity test tmpdir" for libnice and update PNBLACKLIST.
On Thu, May 5, 2016 at 9:22 AM, Noor, Ahsan wrote:
> From: Noor Ahsan
>
> * Update do_compile_append where it was updating *.pc from ${S} and
> removing staging dir folder. Now we have seperate build folder
> where packages got build. We have to search pc in ${B} not in ${S}.
> After updating the tmp dir pathc is removed from nice.pc file.
> * Remove PNBLACKLIST for libnice and farsight recipes.
>
> Signed-off-by: Noor Ahsan
>
The commit message doesn't meet the OE guidelines (not prefixed by the
recipe being changed) and you don't mention farsight2 in the message at all
-- should that be a separate commit?
--
Christopher Larson
clarson at kergoth dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Maintainer - Tslib
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH] packagegroup-tools-bluetooth.bb: Selects the tools appropriate for the version of bluez being used.
On Mon, May 2, 2016 at 8:54 PM, Ann Thornton wrote:
> Signed-off-by: Ann Thornton
> ---
> .../packagegroups/packagegroup-tools-bluetooth.bb | 40
> ++
> 1 file changed, 40 insertions(+)
> create mode 100644 meta-oe/recipes-connectivity/packagegroups/
> packagegroup-tools-bluetooth.bb
>
> diff --git a/meta-oe/recipes-connectivity/packagegroups/
> packagegroup-tools-bluetooth.bb
> b/meta-oe/recipes-connectivity/packagegroups/
> packagegroup-tools-bluetooth.bb
> new file mode 100644
> index 000..bdab389
> --- /dev/null
> +++ b/meta-oe/recipes-connectivity/packagegroups/
> packagegroup-tools-bluetooth.bb
> @@ -0,0 +1,40 @@
> +# Copyright (C) 2014-2015 Freescale Semiconductor
> +# Released under the MIT license (see COPYING.MIT for the terms)
> +
> +LICENSE = "MIT"
> +LIC_FILES_CHKSUM =
> "file://${COREBASE}/LICENSE;md5=3f40d7994397109285ec7b81fdeb3b58 \
> +
> file://${COREBASE}/meta/COPYING.MIT;md5=3da9cfbcb788c80a0384361b4de20420"
> +
> +SUMMARY = "Set of Bluetooh related tools for inclusion on images"
>
Typo: Bluetooh
> +DESCRIPTION = "Add bluetooth tools based on the version of BlueZ in use.\
> +The tools that have been tested and work the best are pulled in \
> +automatically. The same packagegroup can be used in a recipe without \
> +the need to know which version of BlueZ is in use. \
> +Supports BlueZ4 and BlueZ5."
>
This is more a description of the recipe than the package, but this ends up
in the package. This should describe what the purpose of the package is,
not how it's implemented. "Tools" is also rather vague here. Which tools,
for what purpose?
> +inherit packagegroup
> +
> +BLUEZ4_INSTALL = " \
> +obexftp \
> +"
> +
> +BLUEZ5_INSTALL = " \
> + bluez5-noinst-tools \
> + bluez5-obex \
> + bluez5-testtools \
> + libasound-module-bluez \
> + ${@bb.utils.contains('DISTRO_FEATURES', 'pulseaudio',
> "pulseaudio-module-bluetooth-discover \
> +
> pulseaudio-module-bluetooth-policy \
> +
> pulseaudio-module-bluez5-discover \
> +
> pulseaudio-module-bluez5-device \
> +
> pulseaudio-module-switch-on-connect \
> +
> pulseaudio-module-loopback", \
> + '', d)} \
>
This "_INSTALL" naming is a bit odd -- this isn't an image. These are
runtime dependences, not lists of packages to install. The difference is
subtle, but important. I'd suggest renaming those variables.
+# Install either bluez4 or bluez5 if they are in distro.
> +# Otherwise install nothing.
> +RDEPENDS_${PN} = " \
> + ${@bb.utils.contains('DISTRO_FEATURES', 'bluez5',
> '${BLUEZ5_INSTALL}', "", d)} \
> + ${@bb.utils.contains('DISTRO_FEATURES', 'bluez4',
> '${BLUEZ4_INSTALL}', "", d)} \
> +"
I'd suggest inheriting bluetooth and using the BLUEZ variable which is
already set to 'bluez5' or 'bluez4' based on the distro features.
RDEPENDS_${PN} = "${RDEPENDS_${BLUEZ}}" would work, if you defined the
variables as RDEPENDS_bluez4 and RDEPENDS_bluez5, for example. bitbake can
handle nested variable references. Anyone else want to weigh in on this?
--
Christopher Larson
clarson at kergoth dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Maintainer - Tslib
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-networking][PATCH 2/3] ctdb: rdepend on procps
From: Christopher Larson
ctdbd_wrapper requires pgrep.
Signed-off-by: Christopher Larson
---
meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb | 3 +++
1 file changed, 3 insertions(+)
diff --git a/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb
b/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb
index 66c966d..232546c 100644
--- a/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb
+++ b/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb
@@ -28,6 +28,9 @@ PARALLEL_MAKE = ""
DEPENDS += "popt libtevent libtalloc libldb"
+# ctdbd_wrapper requires pgrep, hence procps
+RDEPENDS_${PN} += "procps"
+
do_configure() {
oe_runconf
}
--
2.8.0
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-networking][PATCH 3/3] ctdb: disable the service by default
From: Christopher Larson
The systemd service is disabled by default, as the service will fail to start
without /etc/ctdb/nodes. If the user supplies this, they can re-enable the
service.
Signed-off-by: Christopher Larson
---
meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb | 4
1 file changed, 4 insertions(+)
diff --git a/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb
b/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb
index 232546c..5d75784 100644
--- a/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb
+++ b/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb
@@ -44,6 +44,10 @@ do_install_append() {
rm -r ${D}/${localstatedir}/run
}
+# The systemd service is disabled by default, as the service will fail to
+# start without /etc/ctdb/nodes. If the user supplies this, they can re-enable
+# the service.
+SYSTEMD_AUTO_ENABLE = "disable"
SYSTEMD_SERVICE_${PN} = "ctdb.service"
# onnode is a shell script with bashisms and bash #!
--
2.8.0
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-networking][PATCH 1/3] ctdb: drop duplicated DESCRIPTION
From: Christopher Larson Signed-off-by: Christopher Larson --- meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb | 4 1 file changed, 4 deletions(-) diff --git a/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb b/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb index d4c0f86..66c966d 100644 --- a/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb +++ b/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb @@ -2,10 +2,6 @@ DESCRIPTION = "CTDB is a cluster implementation of the TDB database \ used by Samba and other projects to store temporary data. If an \ application is already using TDB for temporary data it is very easy \ to convert that application to be cluster aware and use CTDB instead." -DESCRIPTION = "CTDB is a cluster implementation of the TDB database \ -used by Samba and other projects to store temporary data. If an \ -application is already using TDB for temporary data it is very easy \ -to convert that application to be cluster aware and use CTDB instead." HOMEPAGE = "https://ctdb.samba.org/"; LICENSE = "GPL-2.0+ & LGPL-3.0+ & GPL-3.0+" -- 2.8.0 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 3/3] ctdb: disable the service by default
From: Christopher Larson
The systemd service is disabled by default, as the service will fail to start
without /etc/ctdb/nodes. If the user supplies this, they can re-enable the
service.
Signed-off-by: Christopher Larson
---
meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb | 4
1 file changed, 4 insertions(+)
diff --git a/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb
b/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb
index 232546c..5d75784 100644
--- a/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb
+++ b/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb
@@ -44,6 +44,10 @@ do_install_append() {
rm -r ${D}/${localstatedir}/run
}
+# The systemd service is disabled by default, as the service will fail to
+# start without /etc/ctdb/nodes. If the user supplies this, they can re-enable
+# the service.
+SYSTEMD_AUTO_ENABLE = "disable"
SYSTEMD_SERVICE_${PN} = "ctdb.service"
# onnode is a shell script with bashisms and bash #!
--
2.8.0
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 1/3] ctdb: drop duplicated DESCRIPTION
From: Christopher Larson Signed-off-by: Christopher Larson --- meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb | 4 1 file changed, 4 deletions(-) diff --git a/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb b/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb index d4c0f86..66c966d 100644 --- a/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb +++ b/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb @@ -2,10 +2,6 @@ DESCRIPTION = "CTDB is a cluster implementation of the TDB database \ used by Samba and other projects to store temporary data. If an \ application is already using TDB for temporary data it is very easy \ to convert that application to be cluster aware and use CTDB instead." -DESCRIPTION = "CTDB is a cluster implementation of the TDB database \ -used by Samba and other projects to store temporary data. If an \ -application is already using TDB for temporary data it is very easy \ -to convert that application to be cluster aware and use CTDB instead." HOMEPAGE = "https://ctdb.samba.org/"; LICENSE = "GPL-2.0+ & LGPL-3.0+ & GPL-3.0+" -- 2.8.0 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 2/3] ctdb: rdepend on procps
From: Christopher Larson
ctdbd_wrapper requires pgrep.
Signed-off-by: Christopher Larson
---
meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb | 3 +++
1 file changed, 3 insertions(+)
diff --git a/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb
b/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb
index 66c966d..232546c 100644
--- a/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb
+++ b/meta-networking/recipes-support/ctdb/ctdb_2.5.1.bb
@@ -28,6 +28,9 @@ PARALLEL_MAKE = ""
DEPENDS += "popt libtevent libtalloc libldb"
+# ctdbd_wrapper requires pgrep, hence procps
+RDEPENDS_${PN} += "procps"
+
do_configure() {
oe_runconf
}
--
2.8.0
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH] fio: Remove --disable-static from EXTRA_OECONF
I think the convention is to add such workarounds to no-static-libs.inc,
not the recipe, at this time.
On Thu, Feb 18, 2016 at 10:30 AM Fabio Berton
wrote:
> As of commit OE-Core:773c9e18071d71454473dd81aff911104a2e9bc6
> EXTRA_OECONF is appended with the option --disable-static on
> DISABLE_STATIC variable and this cause the error:
>
> DEBUG: Python function sysroot_cleansstate finished
> DEBUG: Executing shell function do_configure
> Bad option --disable-static
> --cpu= Specify target CPU if auto-detect fails
> --cc= Specify compiler to use
> --extra-cflags=Specify extra CFLAGS to pass to compiler
> --build-32bit-win Enable 32-bit build on Windows
> --build-static Build a static fio
> --esx Configure build options for esx
> --enable-gfio Enable building of gtk gfio
> --disable-numa Disable libnuma even if found
> --enable-libhdfs Enable hdfs support
> WARNING: exit code 1 from a shell command.
>
> So, we need to disable this option.
>
> Signed-off-by: Fabio Berton
> ---
> meta-oe/recipes-benchmark/fio/fio_2.2.6.bb | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/meta-oe/recipes-benchmark/fio/fio_2.2.6.bb
> b/meta-oe/recipes-benchmark/fio/fio_2.2.6.bb
> index dca0e64..5bf387c 100644
> --- a/meta-oe/recipes-benchmark/fio/fio_2.2.6.bb
> +++ b/meta-oe/recipes-benchmark/fio/fio_2.2.6.bb
> @@ -25,6 +25,9 @@ SRC_URI = "git://git.kernel.dk/fio.git"
>
> S = "${WORKDIR}/git"
>
> +# avoids build breaks when using no-static-libs.inc
> +DISABLE_STATIC = ""
> +
> EXTRA_OEMAKE = "CC='${CC}' LDFLAGS='${LDFLAGS}'"
>
> do_configure() {
> --
> 2.1.4
>
> --
> ___
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-devel
>
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] Kodi 16rc recipes available
On Fri, Feb 12, 2016 at 9:23 AM Koen Kooi
wrote:
> I’ve managed to get kodi 16rc3 to build (and work!) on armv7a and x86-64.
> The recipes can be found here:
> https://github.com/koenkooi/meta-dominion/tree/master/recipes-multimedia
>
> It needs work before I can submit it properly to meta-multimedia. The
> problems I’ve spotted so far:
>
> 1) No libva in oe-core, it should move from meta-intel to oe-core.
> libva-intel-driver is fine where it is.
> 2) No vdpau in oe-core, it should move from meta-dominion to oe-core
> 3) It runs java in do_configure. Since meta-java only works one day per
> year we’ll just have to install it on the host.
> 4) OpenGLES is broken in kodi, a fork has fixes.
> 5) Wayland is broken in kodi, a fork has fixes.
> 6) No PACKAGECONFIGs yet for weird media accelerations like openmax, imx6,
> etc.
>
> and the big one for me:
>
> 7) no hts add-ons available. pvr.hts needs ${S} from kodi, but uses cmake
> instead of autotools. I might just partially copy/paste cmake*bbclass into
> kodi_git.bb to make this work. Needs more thought and less cmake.
>
> So if you have an x86-64 box with intel graphics, great, everything works
> as expected. On ARM you need working openGL, so that limits you to
> snapdragon/freedreno. That works[1], but no accelerated decode.
> If you have non-intel graphics in x86, have a look at what meta-dominion
> and meta-amd (thanks Kergoth!) do, I have it working on ATI and Nvidia
> cards locally.
>
To make sure credit goes where it's due, I didn't have anything to do with
meta-amd, that was other folks at Mentor, IIRC primarily Sean Hudson and
Drew Moseley. They've done good work on it :)
> Anyway, if people want to help out with getting the recipes upstreamed
> into the right layers: awesome. The biggest hurdle named ffmpeg was removed
> (well, added into OE, but you get what I mean) 16 days ago, so it should be
> relatively plain sailing :)
>
Nice, looks like you've done some great work on this :)
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [PATCHv2][RESEND 2/3] lowpan-tools: add missing flex/bison deps
On Tue, Jan 5, 2016 at 12:54 PM, Martin Jansa wrote: > On Tue, Jan 05, 2016 at 12:15:17PM -0700, Christopher Larson wrote: > > On Wed, Dec 9, 2015 at 10:41 AM, Christopher Larson > > wrote: > > > > > From: Christopher Larson > > > > > > Signed-off-by: Christopher Larson > > > > > > > Looks like this hasn't been applied, any objection? > > No objection from me, meta-networking patches are applied by Joe, that's > why I didn't move it from master-next. Ah, right. Not a problem, just wanted to check on status. Thanks. -- Christopher Larson kergoth at gmail dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [PATCHv2][RESEND 2/3] lowpan-tools: add missing flex/bison deps
On Wed, Dec 9, 2015 at 10:41 AM, Christopher Larson wrote: > From: Christopher Larson > > Signed-off-by: Christopher Larson > Looks like this hasn't been applied, any objection? -- Christopher Larson kergoth at gmail dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [RFC] Delete all .la files?
On Mon, Jan 4, 2016 at 10:28 AM, Burton, Ross wrote: > On 23 December 2015 at 21:41, Burton, Ross wrote: > > > So in the new year I plan on submitting a renamed version of this class > to > > oe-core and adding it to INHERIT_DISTRO. Any comments? > > > > Technicalities time: should this be a new class that is in the default > INHERIT_DISTRO? Or directly integrated into base.bbclass? Does anyone > have a strong opinion either way? I'm inclined to say separate class.. as much as I hate too much class proliferation, base.bbclass does an awful lot of stuff already, arguably too much. *shrug* -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] How to force recipe to be parsed eveey time - dynamic PV value
On Fri, Dec 18, 2015 at 4:51 AM, Kubalski, Lukasz (EXT-Espotel -
PL/Wroclaw) wrote:
> I have problem that recipe is not always parsed.
> Is it way to force it? Like checksum/signature/stamp invalidation?
>
> Problem details:
>I need to have my PV to be generated dynamically e.g like PV =
> "${@my_get_version_func(d)}"
>
> In such case recipe is not parsed every time (checksum doesn't change), PV
> is no reevaluated and even my_get_version_func would return different
> value my image is not rebuilt.
>
By default, inline python is not expanded when included in the metadata
checksumming, it's left as is. If you want the result of the calculation to
be included, you can use vardepvalue:
PV[vardepvalue] = "${PV}"
--
Christopher Larson
clarson at kergoth dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Maintainer - Tslib
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH 0/3] Add bits to save/restore URI headrevs
On Tue, Dec 15, 2015 at 12:49 PM, Martin Jansa wrote: > On Tue, Dec 15, 2015 at 12:37:44PM -0700, Christopher Larson wrote: > > From: Christopher Larson > > > > This is useful to support BB_NO_NETWORK with AUTOREV, by letting someone > ship > > dumped headrevs to the user, who then inherit this class, which ensures > that > > the cached headrevs are used, and upstream is not contacted at parse > time. > > BB_SRCREV_POLICY will be set to "cache" as well, if it's not already > set, as > > otherwise bitbake will contact upstream to update the cached values. > > > > This is helpful when shipping downloads to someone when there's a desire > to > > support both BB_NO_NETWORK and AUTOREV. > > > > The restore_headrevs.bbclass will restore dumped headrevs at config > parse time > > (add to INHERIT), and the oe.headrevs python module provides a function > one > > can call to dump the headrevs. > > Maybe I'm missing something, but what's advantage of shipping something > with AUTOREV which is then used together with BB_NO_NETWORK? > It's quite useful to not have to go through and manually lock down all your kernels recipes when doing a release. All our releases ship with sources and are BB_NO_NETWORK-capable, as that's a requirement for a lot of folks. > We already have some implementation in bitbake which dumps latest > SRCREVs used by all recipes (including AUTOREV) onces so I would expect > that the release will ship this .inc files with locked revisions and if > the user of such release wants to use BB_NO_NETWORK he will also include > that .inc file with locked revisions. > > Is it because it's easier of faster to dump them from the headrefs db > you already ship with release and you don't want to generate .inc from > buildhistory? Actually, this was in my patch queue from before buildhistory existed, and I somehow managed to forget about buildhistory-collect-srcrevs :) Feel free to ignore this, in that case. -- Christopher Larson kergoth at gmail dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 2/3] oe.headrevs: save and restore the bitbake URI headrevs
From: Christopher Larson This module provides functions to save and restore the BB_URI_HEADREVS from/to the persist_data database. The BB_URI_HEADREVS are the mappings between git refs and the revisions they refer to. This is useful to fully support BB_NO_NETWORK when AUTOREV is involved, as we can use the dumped mappings rather than contacting upstream at parse time. Signed-off-by: Christopher Larson --- meta-oe/lib/oe/headrevs.py | 38 ++ 1 file changed, 38 insertions(+) create mode 100644 meta-oe/lib/oe/headrevs.py diff --git a/meta-oe/lib/oe/headrevs.py b/meta-oe/lib/oe/headrevs.py new file mode 100644 index 000..9245300 --- /dev/null +++ b/meta-oe/lib/oe/headrevs.py @@ -0,0 +1,38 @@ +"""Save and restore the bitbake URI headrevs. + +This module provides functions to save and restore the BB_URI_HEADREVS from/to +the persist_data database. The BB_URI_HEADREVS are the mappings between git +refs and the revisions they refer to. + +This is useful to fully support BB_NO_NETWORK when AUTOREV is involved, as we +can use the dumped mappings rather than contacting upstream at parse time. See +also restore_headrevs.bbclass. + +By default, any existing mapping will not be overwritten, as we assume the +user's mappings are more current than any dumped content. +""" + + +def copy_persist_domain(d, domain, other_db_path, restore=False, overwrite=False): +import contextlib + +source_table = bb.persist_data.persist(domain, d) +dest_table = bb.persist_data.SQLTable(other_db_path, domain) +if restore: +source_table, dest_table = dest_table, source_table + +with contextlib.nested(source_table, dest_table): +if overwrite: +dest_table.update(source_table) +else: +for key in source_table: +if key not in dest_table: +dest_table[key] = source_table[key] + + +def dump_headrevs(d, dump_db_path): +copy_persist_domain(d, 'BB_URI_HEADREVS', dump_db_path) + + +def restore_headrevs(d, dump_db_path): +copy_persist_domain(d, 'BB_URI_HEADREVS', dump_db_path, restore=True) -- 2.2.1 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 0/3] Add bits to save/restore URI headrevs
From: Christopher Larson This is useful to support BB_NO_NETWORK with AUTOREV, by letting someone ship dumped headrevs to the user, who then inherit this class, which ensures that the cached headrevs are used, and upstream is not contacted at parse time. BB_SRCREV_POLICY will be set to "cache" as well, if it's not already set, as otherwise bitbake will contact upstream to update the cached values. This is helpful when shipping downloads to someone when there's a desire to support both BB_NO_NETWORK and AUTOREV. The restore_headrevs.bbclass will restore dumped headrevs at config parse time (add to INHERIT), and the oe.headrevs python module provides a function one can call to dump the headrevs. The following changes since commit 62bfd1f93f8873611e818c7cc8c13a761d629502: wireshark: update to version 2.0.0 (2015-11-30 14:39:59 -0500) are available in the git repository at: git://github.com/kergoth/meta-openembedded.git headrevs https://github.com/kergoth/meta-openembedded/tree/headrevs Christopher Larson (3): oe: add python namespace package oe.headrevs: save and restore the bitbake URI headrevs restore_headrevs.bbclass: add meta-oe/classes/restore_headrevs.bbclass | 32 +++ meta-oe/lib/oe/__init__.py | 2 ++ meta-oe/lib/oe/headrevs.py | 38 3 files changed, 72 insertions(+) create mode 100644 meta-oe/classes/restore_headrevs.bbclass create mode 100644 meta-oe/lib/oe/__init__.py create mode 100644 meta-oe/lib/oe/headrevs.py -- 2.2.1 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 3/3] restore_headrevs.bbclass: add
From: Christopher Larson
If BB_NO_NETWORK=1, this class restores saved bitbake URI headrevs.
This is useful to support BB_NO_NETWORK with AUTOREV, by letting someone ship
dumped headrevs to the user, who then inherit this class, which ensures that
the cached headrevs are used, and upstream is not contacted at parse time.
BB_SRCREV_POLICY will be set to "cache" as well, if it's not already set, as
otherwise bitbake will contact upstream to update the cached values.
This only handles restoring dumped headrevs. Using the python module to dump
to this database will likely be done at release time, elsewhere.
Signed-off-by: Christopher Larson
---
meta-oe/classes/restore_headrevs.bbclass | 32
1 file changed, 32 insertions(+)
create mode 100644 meta-oe/classes/restore_headrevs.bbclass
diff --git a/meta-oe/classes/restore_headrevs.bbclass
b/meta-oe/classes/restore_headrevs.bbclass
new file mode 100644
index 000..0d83428
--- /dev/null
+++ b/meta-oe/classes/restore_headrevs.bbclass
@@ -0,0 +1,32 @@
+# If BB_NO_NETWORK=1, this class restores saved bitbake URI headrevs.
+#
+# This is useful to support BB_NO_NETWORK with AUTOREV, by letting someone
+# ship dumped headrevs to the user, who then inherit this class, which ensures
+# that the cached headrevs are used, and upstream is not contacted at parse
+# time. BB_SRCREV_POLICY will be set to "cache" as well, if it's not already
+# set, as otherwise bitbake will contact upstream to update the cached values.
+#
+# See also lib/oe/headrevs.py.
+#
+# This only handles restoring dumped headrevs. Using the python module to dump
+# to this database will likely be done at release time, elsewhere.
+
+HEADREVS_RESTORE_STAMP ?= "${PERSISTENT_DIR}/headrevs_restored"
+DUMPED_HEADREVS_DB ?= "${COREBASE}/headrevs.db"
+
+python restore_dumped_headrevs() {
+dump_db_path = d.getVar('DUMPED_HEADREVS_DB', True)
+if os.path.exists(dump_db_path) and d.getVar('BB_NO_NETWORK', True) == '1':
+if 'BB_SRCREV_POLICY' not in d:
+d.setVar('BB_SRCREV_POLICY', 'cache')
+
+stamp_path = d.getVar('HEADREVS_RESTORE_STAMP', True)
+if (not os.path.exists(stamp_path) or
+os.path.getmtime(dump_db_path) > os.path.getmtime(stamp_path)):
+import oe.headrevs
+oe.headrevs.restore_headrevs(d, dump_db_path)
+bb.utils.mkdirhier(os.path.dirname(stamp_path))
+open(stamp_path, 'w').close()
+}
+restore_dumped_headrevs[eventmask] = "bb.event.ConfigParsed"
+addhandler restore_dumped_headrevs
--
2.2.1
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 1/3] oe: add python namespace package
From: Christopher Larson Signed-off-by: Christopher Larson --- meta-oe/lib/oe/__init__.py | 2 ++ 1 file changed, 2 insertions(+) create mode 100644 meta-oe/lib/oe/__init__.py diff --git a/meta-oe/lib/oe/__init__.py b/meta-oe/lib/oe/__init__.py new file mode 100644 index 000..3ad9513 --- /dev/null +++ b/meta-oe/lib/oe/__init__.py @@ -0,0 +1,2 @@ +from pkgutil import extend_path +__path__ = extend_path(__path__, __name__) -- 2.2.1 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [PATCHv2][RESEND 1/3] canutils: rdep on iproute2
On Wed, Dec 9, 2015 at 10:41 AM, Christopher Larson wrote: > From: Christopher Larson > > busybox ip fails to configure can interfaces, so we need iproute2 to do so. > See also http://www.armadeus.com/wiki/index.php?title=CAN_bus_Linux_driver > . > > Signed-off-by: Christopher Larson > Note: dropped the talloc/tdb/tevent attr change from this little series, since Jens Rehsack is working on that. -- Christopher Larson kergoth at gmail dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [PATCHv2][RESEND 1/3] canutils: rdep on iproute2
On Wed, Dec 9, 2015 at 10:41 AM, Christopher Larson wrote: > From: Christopher Larson > > busybox ip fails to configure can interfaces, so we need iproute2 to do so. > See also http://www.armadeus.com/wiki/index.php?title=CAN_bus_Linux_driver > . > > Signed-off-by: Christopher Larson > Note: dropped the talloc/tdb/tevent attr change from this little series, since Jens Rehsack is working on that. -- Christopher Larson kergoth at gmail dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [PATCHv2][RESEND 2/3] lowpan-tools: add missing flex/bison deps
From: Christopher Larson
Signed-off-by: Christopher Larson
---
meta-networking/recipes-support/lowpan-tools/lowpan-tools_git.bb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta-networking/recipes-support/lowpan-tools/lowpan-tools_git.bb
b/meta-networking/recipes-support/lowpan-tools/lowpan-tools_git.bb
index f57df5d..b1bd926 100644
--- a/meta-networking/recipes-support/lowpan-tools/lowpan-tools_git.bb
+++ b/meta-networking/recipes-support/lowpan-tools/lowpan-tools_git.bb
@@ -5,7 +5,7 @@ SECTION = "net"
LICENSE = "GPLv2"
LIC_FILES_CHKSUM = "file://COPYING;md5=751419260aa954499f7abaabaa882bbe"
-DEPENDS = "libnl python"
+DEPENDS = "flex-native bison-native libnl python"
PV = "0.3.1+git${SRCPV}"
SRC_URI = "git://git.code.sf.net/p/linux-zigbee/linux-zigbee \
--
2.2.1
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [PATCHv2][RESEND 3/3] gpsd: also support bluez5
From: Christopher Larson
gpsd uses libbluetooth, not the dbus interface, so it can work with bluez5 as
well, as the library is compatible.
Signed-off-by: Christopher Larson
---
meta-oe/recipes-navigation/gpsd/gpsd_3.14.bb | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/meta-oe/recipes-navigation/gpsd/gpsd_3.14.bb
b/meta-oe/recipes-navigation/gpsd/gpsd_3.14.bb
index dae39c2..6ae2f30 100644
--- a/meta-oe/recipes-navigation/gpsd/gpsd_3.14.bb
+++ b/meta-oe/recipes-navigation/gpsd/gpsd_3.14.bb
@@ -34,8 +34,8 @@ SYSTEMD_OESCONS = "${@base_contains('DISTRO_FEATURES',
'systemd', 'true', 'false
export STAGING_INCDIR
export STAGING_LIBDIR
-PACKAGECONFIG ??= "qt ${@base_contains('DISTRO_FEATURES', 'bluetooth',
'${BLUEZ}', '', d)}"
-PACKAGECONFIG[bluez4] = "bluez='true',bluez='false',bluez4"
+PACKAGECONFIG ??= "qt ${@base_contains('DISTRO_FEATURES', 'bluetooth',
'bluez', '', d)}"
+PACKAGECONFIG[bluez] = "bluez='true',bluez='false',${BLUEZ}"
PACKAGECONFIG[qt] = "qt='yes',qt='no',qt4-x11-free"
EXTRA_OESCONS = " \
sysroot=${STAGING_DIR_TARGET} \
--
2.2.1
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [PATCHv2][RESEND 1/3] canutils: rdep on iproute2
From: Christopher Larson
busybox ip fails to configure can interfaces, so we need iproute2 to do so.
See also http://www.armadeus.com/wiki/index.php?title=CAN_bus_Linux_driver.
Signed-off-by: Christopher Larson
---
meta-oe/recipes-extended/socketcan/canutils_4.0.6.bb | 3 +++
1 file changed, 3 insertions(+)
diff --git a/meta-oe/recipes-extended/socketcan/canutils_4.0.6.bb
b/meta-oe/recipes-extended/socketcan/canutils_4.0.6.bb
index c220ead..e1508af 100644
--- a/meta-oe/recipes-extended/socketcan/canutils_4.0.6.bb
+++ b/meta-oe/recipes-extended/socketcan/canutils_4.0.6.bb
@@ -15,3 +15,6 @@ SRC_URI =
"git://git.pengutronix.de/git/tools/canutils.git;protocol=git \
S = "${WORKDIR}/git"
inherit autotools pkgconfig
+
+# Busybox ip doesn't support can interface configuration, use the real thing
+RDEPENDS_${PN} += "iproute2"
--
2.2.1
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH] talloc: rename to libtalloc and cleanup dependency mess
On Fri, Nov 13, 2015 at 10:08 AM, Jens Rehsack wrote:
> > Am 13.11.2015 um 18:01 schrieb Christopher Larson :
> >
> > On Fri, Nov 13, 2015 at 9:56 AM, Christopher Larson
> wrote:
> > On Fri, Nov 13, 2015 at 8:53 AM, Jens Rehsack wrote:
> > To avoid errors when building dev-images (talloc-dev is missing), and to
> avoid
> > insane empty packages for that, rename talloc -> libtalloc as libtevent
> shows.
> >
> > With that, remove dependencies to attr/xattr and libbsd - unless
> explicitely
> > enabled via PACKAGECONFIG.
> >
> > Signed-off-by: Jens Rehsack
> >
> > Our recipe names should be named based on the upstream project name, not
> the library they happen to ship.
> >
> > Instead, either we could we should kill all the libtalloc bits in the
> PACKAGES var, and let the debian shlib renaming handle renaming talloc* to
> libtalloc*, or add a talloc-dev that pulls in libtalloc-dev & pytalloc-dev.
> >
> > I just realized this may sound critical, but that wasn't intended. The
> other changes look great, thanks for your work on making those dependencies
> optional, it's good stuff.
>
> In that case - please take me by the hand and explain a bit more detailed
>
> > Instead, either we could we should kill all the libtalloc bits in the
> PACKAGES var, and let the debian shlib renaming handle renaming talloc* to
> libtalloc*
>
> since I have no clue what you try to guide me.
I'm not able to reproduce any error when building an image. Add libtalloc
and dev-pkgs, and it builds just fine. You can't add talloc, as there is no
talloc package emitted.
That said, what I was suggesting was something along the lines of
https://gist.github.com/kergoth/ead0e1d2682a9e9fcd59 -- debian.bbclass will
automatically rename the ${PN}, ${PN}-dev, and ${PN}-dbg packages based on
the library soname, so will end up named libtalloc* anyway if you use it.
Since the main packages weren't emitted at all, it was pointless complexity
in the packaging to add libtalloc* explicitly.
--
Christopher Larson
kergoth at gmail dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Maintainer - Tslib
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [OE-core] Fwd: Welcome to the "Openemebdded-architecture" mailing list
On Tue, Nov 17, 2015 at 8:01 AM, Philip Balister wrote: > During OEDEM in Dublin we talked about creating a list for high level > architecture discussions about the OpenEmbedded Build System. > > Bring your big ideas to the list, discuss them, and bring the resources > to implement them. > > We did discuss list proliferation, filtering patches, etc and still > decided the best solution was a new list. > > Philip > > > Forwarded Message > Subject: Welcome to the "Openemebdded-architecture" mailing list > Typo. Openemebdded :) -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH] talloc: rename to libtalloc and cleanup dependency mess
On Fri, Nov 13, 2015 at 9:56 AM, Christopher Larson wrote: > On Fri, Nov 13, 2015 at 8:53 AM, Jens Rehsack wrote: > >> To avoid errors when building dev-images (talloc-dev is missing), and to >> avoid >> insane empty packages for that, rename talloc -> libtalloc as libtevent >> shows. >> >> With that, remove dependencies to attr/xattr and libbsd - unless >> explicitely >> enabled via PACKAGECONFIG. >> >> Signed-off-by: Jens Rehsack >> > > Our recipe names should be named based on the upstream project name, not > the library they happen to ship. > > Instead, either we could we should kill all the libtalloc bits in the > PACKAGES var, and let the debian shlib renaming handle renaming talloc* to > libtalloc*, or add a talloc-dev that pulls in libtalloc-dev & pytalloc-dev. > I just realized this may sound critical, but that wasn't intended. The other changes look great, thanks for your work on making those dependencies optional, it's good stuff. -- Christopher Larson kergoth at gmail dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH] talloc: rename to libtalloc and cleanup dependency mess
On Fri, Nov 13, 2015 at 8:53 AM, Jens Rehsack wrote: > To avoid errors when building dev-images (talloc-dev is missing), and to > avoid > insane empty packages for that, rename talloc -> libtalloc as libtevent > shows. > > With that, remove dependencies to attr/xattr and libbsd - unless > explicitely > enabled via PACKAGECONFIG. > > Signed-off-by: Jens Rehsack > Our recipe names should be named based on the upstream project name, not the library they happen to ship. Instead, either we could we should kill all the libtalloc bits in the PACKAGES var, and let the debian shlib renaming handle renaming talloc* to libtalloc*, or add a talloc-dev that pulls in libtalloc-dev & pytalloc-dev. -- Christopher Larson kergoth at gmail dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH 2/5] talloc, libtevent, libtdb: depend on attr
On Fri, Nov 13, 2015 at 2:38 AM, Jens Rehsack wrote:
> > Am 12.11.2015 um 21:17 schrieb Christopher Larson :
> >
> > From: Christopher Larson
> >
> > These will depend on libattr if it's available, and we need deterministic
> > builds. This fixes a build-deps failure which occurs when attr is built
> before
> > them. There's currently no configuration option to explicitly control
> this
> > dependency, and I don't know enough about waf to add such an option, so
> > unconditionally dep on it for now.
> >
> > Signed-off-by: Christopher Larson
> > ---
> > meta-oe/recipes-support/libtdb/libtdb_1.3.0.bb| 1 +
> > meta-oe/recipes-support/libtevent/libtevent_0.9.21.bb | 2 +-
> > meta-oe/recipes-support/talloc/talloc_2.1.1.bb| 1 +
> > 3 files changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/meta-oe/recipes-support/libtdb/libtdb_1.3.0.bb
> b/meta-oe/recipes-support/libtdb/libtdb_1.3.0.bb
> > index 0b9e1f2..e349187 100644
> > --- a/meta-oe/recipes-support/libtdb/libtdb_1.3.0.bb
> > +++ b/meta-oe/recipes-support/libtdb/libtdb_1.3.0.bb
> > @@ -1,6 +1,7 @@
> > SUMMARY = "The tdb library"
> > SECTION = "libs"
> > LICENSE = "LGPL-3.0+ & GPL-3.0+"
> > +DEPENDS += "attr"
> >
> > LIC_FILES_CHKSUM =
> "file://${COREBASE}/meta/files/common-licenses/LGPL-3.0;md5=bfccfe952269fff2b407dd11f2f3083b
> \
> >
>
> file://${COREBASE}/meta/files/common-licenses/GPL-3.0;md5=c79ff39f19dfec6d293b95dea7b07891"
> > diff --git a/meta-oe/recipes-support/libtevent/libtevent_0.9.21.bb
> b/meta-oe/recipes-support/libtevent/libtevent_0.9.21.bb
> > index f297f30..c37eeb3 100644
> > --- a/meta-oe/recipes-support/libtevent/libtevent_0.9.21.bb
> > +++ b/meta-oe/recipes-support/libtevent/libtevent_0.9.21.bb
> > @@ -3,7 +3,7 @@ HOMEPAGE = "http://tevent.samba.org";
> > SECTION = "libs"
> > LICENSE = "LGPLv3+"
> >
> > -DEPENDS += "talloc libcap"
> > +DEPENDS += "talloc libcap attr"
> > RDEPENDS_${PN} += "libtalloc"
> > RDEPENDS_python-tevent = "python"
> >
> > diff --git a/meta-oe/recipes-support/talloc/talloc_2.1.1.bb
> b/meta-oe/recipes-support/talloc/talloc_2.1.1.bb
> > index 89c390f..7931374 100644
> > --- a/meta-oe/recipes-support/talloc/talloc_2.1.1.bb
> > +++ b/meta-oe/recipes-support/talloc/talloc_2.1.1.bb
> > @@ -2,6 +2,7 @@ SUMMARY = "Hierarchical, reference counted memory pool
> system with destructors"
> > HOMEPAGE = "http://talloc.samba.org";
> > SECTION = "libs"
> > LICENSE = "LGPL-3.0+ & GPL-3.0+"
> > +DEPENDS += "attr"
> >
> > SRC_URI = "http://samba.org/ftp/${BPN}/${BPN}-${PV}.tar.gz";
> > LIC_FILES_CHKSUM =
> "file://${COREBASE}/meta/files/common-licenses/LGPL-3.0;md5=bfccfe952269fff2b407dd11f2f3083b
> \
> > --
> > 2.2.1
>
> RDEPENDS is wrong for deterministic builds, since they leave a race
> condition open that attr is build parallel to libtevent
> and 2nd build is different to first build.
>
> If you want, I handle that better in a patchy way I discussed with JaMa
> yesterday in #yocto (patch check out depending on PACKAGECONFIG)
I don't believe anyone said anything about rdepending on attr. This patch
certainly doesn't, it adds to DEPENDS. But as was said earlier in the
thread, a way to explicitly enable/disable the dep in the underlying
buildsystem would be better.
--
Christopher Larson
kergoth at gmail dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Maintainer - Tslib
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH 2/5] talloc, libtevent, libtdb: depend on attr
On Thu, Nov 12, 2015 at 7:48 PM, Huang, Jie (Jackie) < jackie.hu...@windriver.com> wrote: > > -Original Message- > > From: openembedded-devel-boun...@lists.openembedded.org [mailto: > openembedded-devel- > > boun...@lists.openembedded.org] On Behalf Of Christopher Larson > > Sent: Friday, November 13, 2015 4:17 AM > > To: openembedded-devel@lists.openembedded.org > > Cc: Christopher Larson > > Subject: [oe] [meta-oe][PATCH 2/5] talloc,libtevent,libtdb: depend on > attr > > > > From: Christopher Larson > > > > These will depend on libattr if it's available, and we need deterministic > > builds. This fixes a build-deps failure which occurs when attr is built > before > > them. There's currently no configuration option to explicitly control > this > > dependency, and I don't know enough about waf to add such an option, so > > unconditionally dep on it for now. > > I added depends on libcap for libtevent for the same reason days ago, but > it really concern > me now, the 'libreplace' in these waf based packages check many packages(I > see > libbsd, libcap, libacl for now) existence then build with them if found, > and there > is no configure option to disable them, does anyone has a better idea how > to > make waf build in deterministic way? Maybe we should disable the built-in > libraries 'libreplace' or add a PACKAGECONFIG for it? I don't think PACKAGECONFIG will help unless we have a way to explicitly disable or enable the dependency with waf, otherwise it'd still be non-deterministic. But I think that would be ideal, that's how we handle these things in autoconf, add an argument to let us explicitly enable/disable it. Hopefully someone else on the list knows enough about waf to be able to do so. -- Christopher Larson kergoth at gmail dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 4/5] ctdb: rdep on bash
From: Christopher Larson
onnode is a shell script with bashisms and bash #!.
Signed-off-by: Christopher Larson
---
meta-oe/recipes-support/ctdb/ctdb_2.5.1.bb | 3 +++
1 file changed, 3 insertions(+)
diff --git a/meta-oe/recipes-support/ctdb/ctdb_2.5.1.bb
b/meta-oe/recipes-support/ctdb/ctdb_2.5.1.bb
index 3731efd..5718c5f 100644
--- a/meta-oe/recipes-support/ctdb/ctdb_2.5.1.bb
+++ b/meta-oe/recipes-support/ctdb/ctdb_2.5.1.bb
@@ -41,3 +41,6 @@ do_install_append() {
SYSTEMD_SERVICE_${PN} = "ctdb.service"
FILES_${PN} += "/run"
+
+# onnode is a shell script with bashisms and bash #!
+RDEPENDS_${PN} += "bash"
--
2.2.1
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 2/5] talloc,libtevent,libtdb: depend on attr
From: Christopher Larson
These will depend on libattr if it's available, and we need deterministic
builds. This fixes a build-deps failure which occurs when attr is built before
them. There's currently no configuration option to explicitly control this
dependency, and I don't know enough about waf to add such an option, so
unconditionally dep on it for now.
Signed-off-by: Christopher Larson
---
meta-oe/recipes-support/libtdb/libtdb_1.3.0.bb| 1 +
meta-oe/recipes-support/libtevent/libtevent_0.9.21.bb | 2 +-
meta-oe/recipes-support/talloc/talloc_2.1.1.bb| 1 +
3 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/meta-oe/recipes-support/libtdb/libtdb_1.3.0.bb
b/meta-oe/recipes-support/libtdb/libtdb_1.3.0.bb
index 0b9e1f2..e349187 100644
--- a/meta-oe/recipes-support/libtdb/libtdb_1.3.0.bb
+++ b/meta-oe/recipes-support/libtdb/libtdb_1.3.0.bb
@@ -1,6 +1,7 @@
SUMMARY = "The tdb library"
SECTION = "libs"
LICENSE = "LGPL-3.0+ & GPL-3.0+"
+DEPENDS += "attr"
LIC_FILES_CHKSUM =
"file://${COREBASE}/meta/files/common-licenses/LGPL-3.0;md5=bfccfe952269fff2b407dd11f2f3083b
\
file://${COREBASE}/meta/files/common-licenses/GPL-3.0;md5=c79ff39f19dfec6d293b95dea7b07891"
diff --git a/meta-oe/recipes-support/libtevent/libtevent_0.9.21.bb
b/meta-oe/recipes-support/libtevent/libtevent_0.9.21.bb
index f297f30..c37eeb3 100644
--- a/meta-oe/recipes-support/libtevent/libtevent_0.9.21.bb
+++ b/meta-oe/recipes-support/libtevent/libtevent_0.9.21.bb
@@ -3,7 +3,7 @@ HOMEPAGE = "http://tevent.samba.org";
SECTION = "libs"
LICENSE = "LGPLv3+"
-DEPENDS += "talloc libcap"
+DEPENDS += "talloc libcap attr"
RDEPENDS_${PN} += "libtalloc"
RDEPENDS_python-tevent = "python"
diff --git a/meta-oe/recipes-support/talloc/talloc_2.1.1.bb
b/meta-oe/recipes-support/talloc/talloc_2.1.1.bb
index 89c390f..7931374 100644
--- a/meta-oe/recipes-support/talloc/talloc_2.1.1.bb
+++ b/meta-oe/recipes-support/talloc/talloc_2.1.1.bb
@@ -2,6 +2,7 @@ SUMMARY = "Hierarchical, reference counted memory pool system
with destructors"
HOMEPAGE = "http://talloc.samba.org";
SECTION = "libs"
LICENSE = "LGPL-3.0+ & GPL-3.0+"
+DEPENDS += "attr"
SRC_URI = "http://samba.org/ftp/${BPN}/${BPN}-${PV}.tar.gz";
LIC_FILES_CHKSUM =
"file://${COREBASE}/meta/files/common-licenses/LGPL-3.0;md5=bfccfe952269fff2b407dd11f2f3083b
\
--
2.2.1
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 3/5] lowpan-tools: add missing flex/bison deps
From: Christopher Larson
Signed-off-by: Christopher Larson
---
meta-networking/recipes-support/lowpan-tools/lowpan-tools_git.bb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta-networking/recipes-support/lowpan-tools/lowpan-tools_git.bb
b/meta-networking/recipes-support/lowpan-tools/lowpan-tools_git.bb
index f57df5d..b1bd926 100644
--- a/meta-networking/recipes-support/lowpan-tools/lowpan-tools_git.bb
+++ b/meta-networking/recipes-support/lowpan-tools/lowpan-tools_git.bb
@@ -5,7 +5,7 @@ SECTION = "net"
LICENSE = "GPLv2"
LIC_FILES_CHKSUM = "file://COPYING;md5=751419260aa954499f7abaabaa882bbe"
-DEPENDS = "libnl python"
+DEPENDS = "flex-native bison-native libnl python"
PV = "0.3.1+git${SRCPV}"
SRC_URI = "git://git.code.sf.net/p/linux-zigbee/linux-zigbee \
--
2.2.1
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 5/5] gpsd: also support bluez5
From: Christopher Larson
gpsd uses libbluetooth, not the dbus interface, so it can work with bluez5 as
well, as the library is compatible.
Signed-off-by: Christopher Larson
---
meta-oe/recipes-navigation/gpsd/gpsd_3.14.bb | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/meta-oe/recipes-navigation/gpsd/gpsd_3.14.bb
b/meta-oe/recipes-navigation/gpsd/gpsd_3.14.bb
index dae39c2..6ae2f30 100644
--- a/meta-oe/recipes-navigation/gpsd/gpsd_3.14.bb
+++ b/meta-oe/recipes-navigation/gpsd/gpsd_3.14.bb
@@ -34,8 +34,8 @@ SYSTEMD_OESCONS = "${@base_contains('DISTRO_FEATURES',
'systemd', 'true', 'false
export STAGING_INCDIR
export STAGING_LIBDIR
-PACKAGECONFIG ??= "qt ${@base_contains('DISTRO_FEATURES', 'bluetooth',
'${BLUEZ}', '', d)}"
-PACKAGECONFIG[bluez4] = "bluez='true',bluez='false',bluez4"
+PACKAGECONFIG ??= "qt ${@base_contains('DISTRO_FEATURES', 'bluetooth',
'bluez', '', d)}"
+PACKAGECONFIG[bluez] = "bluez='true',bluez='false',${BLUEZ}"
PACKAGECONFIG[qt] = "qt='yes',qt='no',qt4-x11-free"
EXTRA_OESCONS = " \
sysroot=${STAGING_DIR_TARGET} \
--
2.2.1
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH 1/5] canutils: rdep on iproute2
From: Christopher Larson
busybox ip fails to configure can interfaces, so we need iproute2 to do so.
See also http://www.armadeus.com/wiki/index.php?title=CAN_bus_Linux_driver.
Signed-off-by: Christopher Larson
---
meta-oe/recipes-extended/socketcan/canutils_4.0.6.bb | 3 +++
1 file changed, 3 insertions(+)
diff --git a/meta-oe/recipes-extended/socketcan/canutils_4.0.6.bb
b/meta-oe/recipes-extended/socketcan/canutils_4.0.6.bb
index c220ead..e1508af 100644
--- a/meta-oe/recipes-extended/socketcan/canutils_4.0.6.bb
+++ b/meta-oe/recipes-extended/socketcan/canutils_4.0.6.bb
@@ -15,3 +15,6 @@ SRC_URI =
"git://git.pengutronix.de/git/tools/canutils.git;protocol=git \
S = "${WORKDIR}/git"
inherit autotools pkgconfig
+
+# Busybox ip doesn't support can interface configuration, use the real thing
+RDEPENDS_${PN} += "iproute2"
--
2.2.1
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH 1/3] samba: fix dependencies and QA issues
On Tue, Nov 10, 2015 at 2:22 PM, Jens Rehsack wrote:
> + ${@base_contains('DISTRO_FEATURES', 'sysvinit', 'lsb',
> '', d)} \
>
This really requires lsb to able build, for all sysvinit distros? That
seems unlikely. Was this intended to be based on the lsb distro feature?
Also, why add lsb to rdepends independently of the packageconfig? I'd think
using the packageconfig's 4th field for rdepends, if necessary, would work
better, unless I'm missing something?
--
Christopher Larson
clarson at kergoth dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Maintainer - Tslib
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH] samba: systemv-init script must be executable
On Wed, Nov 4, 2015 at 11:09 AM, Jens Rehsack wrote: > Fix insane permissions installing SystemV init-script for samba4 to allow > executing it correctly. > > Signed-off-by: Jens Rehsack > The /etc/rc our sysvinit package provides will source the startup script rather than executing it if its name ends in .sh, so I don't really see why this would be necessary. -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH] talloc: remove default packages from the packages list since empty
On Sat, Oct 3, 2015 at 11:30 AM, Andrei Kholodnyi <
andrei.kholod...@gmail.com> wrote:
> populate_sdk fails with following message:
> error: talloc-dev-2.1.1-r0 requires talloc = 2.1.1-r0
>
> these packages are empty,remove them from the package list
> ---
> meta-oe/recipes-support/talloc/talloc_2.1.1.bb | 5 +
> 1 file changed, 1 insertion(+), 4 deletions(-)
>
> diff --git a/meta-oe/recipes-support/talloc/talloc_2.1.1.bb
> b/meta-oe/recipes-support/talloc/talloc_2.1.1.bb
> index 89c390f..0c98bb6 100644
> --- a/meta-oe/recipes-support/talloc/talloc_2.1.1.bb
> +++ b/meta-oe/recipes-support/talloc/talloc_2.1.1.bb
> @@ -20,12 +20,9 @@ EXTRA_OECONF += "--disable-rpath \
> --with-libiconv=${STAGING_DIR_HOST}${prefix}\
> "
>
> -PACKAGES += "libtalloc libtalloc-dbg libtalloc-dev pytalloc pytalloc-dbg
> pytalloc-dev"
> +PACKAGES = "libtalloc libtalloc-dbg libtalloc-dev pytalloc pytalloc-dbg
> pytalloc-dev"
>
> #ALLOW_EMPTY_${PN} = "1"
> -FILES_${PN} = ""
> -FILES_${PN}-dev = ""
> -FILES_${PN}-dbg = ""
>
Question, why does this use libtalloc as the package names rather than the
default of talloc? In the common case of the debian bbclass inherited,
they'll get renamed based on the library soname anyway. Also, overriding
the PACKAGES list isn't ideal, since it means future changes to the default
value of PACKAGES will not be picked up here. Also, this packaging is worse
than default, as it doesn't have the -staticdev package.
--
Christopher Larson
clarson at kergoth dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Maintainer - Tslib
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [PATCH] plymouth: building documatation sometimes fails to download so disable it
On Sun, Sep 20, 2015 at 8:14 AM, Khem Raj wrote:
> On Sun, Sep 20, 2015 at 8:07 AM, Andreas Müller
> wrote:
> > | I/O error : Attempt to load network entity
> http://docbook.sourceforge.net/release/xsl/current/manpages/docbook.xsl
> > | warning: failed to load external entity "
> http://docbook.sourceforge.net/release/xsl/current/manpages/docbook.xsl";
> > | cannot parse
> http://docbook.sourceforge.net/release/xsl/current/manpages/docbook.xsl
> > | Makefile:622: recipe for target 'plymouth.1' failed
> >
>
> this is ok
>
> > Signed-off-by: Andreas Müller
> > ---
> > meta-oe/recipes-core/plymouth/plymouth_0.9.2.bb | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/meta-oe/recipes-core/plymouth/plymouth_0.9.2.bb
> b/meta-oe/recipes-core/plymouth/plymouth_0.9.2.bb
> > index b8c82ed..1fb4443 100644
> > --- a/meta-oe/recipes-core/plymouth/plymouth_0.9.2.bb
> > +++ b/meta-oe/recipes-core/plymouth/plymouth_0.9.2.bb
> > @@ -19,7 +19,7 @@ SRC_URI = "
> http://www.freedesktop.org/software/plymouth/releases/${BPN}-${PV}.ta
> > SRC_URI[md5sum] = "ff420994deb7ea203df678df92e7ab7d"
> > SRC_URI[sha256sum] =
> "2f0ce82042cf9c7eadd2517a1f74c8a85fa8699781d9f294a06eade29fbed57f"
> >
> > -EXTRA_OECONF += " --enable-shared --disable-static --disable-gtk \
> > +EXTRA_OECONF += " --enable-shared --disable-static --disable-gtk
> --disable-documentation\
>
At some point we really need to add packageconfigs for all the
documentation generation options and then add distro feature(s) to control
the default state of them. Not saying we need to do that now, but
eventually.. I think someone had started working on getting docbook native
bits working recently..
--
Christopher Larson
clarson at kergoth dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Maintainer - Tslib
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-networking][PATCH] netmap: fix multiple build issues.
On Tue, Sep 8, 2015 at 11:54 AM, Armin Kuster wrote: > +++ > b/meta-networking/recipes-kernel/netmap/files/0001-testmmap-fix-compile-issue-with-gcc-5.x.patch > @@ -0,0 +1,33 @@ > +upstream Status: pending > s/upstream Status: pending/Upstream-Status: Pending/ -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] Where is libunique?
On Mon, Sep 7, 2015 at 8:09 AM, Martin Jansa wrote: > On Mon, Sep 07, 2015 at 03:33:09PM +0300, Alexander Kanavin wrote: > > On 09/07/2015 12:13 PM, Mike Looijmans wrote: > > > Current meta-openembedded master fails to build with errors like this: > > > > > > ERROR: Nothing PROVIDES 'libunique' (but > > > .../meta-oe/meta-gnome/recipes-gnome/gnome-disk-utility/ > gnome-disk-utility_2.32.0.bb > > > DEPENDS on or otherwise requires it). > > > > > > Searching for 'unique' in meta-openembedded and openembedded-core > > > reveiled nothing that would provide it, there's no recipe with 'unique' > > > in its name, nor any recipe that mentions the word 'unique' provides > it, > > > even though I found various recipes that DEPEND on it. > > > > > > Where is libunique? > > > > You should check not just the current state of master branches of > > various layers, but also their commit history. > > > > libunique was just removed from oe-core and a patch was sent to add it > > to meta-oe. There is no way to keep the removal/addition perfectly in > > sync because the two layers are kept in separate repos and have > > different maintainers. > > This case can be prevented by adding it to meta-oe before it's removed > from oe-core (just like when someone is migrating recipes from meta-oe > to oe-core, they are first added to oe-core and then removed from > meta-oe). > Indeed, it should always be added before it's removed, then the worst case is duplication rather than the recipe missing. -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH] iw: remove, this is now in oe-core
From: Christopher Larson
Signed-off-by: Christopher Larson
---
sh-don-t-use-git-describe-for-versioning.patch | 43 --
meta-oe/recipes-connectivity/iw/iw_4.1.bb | 23
2 files changed, 66 deletions(-)
delete mode 100644
meta-oe/recipes-connectivity/iw/iw/0001-iw-version.sh-don-t-use-git-describe-for-versioning.patch
delete mode 100644 meta-oe/recipes-connectivity/iw/iw_4.1.bb
diff --git
a/meta-oe/recipes-connectivity/iw/iw/0001-iw-version.sh-don-t-use-git-describe-for-versioning.patch
b/meta-oe/recipes-connectivity/iw/iw/0001-iw-version.sh-don-t-use-git-describe-for-versioning.patch
deleted file mode 100644
index e64dd0a..000
---
a/meta-oe/recipes-connectivity/iw/iw/0001-iw-version.sh-don-t-use-git-describe-for-versioning.patch
+++ /dev/null
@@ -1,43 +0,0 @@
-From 5310abba864cfe3a8b65af130729447604190b29 Mon Sep 17 00:00:00 2001
-From: Koen Kooi
-Date: Tue, 29 Nov 2011 17:03:27 +0100
-Subject: [PATCH] iw: version.sh: don't use git describe for versioning
-
-It will detect top-level git repositories like the Angstrom setup-scripts and
break.
-
-Upstream-Status: Unknown
-
-Signed-off-by: Koen Kooi
- version.sh | 16 +---
- 1 file changed, 1 insertion(+), 15 deletions(-)
-
-diff --git a/version.sh b/version.sh
-index 11d124b..5d423c4 100755
a/version.sh
-+++ b/version.sh
-@@ -3,21 +3,7 @@
- VERSION="3.15"
- OUT="$1"
-
--if [ -d .git ] && head=`git rev-parse --verify HEAD 2>/dev/null`; then
-- git update-index --refresh --unmerged > /dev/null
-- descr=$(git describe)
--
-- # on git builds check that the version number above
-- # is correct...
-- [ "${descr%%-*}" = "v$VERSION" ] || exit 2
--
-- v="${descr#v}"
-- if git diff-index --name-only HEAD | read dummy ; then
-- v="$v"-dirty
-- fi
--else
-- v="$VERSION"
--fi
-+v="$VERSION"
-
- echo '#include "iw.h"' > "$OUT"
- echo "const char iw_version[] = \"$v\";" >> "$OUT"
---
-1.7.7.3
diff --git a/meta-oe/recipes-connectivity/iw/iw_4.1.bb
b/meta-oe/recipes-connectivity/iw/iw_4.1.bb
deleted file mode 100644
index e1e7c12..000
--- a/meta-oe/recipes-connectivity/iw/iw_4.1.bb
+++ /dev/null
@@ -1,23 +0,0 @@
-SUMMARY = "nl80211 based CLI configuration utility for wireless devices"
-DESCRIPTION = "iw is a new nl80211 based CLI configuration utility for \
-wireless devices. It supports almost all new drivers that have been added \
-to the kernel recently. "
-HOMEPAGE = "http://wireless.kernel.org/en/users/Documentation/iw";
-SECTION = "base"
-LICENSE = "BSD"
-LIC_FILES_CHKSUM = "file://COPYING;md5=878618a5c4af25e9b93ef0be1a93f774"
-
-DEPENDS = "libnl pkgconfig"
-
-SRC_URI = "http://www.kernel.org/pub/software/network/iw/${BP}.tar.gz \
-
file://0001-iw-version.sh-don-t-use-git-describe-for-versioning.patch \
-"
-
-SRC_URI[md5sum] = "68c282285c71c956069957e9ca10a6a7"
-SRC_URI[sha256sum] =
"14bfc627b37f7f607e4ffa63a70ded15fa2ea85177f703cb17d7fe36f9c8f33d"
-
-EXTRA_OEMAKE = ""
-
-do_install() {
-oe_runmake DESTDIR=${D} install
-}
--
2.2.1
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] How to create machine/customer specific ntp.conf files
On Thu, Jul 2, 2015 at 12:55 PM, Philip Balister wrote: > I'm going round and round thinking about creating custom ntp.conf files. > > The existing recipe includes an ntp.conf file. My question is what is > the best way to replace the on from the recipe with a machine specific > config. > > There are a couple of config issues: > > 1) hardware may want to sync with an on board gps > 2) A vendor may want to use their own ntp servers. > > Obviously, these need to apply only to specific machines and not all > builds from the same set of layers. > > Any ideas how to do this without making the entire ntp package machine > specific? > > Should we split the conf file into a separate recipe, sort of like gpsd? > Afaik that’d be the best way to handle it, iirc that’s how connman and the like handle it too. Also remember to add it to SIGGEN_EXCLUDERECIPES_ABISAFE in meta/conf/layer.conf if you go that route, so stuff depending on it doesn’t get unnecessarily rebuilt when it changes. -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [PATCH] ntimed: add recipe to meta-networking (git version)
On Mon, Jun 1, 2015 at 1:05 PM, Khem Raj wrote: > On Sat, May 30, 2015 at 5:31 PM, Christopher Larson > wrote: > > Ntimed is an unreleased ntpd replacement being sponsored by the Linux > > Foundation. Currently it only includes a work-in-progress client, but for > > future use this recipe emits an ntimed-client package and an ntimed meta > > package which will pull in client and server. > > > > should it have RCONFLICTS with ntp ? Probably, I’ll look into that as well as see about adding a service file & startup script, and re-submit after testing that. Thanks. -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [PATCH] ntimed: add recipe to meta-networking (git version)
Ntimed is an unreleased ntpd replacement being sponsored by the Linux
Foundation. Currently it only includes a work-in-progress client, but for
future use this recipe emits an ntimed-client package and an ntimed meta
package which will pull in client and server.
Signed-off-by: Christopher Larson
---
.../ntimed/ntimed/use-ldflags.patch| 16 +
.../recipes-support/ntimed/ntimed_git.bb | 41 ++
2 files changed, 57 insertions(+)
create mode 100644
meta-networking/recipes-support/ntimed/ntimed/use-ldflags.patch
create mode 100644 meta-networking/recipes-support/ntimed/ntimed_git.bb
diff --git a/meta-networking/recipes-support/ntimed/ntimed/use-ldflags.patch
b/meta-networking/recipes-support/ntimed/ntimed/use-ldflags.patch
new file mode 100644
index 000..87a7778
--- /dev/null
+++ b/meta-networking/recipes-support/ntimed/ntimed/use-ldflags.patch
@@ -0,0 +1,16 @@
+Obey LDFLAGS.
+
+Signed-off-by: Christopher Larson
+Upstream-Status: Pending
+
+--- Ntimed.orig/configure 2015-05-30 11:57:59.927796993 -0700
Ntimed/configure 2015-05-30 11:58:26.143948894 -0700
+@@ -142,7 +142,7 @@
+
+ echo
+ echo "ntimed-client:${l}"
+- echo " \${CC} \${CFLAGS} -o ntimed-client ${l} -lm"
++ echo " \${CC} \${CFLAGS} -o ntimed-client ${l} \${LDFLAGS} -lm"
+ echo
+ echo "clean:"
+ echo " rm -f ${l} ntimed-client"
diff --git a/meta-networking/recipes-support/ntimed/ntimed_git.bb
b/meta-networking/recipes-support/ntimed/ntimed_git.bb
new file mode 100644
index 000..0990541
--- /dev/null
+++ b/meta-networking/recipes-support/ntimed/ntimed_git.bb
@@ -0,0 +1,41 @@
+SUMMARY = "Network time synchronization software, NTPD replacement"
+DESCRIPTION = "This is a preview/early-access/alpha/buzzword-of-the-times \
+release of a new FOSS project written to gradually take over the world of \
+networked timekeeping."
+HOMEPAGE = "https://github.com/bsdphk/Ntimed";
+SECTION = "console/network"
+
+LICENSE = "BSD-2-Clause"
+LIC_FILES_CHKSUM =
"file://main.c;startline=2;endline=24;md5=eda11d21005319bb76cbb6f911f0f66d"
+
+SRC_URI = "git://github.com/bsdphk/Ntimed \
+ file://use-ldflags.patch"
+
+PV = "0.0+git${SRCPV}"
+SRCREV = "db0abbb4c80f2ecef6bc5d9639bca5bea28532a2"
+
+S = "${WORKDIR}/git"
+
+EXTRA_OEMAKE = "\
+'CC=${CC}' \
+'CFLAGS=${CFLAGS}' \
+'LDFLAGS=${LDFLAGS}' \
+"
+
+do_configure () {
+sh ./configure
+}
+
+do_compile () {
+oe_runmake
+}
+
+do_install () {
+install -D -m 0755 ntimed-client ${D}${sbindir}/ntimed-client
+}
+
+ALLOW_EMPTY_${PN} = "1"
+RDEPENDS_${PN} += "ntimed-client"
+
+PACKAGE_BEFORE_PN += "ntimed-client"
+FILES_ntimed-client = "${sbindir}/ntimed-client"
--
2.2.1
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] meta-selinux
On Wed, Feb 11, 2015 at 8:53 AM, dpquigl wrote: > I'm working on OpenXT and it makes use of the meta-selinux repo hosted > by the yocto project. I'm trying to use it with a base openembedded core > and its not in sync with oe-core because its based on pokey. This made > me think of two questions. 1) Why is this not in OE core since so many > packages in core can potentially have SELinux support enabled and 2) if > its not supposed to be in core where should turning on SELinux support > in a recipe go? For example coreutils can have SELinux support enabled. > Currently this is in meta-selinux as a bbappend to the coreutils > package. This works out because its always going to be there. However > there is also a bbappend for an LXC recipe. LXC isn't in core which > means it has a dependency on a layer not in core. > This is a bug in the layer. It's fairly trivial to construct a layer in such a way that you can have per-layer bbappends that are only applied when that layer exists. This is likely the approach meta-selinux should take to address this implicit dependency upon meta-virtualization. That said, I think most folks would be open to PACKAGECONFIGs for selinux capability going into the main recipes, as that's not an invasive change, nor a patch, but just a tweak in configuration. -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH] rng-tools: obey the sysvinit distro feature
From: Christopher Larson
Signed-off-by: Christopher Larson
---
meta-oe/recipes-support/rng-tools/rng-tools_4.bb | 22 --
1 file changed, 16 insertions(+), 6 deletions(-)
diff --git a/meta-oe/recipes-support/rng-tools/rng-tools_4.bb
b/meta-oe/recipes-support/rng-tools/rng-tools_4.bb
index f10fad1..0b40fc5 100644
--- a/meta-oe/recipes-support/rng-tools/rng-tools_4.bb
+++ b/meta-oe/recipes-support/rng-tools/rng-tools_4.bb
@@ -10,16 +10,26 @@ SRC_URI =
"http://heanet.dl.sourceforge.net/sourceforge/gkernel/${BP}.tar.gz \
SRC_URI[md5sum] = "ae89dbfcf08bdfbea19066cfbf599127"
SRC_URI[sha256sum] =
"b71bdfd4222c05e8316001556be90e1606f2a1bac3efde60153bd84e873cc195"
+# As the recipe doesn't inherit systemd.bbclass, we need to set this variable
+# manually to avoid unnecessary postinst/preinst generated.
+python () {
+if not bb.utils.contains('DISTRO_FEATURES', 'sysvinit', True, False, d):
+d.setVar("INHIBIT_UPDATERCD_BBCLASS", "1")
+}
+
inherit autotools update-rc.d
do_install_append() {
-install -d "${D}${sysconfdir}/init.d"
-install -m 0755 ${WORKDIR}/init ${D}${sysconfdir}/init.d/rng-tools
-sed -i -e 's,/etc/,${sysconfdir}/,' -e 's,/usr/sbin/,${sbindir},' \
-${D}${sysconfdir}/init.d/rng-tools
+# Only install the init script when 'sysvinit' is in DISTRO_FEATURES.
+if ${@bb.utils.contains('DISTRO_FEATURES', 'sysvinit', 'true', 'false',
d)}; then
+install -d "${D}${sysconfdir}/init.d"
+install -m 0755 ${WORKDIR}/init ${D}${sysconfdir}/init.d/rng-tools
+sed -i -e 's,/etc/,${sysconfdir}/,' -e 's,/usr/sbin/,${sbindir},' \
+${D}${sysconfdir}/init.d/rng-tools
-install -d "${D}${sysconfdir}/default"
-install -m 0644 ${WORKDIR}/default ${D}${sysconfdir}/default
+install -d "${D}${sysconfdir}/default"
+install -m 0644 ${WORKDIR}/default ${D}${sysconfdir}/default
+fi
}
INITSCRIPT_NAME = "rng-tools"
--
1.8.3.4
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH v2] rarpd : add new recipe
On Tue, Nov 18, 2014 at 2:55 AM, Li xin wrote: > RARP (Reverse Address Resolution Protocol) is a protocol which allows > individual devices on an IP network to get their own IP addresses from > the RARP server. Some machines (e.g. SPARC boxes) use this protocol > instead of e.g. DHCP to query their IP addresses during network bootup. > > Linux kernels up to 2.2 used to provide a kernel daemon for this > service,but since 2.3 kernels it is served by this userland daemon. > You should install rarpd if you want to set up a RARP server on your > network. > > Signed-off-by: Li Xin > Should this go in meta-networking? -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [OE-core] RFC: automatic -dbg splitting
On Tue, Oct 7, 2014 at 3:25 AM, Burton, Ross wrote:
> On 11 June 2014 20:10, Burton, Ross wrote:
> > I know there are recipes floating around that have multiple -dbg
> > packages, but I can't see a great rationale for that. I also can't
> > see a reason why you'd want debug objects/sources outside of the -dbg
> > package, or something that wasn't a debug object or source inside the
> > -dbg package.
>
> So last night I resurrected this patch and did an oe-core world build
> with a small change to clear FILES_${PN}-dbg before automatically
> populating it. The results were interesting and invalidated my two
> assumptions:
>
> 1) Adding files that are not debug symbols or sources to -dbg packages
> is done: both gcc and glib add Python code for gdb to execute and the
> -dbg package is the obvious location for it.
>
> 2) A recipe having multiple -dbg packages is sensible for eg Qt4,
> which if all the debug information is put into a single package
> results in a 367M package, expanding to 1.1G on disk.
>
> So I've a proposed change to the logic:
> * If there are multiple names in PACKAGES that match *-dbg, then don't
> do automatic debug packaging
> * The default value for FILES_PN-dbg in bitbake.conf should be empty,
> and recipes can extend it as required (e.g. gdb Python scripts) with
> the automatic debug packaging extending FILES_PN-dbg.
For what it's worth, this behavior seems quite sane to me.
--
Christopher Larson
clarson at kergoth dot com
Founder - BitBake, OpenEmbedded, OpenZaurus
Maintainer - Tslib
Senior Software Engineer, Mentor Graphics
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH] lmbench: add lmbench-exception LICENSE
On Mon, Aug 25, 2014 at 11:54 AM, Khem Raj wrote: > On 14-08-25 23:38:31, Yasir-Khan wrote: > > Signed-off-by: Christopher Larson > > Signed-off-by: Yasir-Khan > > --- > > meta-oe/licenses/GPL-2.0-with-lmbench-restriction | 108 > > > Is there some SPDX compliant naming convention for this Afaik this is already the convention, see e.g. GPL-2.0-with-bison-exception on https://spdx.org/licenses/. Of course, this specific exception isn't in the SPDX license list, but it follows suit. -Chris -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-oe][PATCH] lmbench: if a machine config file exists, install it
On Tue, Aug 19, 2014 at 5:07 AM, Martin Jansa
wrote:
> On Tue, Aug 19, 2014 at 04:49:24PM +0500, Yasir-Khan wrote:
> > Signed-off-by: Christopher Larson
> > Signed-off-by: Yasir-Khan
> > ---
> > .../recipes-benchmark/lmbench/lmbench_3.0-a9.bb|7 +++
> > 1 file changed, 7 insertions(+)
> >
> > diff --git a/meta-oe/recipes-benchmark/lmbench/lmbench_3.0-a9.bb
> b/meta-oe/recipes-benchmark/lmbench/lmbench_3.0-a9.bb
> > index 41db150..a3873a6 100644
> > --- a/meta-oe/recipes-benchmark/lmbench/lmbench_3.0-a9.bb
> > +++ b/meta-oe/recipes-benchmark/lmbench/lmbench_3.0-a9.bb
> > @@ -60,6 +60,13 @@ do_install () {
> > ${D}${bindir}/lmbench-run
> > install -m 0755 ${S}/scripts/lmbench ${D}${bindir}
> > install -m 0755 ${S}/scripts/* ${D}${datadir}/lmbench/scripts
> > +
> > +# If the machine installs a config file, install it to the
> appropriate place
> > +if [ -f ${WORKDIR}/CONFIG.${MACHINE} ]; then
> > +install -D 0644 ${WORKDIR}/CONFIG.${MACHINE} \
> > +${D}/${datadir}/lmbench
> /bin/${TARGET_PREFIX}/CONFIG.${MACHINE}
> > +fi
>
> that makes lmbench MACHINE_ARCH and PACKAGE_ARCH isn't set that way.
Any objection to creating an lmbench-config package, either in lmbench or
in its own recipe, making that machine specific, and rrecommending it from
lmbench? IIRC that's how we usually handle such things (e.g. alsa-state
pulls in alsa-states which is usually empty, but which bsps can append to
pull in the machine specific configuration).
-Chris
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [yocto] toybox in OpenEmbedded
On Mon, Aug 18, 2014 at 6:44 AM, Paul Barker wrote: > I've been wanting to try out toybox within OpenEmbedded for quite a while > and > have finally found time to put a recipe together. Toybox is a BSD-licensed > alternative to Busybox which may be of interest to a few people. It's > still a > work in progress but it already covers a large number of the basic > commands a > system needs. It should tie in well with the recent work Khem has done to > enable > the use of musl in OpenEmbedded as toybox and musl are often used together > (e.g. > Aboriginal Linux distro). > > My patches are currently against openembedded-core for convenience but this > probably belongs in meta-oe for now. I've done some basic testing on > qemuarm but > nothing significant. toybox builds and installs correctly though and I've > set > ALTERNATIVE_PRIORITY = 60 so that it takes precedence over busybox tools. > > Removing toybox after it has been installed currently bricks the system > though > as the postrm script runs after '/bin/toybox' has been removed (so things > like > '/bin/ln' become broken links and the update-alternatives script can't > run). > One option for that might be to try the C implementation of alternatives that's provided in chkconfig, rather than cworth's shell version. I haven't tested it in that context, but I doubt it runs so many external tools. -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH v2 3/5] dnsmasq: move parallel make patch to .bb, add upstream-status
From: Christopher Larson
Patches really belong in the recipe, not the .inc, given patches are generally
version-bound.
Signed-off-by: Christopher Larson
---
meta-networking/recipes-support/dnsmasq/dnsmasq.inc | 1 -
meta-networking/recipes-support/dnsmasq/dnsmasq_2.68.bb | 2 ++
meta-networking/recipes-support/dnsmasq/files/parallel-make.patch | 1 +
3 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/meta-networking/recipes-support/dnsmasq/dnsmasq.inc
b/meta-networking/recipes-support/dnsmasq/dnsmasq.inc
index 7037cb1..92c5367 100644
--- a/meta-networking/recipes-support/dnsmasq/dnsmasq.inc
+++ b/meta-networking/recipes-support/dnsmasq/dnsmasq.inc
@@ -11,7 +11,6 @@ SRC_URI =
"http://www.thekelleys.org.uk/dnsmasq/${@['archive/', ''][float(d.getV
file://init \
file://dnsmasq.conf \
file://dnsmasq.service \
- file://parallel-make.patch \
"
inherit pkgconfig update-rc.d systemd
diff --git a/meta-networking/recipes-support/dnsmasq/dnsmasq_2.68.bb
b/meta-networking/recipes-support/dnsmasq/dnsmasq_2.68.bb
index b1bef7a..6caa92a 100644
--- a/meta-networking/recipes-support/dnsmasq/dnsmasq_2.68.bb
+++ b/meta-networking/recipes-support/dnsmasq/dnsmasq_2.68.bb
@@ -1,4 +1,6 @@
require dnsmasq.inc
+SRC_URI += "file://parallel-make.patch"
+
SRC_URI[dnsmasq-2.68.md5sum] = "6f8351ca0901f248efdb81532778d2ef"
SRC_URI[dnsmasq-2.68.sha256sum] =
"402019d61f53f1ccc04cc6218719fd3b158ea3fca9a83ab55471f981a2097c2f"
diff --git a/meta-networking/recipes-support/dnsmasq/files/parallel-make.patch
b/meta-networking/recipes-support/dnsmasq/files/parallel-make.patch
index 3d1040c..6d5b722 100644
--- a/meta-networking/recipes-support/dnsmasq/files/parallel-make.patch
+++ b/meta-networking/recipes-support/dnsmasq/files/parallel-make.patch
@@ -5,6 +5,7 @@ can't rely on that order when doing a paralllel make build, so
add an explicit
rule to enforce that order.
Signed-off-by: Christopher Larson
+Upstream-status: Pending
--- dnsmasq-2.68.orig/Makefile
+++ dnsmasq-2.68/Makefile
--
1.8.3.4
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH v2 5/5] dnsmasq: add packageconfigs for idn, conntrack, lua
From: Christopher Larson
Signed-off-by: Christopher Larson
---
meta-networking/recipes-support/dnsmasq/dnsmasq.inc | 14 ++
1 file changed, 10 insertions(+), 4 deletions(-)
diff --git a/meta-networking/recipes-support/dnsmasq/dnsmasq.inc
b/meta-networking/recipes-support/dnsmasq/dnsmasq.inc
index 92c5367..1335245 100644
--- a/meta-networking/recipes-support/dnsmasq/dnsmasq.inc
+++ b/meta-networking/recipes-support/dnsmasq/dnsmasq.inc
@@ -18,10 +18,16 @@ inherit pkgconfig update-rc.d systemd
INITSCRIPT_NAME = "dnsmasq"
INITSCRIPT_PARAMS = "defaults"
-PACKAGECONFIG ??= ""
-PACKAGECONFIG[dbus] = "COPTS=-DHAVE_DBUS,,dbus"
-EXTRA_OEMAKE = "${EXTRA_OECONF}"
-EXTRA_OEMAKE += "\
+PACKAGECONFIG ?= ""
+PACKAGECONFIG[dbus] = ",,dbus"
+PACKAGECONFIG[idn] = ",,libidn"
+PACKAGECONFIG[conntrack] = ",,libnetfilter-conntrack"
+PACKAGECONFIG[lua] = ",,lua"
+EXTRA_OEMAKE = "\
+'COPTS=${@base_contains('PACKAGECONFIG', 'dbus', '-DHAVE_DBUS', '', d)} \
+ ${@base_contains('PACKAGECONFIG', 'idn', '-DHAVE_IDN', '', d)} \
+ ${@base_contains('PACKAGECONFIG', 'conntrack', '-DHAVE_CONNTRACK',
'', d)} \
+ ${@base_contains('PACKAGECONFIG', 'lua', '-DHAVE_LUASCRIPT', '',
d)}' \
'CFLAGS=${CFLAGS}' \
'LDFLAGS=${LDFLAGS}' \
"
--
1.8.3.4
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH v2 1/5] dnsmasq: add missing inherit pkgconfig
From: Christopher Larson
dnsmasq runs pkg-config to get cflags/libs for its dependencies.
Signed-off-by: Christopher Larson
---
meta-networking/recipes-support/dnsmasq/dnsmasq.inc | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta-networking/recipes-support/dnsmasq/dnsmasq.inc
b/meta-networking/recipes-support/dnsmasq/dnsmasq.inc
index 1955d6f..71a491f 100644
--- a/meta-networking/recipes-support/dnsmasq/dnsmasq.inc
+++ b/meta-networking/recipes-support/dnsmasq/dnsmasq.inc
@@ -14,7 +14,7 @@ SRC_URI =
"http://www.thekelleys.org.uk/dnsmasq/${@['archive/', ''][float(d.getV
file://parallel-make.patch \
"
-inherit update-rc.d systemd
+inherit pkgconfig update-rc.d systemd
INITSCRIPT_NAME = "dnsmasq"
INITSCRIPT_PARAMS = "defaults"
--
1.8.3.4
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH v2 2/5] dnsmasq: obey CFLAGS & LDFLAGS
From: Christopher Larson
Signed-off-by: Christopher Larson
---
meta-networking/recipes-support/dnsmasq/dnsmasq.inc | 4
1 file changed, 4 insertions(+)
diff --git a/meta-networking/recipes-support/dnsmasq/dnsmasq.inc
b/meta-networking/recipes-support/dnsmasq/dnsmasq.inc
index 71a491f..7037cb1 100644
--- a/meta-networking/recipes-support/dnsmasq/dnsmasq.inc
+++ b/meta-networking/recipes-support/dnsmasq/dnsmasq.inc
@@ -22,6 +22,10 @@ INITSCRIPT_PARAMS = "defaults"
PACKAGECONFIG ??= ""
PACKAGECONFIG[dbus] = "COPTS=-DHAVE_DBUS,,dbus"
EXTRA_OEMAKE = "${EXTRA_OECONF}"
+EXTRA_OEMAKE += "\
+'CFLAGS=${CFLAGS}' \
+'LDFLAGS=${LDFLAGS}' \
+"
do_compile_append() {
# build dhcp_release
--
1.8.3.4
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH v2 4/5] dnsmasq: look for lua.pc, not lua5.1.pc
From: Christopher Larson
Signed-off-by: Christopher Larson
---
.../recipes-support/dnsmasq/dnsmasq/lua.patch| 16
.../dnsmasq/{files => dnsmasq}/parallel-make.patch | 0
meta-networking/recipes-support/dnsmasq/dnsmasq_2.68.bb | 5 -
3 files changed, 20 insertions(+), 1 deletion(-)
create mode 100644 meta-networking/recipes-support/dnsmasq/dnsmasq/lua.patch
rename meta-networking/recipes-support/dnsmasq/{files =>
dnsmasq}/parallel-make.patch (100%)
diff --git a/meta-networking/recipes-support/dnsmasq/dnsmasq/lua.patch
b/meta-networking/recipes-support/dnsmasq/dnsmasq/lua.patch
new file mode 100644
index 000..aac7b97
--- /dev/null
+++ b/meta-networking/recipes-support/dnsmasq/dnsmasq/lua.patch
@@ -0,0 +1,16 @@
+Upstream-status: Inappropriate [OE specific]
+Signed-off-by: Christopher Larson
+
+--- dnsmasq-2.68.orig/Makefile 2014-08-11 09:05:17.0 -0700
dnsmasq-2.68/Makefile 2014-08-11 09:05:51.930648617 -0700
+@@ -57,8 +57,8 @@
+ idn_libs =`echo $(COPTS) | $(top)/bld/pkg-wrapper HAVE_IDN $(PKG_CONFIG)
--libs libidn`
+ ct_cflags = `echo $(COPTS) | $(top)/bld/pkg-wrapper HAVE_CONNTRACK
$(PKG_CONFIG) --cflags libnetfilter_conntrack`
+ ct_libs = `echo $(COPTS) | $(top)/bld/pkg-wrapper HAVE_CONNTRACK
$(PKG_CONFIG) --libs libnetfilter_conntrack`
+-lua_cflags = `echo $(COPTS) | $(top)/bld/pkg-wrapper HAVE_LUASCRIPT
$(PKG_CONFIG) --cflags lua5.1`
+-lua_libs =`echo $(COPTS) | $(top)/bld/pkg-wrapper HAVE_LUASCRIPT
$(PKG_CONFIG) --libs lua5.1`
++lua_cflags = `echo $(COPTS) | $(top)/bld/pkg-wrapper HAVE_LUASCRIPT
$(PKG_CONFIG) --cflags lua`
++lua_libs =`echo $(COPTS) | $(top)/bld/pkg-wrapper HAVE_LUASCRIPT
$(PKG_CONFIG) --libs lua`
+ sunos_libs = `if uname | grep SunOS >/dev/null 2>&1; then echo -lsocket
-lnsl -lposix4; fi`
+ version = -DVERSION='\"`$(top)/bld/get-version $(top)`\"'
+
diff --git a/meta-networking/recipes-support/dnsmasq/files/parallel-make.patch
b/meta-networking/recipes-support/dnsmasq/dnsmasq/parallel-make.patch
similarity index 100%
rename from meta-networking/recipes-support/dnsmasq/files/parallel-make.patch
rename to meta-networking/recipes-support/dnsmasq/dnsmasq/parallel-make.patch
diff --git a/meta-networking/recipes-support/dnsmasq/dnsmasq_2.68.bb
b/meta-networking/recipes-support/dnsmasq/dnsmasq_2.68.bb
index 6caa92a..61b94a0 100644
--- a/meta-networking/recipes-support/dnsmasq/dnsmasq_2.68.bb
+++ b/meta-networking/recipes-support/dnsmasq/dnsmasq_2.68.bb
@@ -1,6 +1,9 @@
require dnsmasq.inc
-SRC_URI += "file://parallel-make.patch"
+SRC_URI += "\
+file://parallel-make.patch \
+file://lua.patch \
+"
SRC_URI[dnsmasq-2.68.md5sum] = "6f8351ca0901f248efdb81532778d2ef"
SRC_URI[dnsmasq-2.68.sha256sum] =
"402019d61f53f1ccc04cc6218719fd3b158ea3fca9a83ab55471f981a2097c2f"
--
1.8.3.4
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH v2 0/5] Various dnsmasq fixes/improvements
From: Christopher Larson
This is rebased on current master. The parallel make fix's commit message is
adjusted to match what it does, and the lua patch is fixed.. It seems
something inadvertantly trimmed trailing whitespace out of the patch, oops.
The following changes since commit 4d4ab8fd320b58323033c12cf705e7b3ce10d659:
integrate fcgi-2.4.0 and add OML license file (2014-08-10 17:23:22 +0100)
are available in the git repository at:
https://github.com/kergoth/meta-openembedded dnsmasq-fixes
for you to fetch changes up to 2e84f59401a5ce57ba17841ff4ddb01cffeb1145:
dnsmasq: add packageconfigs for idn, conntrack, lua (2014-08-11 09:07:14
-0700)
Christopher Larson (5):
dnsmasq: add missing inherit pkgconfig
dnsmasq: obey CFLAGS & LDFLAGS
dnsmasq: move parallel make patch to .bb, add upstream-status
dnsmasq: look for lua.pc, not lua5.1.pc
dnsmasq: add packageconfigs for idn, conntrack, lua
meta-networking/recipes-support/dnsmasq/dnsmasq.inc | 19 ++-
.../recipes-support/dnsmasq/dnsmasq/lua.patch | 16
.../dnsmasq/{files => dnsmasq}/parallel-make.patch| 1 +
.../recipes-support/dnsmasq/dnsmasq_2.68.bb | 5 +
4 files changed, 36 insertions(+), 5 deletions(-)
create mode 100644 meta-networking/recipes-support/dnsmasq/dnsmasq/lua.patch
rename meta-networking/recipes-support/dnsmasq/{files =>
dnsmasq}/parallel-make.patch (96%)
--
1.8.3.4
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
