On 15 June 2016 at 09:27, Anicic Damir (PSI) wrote:
> The problem seems to be missing qemu-ppc64, see down "exit code: 127".
> qemu-* executables ( tmp/sysroots/x86_64-linux/usr/bin/qemu-* ) are built,
> but qemu-ppc64 is missing.
>
Fixing this is trivial, add ppc64 to QEMU_TARGETS in qemu.inc.
On 15 June 2016 at 10:31, Burton, Ross wrote:
> Fixing this is trivial, add ppc64 to QEMU_TARGETS in qemu.inc.
>
I'm wrong - this is more complicated as there's numerous options for
ppc64. I've filed https://bugzilla.yoctoproject.org/show_bug.cgi?id=9778
to track thi
On 16 June 2016 at 07:33, Anicic Damir (PSI) wrote:
> I do not find meta-qt3 & meta-qt4 in Yocto 2.1
>
That's because they're not included as in general they are not needed.
https://layers.openembedded.org/layerindex/branch/master/layer/meta-qt3/
https://layers.openembedded.org/layerindex/branc
On 16 June 2016 at 14:53, Jussi Kukkonen wrote:
> Now, the default configuration is to build only support for Gtk+3 but it
> is possible to support also Gtk+2 by adding "gtk2-im" into the
> matchbox-keyboard PACKAGECONFIG. This is not done by default to prevent the
> keyboard from bringing gtk+ i
The repository is yocto-docs on git.yoctoproject.org.
To submit a patch sent it to yocto@ and it's probably best to CC Scott too.
Thanks,
Ross
On 17 June 2016 at 11:04, Woronicz, Bartosz (Nokia - PL/Wroclaw) <
bartosz.woron...@nokia.com> wrote:
> I would like to push some fixes to docs how can
On 20 June 2016 at 18:47, Michael Halstead
wrote:
> OpenEmbedded is nearly ready to upgrade to a new version of Patchwork.
> The new version offers several new features that will allow OpenEmbedded
> to improve developer workflow and automated checks on incoming patches.
>
Woohoo!
Thanks Michae
On 22 June 2016 at 15:34, Rajasekaran, Monica <
monica.rajaseka...@us.fujitsu.com> wrote:
> |diff --git
> data/users/mrajase/l100_label_based_r100poc/repo/poky/build/tmp/work/core2-64-enea-linux/openvswitch/2.0.0-r0/openvswitch-2.0.0/ofproto/ofproto.c
> data/users/mrajase/extra/ofproto.c
>
> |inde
On 22 June 2016 at 19:25, Rajasekaran, Monica <
monica.rajaseka...@us.fujitsu.com> wrote:
> After applying the patches, I brought up my openvswitch which works
> correctly alongside the controller but I don’t see any of the print
> statements I added to view the changes inside. Is there a way to c
On 22 June 2016 at 17:43, Patrick Williams wrote:
> The current poky fails when building on a non-x86 platform with a
> message like:
>
> uninative selected but not configured, please set
> UNINATIVE_CHKSUM[ppc64le]
>
> We have a number of developers that use ppc64le systems to do work on,
>
On 22 June 2016 at 21:00, Burton, Ross wrote:
> On 22 June 2016 at 17:43, Patrick Williams wrote:
>
>> The current poky fails when building on a non-x86 platform with a
>> message like:
>>
>> uninative selected but not configured, please set
>> UNINA
<
monica.rajaseka...@us.fujitsu.com> wrote:
> The folders in tmp/work on my source machine don’t have today’s date as
> last modified.
>
>
>
> Hmm… I don’t know what might be the mistake with applying the patch.
>
>
>
> *Thanks,*
>
> *Monica *
>
>
>
&
Hi,
On 24 June 2016 at 11:41, Kumar, Shrawan wrote:
> Is there a way to add a capability to a binary (cap_net_raw+ep),into a
> recipe?
>
>
>
> Example :
>
> do_install() {
>
>install -d ${D}${bindir}
>
>install -m 0755 helloworld ${D}${bindir}
>
>install -d $
${D}/lib/systemd/system/
>
> setcap cap_net_raw+ep ${D}${bindir}/helloworld
>
>
>
> }
>
>
>
> Thanks and Regards
>
> Shrawan
>
>
>
> *From:* Burton, Ross [mailto:ross.bur...@intel.com]
> *Sent:* Friday, June 24, 2016 5:09 PM
> *To:*
I just discovered that this was never merged, Peter can you review it?
Ross
On 3 May 2016 at 14:18, George McCollister
wrote:
> The setcap utility supplied by libcap is used to set capabilities on a
> file. Before setting a file's capabilities with cap_set_file() (which uses
> setxattr()) it ca
On 24 June 2016 at 13:31, Kumar, Shrawan wrote:
> I am using Yocto 2.0.2
>
>
>
> Thanks and Regards
>
> Shrawan
>
>
>
> *From:* Burton, Ross [mailto:ross.bur...@intel.com]
> *Sent:* Friday, June 24, 2016 5:56 PM
>
> *To:* Kumar, Shrawan
> *Cc:* y
Yes, x32 is built and tested on the autobuilder daily.
Ross
On 26 June 2016 at 01:39, Paul D. DeRocco wrote:
> Has anyone successfully used Yocto to build a Linux system for a 64-bit
> Intel processor using the -mx32 compiler option, forcing all long and
> pointer types to 32 bits? I've got an
On 23 June 2016 at 17:18, Patrick Williams wrote:
> Thanks. We will give it a try. I suspected there might be more to it than
> that
> based on this line:
>
> UNINATIVE_LOADER ?=
> "${STAGING_DIR}-uninative/${BUILD_ARCH}-linux/lib/${@bb.utils.contains('BUILD_ARCH',
> 'x86_64', 'ld-linux-x86-64.s
On 28 June 2016 at 06:56, Zhenhua Luo wrote:
> [Luo Zhenhua-B19537] gcc can't be found after sourcing the SDK env script.
> but the /usr/bin is in the PATH env variable.
>
You *do* have a gcc in /usr/bin, right?
Ross
--
___
yocto mailing list
yocto@y
On 28 June 2016 at 16:38, Austin Kim wrote:
> I'm wondering how to copy only stripped binary and leave debug symbols on
> the host build machine. I'm running on the OS on a very tight HDD space so
> I cannot put debug symbols into the image. So I tried
>
>
> PACKAGE_DEBUG_SPLIT_STYLE = 'debug-wit
On 29 June 2016 at 05:10, Kumar, Shrawan wrote:
> Could your update on my issues ?
>
>
The patch is probably against a newer version. If you're unable to switch
to the latest version of pseudo to debug this then you can run setcap in a
postinst.
Ross
--
___
On 29 June 2016 at 15:43, Daniel. wrote:
> It seems that Yocto inspect the first line of scripts to grab this
> dependency but fails to map /usr/bin/python to python-core (which is
> already installed). How to make this work right?!
>
Add a RDEPENDS_${PN} += "python-core python-modules".
Ross
-
On 30 June 2016 at 10:40, Gujulan Elango, Hari Prasath (H.) <
hguju...@visteon.com> wrote:
> The dlt-daemon package build directory at
> tmp/work/aarch64-poky-linux/dlt-daemon/2.11.1-r0/image
> has the following build output where I couldn't see any directory named
> lib64. Probably this is the r
On 1 July 2016 at 15:03, Mathieu Allard wrote:
> I think that the main issue here is that the pkg_postinst function runs
> its action at the rootfs creation time, and not on the target as advised by
> Ross.
>
Yes, as I said in the first suggestion you'll need to ensure this runs on
the target (c
On 1 July 2016 at 17:37, Daniel. wrote:
> I read that python can't execute multiple threads simultaneously, but
> that wouldn't be a problem if each python task is executed at its own
> interpreter instance (process). This last statement is what I don't
> really know, some experts on Yocto's inte
On 2 July 2016 at 03:12, Takashi Matsuzawa
wrote:
> There seems to be PyPy, Stackless Python, etc. but I am not sure they can
> be tried 'in-place' to see if they work faster.
>
In the context of a bitbake build, the overhead of Python itself is utterly
insignificant in the context of doing the c
On 2 July 2016 at 10:22, Kumar, Shrawan wrote:
> Can someone review the attached recipe and help solve the problem
> statement ?
>
As has been said, you need to ensure the postinst is delayed so it runs on
the target and not on the build host.
http://www.yoctoproject.org/docs/2.1/dev-manual/de
On 2 July 2016 at 11:37, Kumar, Shrawan wrote:
> Post-installation scripts run immediately after installing a package on
> the target *or during image creation when a package is included in an
> image*.
>
> Does it not mean that we can set the file attributes(setcap) during
> image creation ?
>
On 3 July 2016 at 18:45, Robert P. J. Day wrote:
> should i submit a patch that just deletes the PREFERRED_VERSION
> line?
>
Yes, that seems like the right thing to do.
Thanks,
Ross
--
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoc
On 4 July 2016 at 19:10, Marcelo E. Magallon
wrote:
> That is true, but I've seen cases where the start up time of Python (or
> the time it takes bitbake to start "doing work") is *perceptually* much
> larger than the work done by the recipe, or at least it looks like it
> represents a sigficant
On 5 July 2016 at 10:20, Chandra Mishra (chanmish)
wrote:
> * -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed -static -lnsl -lpthread
> -lm -lrt*
>
> */sysroots/corei7-64-poky-linux/usr/lib/../lib/librt.a(shm_open.o):
> In function `shm_open': *
>
> * PATH>/build/tmp/w
On 5 July 2016 at 11:06, Chandra Mishra (chanmish)
wrote:
> Thanks Ross for your suggestion. It worked for me. But in the past it
> worked for me with old sequence but different gcc cross toolchain, so not
> thinking of changing the order.
>
>
Newer toolchain may have different inter-dependencies
On 5 July 2016 at 14:10, Daniel. wrote:
> I faced problems with symbols being droped if -l option is passed
> after -Wl,--as-needed. To circumvent this I changed my Makefiles to
> add link options prior $(LDFLAGS) expansion.
>
Note that --as-needed only applies to links specified after the optio
On 6 July 2016 at 10:39, Anders Darander wrote:
> > CONFFILES_${PN} += "${sysconfdir}/init.d/can_if"
>
> If this is the complete recipe, you never install can_if...
>
Which means the package is empty, which means it doesn't get generated,
which explains why you can't add it to an image.
Ross
--
On 6 July 2016 at 15:23, Gary Thomas wrote:
> Dependency on checksum of file
> MIPS-GAS-Fix-an-ISA-override-not-lifting-ABI-restrictions.patch was removed
>
As this show binutils changing its SRC_URI this suggests that you did
something in between the two runs, or those hashes are not the releva
On 7 July 2016 at 14:24, wrote:
> I want to install my software packages to the usr/local/bin dir. That is
> why I need to change the prefix form usr to usr/local. How to archive that?
>
> The software is build through a cmake. Is there somethingt I need to think
> extra for setting the prefix?
On 11 July 2016 at 11:51, Jussi Kukkonen wrote:
> Add an internal copy of xsettings-client into libmatchbox: it was
> never meant to be a shared library.
>
> Other commits avoid deprecated API use or fix problems found by
> modern gcc with -Wall.
>
Looks good, feel free to push.
Ross
--
__
On 12 July 2016 at 10:03, Esponde, Joel wrote:
> Hi,
>
>
>
> Some days ago, we updated poky layer from fido to krogoth and since then
> we lost the sftp feature when using dropbear instead of openssh.
>
> A commit has been done in the master branch to fix this issue:
>
>
> http://git.yoctoproject
On 12 July 2016 at 13:11, Dominic Sacré wrote:
> The patch works fine with krogoth, and I originally sent it with both
> master and krogoth in the subject line:
>
> https://patchwork.openembedded.org/patch/123485/
>
> Should I resend the patch for krogoth?
>
Okay, my mistake. Just ping the orig
On 13 July 2016 at 08:37, Christian Fuchs wrote:
> From the error message I guess that the perl library for RRD is missing in
> sysroot. The library is built as part of the rrdtool recipe, but it is
> moved to a separate package called "rrdtool-perl". How can I tell bitbake
> to install this pack
+1
On 13 July 2016 at 14:00, Jussi Kukkonen wrote:
> Signed-off-by: Jussi Kukkonen
> ---
> applet.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/applet.c b/applet.c
> index 699a920..d403619 100644
> --- a/applet.c
> +++ b/applet.c
> @@ -8,7 +8,7 @@
>
> #include
Looks good!
On 13 July 2016 at 14:12, Jussi Kukkonen wrote:
> Port all remaining applets to ScalingImage2.
> Also fix a bug in ScalingImage2.
>
> The following changes since commit
> 145a8075c66952dad9ab5f36d944c78c1ae3f4ab:
>
> Release 2.10, update contact info (2016-06-02 15:08:14 +0300)
>
>
On 14 July 2016 at 11:22, wrote:
> I want to create a recipe for an application. This application is
> generated through cmake. In the cmake file "CMakeLists.txt" is the passage:
> ##
> include(FindPythonInterp)
> include(FindPythonLibs)
>
> add_custom_command(OUTPUT "${CMAKE_BIN
On 15 July 2016 at 08:43, Rudolf Streif wrote:
> Devtool works reasonably well with repos and packaged source tarballs. We
> did a test with a source RPM which resulted in a simple stub recipe
> although the source rpm contained an autotooled package. devtool did not
> even make an attempt to unp
On 18 July 2016 at 14:55, Yile Ku wrote:
> I am getting this error when building for PowerPC 64.
>
>
> The file:
> meta-sdr/recipes-core/gnuradio/gnuradio_git.bb
>
> says to include the libraries
> FILES_libvolk = "${libdir}/libvolk.so*"
> FILES_libvolk-dbg = "${libdir}/.debug/libvolk.so*"
>
> wh
On 20 July 2016 at 18:30, Robert Berger
wrote:
> I was wondering what's the process if someone wanted to use, say Jethro
> or Krogoth, but with different/custom versions of gcc, binutils, glibc
> e.g. to cook some syscalls and to compile ancient Linux kernels.
>
> I came across tcmode-default.inc
On 21 July 2016 at 22:51, Vijayakumar Badiger
wrote:
> /local/mnt/workspace/c_vbadig/LV/poky/build/tmp-glibc/work/aarch64-poky-linux/system-core/git-r17/system/core/libsync/sync.c:24:24:
> fatal error: linux/sync.h: No such file or directory
> compilation terminated.
> make[2]: *** [libsync_la-sy
On 26 July 2016 at 15:09, Gary Thomas wrote:
> I have a build host that I've used for years. It is not
> possible to [directly] update python3 on this system to
> one which is identified as suitable for use with bitbake.
>
> To try and work around this, I created a meta-toolchain SDK
> for my ta
On 26 July 2016 at 15:57, Gary Thomas wrote:
> Yes, this is just what I needed (except I had to build my own because my
> box is i686)
>
Oh good, next time we have a bug which only appears on 32-bit build hosts
I'll ping you. :)
> Looks like this SDK doesn't actually contain a full copy of Pyt
On 27 July 2016 at 22:53, Vijayakumar Badiger
wrote:
> We have a requirement here to set up the perf tool through our Yocto frame
> work.
> As Perf source comes with linux kernel under tools/perf ,so I added this
> perf tool to be installed
> as part of my rootfs image.
>
> But when I build the i
On 1 August 2016 at 13:01, Kyle Russell wrote:
> scripts/contrib/python/generate-manifest-3.4.py | 10 +-
>
Sorry, but you need to edit the script and then run the script, so the
patch contains both the script and inc file changes.
Ross
--
__
On 30 July 2016 at 09:41, Gary Thomas wrote:
> I'm trying to isolate a problem that showed up in my builds
> in the last month. Little has changed in the sources, but
> the one big change was I moved from GCC/4.9 to GCC/5.x On
> the surface, this seemed to be a non-consequence, but I have
> one
On 1 August 2016 at 15:26, Gary Thomas wrote:
> Nothing else changed, I'm sure.
>
*Something* changed, although it may be something that shouldn't have
caused a rebuild (at which point printdiff is useful to find that out).
Ross
--
___
yocto mailing
On 2 August 2016 at 08:56, Richard Zhang wrote:
> checking for xml2-config... yes
>
> ERROR: /usr/bin/xml2-config should not be used, use an alternative such as
> pkg-config
>
> ERROR: /usr/bin/xml2-config should not be used, use an alternative such as
> pkg-config
>
> checking for xmlDOMWrapReco
On 3 August 2016 at 20:34, Joel Zhou wrote:
> WARNING: quilt-native-0.64-r0 do_populate_lic: Could not copy license file
> /home/joelz/poky2.1/poky/build/tmp/work/x86_64-linux/quilt-native/0.64-r0/quilt-0.64/COPYING
> to
> /home/joelz/poky2.1/poky/build/tmp/work/x86_64-linux/quilt-native/0.64-r0/
On 3 August 2016 at 22:09, Philippe Coval
wrote:
> +INSANE_SKIP_${PN}-dev = "ldflags"
> +INSANE_SKIP_${PN}-plugins-samples = "ldflags"
> +INSANE_SKIP_${PN}-resource = "ldflags"
> +INSANE_SKIP_${PN}-resource-samples = "ldflags"
> +INSANE_SKIP_${PN}-service = "ldflags"
> +INSANE_SKIP_${PN}-service-
On 4 August 2016 at 09:46, Richard Zhang wrote:
> Which tmp dir should I wipe out?Thanks
>
>
The tmp/ inside your build directory. If it happens again, then libnetconf
is doing something very wrong.
Ross
--
___
yocto mailing list
yocto@yoctoproject.o
On 8 August 2016 at 16:59, Richard Purdie <
richard.pur...@linuxfoundation.org> wrote:
> FWIW, I think the connmand segfault on mips and powerpc is hinted at
> with this bug:
>
> https://bugs.launchpad.net/ubuntu/+source/lua5.3/+bug/1570055
>
> connmand is a binary linked with a linker version-scr
On 12 August 2016 at 09:46, wrote:
>
> If have a simple problem. On one system a program is started as a daemon
> and on the other one not. In my recipe the program is build. by using
> update-rc.d I create the init script for systemV. The code looks like:
The easy solution here is to build two
On 11 August 2016 at 22:14, Kyle Russell wrote:
> In case anyone is interested, I posted an update for both the script and
> .inc to openembedded-core (which now has master at python 3.5), but it
> doesn't seem to have gone anywhere yet.
>
Sorry that got stalled because the patch was mangled by
On 12 August 2016 at 11:50, Robert P. J. Day wrote:
> built mono for the first time, and a significant amount of time was
> spent compiling mono-native. if i already have mono on my build host,
> is it worth checking if i can take advantage of it?
>
sstate obviously means if it doesn't change
On 17 August 2016 at 07:49, wrote:
> In my case I have a software which do logging. The amount of the logs
> depends on type of the build (Release skips all debug messages). Let's
> assume that my recipe for the software(myApp) is called myApp.bb. It is
> downloaded from an external git source an
On 19 August 2016 at 02:50, Richard Zhang wrote:
> After built an ‘intel-corei7-64’ image with Toaster, I try to run a
> runqemu, it can’t run.
>
> ERROR: unable to classify arg [intel-corei7-64].
>
> Does intel-corei7-64 could be run with runqemu?
>
>
Currently, not out of the box, although it s
On 19 August 2016 at 09:42, Maxin B. John wrote:
> (A quick grep in meta-qt4 for PACKAGECONFIG_CONFARGS returned zero results)
>
FWIW it's a relatively new feature: when PACKAGECONFIG is evaluated it
writes the enabled/disabled values to PACKAGECONFIG_CONFARGS, which classes
can then use to set
On 2 September 2016 at 10:37, pirmin.weis...@t-online.de <
pirmin.weis...@t-online.de> wrote:
> One function is not working on the target system: The automatic popup of
> the virtual keyboard when I select a text entry field (e.g. by touching it).
>
> Display and Touch are working fine, internet c
On 8 September 2016 at 15:36, nick83ola wrote:
> it gives me an error because gtk+ requires it and python-pygst is a
> dependance of python-gst
>
Looking at the pygtk configure you might be able to change the pygtk recipe
to not pull in the gtk+ and so on dependencies if X11 isn't in
DISTRO_FEAT
On 8 September 2016 at 23:06, Siegel, Jeffrey (Nokia - US) <
jeffrey.sie...@nokia.com> wrote:
> ‘bb’ Python source contains shebang path ‘/usr/bin/env python3’ , which
> means it expects to be run by a Python 3.x interpreter. But then it imports
> python code from bitbake/lib, and bitbake/Yocto us
This is because your distro has enabled the security options, but
native-frameworks has bad link lines that work without security flags more
via luck than anything else.
You'll need to fix the linking order used to build libbinder.
Ross
On 9 September 2016 at 08:15, BHARATH RAJ wrote:
> I am n
On 10 September 2016 at 10:35, Szymon Guza wrote:
> i want to prepare a yocto distribution with openCV and C++ for Xilinx
> zedboard. Do you have any tips for me how to do it? I wanted to take a one
> of the core-images with user interfece and add some packages i need.
>
You've pretty much got i
On 12 September 2016 at 15:48, William Mills wrote:
> 5) Can Yocto [cross-]build snaps or flatpaks?
>
Just because it's interesting, note that the standard freedesktop.org
flatpak runtime is in fact bootstrapped through a custom OE distro.
Ross
--
__
On 12 September 2016 at 19:22, Matthew Stanger
wrote:
> I'm using Yocto 1.7.1 I'd like built programs to have the build path
> striped down. The goal is if I'm looking at some programs log after cross
> compiling I want this:
>
> src/program.c:388
>
> where I currently get this:
>
> /usr/local/je
On 13 September 2016 at 08:08, wrote:
>
> ??? do_fetch_append () {
> ???extract ${S}/example.tar.gz
> ??? }
> ??? do_install () {
> ???copy ${S}/ ${D}/opt/
> ??? }
It's a tarball so use "tar" to extract, and you'll need to mkdir ${D}/opt
before copying. but that's barely just p
On 14 September 2016 at 08:52, Anicic Damir (PSI)
wrote:
> Fetcher failure for URL: 'https://www.example.com/'. URL
> https://www.example.com/ doesn't work. Please ensure your network is
> configured correctly.
>
This is the "is networking correctly configured" test, and if it fails in
Yocto
On 14 September 2016 at 13:42, Anicic Damir (PSI)
wrote:
> wget https://www.exampe.com/
> --2016-09-14 14:42:39-- https://www.exampe.com/
> Resolving www.exampe.com... 69.172.201.153
> Connecting to www.exampe.com|69.172.201.153|:443... connected.
> Unable to establish SSL connection.
>
"exampl
On 14 September 2016 at 15:48, Anicic Damir (PSI)
wrote:
> it turns out that something HAS changed in my system/libraries (but not
> Python itself),
> influencing python urllib2 to fail, although wget, chrome, firefox, ...
> worked
>
> Fortunately we have another python version installed, and wit
On 14 September 2016 at 22:44, Nadikuda, Lavanya <
lavanya.nadik...@us.fujitsu.com> wrote:
> Could you please let me know what the resolution is for /lib64 directory
> not found in yocto.
>
>
If you're bothered by this, you can change your base_libdir to be /lib64
instead of /lib.
Ross
--
__
On 15 September 2016 at 15:51, Nadikuda, Lavanya <
lavanya.nadik...@us.fujitsu.com> wrote:
> What if I want both the directories.
>
>
So Set libdir to one, install a symlink for the other.
Ross
--
___
yocto mailing list
yocto@yoctoproject.org
https://l
On 16 September 2016 at 06:49, Martin Bachmann wrote:
> In my recipe I've added RDEPENDS += "opencv " and yocto installs the
> libraries in the target rootfs. But now i'd like the libraries compiled
> natively and installed to x86_x64-linux rootfs. This would allow me to
> unittest my application
On 16 September 2016 at 13:40, Chris Trobridge
wrote:
> The install line in the makefile is "cp -af $(APP_LIBXX_FILES)
> $(DESTDIR)$(libdir)/", so I am at a bit of a loss as this should preserve
> ownership.
>
It does - the build happens as you (uid 1000) so the file gets installed as
1000. Thi
On 28 September 2016 at 14:40, wrote:
> When populate_sysroot is called in myweb.bb finishes with the error:
>
> myweb-1.0-r0 do_populate_sysroot: The recipe myweb is trying to install
> files into a shared area when those files already exist. Those files and
> their manifest location are:
>/
On 28 September 2016 at 14:50, wrote:
> do_install () {
> install -d ${D}/usr/share/apache2/htdocs/
> cp -r ${S}/myweb/* ${D}/usr/share/apache2/htdocs/
> }
> FILES_${PN} += "/usr/share/apache2/htdocs/*"
Just put the files somewhere else, like /var/www/myweb/. You'll need to
con
On 29 September 2016 at 15:46, Karim ATIKI wrote:
> I have generated a core-image-x11 image for a Toradex Colibri T20 board.
> I aim to compile and execute our application that normally runs under X11
> + OpenGL.
>
What sort of GPU does this board have, and what GL drivers are you using?
Mesa sh
On 5 October 2016 at 13:58, Ioan-Adrian Ratiu wrote:
> (the reason why the lib is moved in the first place is to avoid a QA issue
> because there's a risk for /usr to be on another partition)
>
I'm not a meta-selinux maintainer but this is where I ask is there actually
a risk, or is this somethi
On 11 October 2016 at 06:18, Dinh Nguyen (dinhn) wrote:
> ERROR: QA Issue: Architecture did not match (20 to 62) on
> /work/ppce500v2-poky-linux-gnuspe/c-mlib/1.1-r0/packages-split/c-mlib/usr/bin/xxx
> [arch]
>
You can look up the architectures at
https://en.wikipedia.org/wiki/Executable_and_Li
On 13 October 2016 at 09:49, Contrib Open Source <
contrib.open.sou...@gmail.com> wrote:
> it seems that license.manifest file describes application/lib/etc.
> source code license,
> but what about Yocto recipe license itself ?
> So what is the license of a Yocto recipe ?
> How to know it ?
> If a
On 14 October 2016 at 06:23, Contrib Open Source <
contrib.open.sou...@gmail.com> wrote:
> Here "metadata" stands for "oe-core recipe", right ?
>
"metadata" includes all the recipes and classes in oe-core.
> What if we extend (*.bbappend) a GPLv2 recipe: is it "contaminated" or not
> ?
>
Presu
On 12 October 2016 at 18:58, Michael Callahan
wrote:
> I just moved my project from Fido to Krogoth and now I have two
> pythons (2.7 and 3). This bumps the size of my image by approximately
> 25%. What would be the most reasonable way to get back to just one
> python? Ideally I'd like a magic
What tool is generating the code? It probably shouldn't be using the
full path of the header...
Ross
On 10 April 2018 at 09:29, Måns Zigher wrote:
> Hi,
>
> I am having some problems with one of my recipes when using the SRCPV the
> get_srcrev is returning "AUTOINC+" + rev. This will cause prob
s Zigher wrote:
> I believe dbus codegen is the tools generating the code.
>
> Br
> Mans Zigher
>
> On Tue, Apr 10, 2018, 14:44 Burton, Ross wrote:
>>
>> What tool is generating the code? It probably shouldn't be using the
>> full path of the header...
>&g
(1) is basically exactly what images do: they depend on do_deploy of
all the packages that goes into them. The pkgdata tree in tmp/ will
tell you (2), (3) is just shell operations, and (4) is trivial.
Ross
On 9 April 2018 at 16:31, Sebastian Rohde
wrote:
> Hello everybody,
>
> in my setup opkg
If your project also includes meta-poky then poky.conf will contain a
DISTRO_VERSION assignment. However,
https://wiki.yoctoproject.org/wiki/Releases indicates that bitbake
1.32 was part of 2.2/morty.
Ross
On 12 April 2018 at 10:44, Richard Collins
wrote:
> I've inherited a Yocto project that I
I think the message here is that we need to embed the oe-core version
in the metadata somewhere...
On 12 April 2018 at 13:11, Richard Collins
wrote:
> I get the following for DISTRO.
>
> DISTRO= "yogurt"
> DISTRO_VERSION= "BSP-Yocto-RK3288-PD17.1.1"
>
> Can't find any reference to
That is an excellent suggestion, yes.
Ross
On 12 April 2018 at 13:54, Martin Jansa wrote:
> Cannot LAYERSERIES_CORENAMES added in rocko be used for that?
>
> On Thu, Apr 12, 2018 at 2:47 PM, Burton, Ross wrote:
>>
>> I think the message here is that we need to embed the
Is it feasible to fix the original patch?
Ross
On 13 April 2018 at 00:55, Juro Bystricky wrote:
> The commit "qemu: Bump to version 2.11.0" in oe-core broke the
> build of qemu for mingw, due to using "socketpair", which is
> not supported by mingw. "socketpair" is used in a local patch,
> not i
2
> I'll do it gain. Hopefully it will last longer this time.
>
>
> ____
> From: Burton, Ross [ross.bur...@intel.com]
> Sent: Friday, April 13, 2018 5:50 AM
> To: Bystricky, Juro
> Cc: Yocto-mailing-list; Juro Bystricky; Alistair Francis
>
On 10 April 2018 at 18:49, Aaron M. Biver wrote:
> The problem that myapp-native is trying to include mymodule-native. This
> complains:
Just make the dependency on mymodule target-specific. Lots of ways
depending on what the dependencies are.
Either clear the native depends:
DEPENDS = "myapp
If you've an i5 why are you running a 32-bit OS? Just so you're
aware, it's practically impossible to build webkit on a 32-bit host
because there just isn't enough memory available.
Ross
On 17 April 2018 at 10:50, Stéphane Ancelot wrote:
>
>
> Le 17/04/2018 à 09:08, Iván Castell a écrit :
>
> R
On 17 April 2018 at 15:03, Khem Raj wrote:
> On Tue, Apr 17, 2018 at 3:13 AM, Burton, Ross wrote:
>> If you've an i5 why are you running a 32-bit OS? Just so you're
>> aware, it's practically impossible to build webkit on a 32-bit host
>> because ther
On 18 April 2018 at 12:04, Robert Yang wrote:
> fixup! update: don't stop on unsatisfied layer dependencies
Pretty sure this isn't what you intended.
Ross
--
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto
This is the wrong list, please submit the patch to
openembedded-c...@lists.openembedded.org. Also the patches, not just
a link to the patches.
Ross
On 18 April 2018 at 05:14, Vignesh Rajendran (RBEI/ECF3)
wrote:
> Hi Yocto,
>
>
>
> I have made some changes to fix missing debug sources in -dbg p
Pushed, thanks.
Ross
On 21 April 2018 at 02:30, Peter Kjellerstedt
wrote:
> Most shell scripts have '#!/bin/{sh,bash}' on the first line of the
> script, which triggers RPM to automatically add a runtime dependency
> on that path for any package that contains shell scripts. However,
> when the
801 - 900 of 1567 matches
Mail list logo
