Date   

[PATCH] qt-x11-free-native: Use private libpng

Saul Wold <sgw@...>
 

Ensure the 3rdparty version of libpng headers are found first, otherwise
the sysroot libpng15 header are found which causes a mis-match issue.

Signed-off-by: Saul Wold <sgw@...>
---
recipes-qt3/qt3/qt-x11-free-native_3.3.5.bb | 1 +
1 file changed, 1 insertion(+)

diff --git a/recipes-qt3/qt3/qt-x11-free-native_3.3.5.bb b/recipes-qt3/qt3/qt-x11-free-native_3.3.5.bb
index 2d87d26..b18ea0d 100644
--- a/recipes-qt3/qt3/qt-x11-free-native_3.3.5.bb
+++ b/recipes-qt3/qt3/qt-x11-free-native_3.3.5.bb
@@ -40,6 +40,7 @@ inherit native
do_configure() {
echo "yes" | ./configure -prefix ${prefix} ${QT_CONFIG_FLAGS} -fast -L ${STAGING_LIBDIR} \
-I ${STAGING_INCDIR_NATIVE}
+ sed -i -e '\:obj/release-shared-mt/qpngio.o\ kerne:s:$(CXXFLAGS):-I3rdparty/libpng $(CXXFLAGS):' src/Makefile
}

do_compile() {
--
1.7.9.5


cannot bitbake a java runtime

Tim Coote <tim+yoctoproject.org@...>
 

Hullo
I'm trying to get a java runtime on a qemu arm image. I'm following this: http://bit.ly/UW7fVs, which I think is the right thing to do. However, when I get to:

bitbake openjdk-6

I get the following error:

ERROR: Nothing PROVIDES 'giflib' (but /opt/poky/poky-danny-8.0/meta-java/recipes-core/openjdk/openjdk-6_6b24-1.11.5.bb DEPENDS on or otherwise requires it)
ERROR: Required build target 'openjdk-6' has no buildable providers.
Missing or unbuildable dependency chain was: ['openjdk-6', 'giflib']


Am I using the wrong documentation, or is there a missing package?

Any pointers welcome.

Tim


Re: I think my kernel image just went back in time...

Tomas Frydrych <tf+lists.yocto@...>
 

On 09/12/12 23:17, Chris Tapp wrote:
In tmp/deploy/images:

1) bzImage had a timestamp of 2012-12-09 21:49 and the file ended
-20121209214459.bin

2) Cleaned the image and the task it contained (to pick up some
changes on rebuild) and deleted the image files (.hddimg, etc.) in
tmp/deploy/images making sure to leave the kernel files intact.
Did you cleansstate it? I find that without cleaning the state kernel
image gets always pulled out of sstate rather than properly rebuilt.

Tomas



--
http://sleepfive.com


Re: [meta-freescale] uboot i.MX6

Daiane Angolini <daiane.list@...>
 

On Fri, Dec 21, 2012 at 3:34 PM, McClintock Matthew-B29882
<B29882@...> wrote:
On Thu, Dec 6, 2012 at 4:24 AM, Tim Verstraete
<Tim.Verstraete@...> wrote:
I have build the i.MX6 with danny release of yocto and latest freescale meta
using the core-image-minimal image for imx6qsabresd target machine è result
is build has done everything BUT I when I upload my uboot to my sd card I
get nothing. When I change uboot to the image from the freescale website
everything runs …
What´s your board? Are you using imx6 Sabre SD?
I´ve been working with imx6qsabresd last week and I always boot the board.

Let´s understand why you facing this behavior.


Daiane


Re: uboot i.MX6

McClintock Matthew-B29882 <B29882@...>
 

Adding meta-freescale.

-M

On Thu, Dec 6, 2012 at 4:24 AM, Tim Verstraete
<Tim.Verstraete@...> wrote:
Hi everybody,



I have build the i.MX6 with danny release of yocto and latest freescale meta
using the core-image-minimal image for imx6qsabresd target machine è result
is build has done everything BUT I when I upload my uboot to my sd card I
get nothing. When I change uboot to the image from the freescale website
everything runs …



So this means the kernel and root file system are just fine, only uboot is
not ok. My uboot is about half the size of the freescale uboot.



Do I need to do anything special for uboot to run on my device?



Thanks in advance,



Kind regards,



Tim




_______________________________________________
yocto mailing list
yocto@...
https://lists.yoctoproject.org/listinfo/yocto


Re: I think my kernel image just went back in time...

McClintock Matthew-B29882 <B29882@...>
 

On Sun, Dec 9, 2012 at 5:17 PM, Chris Tapp <opensource@...> wrote:
In tmp/deploy/images:

1) bzImage had a timestamp of 2012-12-09 21:49 and the file ended -20121209214459.bin

2) Cleaned the image and the task it contained (to pick up some changes on rebuild) and deleted the image files (.hddimg, etc.) in tmp/deploy/images making sure to leave the kernel files intact.

3) Rebuilt the image. The bzImage now has a timestamp of 2012-12-09 11:13 and the name ends -20121209105420.bin

This looks to me like the kernel image has reverted to an earlier one! It is certainly behaving as if this is the case, because a kernel option that I added is no longer in the deployed image...
Did it redeploy from an older sstate-cache?

-M


Re: [meta-freescale] hiawatha-8.6 using Yocto

Eric Bénard <eric@...>
 

Hi,

Le Fri, 21 Dec 2012 21:41:38 +0530,
GOPIKRISHNAN S <gopikrishnans@...> a écrit :
I was trying to install hiawatha on my freescale p2041rdb. But it is using
cmake to configure and install.
How to write recipe for this with cmake. I have writted recipe for packages
with autotools, and make file.
Any Idea???
by using meta-openembedded/meta-webserver you can get hiawatha running
in a few minutes :
http://cgit.openembedded.org/meta-openembedded/tree/meta-webserver/recipes-httpd/hiawatha

Best regards
Eric


Re: hiawatha-8.6 using Yocto

Burton, Ross <ross.burton@...>
 

On 21 December 2012 16:11, GOPIKRISHNAN S <gopikrishnans@...> wrote:
I was trying to install hiawatha on my freescale p2041rdb. But it is using
cmake to configure and install.
How to write recipe for this with cmake. I have writted recipe for packages
with autotools, and make file.
Any Idea???
Basically, instead of "inherit autotools", use "inherit cmake".

In oe-core have a look at libmusicbrainz or libproxy for examples.

Ross


hiawatha-8.6 using Yocto

GOPIKRISHNAN S <gopikrishnans@...>
 

Hi,

I was trying to install hiawatha on my freescale p2041rdb. But it is using cmake to configure and install.
How to write recipe for this with cmake. I have writted recipe for packages with autotools, and make file.
Any Idea???


Reg,
Gopi Krishnan S


Re: Set hostname on image?

Tomas Frydrych <tf+lists.yocto@...>
 

On 21/12/12 15:04, Jonas Jonsson L wrote:
The installed (/etc/hostname) file is created & installed in the
base-files recipe, but I don't understand how to set/use the
hostname-variable in that recipe so that it becomes what I want it to be
(unless I start editing that recipe ....).

Is there a way to modify the hostname from my 'local.conf' file?
You will not be able to override it in local.conf because the 'hostname'
variable in the recipe is not weakly assigned. But you should be able to
override it in a base-files.bbappend.

Tomas


Set hostname on image?

Jonas Jonsson L <jonas.l.jonsson@...>
 

Hi!
 
This is probably a very stupid question, but I don't understand how to set the hostname of the system that I'm building (I'm building a 'core-image-minimal'), unless I
 
A) write a /etc/hostname file to the image, or
B) write a script that updates the hostname during startup
 
The installed (/etc/hostname) file is created & installed in the base-files recipe, but I don't understand how to set/use the hostname-variable in that recipe so that it becomes what I want it to be (unless I start editing that recipe ....).
 
Is there a way to modify the hostname from my 'local.conf' file?
 
Best regards, Merry Christmas & Happy New Year!
 
Jonas
 
 


Can't build systemd from meta-systemd

Jerrod Peach <peachj@...>
 

All,

I'm getting a failure in do_configure when trying to build systemd from meta-systemd: 

<my path>/poky/build/tmp/sysroots/x86_64-linux/usr/bin/i586-poky-linux/../../lib/i586-poky-linux/gcc/i586-poky-
linux/4.7.2/include-fixed/limits.h:169:61: error: no include path in which to search for limits.h

(You can find a little more context at http://pastebin.com/3EeHMTtG, if it'll help)

Thoughts as to what's going wrong?

Kind regards,

Jerrod


Tasks must be python executable or both python and shell executable?

Biao <huanmateme@...>
 

Hi All,

There is a description in the manual of the bitbake,
#1 "In BitBake, each step that needs to be run for a given .bb is known as a task. There is a command 
addtask to add new tasks (must be a defined Python executable metadata and must start with do_) and describe intertask dependencies."
#2 But i saw a lot of tasks is shell executable. 
#1 and #2 seem contradict with each other, what is the real story?

Thanks,
Biao


Where does the python's output to?

Biao <huanmateme@...>
 

Hi All,

I have a task in a dum.bb file, after running the task with -vDDD, i can not find where the "I am the python output" goes, it is not in the terminal, not in the log file. Where does the sys.__stdout__.write output to?

/*----code snap begins---*/
python do_dm2 () {
        import sys
        sys.__stdout__.write("I am the python output\n")
}
addtask dm2
/*----code snap ends---*/

Best Regards,
Biao


Re: [PATCH 4/5] yocto-layer: add 'layer' template data

Tom Zanussi <tom.zanussi@...>
 

On Thu, 2012-12-20 at 09:29 -0700, Gary Thomas wrote:
On 2012-12-20 09:23, Tom Zanussi wrote:
On Wed, 2012-12-19 at 16:40 -0800, Darren Hart wrote:

On 12/17/2012 09:51 AM, tom.zanussi@... wrote:
From: Tom Zanussi <tom.zanussi@...>

Add a 'layer' target containing all the data that will be used to
generate a generic yocto layer.

Signed-off-by: Tom Zanussi <tom.zanussi@...>
---
Great stuff Tom, only one nit below:

new file mode 100644
index 0000000..90ce90e
--- /dev/null
+++ "b/scripts/lib/bsp/substrate/target/arch/layer/{{ if create_example_recipe == \"y\": }} recipes-example/example/{{=example_recipe_name}}-0.1/helloworld.c"
@@ -0,0 +1,6 @@
+#include <stdio.h>
+
+int main(int argc, char **argv)
+{
+ printf("Hello World!\n");
If you specify a return type of int, you should be returning one:

return 0;

:-)
Hmm, I don't think that's actually necessary for main() - if you don't
specify a return value for main(), it should already implicitly return 0
IIRC...
Not true - that would be compiler specific and should not be relied on.
Always best to be explicit (plus it will eliminate a GCC warning!)
Yeah, true, and definitely can't hurt in any case.

Tom


Tom

+}

Looks good otherwise.


Re: [PATCH 4/5] yocto-layer: add 'layer' template data

Gary Thomas
 

On 2012-12-20 09:23, Tom Zanussi wrote:
On Wed, 2012-12-19 at 16:40 -0800, Darren Hart wrote:

On 12/17/2012 09:51 AM, tom.zanussi@... wrote:
From: Tom Zanussi <tom.zanussi@...>

Add a 'layer' target containing all the data that will be used to
generate a generic yocto layer.

Signed-off-by: Tom Zanussi <tom.zanussi@...>
---
Great stuff Tom, only one nit below:

new file mode 100644
index 0000000..90ce90e
--- /dev/null
+++ "b/scripts/lib/bsp/substrate/target/arch/layer/{{ if create_example_recipe == \"y\": }} recipes-example/example/{{=example_recipe_name}}-0.1/helloworld.c"
@@ -0,0 +1,6 @@
+#include <stdio.h>
+
+int main(int argc, char **argv)
+{
+ printf("Hello World!\n");
If you specify a return type of int, you should be returning one:

return 0;

:-)
Hmm, I don't think that's actually necessary for main() - if you don't
specify a return value for main(), it should already implicitly return 0
IIRC...
Not true - that would be compiler specific and should not be relied on.
Always best to be explicit (plus it will eliminate a GCC warning!)


Tom

+}

Looks good otherwise.
--
------------------------------------------------------------
Gary Thomas | Consulting for the
MLB Associates | Embedded world
------------------------------------------------------------


Re: [PATCH 4/5] yocto-layer: add 'layer' template data

Tom Zanussi <tom.zanussi@...>
 

On Wed, 2012-12-19 at 16:40 -0800, Darren Hart wrote:

On 12/17/2012 09:51 AM, tom.zanussi@... wrote:
From: Tom Zanussi <tom.zanussi@...>

Add a 'layer' target containing all the data that will be used to
generate a generic yocto layer.

Signed-off-by: Tom Zanussi <tom.zanussi@...>
---
Great stuff Tom, only one nit below:

new file mode 100644
index 0000000..90ce90e
--- /dev/null
+++ "b/scripts/lib/bsp/substrate/target/arch/layer/{{ if create_example_recipe == \"y\": }} recipes-example/example/{{=example_recipe_name}}-0.1/helloworld.c"
@@ -0,0 +1,6 @@
+#include <stdio.h>
+
+int main(int argc, char **argv)
+{
+ printf("Hello World!\n");
If you specify a return type of int, you should be returning one:

return 0;

:-)
Hmm, I don't think that's actually necessary for main() - if you don't
specify a return value for main(), it should already implicitly return 0
IIRC...

Tom

+}

Looks good otherwise.


Re: How to update Git clone when rebuilding

Tino Breddin <tb@...>
 

Hi Jerrod,

That looks promising. Will try that once I can build again.

Thanks,
Tino

On 12/20/2012 04:38 PM, Jerrod Peach wrote:
Tino,

SRCREV = "${AUTOREV}"

You can read a little more about that in these two docs:

http://www.yoctoproject.org/docs/current/poky-ref-manual/poky-ref-manual.html#ref-variables-glossary
http://www.yoctoproject.org/docs/1.3/dev-manual/dev-manual.html#platdev-appdev-srcrev

Kind regards,

Jerrod


On Thu, Dec 20, 2012 at 10:34 AM, Tino Breddin <tb@...> wrote:

Hi,

I've noticed that bitbake won't update a Git checkout of an
application referenced through a recipe when running a follow-up build.

Is there some way to tell bitbake to check whether the remote Git
repository provides updates which should trigger a rebuild of the
application?

Cheers,
Tino
_______________________________________________
yocto mailing list
yocto@...
https://lists.yoctoproject.org/listinfo/yocto
--
Tino Breddin
Technical Director

email: tb@...
mobile: +49-151-27054995

------------------- enabling your networks -------------------

Travelping GmbH phone: +49-391-819099229
Roentgenstr. 13 fax: +49-391-819099299
D-39108 Magdeburg email: info@...
GERMANY web: http://www.travelping.com


Company Registration: Amtsgericht Stendal Reg No.: HRB 10578
Geschaeftsfuehrer: Holger Winkelmann | VAT ID No.: DE236673780
--------------------------------------------------------------


Re: [PATCH 0/5] RFC: new 'yocto-layer' tool for creating generic Yocto layers

Tom Zanussi <tom.zanussi@...>
 

On Wed, 2012-12-19 at 16:34 -0800, Darren Hart wrote:
On 12/17/2012 09:51 AM, tom.zanussi@... wrote:
From: Tom Zanussi <tom.zanussi@...>

Since I've been doing kind of similar work lately for the 'custom kernel'
support for yocto-bsp and have gotten several requests lately (mainly
from Darren in support of the new kernel documentation) for something
like this, I decided to just go ahead and try to quickly implement a
general-purpose layer-generation tool based on the BSP-generation code
used in the yocto-bsp tool.

There's actually an enhancement request bug for this already in the
Yocto bugzilla, but it doesn't contain many details:

Bug 3094 - Add a layer generation tool:

https://bugzilla.yoctoproject.org/show_bug.cgi?id=3094

So anyway, this is my initial take on what that bug means. It's an RFC for
a couple reasons, first that I'd like to get comments on whether this is
actually what's needed for a 'layer generation tool', and second, because I
still need to do some cleanup of the refactored code, but don't want to
waste much time on it if it's not really what's needed.

Below are a couple examples of how it's used - you really have to try it
yourself to see what's generated - I don't want to post tarballs or such on
the list and it's simple to generate and look at the layers.

The first case is just a very simple layer with a layer.conf and README -
basically the simplest layer you can create, and which exists mainly because
even that is easy to get wrong. As with the yocto-bsp tool, the script
queries the user for a couple items, here we just take the defaults, which
are a priority of 6 for the layer and no other components such as example
recipes:

[trz@empanada build]$ yocto-layer create simplestlayer
Please enter the layer priority you'd like to use for the layer: [default: 6]
Would you like to have an example recipe created? (y/n) [default: n]
Would you like to have an example bbappend file created? (y/n) [default: n]

Can these options be specified on the cli? It would be nice for
documentation to not have to list all the questions to describe how to
accomplish a specific task such as this. e.g.

[trz@empanada build]$ yocto-layer create simplestlayer -p 6 --no-recipe
--no-bbappend

Or possibly having defaults to those questions accepted by default
unless [-i|--interactive] is specified?
Well, you can do it all from the command-line if you use the -i command
(where -i doesn't mean interactive, just the opposite in fact) with a
JSON file:

[trz@empanada build]$ yocto-layer create jtest -i yocto-layer-input.json

With the contents of yocto-layer-input.json being values for the
properties you get from 'yocto-layer list properties':

{"layer_priority":"6","create_example_recipe":"n","create_example_bbappend":"n","example_recipe_name":"example","example_bbappend_name":"example","example_bbappend_version":"0.1"}

Obviously this isn't designed for humans to use directly - the purpose
is to allow other tools to use the Yocto BSP tools but avoid having to
display the default text-based interface. ADT uses this for instance to
create an Eclipse UI for yocto-bsp.

I could add specific command-line options for each of those properties,
or a subset if that makes sense, but the templating engine and the
resulting tools were meant to be 'user-friendly' and 'menu-driven' for
complete novices who don't necessarily want to read documentation.

I'm thinking that most users probably won't use these tools very often,
so providing a complete command-line equivalent encompassing all the
options would be more trouble than it's worth. It might make sense
though to do it for the case of the 'simplest layer' which would mean
just adding a 'priority' option in addition to the layer name already
specified on the command line.

empanada :-) Mmmmmm
Yeah, that's what happens when you install a new system just before
dinner. ;-)


New layer created in meta-simplestlayer.

Don't forget to add it to your BBLAYERS (for details see meta-simplestlayer\README).

[trz@empanada build]$ find .
.
./meta-simplestlayer
./meta-simplestlayer/conf
./meta-simplestlayer/conf/layer.conf
./meta-simplestlayer/README
./meta-simplestlayer/COPYING.MIT

It might make sense to create a recipes/ directory as well, just to
support convention.
Not sure what you mean - if you say 'y' to either the recipe or the
bbappend, it will create a recipes-example/ dir to contain them. Do you
mean something else?



In the second case, we tell the tool that we do want an example .bb and and
an example .bbappend. We're queried for the recipe name that we want our
recipe to have, and for the .bbappend, the name of the base recipe and its
version. Below you can see the files it generates - please look at the files
themselves to see the contents. For the recipe example, it generates a recipe
based on the example in the Yocto manual, and my own helloworld.c code
(untested so far which is also why this is an RFC). For the .bbappend example,
it just creates an empty .patch file with some instructions on what to do to
modify the parent recipe with a patch:

[trz@empanada build]$ yocto-layer create mylayer
Please enter the layer priority you'd like to use for the layer: [default: 6]
Would you like to have an example recipe created? (y/n) [default: n] y
Please enter the name you'd like to use for your example recipe: [default: example] flork
Would you like to have an example bbappend file created? (y/n) [default: n] y
Please enter the name you'd like to use for your bbappend file: [default: example] chork
Please enter the version number you'd like to use for your bbappend file (this should match the recipe you're appending to): [default: 0.1] 0.22.3

Wow, that's quite a bit more than I was asking for... I can imagine that
some folks my find that useful.



New layer created in meta-mylayer.

Don't forget to add it to your BBLAYERS (for details see meta-mylayer\README).

[trz@empanada build]$ find .
.
./meta-mylayer
./meta-mylayer/recipes-example
./meta-mylayer/recipes-example/example
./meta-mylayer/recipes-example/example/flork-0.1
./meta-mylayer/recipes-example/example/flork-0.1/helloworld.c
./meta-mylayer/recipes-example/example/flork-0.1/example.patch
./meta-mylayer/recipes-example/example/flork_0.1.bb
./meta-mylayer/conf
./meta-mylayer/conf/layer.conf
./meta-mylayer/recipes-example-bbappend
./meta-mylayer/recipes-example-bbappend/example-bbappend
./meta-mylayer/recipes-example-bbappend/example-bbappend/chork-0.22.3
./meta-mylayer/recipes-example-bbappend/example-bbappend/chork-0.22.3/example.patch
./meta-mylayer/recipes-example-bbappend/example-bbappend/chork_0.22.3.bbappend
./meta-mylayer/README
./meta-mylayer/COPYING.MIT

Please try it and look at the generated code - any comments or suggestions
are welcome.

In concept, looks pretty good to me, and a natural refactoring of
yocto-bsp. I'll review the patches independently.
Thanks for reviewing them!

Tom

Thanks,

Darren


Thanks,

Tom

The following changes since commit da598d4f0fcf6faa62055084a51d70d735399d33:

eglibc: fix checksums of fetched patches (2012-12-13 18:02:21 +0000)

are available in the git repository at:

git://git.yoctoproject.org/poky-contrib.git tzanussi/yocto-layer-rfc
http://git.yoctoproject.org/cgit/cgit.cgi/poky-contrib/log/?h=tzanussi/yocto-layer-rfc

Tom Zanussi (5):
scripts/lib/bsp/engine.py: add yocto_layer_create()
yocto-layer: new script
yocto-layer: add help/usage
yocto-layer: add 'layer' template data
scripts/lib/bsp/engine.py: refactor bsp-creation code

scripts/lib/bsp/engine.py | 86 +++++++-
scripts/lib/bsp/help.py | 228 +++++++++++++++++++++
.../bsp/substrate/target/arch/layer/COPYING.MIT | 17 ++
scripts/lib/bsp/substrate/target/arch/layer/README | 64 ++++++
.../substrate/target/arch/layer/conf/layer.conf | 10 +
.../target/arch/layer/layer-questions.noinstall | 14 ++
.../example.patch" | 12 ++
..._name}}_{{=example_bbappend_version}}.bbappend" | 8 +
.../{{=example_recipe_name}}-0.1/example.patch" | 12 ++
.../{{=example_recipe_name}}-0.1/helloworld.c" | 6 +
.../example/{{=example_recipe_name}}_0.1.bb" | 23 +++
scripts/yocto-layer | 142 +++++++++++++
12 files changed, 612 insertions(+), 10 deletions(-)
create mode 100644 scripts/lib/bsp/substrate/target/arch/layer/COPYING.MIT
create mode 100644 scripts/lib/bsp/substrate/target/arch/layer/README
create mode 100644 scripts/lib/bsp/substrate/target/arch/layer/conf/layer.conf
create mode 100644 scripts/lib/bsp/substrate/target/arch/layer/layer-questions.noinstall
create mode 100644 "scripts/lib/bsp/substrate/target/arch/layer/{{ if create_example_bbappend == \"y\": }} recipes-example-bbappend/example-bbappend/{{=example_bbappend_name}}-{{=example_bbappend_version}}/example.patch"
create mode 100644 "scripts/lib/bsp/substrate/target/arch/layer/{{ if create_example_bbappend == \"y\": }} recipes-example-bbappend/example-bbappend/{{=example_bbappend_name}}_{{=example_bbappend_version}}.bbappend"
create mode 100644 "scripts/lib/bsp/substrate/target/arch/layer/{{ if create_example_recipe == \"y\": }} recipes-example/example/{{=example_recipe_name}}-0.1/example.patch"
create mode 100644 "scripts/lib/bsp/substrate/target/arch/layer/{{ if create_example_recipe == \"y\": }} recipes-example/example/{{=example_recipe_name}}-0.1/helloworld.c"
create mode 100644 "scripts/lib/bsp/substrate/target/arch/layer/{{ if create_example_recipe == \"y\": }} recipes-example/example/{{=example_recipe_name}}_0.1.bb"
create mode 100755 scripts/yocto-layer


Re: How to update Git clone when rebuilding

Jerrod Peach <peachj@...>
 


On Thu, Dec 20, 2012 at 10:34 AM, Tino Breddin <tb@...> wrote:
Hi,

I've noticed that bitbake won't update a Git checkout of an
application referenced through a recipe when running a follow-up build.

Is there some way to tell bitbake to check whether the remote Git
repository provides updates which should trigger a rebuild of the
application?

Cheers,
Tino
_______________________________________________
yocto mailing list
yocto@...
https://lists.yoctoproject.org/listinfo/yocto