Signed-off-by: Markus Volk
---
meta/recipes-support/libical/libical_3.0.12.bb | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/meta/recipes-support/libical/libical_3.0.12.bb
b/meta/recipes-support/libical/libical_3.0.12.bb
index b2b2211b9b..614a76ba5b 100644
--- a/meta/r
On 1/20/22 9:47 PM, Peter Kjellerstedt wrote:
-Original Message-
From: openembedded-core@lists.openembedded.org On Behalf Of Robert Yang
Sent: den 20 januari 2022 08:09
To: openembedded-core@lists.openembedded.org
Cc: jupiter@gmail.com
Subject: [OE-core] [PATCH 1/1] bitbake.conf: A
On 1/20/22 9:38 PM, Richard Purdie wrote:
On Wed, 2022-01-19 at 23:09 -0800, Robert Yang wrote:
The NIS or icecc can't work when task network is dissable, add BB_TASK_NETWORK
to enable network globally for such exceptions.
Note, enable nscd on the build machine might be a solution, but that i
On 1/20/22 9:31 PM, Quentin Schulz wrote:
Hi Robert,
On 1/20/22 08:09, Robert Yang wrote:
The NIS or icecc can't work when task network is dissable, add BB_TASK_NETWORK
to enable network globally for such exceptions.
Note, enable nscd on the build machine might be a solution, but that isn't
Markus,
Thanks for the patch, seems sensible to me but please resend to:
openembedded-de...@lists.openembedded.org
as documented in the README:
$ grep send-email meta-gnome/README
git send-email -M -1 --to openembedded-de...@lists.openembedded.org
--subject-prefix='meta-gnome][PATCH'
O
On Fri, 21 Jan 2022 at 01:01, Markus Volk wrote:
> This needs to be conditionally enabled similar to webkitgtk recipe:
> ${@bb.utils.contains('GI_DATA_ENABLED', 'True',
> '-DENABLE_INTROSPECTION=ON', '-DENABLE_INTROSPECTION=OFF', d)} \
>
> Most Gnome Projects meanwhile consider go
This needs to be conditionally enabled similar to webkitgtk recipe:
${@bb.utils.contains('GI_DATA_ENABLED', 'True',
'-DENABLE_INTROSPECTION=ON', '-DENABLE_INTROSPECTION=OFF', d)} \
Most Gnome Projects meanwhile consider gobject-introspection to be
mandatory. Are you sure this n
On Thu, 20 Jan 2022 at 23:55, Richard Purdie <
richard.pur...@linuxfoundation.org> wrote:
> Disabling shared libraries because there is a makefile bug doesn't seem
> like the
> right solution :(.
>
> Better might be disabling parallelism in make install. It raises questions
> about
> why it is lin
On Fri, 21 Jan 2022 at 00:41, Markus Volk wrote:
> +EXTRA_OECMAKE += "-DGOBJECT_INTROSPECTION=ON"
>
This needs to be conditionally enabled similar to webkitgtk recipe:
${@bb.utils.contains('GI_DATA_ENABLED', 'True',
'-DENABLE_INTROSPECTION=ON', '-DENABLE_INTROSPECTION=OFF', d)} \
Gnome 'folks' backends depend on python3-dbusmock-native
Signed-off-by: Markus Volk
---
meta/recipes-devtools/python/python3-dbusmock_0.25.0.bb | 2 ++
1 file changed, 2 insertions(+)
diff --git a/meta/recipes-devtools/python/python3-dbusmock_0.25.0.bb
b/meta/recipes-devtools/python/python3-db
Signed-off-by: Markus Volk
---
meta/recipes-support/libical/libical_3.0.12.bb | 18 +-
1 file changed, 17 insertions(+), 1 deletion(-)
diff --git a/meta/recipes-support/libical/libical_3.0.12.bb
b/meta/recipes-support/libical/libical_3.0.12.bb
index b2b2211b9b..fef569ea64 100644
On Thu, 2022-01-20 at 23:22 +0100, Alexander Kanavin wrote:
> It was found to cause races in make install, as observed in
> https://bugzilla.yoctoproject.org/show_bug.cgi?id=14695
>
> [YOCTO #14695]
>
> Signed-off-by: Alexander Kanavin
> ---
> meta/recipes-devtools/ruby/ruby_3.1.0.bb | 2 +-
>
It was found to cause races in make install, as observed in
https://bugzilla.yoctoproject.org/show_bug.cgi?id=14695
[YOCTO #14695]
Signed-off-by: Alexander Kanavin
---
meta/recipes-devtools/ruby/ruby_3.1.0.bb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta/recipes-devto
From: Marek Vasut
Commit 4efdcc1090 ("weston: Use systemd notify,") has non-trivial to
backport dependencies without which it cannot work, revert backport.
In oe-core dunfell, weston is still started using /usr/bin/weston-start
script in meta/recipes-graphics/wayland/weston-init/weston@.service
From: Richard Purdie
This was being provided by other pieces of the dependency chain but is
specifically required by configure and could fail if those pieces come
from sstate. Fix such builds by adding the missing dependency.
Signed-off-by: Richard Purdie
(cherry picked from commit ba2587beb2a3
From: Bruce Ashfield
The 5.16 kernel introduced mandatory schema checking on any dtb file
built through the kernel.
That funcionality is provided via python3-dt-schema.
The dependencies to enable that functionality is not small, and may
not always be desired (in particular on architectures that
From: Bruce Ashfield
Updating linux-yocto/5.4 to the latest korg -stable release that comprises
the following commits:
b7f70762d158 Linux 5.4.172
f415409551b0 staging: greybus: fix stack size warning with UBSAN
65c2e7176f77 drm/i915: Avoid bitwise vs logical OR warning in
snb_wm_lat
From: Bruce Ashfield
Updating linux-yocto/5.4 to the latest korg -stable release that comprises
the following commits:
0a4ce4977bbe Linux 5.4.171
0101f118529d mISDN: change function names to avoid conflicts
34821931e18e atlantic: Fix buff_ring OOB in aq_ring_rx_clean
44065cc11797
From: Bruce Ashfield
Updating linux-yocto/5.4 to the latest korg -stable release that comprises
the following commits:
047dedaa38ce Linux 5.4.170
2c3920c58e03 perf script: Fix CPU filtering of a script's switch events
fe5838c22b98 net: fix use-after-free in tw_timer_handler
46556c4ecd63 Input: s
From: Bruce Ashfield
Updating linux-yocto/5.4 to the latest korg -stable release that comprises
the following commits:
4ca2eaf1d477 Linux 5.4.169
48c76fc53582 phonet/pep: refuse to enable an unbound pipe
a5c6a13e9056 hamradio: improve the incomplete fix to avoid NPD
ef5f7bfa19e3 hamradio: defer
From: Kai Kang
Backport patch to fix CVE-2020-23903.
CVE: CVE-2020-23903
Signed-off-by: Kai Kang
Signed-off-by: Richard Purdie
(cherry picked from commit b8f56e5e9eef32c1e01742f913e205d93548de1f)
Signed-off-by: Steve Sakoman
---
.../speex/speex/CVE-2020-23903.patch | 30 +++
In doProlog in xmlparse.c in Expat (aka libexpat) before 2.4.3, an
integer overflow exists for m_groupSize.
Backport patch from:
https://github.com/libexpat/libexpat/pull/538/commits/85ae9a2d7d0e9358f356b33977b842df8ebaec2b
CVE: CVE-2021-46143
Signed-off-by: Steve Sakoman
---
.../expat/expat/CV
In Expat (aka libexpat) before 2.4.3, a left shift by 29 (or more)
places in the storeAtts function in xmlparse.c can lead to realloc
misbehavior (e.g., allocating too few bytes, or only freeing memory).
Backport patch from:
https://github.com/libexpat/libexpat/pull/534/commits/0adcb34c49bee5b19bd
xmlparse.c has multiple integer overflows. The involved functions are:
- addBinding (CVE-2022-22822)
- build_model (CVE-2022-22823)
- defineAttribute (CVE-2022-22824)
- lookup (CVE-2022-22825)
- nextScaffoldPart (CVE-2022-22826)
- storeAtts (CVE-2022-22827)
Backport patch from:
https://github.com
Please review this set of patches for dunfell and have comments back by end
of day Monday.
Passed a-full on autobuilder:
https://autobuilder.yoctoproject.org/typhoon/#/builders/83/builds/3141
The following changes since commit 01f256bc72fb45c80b6a6c77506bc4c375965a3a:
glibc: Add fix for data
Interesting, I thought the image-prelink class had been removed completely, but
apparently it was only the references to it in local.conf.sample that was
removed.
Anyway, if you are going to do that change, I believe it is better to use
bb.data.inherits_class() to see if the image-prelink class
Not all of these are ffmpeg caused, e.g. picking a few random items shows
they are new meson issues:
https://errors.yoctoproject.org/Errors/Details/622936/
https://errors.yoctoproject.org/Errors/Details/622943/
I think we should also check whether something is abandonware, e.g.
minidlna had its l
On Thu, 2022-01-20 at 19:43 +0100, Michael Opdenacker wrote:
> Hi Alex
>
> On 1/20/22 4:28 PM, Alexander Kanavin wrote:
> > I think this should cover aarch64 on aarch64 too :) And presence of
> > (and ability to open) /dev/kvm is probably a better check than looking
> > through cpuinfo?
>
> Thank
On Thu, 2022-01-20 at 12:41 -0500, bkyleruss...@gmail.com wrote:
> Yes, we do use prelink. I think our use case primarily benefits from CoW
> memory savings, rather than load times. Of course, GCCPIE can be overridden
> in the distro layer, but seeing as image-prelink.bbclass still exists
> upstr
On Thu, Jan 20, 2022 at 12:28 PM Marek Vasut wrote:
>
> On 1/20/22 21:15, Khem Raj wrote:
> > there are bunch of failures in meta-openembedded as well see
> >
> > https://errors.yoctoproject.org/Errors/Build/138784/
>
> So, how shall we proceed here ?
>
> I suspect Alex will send a patch for the
sorry, missed that. I'll send a v2 patch for it
Am 20.01.22 um 21:28 schrieb Khem Raj:
I posted the link with error message the specific error is
Run-time dependency libxml-2.0 found: NO (tried pkgconfig)
../folks-0.15.4/meson.build:126:2: ERROR: Dependency "libxml-2.0" not
found, tried pkgcon
I posted the link with error message the specific error is
Run-time dependency libxml-2.0 found: NO (tried pkgconfig)
../folks-0.15.4/meson.build:126:2: ERROR: Dependency "libxml-2.0" not
found, tried pkgconfig
On Thu, Jan 20, 2022 at 12:22 PM Markus Volk wrote:
>
> mmmh. Whats the error messag
On 1/20/22 21:15, Khem Raj wrote:
there are bunch of failures in meta-openembedded as well see
https://errors.yoctoproject.org/Errors/Build/138784/
So, how shall we proceed here ?
I suspect Alex will send a patch for the problem detected in oe-core and
then oe-core with ffmpeg 5.0 won't be
mmmh. Whats the error message ?
I can build folks like this for aarch64 and x86-64
Am 20.01.22 um 21:11 schrieb Khem Raj:
+SUMMARY = "Folks is a contact aggregation library."
+LICENSE = "LGPLv2.1"
+LIC_FILES_CHKSUM ="file://COPYING;md5=2d5025d4aa3495befef8f17206a5b0a1"
+
+DEPENDS = " \
+gli
there are bunch of failures in meta-openembedded as well see
https://errors.yoctoproject.org/Errors/Build/138784/
On Tue, Jan 18, 2022 at 12:48 PM Alexander Kanavin
wrote:
>
> This update breaks gst-libav, and I have a patch in the works that fixes that
> too :)
>
> Alex
>
> On Tue, 18 Jan 202
fails to configure
https://errors.yoctoproject.org/Errors/Details/622929/
On Thu, Jan 20, 2022 at 7:18 AM Markus Volk wrote:
>
> Signed-off-by: Markus Volk
> ---
> .../folks/folks_0.15.4.bb | 23 +++
> 1 file changed, 23 insertions(+)
> create mode 100644 m
On Thu, 20 Jan 2022 at 19:43, Michael Opdenacker <
michael.opdenac...@bootlin.com> wrote:
> However, my abilities to build and test on aarch64 are limited. At least
> this change would take me much more time to develop and is more likely
> to cause regressions.
>
Checking for readability of /dev/
This isn't correct, as recipe's own sysroot is populated by
prepare_recipe_sysroot() of the recipe.
I think the real correct fix is
addtask do_make_icudata before do_configure after do_patch
do_prepare_recipe_sysroot
You need to tweak the line at the end of the recipe.
Alex
On Thu, 20 Jan 2022
Hi Alex
On 1/20/22 4:28 PM, Alexander Kanavin wrote:
> I think this should cover aarch64 on aarch64 too :) And presence of
> (and ability to open) /dev/kvm is probably a better check than looking
> through cpuinfo?
Thanks for the review!
However, my abilities to build and test on aarch64 are lim
thanks for the patches, they should be sent to oe-devel mailing list.
On 1/20/22 7:18 AM, Markus Volk wrote:
Signed-off-by: Markus Volk
---
.../recipes-connectivity/geary/geary_40.0.bb | 54 +++
1 file changed, 54 insertions(+)
create mode 100644 meta-gnome/recipes-connect
This updates the QEMU sounds options for x86 emulation,
when "runqemu" is called with the "audio" argument,
to fix the below error:
runqemu - ERROR - Failed to run qemu: qemu-system-x86_64: warning: '-soundhw
ac97' is deprecated, please use '-device AC97' instead
Signed-off-by: Michael Opdenacke
Yes, we do use prelink. I think our use case primarily benefits from CoW
memory savings, rather than load times. Of course, GCCPIE can be
overridden in the distro layer, but seeing as image-prelink.bbclass still
exists upstream, the default definition should support configurations that
choose to
The make_icudata task is set before the configure task then the
dependencies for this task are not populate yet.
Fixed it by setting the deptask flag to do_populate_sysroot. With this, the
do_populate_sysroot task of each item in DEPENDS must complete before
do_make_icudata can execute.
Signed-of
On Thu, 20 Jan 2022 16:09:25 +0100
Alexander Kanavin wrote:
> > Could you be more explicit or use an example?
> >
>
> I think you need something like
>
> do_make_icudata[deptask] = "do_prepare_recipe_sysroot"
Oh, I didn't know this deptask flag, thanks to point me this out.
I will send a v2
I think this should cover aarch64 on aarch64 too :) And presence of (and
ability to open) /dev/kvm is probably a better check than looking through
cpuinfo?
Alex
On Thu, 20 Jan 2022 at 16:21, Michael Opdenacker <
michael.opdenac...@bootlin.com> wrote:
> This automatically turns on the "kvm" optio
This automatically turns on the "kvm" option when emulating
an x86 system on x86 CPUs with VT capability.
On an Intel i7-5600U CPU at 2.60GHz, using the "kvm"
option is at least 4x faster, booting "core-image-minimal" for qemux86-64.
The performance difference can even be bigger for larger systems
Signed-off-by: Markus Volk
---
.../recipes-connectivity/geary/geary_40.0.bb | 54 +++
1 file changed, 54 insertions(+)
create mode 100644 meta-gnome/recipes-connectivity/geary/geary_40.0.bb
diff --git a/meta-gnome/recipes-connectivity/geary/geary_40.0.bb
b/meta-gnome/recipes-c
Signed-off-by: Markus Volk
---
.../folks/folks_0.15.4.bb | 23 +++
1 file changed, 23 insertions(+)
create mode 100644 meta-gnome/recipes-connectivity/folks/folks_0.15.4.bb
diff --git a/meta-gnome/recipes-connectivity/folks/folks_0.15.4.bb
b/meta-gnome/recip
On Thu, 20 Jan 2022 at 16:03, Köry Maincent
wrote:
> Hello Alexander,
>
> On Thu, 20 Jan 2022 15:37:26 +0100
> Alexander Kanavin wrote:
>
> > Can we rather make a dependency on the task that prepares the sysroot,
> > instead of listing the dependencies one by one (with anonymous python
> too!)?
Hello Alexander,
On Thu, 20 Jan 2022 15:37:26 +0100
Alexander Kanavin wrote:
> Can we rather make a dependency on the task that prepares the sysroot,
> instead of listing the dependencies one by one (with anonymous python too!)?
Thanks for the review,
I do not understand your comments as the ma
You need to set your sstate cache directory to somewhere outside of the
build directory. Then the sstate cache will be reused between selftest
runs. The default behavior is to delete the selftest build directory if the
selftest succeeded, but the cache will ensure that no long rebuilds will
be tri
Hi everyone,
I do have some questions about a standard workflow people usually
follow when adding new test sets to meta/lib/oeqa/selftest/cases/.
When oe-selftest is invoked for example this way:
oe-selftest --run-tests wic.Wic2.test_offset --machine random
It usually creates a new build dir, t
Can we rather make a dependency on the task that prepares the sysroot,
instead of listing the dependencies one by one (with anonymous python too!)?
Alex
On Thu, 20 Jan 2022 at 15:32, Kory Maincent
wrote:
> The make_icudata task is set before the configure task then the
> dependencies for this t
The make_icudata task is set before the configure task then the
dependencies for this task are not populate yet.
Fixed it by setting autoconf, automake and ccache native dependencies
to this task.
Signed-off-by: Kory Maincent
---
meta/recipes-support/icu/icu_70.1.bb | 8
1 file changed
> -Original Message-
> From: openembedded-core@lists.openembedded.org c...@lists.openembedded.org> On Behalf Of Robert Yang
> Sent: den 20 januari 2022 08:09
> To: openembedded-core@lists.openembedded.org
> Cc: jupiter@gmail.com
> Subject: [OE-core] [PATCH 1/1] bitbake.conf: Add BB_TAS
On Wed, 2022-01-19 at 23:09 -0800, Robert Yang wrote:
> The NIS or icecc can't work when task network is dissable, add BB_TASK_NETWORK
> to enable network globally for such exceptions.
>
> Note, enable nscd on the build machine might be a solution, but that isn't
> reliable since it depends on whe
Hi Robert,
On 1/20/22 08:09, Robert Yang wrote:
The NIS or icecc can't work when task network is dissable, add BB_TASK_NETWORK
to enable network globally for such exceptions.
Note, enable nscd on the build machine might be a solution, but that isn't
reliable since it depends on whether the netw
Upstream-Status: Backport
[https://sourceware.org/git/?p=glibc.git;a=commit;h=226b46770c82899b555986583294b049c6ec9b40]
Upstream-Status: Backport
[https://sourceware.org/git/?p=glibc.git;a=commit;h=ef972a4c50014a16132b5c75571cfb6b30bef136]
Signed-off-by: pgowda
---
.../glibc/glibc/0001-CVE-202
Upstream-Status: Backport
[https://sourceware.org/git/?p=glibc.git;a=commit;h=e368b12f6c16b6888dda99ba641e999b9c9643c8]
Upstream-Status: Backport
[https://sourceware.org/git/?p=glibc.git;a=commit;h=f545ad4928fa1f27a3075265182b38a4f939a5f7]
Signed-off-by: pgowda
---
.../glibc/glibc/0001-CVE-202
I think we could address this properly by adding a 'package-database' image
feature?
Alex
On Thu, 20 Jan 2022 at 09:39, Jia, Hongxu wrote:
> On 1/20/22 14:34, Diego Santa Cruz wrote:
>
> [Please note: This e-mail is from an EXTERNAL e-mail address]
>
> Hi,
>
>
>
> We do exactly that in our imag
On 1/20/22 14:34, Diego Santa Cruz wrote:
[Please note: This e-mail is from an EXTERNAL e-mail address]
Hi,
We do exactly that in our images, keep rpm but avoid dnf. We do it by adding
this to the image recipes (via in inc file).
ROOTFS_PKGMANAGE_remove = "dnf"
ROOTFS_POSTUNINSTALL_COMMAND += "s
I think we pretty much abandoned prelink at this point, are you using it
and do you see the benefits?
Alex
On Thu, 20 Jan 2022 at 04:30, wrote:
> Since a prelinked rootfs is in conflict with PIE, don't attempt the latter
> if the image enables prelink.
> ---
> meta/conf/distro/include/security
62 matches
Mail list logo
