[OE-core] [nanbield] git command included in buildtools-extended does not work with https authentication

[Christian Eggers]

   <= Recv header: Date: Tue, 07 May 2024 
12:35:38 GMT
12:35:38.774957 http.c:767  <= Recv header, 41 bytes 
(0x0029)
12:35:38.774963 http.c:779  <= Recv header: Content-Type: 
text/plain; charset=utf-8
12:35:38.774973 http.c:767  <= Recv header, 21 bytes 
(0x0015)
12:35:38.774977 http.c:779  <= Recv header: Content-Length: 270
12:35:38.774980 http.c:767  <= Recv header, 24 bytes 
(0x0018)
12:35:38.774982 http.c:779  <= Recv header: Connection: keep-alive
12:35:38.774987 http.c:767  <= Recv header, 25 bytes 
(0x0019)
12:35:38.774989 http.c:779  <= Recv header: Cache-Control: no-cache
12:35:38.774991 http.c:767  <= Recv header, 14 bytes 
(0x000e)
12:35:38.774996 http.c:779  <= Recv header: Vary: Accept
12:35:38.774999 http.c:767  <= Recv header, 40 bytes 
(0x0028)
12:35:38.775001 http.c:779  <= Recv header: WWW-Authenticate: Basic 
realm="GitLab"
12:35:38.775007 http.c:767  <= Recv header, 33 bytes 
(0x0021)
12:35:38.775009 http.c:779  <= Recv header: X-Content-Type-Options: 
nosniff
12:35:38.775015 http.c:767  <= Recv header, 28 bytes 
(0x001c)
12:35:38.775018 http.c:779  <= Recv header: X-Download-Options: 
noopen
12:35:38.775020 http.c:767  <= Recv header, 29 bytes 
(0x001d)
12:35:38.775025 http.c:779  <= Recv header: X-Frame-Options: 
SAMEORIGIN
12:35:38.775028 http.c:767  <= Recv header, 78 bytes 
(0x004e)
12:35:38.775033 http.c:779  <= Recv header: X-Gitlab-Meta: 
{"correlation_id":"XX","version":"1"}
12:35:38.775036 http.c:767  <= Recv header, 41 bytes 
(0x0029)
12:35:38.775041 http.c:779  <= Recv header: 
X-Permitted-Cross-Domain-Policies: none
12:35:38.775044 http.c:767  <= Recv header, 42 bytes 
(0x002a)
12:35:38.775046 http.c:779  <= Recv header: X-Request-Id: 
XX
12:35:38.775051 http.c:767  <= Recv header, 21 bytes 
(0x0015)
12:35:38.775053 http.c:779  <= Recv header: X-Runtime: 0.021434
12:35:38.775056 http.c:767  <= Recv header, 21 bytes 
(0x0015)
12:35:38.775061 http.c:779  <= Recv header: X-Xss-Protection: 0
12:35:38.775064 http.c:767  <= Recv header, 02 bytes 
(0x0002)
12:35:38.775071 http.c:779  <= Recv header:
12:35:38.775087 http.c:820  == Info: Connection #0 to host 
git.mycompany.com left intact
remote: HTTP Basic: Access denied. The provided password or token is incorrect 
or your account has 2FA enabled and you must use a personal access token 
instead of a password. See 
https://git.mycompany.com/help/topics/git/troubleshooting_git#error-on-git-fetch-http-basic-access-denied
fatal: Authentication failed for 'https://git.mycompany.com/myrepo.git/'




___
Christian
Eggers
Software Engineer
​
ARRI
Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Arriweg 17,
83071
Stephanskirchen
www.arri.com

+49 8036 3009-3118
cegg...@arri.de

​

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Matthias Erb (Chairman); Lars Weyer; Walter Trauninger

​

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#199093): 
https://lists.openembedded.org/g/openembedded-core/message/199093
Mute This Topic: https://lists.openembedded.org/mt/105959618/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] kirkstone boost upgrade?

[Christian Eggers]

On Friday, 21 July 2023, 06:49:24 CEST, Thomas Roos via lists.openembedded.org 
wrote:
> Hi,
> we are hitting an issue with boost version 1.78 and like to ask if and how 
> there is a boost upgrade on kirkstone branch?

Another option:

Yesterday there were two posts on the d...@yoctoproject.org list about "mixin" 
layers [1][2].
There is already an own repository for this [3].

So the best option seems to be creating a branch like "kirkstone/boost" (for 
latest boost
release) or "kirkstone/boost-1.82" (for a specific boost release).

It looks like patches for the meta-lts-mixin layer should be posted to [4]. 
Searching for
"mixin" on the web interface shows some example messages. If you could put me 
on CC, I am
happy to test your patches.

regards
Christian


[1] https://lists.yoctoproject.org/g/docs/message/4073
[2] https://lists.yoctoproject.org/g/docs/message/4074
[3] https://git.yoctoproject.org/meta-lts-mixins/
[4] https://lists.yoctoproject.org/g/yocto/topics




_______
Christian
Eggers
Software Engineer
​
ARRI
Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Arriweg 17,
83071
Stephanskirchen
www.arri.com

+49 8036 3009-3118
cegg...@arri.de

​
Get all the latest information from www.arri.com, Facebook, Twitter, Instagram, 
LinkedIn and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Matthias Erb (Chairman); Lars Weyer; Stephan Schenk; 
Walter Trauninger

​

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#184670): 
https://lists.openembedded.org/g/openembedded-core/message/184670
Mute This Topic: https://lists.openembedded.org/mt/100271099/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] kirkstone boost upgrade?

[Christian Eggers]

Hi Thomas,

On Friday, 21 July 2023, 06:49:24 CEST, Thomas Roos via lists.openembedded.org 
wrote:
> Hi,
> we are hitting an issue with boost version 1.78 and like to ask if and how 
> there is a boost upgrade on kirkstone branch?

I've a similar situation and decided to update boost by copying the current 
recipes from
oe-core master branch into my project's local layer (which works fine for me).

I would appreciate having a newer version in kirkstone, but this may be against 
the
policy of the LTS branch.

regards
Christian




___
Christian
Eggers
Software Engineer
​
ARRI
Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Arriweg 17,
83071
Stephanskirchen
www.arri.com

+49 8036 3009-3118
cegg...@arri.de

​
Get all the latest information from www.arri.com, Facebook, Twitter, Instagram, 
LinkedIn and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Matthias Erb (Chairman); Lars Weyer; Stephan Schenk; 
Walter Trauninger

​

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#184669): 
https://lists.openembedded.org/g/openembedded-core/message/184669
Mute This Topic: https://lists.openembedded.org/mt/100271099/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] [PATCH v2 3/3] bitbake.conf: add debug symbol for sdk

[Christian Eggers]

Hi Max,

I use the following lines in my local.conf file (on kirkstone). Maybe that this 
is not
optimal as it may cause silent side effects in case the original values for 
BUILDSDK_OPTIMIZATION/_CFLAGS/_CXXFALGS will change same day.

# Build -nativesdk packages with debug info (required for generating -dbg/-src 
packages)
# BUILDSDK_OPTIMIZATION is based on BUILD_OPTIMIZATION in bitbake.conf
BUILDSDK_OPTIMIZATION = "${@oe.utils.vartrue('DEBUG_BUILD', '-Og', '-O2', d)} 
-g -feliminate-unused-debug-types -pipe"
BUILDSDK_CFLAGS = "${BUILDSDK_CPPFLAGS} ${BUILDSDK_OPTIMIZATION} 
${DEBUG_PREFIX_MAP}"
BUILDSDK_CXXFLAGS = "${BUILDSDK_CFLAGS}"


Additionally I have to manually select all available nativesdk-*-dbg/-src 
packages in my
nativesdk-packagegroup-sdk-host.bbappend file. Unfortunately these packages are 
not added
automatically to my SDK image (SDKIMAGE_FEATURES only applies for target 
packages, not for
nativesdk packages). Maybe you'll find a better solution for that.

In my project I have added BBCLASSEXTEND="nativesdk" to many bbappends in order 
to have most
libraries I use on my target also for nativesdk (I can compile my embedded 
application also
for nativesdk then). As debugging performance using gdbserver is quite poor on 
my system,
developing my application with nativesdk is a welcome alternative (that's  why 
I also have
enabled debug information nativesdk).

But from some previous comments on the OE list it seems that nativesdk is not 
really intended
for this purpose. But what is the alternative? Having two SDKs?

regards
Christian



On Friday, 23 June 2023, 16:58:56 CEST, Maxime Roussin-Bélanger wrote:
> On Fri, Jun 23, 2023 at 3:50 AM Alexander Kanavin
>  wrote:
> >
> > But then you can set the parameters as you want them in your local
> > configuration. Setting -g globally in the absence of DEBUG_BUILD needs
> > a use case that applies to most users.
> >
> > Alex
> 
> If user wants to install nativesdk dgb/src package you need to add that flag,
> so instead of modifying the `BUILDSDK_CPPFLAGS`, it should probably be
> documented.
> 
> Do you have any pointers where it could be documented?
> 
> Max.
> 
> >
> > On Thu, 22 Jun 2023 at 19:11, Maxime Roussin-Bélanger
> >  wrote:
> > >
> > > Hi Alex,
> > >
> > > I don't want a complete debug build. I want to keep optimization and
> > > have the debug symbols. I don't think I can achieve that with
> > > DEBUG_BUILD.
> > >
> > > Max
> > >
> > > On Thu, Jun 22, 2023 at 12:30 PM Alexander Kanavin
> > >  wrote:
> > > >
> > > > This change is unnecessary, as there is already a mechanism. Just
> > > > slightly further down you can see that BUILDSDK_CFLAGS pulls in
> > > > BUILD_OPTIMIZATION, which sets -g if you enable DEBUG_BUILD in your
> > > > config.
> > > >
> > > > Alex
> > > >
> > > > On Thu, 22 Jun 2023 at 18:20, Maxime Roussin-Bélanger
> > > >  wrote:
> > > > >
> > > > > From: Maxime Roussin-Belanger 
> > > > >
> > > > > Without adding debug flag nativesdk dbg package is not packaged since
> > > > > there is no debugging information
> > > > >
> > > > > Signed-off-by: Maxime Roussin-Belanger 
> > > > > 
> > > > > ---
> > > > > v2: New commit
> > > > >
> > > > >  meta/conf/bitbake.conf | 2 +-
> > > > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > > > >
> > > > > diff --git a/meta/conf/bitbake.conf b/meta/conf/bitbake.conf
> > > > > index 9d9b268858..0635affbc1 100644
> > > > > --- a/meta/conf/bitbake.conf
> > > > > +++ b/meta/conf/bitbake.conf
> > > > > @@ -601,7 +601,7 @@ PATCHRESOLVE = "noop"
> > > > >  ##
> > > > >
> > > > >  export BUILD_CPPFLAGS = "-isystem${STAGING_INCDIR_NATIVE}"
> > > > > -BUILDSDK_CPPFLAGS = ""
> > > > > +BUILDSDK_CPPFLAGS ?= "-g"
> > > > >  export CPPFLAGS = "${TARGET_CPPFLAGS}"
> > > > >  TARGET_CPPFLAGS = ""
> > > > >
> > > > > --
> > > > > 2.36.0
> > > > >
> > > > >
> > > > > 
> > > > >
> 





-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#183334): 
https://lists.openembedded.org/g/openembedded-core/message/183334
Mute This Topic: https://lists.openembedded.org/mt/99701056/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] [meta-oe][PATCH] boost-url: remove recipe

[Christian Eggers]

On Friday, 6 January 2023, 11:53:20 CET, Ross Burton wrote:
> This should be posted to openembedded-devel@, not -core@.

I already recognized this myself a few seconds after sending that
email.  Meanwhile this patch has been integrated into the master branch
of meta-openembedded, so it's time to add the other patch to oe-core.

regards
Christian


> 
> Ross
> 
> > On 5 Jan 2023, at 13:08, Christian Eggers via lists.openembedded.org 
> >  wrote:
> > 
> > The boost recipe in oe-core has recently been updated to boost-1.81.0.
> > Starting with this release, boost-url is included in the boost
> > distribution.
> > 
> > As a prerequisite for creating the -url sub package from the boost
> > recipe, the boost-url recipe should be removed in order to avoid
> > conflicts with installed files.
> > 
> > Signed-off-by: Christian Eggers 
> > ---
> > .../boost-url/boost-url_git.bb| 20 ---
> > 1 file changed, 20 deletions(-)
> > delete mode 100644 meta-oe/recipes-devtools/boost-url/boost-url_git.bb
> > 
> > diff --git a/meta-oe/recipes-devtools/boost-url/boost-url_git.bb 
> > b/meta-oe/recipes-devtools/boost-url/boost-url_git.bb
> > deleted file mode 100644
> > index 14da3ca3f1af..
> > --- a/meta-oe/recipes-devtools/boost-url/boost-url_git.bb
> > +++ /dev/null
> > @@ -1,20 +0,0 @@
> > -DESCRIPTION = "Boost.URL is a library for manipulating Uniform Resource 
> > Identifiers (URI) and Locators (URL)"
> > -HOMEPAGE = "https://github.com/CPPAlliance/url;
> > -SECTION = "libs"
> > -LICENSE = "BSL-1.0"
> > -LIC_FILES_CHKSUM = 
> > "file://LICENSE_1_0.txt;md5=e4224ccaecb14d942c71d31bef20d78c"
> > -
> > -SRC_URI = 
> > "git://github.com/CPPAlliance/url.git;branch=develop;protocol=https"
> > -
> > -SRCREV = "d740a92d38e3a8f4d5b2153f53b82f1c98e312ab"
> > -
> > -S = "${WORKDIR}/git"
> > -
> > -DEPENDS = "boost"
> > -
> > -BBCLASSEXTEND = "native nativesdk"
> > -
> > -do_install() {
> > -mkdir -p ${D}/${includedir}
> > -cp -r ${S}/include/** ${D}/${includedir}/
> > -}
> 
> 





-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#175670): 
https://lists.openembedded.org/g/openembedded-core/message/175670
Mute This Topic: https://lists.openembedded.org/mt/96071097/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH] boost: add url lib

[Christian Eggers]

Boost.URL is a library for manipulating Uniform Resource Identifiers
(URIs) and Locators (URLs).

Signed-off-by: Christian Eggers 
---
This patch depends on removal of the boost-url recipe in meta-openembedded
https://lists.openembedded.org/g/openembedded-devel/message/100414

 meta/recipes-support/boost/boost.inc | 1 +
 1 file changed, 1 insertion(+)

diff --git a/meta/recipes-support/boost/boost.inc 
b/meta/recipes-support/boost/boost.inc
index 4263e645de1c..77105aa1fa13 100644
--- a/meta/recipes-support/boost/boost.inc
+++ b/meta/recipes-support/boost/boost.inc
@@ -40,6 +40,7 @@ BOOST_LIBS = "\
thread \
timer \
type_erasure \
+   url \
wave \
"
 
-- 
2.35.3


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#175523): 
https://lists.openembedded.org/g/openembedded-core/message/175523
Mute This Topic: https://lists.openembedded.org/mt/96071174/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [meta-oe][PATCH] boost-url: remove recipe

[Christian Eggers]

The boost recipe in oe-core has recently been updated to boost-1.81.0.
Starting with this release, boost-url is included in the boost
distribution.

As a prerequisite for creating the -url sub package from the boost
recipe, the boost-url recipe should be removed in order to avoid
conflicts with installed files.

Signed-off-by: Christian Eggers 
---
 .../boost-url/boost-url_git.bb| 20 ---
 1 file changed, 20 deletions(-)
 delete mode 100644 meta-oe/recipes-devtools/boost-url/boost-url_git.bb

diff --git a/meta-oe/recipes-devtools/boost-url/boost-url_git.bb 
b/meta-oe/recipes-devtools/boost-url/boost-url_git.bb
deleted file mode 100644
index 14da3ca3f1af..
--- a/meta-oe/recipes-devtools/boost-url/boost-url_git.bb
+++ /dev/null
@@ -1,20 +0,0 @@
-DESCRIPTION = "Boost.URL is a library for manipulating Uniform Resource 
Identifiers (URI) and Locators (URL)"
-HOMEPAGE = "https://github.com/CPPAlliance/url;
-SECTION = "libs"
-LICENSE = "BSL-1.0"
-LIC_FILES_CHKSUM = 
"file://LICENSE_1_0.txt;md5=e4224ccaecb14d942c71d31bef20d78c"
-
-SRC_URI = "git://github.com/CPPAlliance/url.git;branch=develop;protocol=https"
-
-SRCREV = "d740a92d38e3a8f4d5b2153f53b82f1c98e312ab"
-
-S = "${WORKDIR}/git"
-
-DEPENDS = "boost"
-
-BBCLASSEXTEND = "native nativesdk"
-
-do_install() {
-mkdir -p ${D}/${includedir}
-cp -r ${S}/include/** ${D}/${includedir}/
-}
-- 
2.35.3


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#175522): 
https://lists.openembedded.org/g/openembedded-core/message/175522
Mute This Topic: https://lists.openembedded.org/mt/96071097/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] [PATCH] boost: upgrade 1.80.0 -> 1.81.0

[Christian Eggers]

On Wednesday, 21 December 2022, 03:59:07 CET, wan...@fujitsu.com wrote:
> From: Wang Mingyu 
> 
> 0001-The-std-lib-unary-binary_function-base-classes-are-d.patch
> removed since it's included in 1.81.0
> 
> Signed-off-by: Wang Mingyu 
> ---
>  .../{boost-1.80.0.inc => boost-1.81.0.inc}|  2 +-
>  ...y-binary_function-base-classes-are-d.patch | 34 ---
>  .../{boost_1.80.0.bb => boost_1.81.0.bb}  |  1 -
>  3 files changed, 1 insertion(+), 36 deletions(-)
>  rename meta/recipes-support/boost/{boost-1.80.0.inc => boost-1.81.0.inc} 
> (90%)
>  delete mode 100644 
> meta/recipes-support/boost/boost/0001-The-std-lib-unary-binary_function-base-classes-are-d.patch
>  rename meta/recipes-support/boost/{boost_1.80.0.bb => boost_1.81.0.bb} (75%)
> 
> diff --git a/meta/recipes-support/boost/boost-1.80.0.inc 
> b/meta/recipes-support/boost/boost-1.81.0.inc
> similarity index 90%
> rename from meta/recipes-support/boost/boost-1.80.0.inc
> rename to meta/recipes-support/boost/boost-1.81.0.inc
> index 3ee82eb9b2..291796c736 100644
> --- a/meta/recipes-support/boost/boost-1.80.0.inc
> +++ b/meta/recipes-support/boost/boost-1.81.0.inc
> @@ -12,7 +12,7 @@ BOOST_MAJ = "${@"_".join(d.getVar("PV").split(".")[0:2])}"
>  BOOST_P = "boost_${BOOST_VER}"
>  
>  SRC_URI = 
> "https://boostorg.jfrog.io/artifactory/main/release/${PV}/source/${BOOST_P}.tar.bz2;
> -SRC_URI[sha256sum] = 
> "1e19565d82e43bc59209a168f5ac899d3ba471d55c7610c677d4ccf2c9c500c0"
> +SRC_URI[sha256sum] = 
> "71feeed900fbccca04a3b4f2f84a7c217186f28a940ed8b7ed4725986baf99fa"
>  
>  UPSTREAM_CHECK_URI = "http://www.boost.org/users/download/;
>  UPSTREAM_CHECK_REGEX = "release/(?P.*)/source/"
> diff --git 
> a/meta/recipes-support/boost/boost/0001-The-std-lib-unary-binary_function-base-classes-are-d.patch
>  
> b/meta/recipes-support/boost/boost/0001-The-std-lib-unary-binary_function-base-classes-are-d.patch
> deleted file mode 100644
> index 49603348ee..00
> --- 
> a/meta/recipes-support/boost/boost/0001-The-std-lib-unary-binary_function-base-classes-are-d.patch
> +++ /dev/null
> @@ -1,34 +0,0 @@
> -From f9b55f5a1fab85bf73c95e6372779d6f50f75e84 Mon Sep 17 00:00:00 2001
> -From: jzmaddock 
> -Date: Mon, 11 Jul 2022 18:26:07 +0100
> -Subject: [PATCH] The std lib unary/binary_function base classes are
> - deprecated/removed from libcpp15. Fixes
> - https://github.com/boostorg/container_hash/issues/24.
> -
> -Upstream-Status: Backport 
> [https://github.com/boostorg/config/pull/440/commits/f0af4a9184457939b89110795ae2d293582c5f66]
> -Signed-off-by: Khem Raj 
> 
> - boost/config/stdlib/libcpp.hpp | 9 +
> - 1 file changed, 9 insertions(+)
> -
> -diff --git a/boost/config/stdlib/libcpp.hpp b/boost/config/stdlib/libcpp.hpp
> -index bc8536ead..0e9f2445e 100644
>  a/boost/config/stdlib/libcpp.hpp
> -+++ b/boost/config/stdlib/libcpp.hpp
> -@@ -168,4 +168,13 @@
> - #  define BOOST_NO_CXX14_HDR_SHARED_MUTEX
> - #endif
> - 
> -+#if _LIBCPP_VERSION >= 15000
> -+//
> -+// Unary function is now deprecated in C++11 and later:
> -+//
> -+#if __cplusplus >= 201103L
> -+#define BOOST_NO_CXX98_FUNCTION_BASE
> -+#endif
> -+#endif
> -+
> - //  --- end ---
> --- 
> -2.37.2
> -
> diff --git a/meta/recipes-support/boost/boost_1.80.0.bb 
> b/meta/recipes-support/boost/boost_1.81.0.bb
> similarity index 75%
> rename from meta/recipes-support/boost/boost_1.80.0.bb
> rename to meta/recipes-support/boost/boost_1.81.0.bb
> index c34ab7d703..4b580d078b 100644
> --- a/meta/recipes-support/boost/boost_1.80.0.bb
> +++ b/meta/recipes-support/boost/boost_1.81.0.bb
> @@ -4,5 +4,4 @@ require boost.inc
>  SRC_URI += "file://boost-math-disable-pch-for-gcc.patch \
> 
> file://0001-Don-t-set-up-arch-instruction-set-flags-we-do-that-o.patch \
> file://0001-dont-setup-compiler-flags-m32-m64.patch \
> -   
> file://0001-The-std-lib-unary-binary_function-base-classes-are-d.patch \
> "
> 

There is a new library (boost-url) in boost-1.81.0. So "url" should be added to 
BOOST_LIBS.
Unfortunately this causes a conflict (sorry, I don't remember which type) with 
the standalone
boost-url recipe in meta-openembedded (that recipe should be obsolete now).

What is the correct procedure in this case? I guess that it's not possible to 
atomically remove
the boost-url recipe from meta-openembedded and add "url" to BOOST_LIBS.

regards
Christian




-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#175383): 
https://lists.openembedded.org/g/openembedded-core/message/175383
Mute This Topic: https://lists.openembedded.org/mt/9584/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] kernel: export native PKGCONFIG variables

[Christian Eggers]

Hi Bruce,

on 2021-11-23, you added the following patch:

> 9172d61c57e23682c3d2c25701cbd53c84d01a27
>
> kernel: export native PKGCONFIG variables
>
> In a similar manner to cml1.bbclass, we export the pkg-config
> variables to allow a direct call to pkg-config access to the
> native sysroot versus the target sysroot.
>
> The kernel doesn't use pkg-config for target configuration,
I found some calls in the tools/ directory of the kernel.
Isn't this related to target configuration?

> and has many explicit calls to pkg-config, without the
> possibility of easy override to pkg-config-native.
I just noticed that this has recently changed:

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=d5ea4fece4508bf8e72b659cd22fa4840d8d61e5

This change is in Linux 6.0, but not in 5.15. Probably it
could be backported easily.

>
> The calls to pkg-config could be made cross friendly via
> replacement with make variables, but until that effort is
> undertaken upstream, we need a bridge approach.
>
> In particular, this is required for dtschema validation,
> which is a requirement in kernel 5.16+

regards
Christian




_______
Christian
Eggers
Software Engineer
​
ARRI
Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Arriweg 17,
83071
Stephanskirchen
www.arri.com

+49 8036 3009-3118
cegg...@arri.de

​
Get all the latest information from www.arri.com, Facebook, Twitter, Instagram, 
LinkedIn and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Matthias Erb (Chairman); Dr. Michael Neuhäuser; Stephan 
Schenk; Walter Trauninger

​

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#174343): 
https://lists.openembedded.org/g/openembedded-core/message/174343
Mute This Topic: https://lists.openembedded.org/mt/95511690/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH] linux-firmware: split rtl8761 firmware

[Christian Eggers]

Realtek Bluetooth devices require binary firmware files. Package them
separately in order to avoid installing the full linux-firmware package
on embedded devices.

Affected (end user) products (incomplete list):
- TP-Link UB500
- Logilink BT0054

Signed-off-by: Christian Eggers 
---
 .../linux-firmware/linux-firmware_20220913.bb   | 6 ++
 1 file changed, 6 insertions(+)

diff --git a/meta/recipes-kernel/linux-firmware/linux-firmware_20220913.bb 
b/meta/recipes-kernel/linux-firmware/linux-firmware_20220913.bb
index 45c9d0e86123..eea9251a840e 100644
--- a/meta/recipes-kernel/linux-firmware/linux-firmware_20220913.bb
+++ b/meta/recipes-kernel/linux-firmware/linux-firmware_20220913.bb
@@ -235,6 +235,7 @@ PACKAGES =+ "${PN}-ralink-license ${PN}-ralink \
  ${PN}-vt6656-license ${PN}-vt6656 \
  ${PN}-rs9113 ${PN}-rs9116 \
  ${PN}-rtl-license ${PN}-rtl8188 ${PN}-rtl8192cu ${PN}-rtl8192ce 
${PN}-rtl8192su ${PN}-rtl8723 ${PN}-rtl8821 \
+ ${PN}-rtl8761 \
  ${PN}-rtl8168 \
  ${PN}-cypress-license \
  ${PN}-broadcom-license \
@@ -563,6 +564,7 @@ LICENSE:${PN}-rtl8192cu = "Firmware-rtlwifi_firmware"
 LICENSE:${PN}-rtl8192ce = "Firmware-rtlwifi_firmware"
 LICENSE:${PN}-rtl8192su = "Firmware-rtlwifi_firmware"
 LICENSE:${PN}-rtl8723 = "Firmware-rtlwifi_firmware"
+LICENSE:${PN}-rtl8761 = "Firmware-rtlwifi_firmware"
 LICENSE:${PN}-rtl8821 = "Firmware-rtlwifi_firmware"
 LICENSE:${PN}-rtl-license = "Firmware-rtlwifi_firmware"
 LICENSE:${PN}-rtl8168 = "WHENCE"
@@ -588,6 +590,9 @@ FILES:${PN}-rtl8723 = " \
 FILES:${PN}-rtl8821 = " \
   ${nonarch_base_libdir}/firmware/rtlwifi/rtl8821*.bin \
 "
+FILES:${PN}-rtl8761 = " \
+  ${nonarch_base_libdir}/firmware/rtl_bt/rtl8761*.bin \
+"
 FILES:${PN}-rtl8168 = " \
   ${nonarch_base_libdir}/firmware/rtl_nic/rtl8168*.fw \
 "
@@ -598,6 +603,7 @@ RDEPENDS:${PN}-rtl8192cu += "${PN}-rtl-license"
 RDEPENDS:${PN}-rtl8192su = "${PN}-rtl-license"
 RDEPENDS:${PN}-rtl8723 += "${PN}-rtl-license"
 RDEPENDS:${PN}-rtl8821 += "${PN}-rtl-license"
+RDEPENDS:${PN}-rtl8761 += "${PN}-rtl-license"
 RDEPENDS:${PN}-rtl8168 += "${PN}-whence-license"
 
 # For ti-connectivity
-- 
2.35.3


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#171914): 
https://lists.openembedded.org/g/openembedded-core/message/171914
Mute This Topic: https://lists.openembedded.org/mt/94387658/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] [PATCH 0/6] npm.bbclass: work with nodejs 16

[Christian Eggers]

This seems to solve my problem with the quicktype recipe [1]. Thank
you very much for working in this.

[1] 
https://lore.kernel.org/all/CANNYZj-V8O1-USTdtOYKi5hB80QjxQZH2d2=ybj-4y2sm5b...@mail.gmail.com/T/


On Thursday, 19 May 2022, 12:05:50 CEST, Enrico Scholz via 
lists.openembedded.org wrote:
> nodejs 16 changed internal caching significantly which breaks the
> existing npm.bblcass.
> 
> Simulate parts of the npm registry and cache data in the way as
> expected.
> 
> This patchset requires an additional one in the oe-meta layer.
> 
> Enrico Scholz (6):
>   npm: replace 'npm pack' call by 'tar czf'
>   npm: return content of 'package.json' in 'npm_pack'
>   npm: take 'version' directly from 'package.json'
>   npm: disable 'audit' + 'fund'
>   lib:npm_registry: initial checkin
>   npm: use npm_registry to cache package
> 
>  meta/classes/npm.bbclass|  65 +-
>  meta/lib/oe/npm_registry.py | 169 
>  2 files changed, 213 insertions(+), 21 deletions(-)
>  create mode 100644 meta/lib/oe/npm_registry.py
> 
> 





-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#165999): 
https://lists.openembedded.org/g/openembedded-core/message/165999
Mute This Topic: https://lists.openembedded.org/mt/91205437/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH 4/4] librsvg: reenable nativesdk

[Christian Eggers]

Rust is now available for nativesdk. So the nativesdk build for librsvg
can be enabled again.

Signed-off-by: Christian Eggers 
---
 meta/recipes-gnome/librsvg/librsvg_2.52.7.bb | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/recipes-gnome/librsvg/librsvg_2.52.7.bb 
b/meta/recipes-gnome/librsvg/librsvg_2.52.7.bb
index eb420e732cc8..78eb93c6351e 100644
--- a/meta/recipes-gnome/librsvg/librsvg_2.52.7.bb
+++ b/meta/recipes-gnome/librsvg/librsvg_2.52.7.bb
@@ -12,7 +12,7 @@ LIC_FILES_CHKSUM = 
"file://COPYING.LIB;md5=4fbd65380cdd255951079008b364516c \
 
 SECTION = "x11/utils"
 DEPENDS = "cairo gdk-pixbuf glib-2.0 libcroco libxml2 pango"
-BBCLASSEXTEND = "native"
+BBCLASSEXTEND = "native nativesdk"
 
 inherit gnomebase gtk-doc pixbufcache upstream-version-is-even 
gobject-introspection rust vala
 
-- 
2.34.1


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#163631): 
https://lists.openembedded.org/g/openembedded-core/message/163631
Mute This Topic: https://lists.openembedded.org/mt/90017538/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH 3/4] rust-crosssdk: use ${RUST_LIBC} in ${PN}

[Christian Eggers]

TCLIBC is only valid for TARGET, not for HOST or NATIVESDK.
Fixes build of rust-crosssdk if TCLIBC is set to musl.

Signed-off-by: Christian Eggers 
---
 meta/recipes-devtools/rust/rust-crosssdk_1.59.0.bb | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/recipes-devtools/rust/rust-crosssdk_1.59.0.bb 
b/meta/recipes-devtools/rust/rust-crosssdk_1.59.0.bb
index 5929978591fb..6ea8cb09b21b 100644
--- a/meta/recipes-devtools/rust/rust-crosssdk_1.59.0.bb
+++ b/meta/recipes-devtools/rust/rust-crosssdk_1.59.0.bb
@@ -5,4 +5,4 @@ require rust-source.inc
 
 DEPENDS += "virtual/${TARGET_PREFIX}gcc-crosssdk 
virtual/nativesdk-${TARGET_PREFIX}compilerlibs virtual/nativesdk-libc"
 PROVIDES = "virtual/nativesdk-${TARGET_PREFIX}rust"
-PN = "rust-crosssdk-${TUNE_PKGARCH}-${TCLIBC}"
+PN = "rust-crosssdk-${TUNE_PKGARCH}-${RUST_LIBC}"
-- 
2.34.1


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#163630): 
https://lists.openembedded.org/g/openembedded-core/message/163630
Mute This Topic: https://lists.openembedded.org/mt/90017460/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH 2/4] machine-sdk: clear ABIEXTENSION for class-crosssdk

[Christian Eggers]

For nativesdk/crosssdk, we have no ABIEXTENSION. Fixes build of
rust-crosssdk if ABIEXTENSION is set toe something like "eabi".

Signed-off-by: Christian Eggers 
---
 meta/conf/machine-sdk/x86_64.conf | 1 +
 1 file changed, 1 insertion(+)

diff --git a/meta/conf/machine-sdk/x86_64.conf 
b/meta/conf/machine-sdk/x86_64.conf
index 8d494b261e0c..2ab607e65c22 100644
--- a/meta/conf/machine-sdk/x86_64.conf
+++ b/meta/conf/machine-sdk/x86_64.conf
@@ -1,2 +1,3 @@
 SDK_ARCH = "x86_64"
+ABIEXTENSION:class-crosssdk = ""
 ABIEXTENSION:class-nativesdk = ""
-- 
2.34.1


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#163629): 
https://lists.openembedded.org/g/openembedded-core/message/163629
Mute This Topic: https://lists.openembedded.org/mt/90017457/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH 1/4] rust-common: override RUST_LIBC for crosssdk

[Christian Eggers]

For nativesdk/crosssdk, always glibc is used. Fixes build of
rust-crosssdk if TCLIBC is set to musl.

Signed-off-by: Christian Eggers 
---
 meta/classes/rust-common.bbclass | 1 +
 1 file changed, 1 insertion(+)

diff --git a/meta/classes/rust-common.bbclass b/meta/classes/rust-common.bbclass
index 65ad677499cb..02a538258af6 100644
--- a/meta/classes/rust-common.bbclass
+++ b/meta/classes/rust-common.bbclass
@@ -16,6 +16,7 @@ RUST_PANIC_STRATEGY ?= "unwind"
 # Native builds are not effected by TCLIBC. Without this, rust-native
 # thinks it's "target" (i.e. x86_64-linux) is a musl target.
 RUST_LIBC = "${TCLIBC}"
+RUST_LIBC:class-crosssdk = "glibc"
 RUST_LIBC:class-native = "glibc"
 
 def determine_libc(d, thing):
-- 
2.34.1


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#163628): 
https://lists.openembedded.org/g/openembedded-core/message/163628
Mute This Topic: https://lists.openembedded.org/mt/90017456/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH] glib-2.0: upgrade 2.70.4 -> 2.72.0

[Christian Eggers]

- 0001-gio-tests-resources.c-comment-out-a-build-host-only-.patch:
  dropped (integrated upstream)

News


* Bugs fixed:
 - #2620 g_time_zone_new_offset() assertion failure if offset >= 25 hours
 - #2538 Various unit test fixes
 - #2542 fuzzing: Fix test failure with G_DISABLE_ASSERT
 - #2547 gprintf: Fix a memory leak with an invalid format in g_vasprintf()
 - #2548 tests: Various fixes to gdbus-auth, gdbus-non-socket, 
gdbus-connection-flush, spawn-multithreaded tests
 - #2551 tests: More flaky test fixes to converter-stream and test-printf
 - #2552 gtlsconnection: fix typo in docs

* Translation updates:
 - Czech
 - French
 - Friulian
 - Hebrew
 - Hungarian
 - Italian
 - Kazakh
 - Polish
 - Romanian
 - Serbian
 - Swedish

Signed-off-by: Christian Eggers 
---
 ...-correctly-when-building-with-mingw3.patch | 32 +++--
 ...degen.py-bump-timeout-to-100-seconds.patch | 26 --
 ...ble-more-tests-while-cross-compiling.patch | 48 ++-
 ...{glib-2.0_2.70.4.bb => glib-2.0_2.72.0.bb} |  3 +-
 4 files changed, 44 insertions(+), 65 deletions(-)
 delete mode 100644 
meta/recipes-core/glib-2.0/glib-2.0/0001-gio-tests-codegen.py-bump-timeout-to-100-seconds.patch
 rename meta/recipes-core/glib-2.0/{glib-2.0_2.70.4.bb => glib-2.0_2.72.0.bb} 
(93%)

diff --git 
a/meta/recipes-core/glib-2.0/glib-2.0/0001-Set-host_machine-correctly-when-building-with-mingw3.patch
 
b/meta/recipes-core/glib-2.0/glib-2.0/0001-Set-host_machine-correctly-when-building-with-mingw3.patch
index 8a6d46df799b..59de3fa969a4 100644
--- 
a/meta/recipes-core/glib-2.0/glib-2.0/0001-Set-host_machine-correctly-when-building-with-mingw3.patch
+++ 
b/meta/recipes-core/glib-2.0/glib-2.0/0001-Set-host_machine-correctly-when-building-with-mingw3.patch
@@ -1,11 +1,10 @@
-From 333809ded70ad4e3470b7134e3fac1a42ff48e61 Mon Sep 17 00:00:00 2001
+From 4f327be49fd6ac5a77da6e48459b37f37a601977 Mon Sep 17 00:00:00 2001
 From: Alexander Kanavin 
 Date: Wed, 13 Feb 2019 15:32:05 +0100
 Subject: [PATCH] Set host_machine correctly when building with mingw32
 
 Upstream-Status: Inappropriate [oe-core specific]
 Signed-off-by: Alexander Kanavin 
-
 ---
  gio/tests/meson.build  | 8 
  glib/tests/meson.build | 2 +-
@@ -14,10 +13,10 @@ Signed-off-by: Alexander Kanavin 
  4 files changed, 9 insertions(+), 6 deletions(-)
 
 diff --git a/gio/tests/meson.build b/gio/tests/meson.build
-index 3a19c82..b762835 100644
+index abe676767c60..34b347815308 100644
 --- a/gio/tests/meson.build
 +++ b/gio/tests/meson.build
-@@ -27,7 +27,7 @@ test_c_args = [
+@@ -27,7 +27,7 @@ if build_machine.system() == 'linux'
endif # libutil.length() > 0
  endif # build_machine.system() == 'linux'
  
@@ -26,7 +25,7 @@ index 3a19c82..b762835 100644
common_gio_tests_deps += [iphlpapi_dep, winsock2, cc.find_library 
('secur32')]
  endif
  
-@@ -148,7 +148,7 @@ else
+@@ -176,7 +176,7 @@ else
  endif
  
  #  Test programs buildable on UNIX only
@@ -35,7 +34,7 @@ index 3a19c82..b762835 100644
gio_tests += {
  'file' : {},
  'gdbus-peer' : {
-@@ -402,7 +402,7 @@ if host_machine.system() != 'windows'
+@@ -434,7 +434,7 @@ if host_machine.system() != 'windows'
  endif # unix
  
  #  Test programs buildable on Windows only
@@ -44,7 +43,7 @@ index 3a19c82..b762835 100644
gio_tests += {'win32-streams' : {}}
  endif
  
-@@ -472,7 +472,7 @@ if cc.get_id() != 'msvc' and cc.get_id() != 'clang-cl'
+@@ -504,7 +504,7 @@ if cc.get_id() != 'msvc' and cc.get_id() != 'clang-cl'
}
  endif
  
@@ -54,10 +53,10 @@ index 3a19c82..b762835 100644
  'gdbus-example-unix-fd-client' : {
'install' : false,
 diff --git a/glib/tests/meson.build b/glib/tests/meson.build
-index 6eb23e8..36eb919 100644
+index a0c64afe6ae9..48407f99569c 100644
 --- a/glib/tests/meson.build
 +++ b/glib/tests/meson.build
-@@ -142,7 +142,7 @@ if glib_conf.has('HAVE_EVENTFD')
+@@ -151,7 +151,7 @@ if glib_conf.has('HAVE_EVENTFD')
}
  endif
  
@@ -67,10 +66,10 @@ index 6eb23e8..36eb919 100644
  glib_tests += {
'gpoll' : {
 diff --git a/meson.build b/meson.build
-index 47f3a5c..7ea7ad1 100644
+index e0b14319fb58..f6756a3c7dfb 100644
 --- a/meson.build
 +++ b/meson.build
-@@ -32,6 +32,9 @@ else
+@@ -54,6 +54,9 @@ else
  endif
  
  host_system = host_machine.system()
@@ -81,15 +80,18 @@ index 47f3a5c..7ea7ad1 100644
  if host_system == 'darwin'
ios_test_code = '''#include 
 diff --git a/tests/meson.build b/tests/meson.build
-index 6741f8f..12fdc90 100644
+index 5ff99a4a8258..8a4dad0ccae7 100644
 --- a/tests/meson.build
 +++ b/tests/meson.build
-@@ -73,7 +73,7 @@ test_extra_programs = {
-   'unicode-collate' : {},
+@@ -53,7 +53,7 @@ test_extra_programs = {
+   'assert-msg-test' : {},
  }
  
 -if host_machine.system() != 'windows'
 +if host_system != 'windows'
tests += {
  'timeloop' : {},
- 'iochannel-test' : {},
+   }
+-- 
+2.34.1
+
diff --git 
a/meta/recipes-core/glib-2.0/glib-2.0/0001-gio-tests-codegen.py-bump-timeout-to-100-seconds.patch
 
b/meta/

Re: [OE-core] [PATCH] glib-2.0: upgrade 2.70.4 -> 2.71.3

[Christian Eggers]

On Wednesday, 9 March 2022, 17:52:49 CET, Alexander Kanavin wrote:
> 2.71.x is a development release, and should not be used. glib follows
> odd-even version scheme.
I already supposed something like this, but I didn't find any
documentation about GLib's versioning scheme... Would be nice if
there were some docs about that (particularly as the GNOME project
moved away from odd/even release ([1], search for "even/odd")).

regards,
Christian
> 
> Alex
> 

[1] https://discourse.gnome.org/t/new-gnome-versioning-scheme/4235



-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#162986): 
https://lists.openembedded.org/g/openembedded-core/message/162986
Mute This Topic: https://lists.openembedded.org/mt/89665588/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH] glib-2.0: upgrade 2.70.4 -> 2.71.3

[Christian Eggers]

- 0001-gio-tests-resources.c-comment-out-a-build-host-only-.patch:
  dropped (integrated upstream)

Signed-off-by: Christian Eggers 
---
 ...-correctly-when-building-with-mingw3.patch | 32 +++--
 ...degen.py-bump-timeout-to-100-seconds.patch | 26 --
 ...ble-more-tests-while-cross-compiling.patch | 48 ++-
 ...{glib-2.0_2.70.4.bb => glib-2.0_2.71.3.bb} |  3 +-
 4 files changed, 44 insertions(+), 65 deletions(-)
 delete mode 100644 
meta/recipes-core/glib-2.0/glib-2.0/0001-gio-tests-codegen.py-bump-timeout-to-100-seconds.patch
 rename meta/recipes-core/glib-2.0/{glib-2.0_2.70.4.bb => glib-2.0_2.71.3.bb} 
(93%)

diff --git 
a/meta/recipes-core/glib-2.0/glib-2.0/0001-Set-host_machine-correctly-when-building-with-mingw3.patch
 
b/meta/recipes-core/glib-2.0/glib-2.0/0001-Set-host_machine-correctly-when-building-with-mingw3.patch
index 8a6d46df799b..59de3fa969a4 100644
--- 
a/meta/recipes-core/glib-2.0/glib-2.0/0001-Set-host_machine-correctly-when-building-with-mingw3.patch
+++ 
b/meta/recipes-core/glib-2.0/glib-2.0/0001-Set-host_machine-correctly-when-building-with-mingw3.patch
@@ -1,11 +1,10 @@
-From 333809ded70ad4e3470b7134e3fac1a42ff48e61 Mon Sep 17 00:00:00 2001
+From 4f327be49fd6ac5a77da6e48459b37f37a601977 Mon Sep 17 00:00:00 2001
 From: Alexander Kanavin 
 Date: Wed, 13 Feb 2019 15:32:05 +0100
 Subject: [PATCH] Set host_machine correctly when building with mingw32
 
 Upstream-Status: Inappropriate [oe-core specific]
 Signed-off-by: Alexander Kanavin 
-
 ---
  gio/tests/meson.build  | 8 
  glib/tests/meson.build | 2 +-
@@ -14,10 +13,10 @@ Signed-off-by: Alexander Kanavin 
  4 files changed, 9 insertions(+), 6 deletions(-)
 
 diff --git a/gio/tests/meson.build b/gio/tests/meson.build
-index 3a19c82..b762835 100644
+index abe676767c60..34b347815308 100644
 --- a/gio/tests/meson.build
 +++ b/gio/tests/meson.build
-@@ -27,7 +27,7 @@ test_c_args = [
+@@ -27,7 +27,7 @@ if build_machine.system() == 'linux'
endif # libutil.length() > 0
  endif # build_machine.system() == 'linux'
  
@@ -26,7 +25,7 @@ index 3a19c82..b762835 100644
common_gio_tests_deps += [iphlpapi_dep, winsock2, cc.find_library 
('secur32')]
  endif
  
-@@ -148,7 +148,7 @@ else
+@@ -176,7 +176,7 @@ else
  endif
  
  #  Test programs buildable on UNIX only
@@ -35,7 +34,7 @@ index 3a19c82..b762835 100644
gio_tests += {
  'file' : {},
  'gdbus-peer' : {
-@@ -402,7 +402,7 @@ if host_machine.system() != 'windows'
+@@ -434,7 +434,7 @@ if host_machine.system() != 'windows'
  endif # unix
  
  #  Test programs buildable on Windows only
@@ -44,7 +43,7 @@ index 3a19c82..b762835 100644
gio_tests += {'win32-streams' : {}}
  endif
  
-@@ -472,7 +472,7 @@ if cc.get_id() != 'msvc' and cc.get_id() != 'clang-cl'
+@@ -504,7 +504,7 @@ if cc.get_id() != 'msvc' and cc.get_id() != 'clang-cl'
}
  endif
  
@@ -54,10 +53,10 @@ index 3a19c82..b762835 100644
  'gdbus-example-unix-fd-client' : {
'install' : false,
 diff --git a/glib/tests/meson.build b/glib/tests/meson.build
-index 6eb23e8..36eb919 100644
+index a0c64afe6ae9..48407f99569c 100644
 --- a/glib/tests/meson.build
 +++ b/glib/tests/meson.build
-@@ -142,7 +142,7 @@ if glib_conf.has('HAVE_EVENTFD')
+@@ -151,7 +151,7 @@ if glib_conf.has('HAVE_EVENTFD')
}
  endif
  
@@ -67,10 +66,10 @@ index 6eb23e8..36eb919 100644
  glib_tests += {
'gpoll' : {
 diff --git a/meson.build b/meson.build
-index 47f3a5c..7ea7ad1 100644
+index e0b14319fb58..f6756a3c7dfb 100644
 --- a/meson.build
 +++ b/meson.build
-@@ -32,6 +32,9 @@ else
+@@ -54,6 +54,9 @@ else
  endif
  
  host_system = host_machine.system()
@@ -81,15 +80,18 @@ index 47f3a5c..7ea7ad1 100644
  if host_system == 'darwin'
ios_test_code = '''#include 
 diff --git a/tests/meson.build b/tests/meson.build
-index 6741f8f..12fdc90 100644
+index 5ff99a4a8258..8a4dad0ccae7 100644
 --- a/tests/meson.build
 +++ b/tests/meson.build
-@@ -73,7 +73,7 @@ test_extra_programs = {
-   'unicode-collate' : {},
+@@ -53,7 +53,7 @@ test_extra_programs = {
+   'assert-msg-test' : {},
  }
  
 -if host_machine.system() != 'windows'
 +if host_system != 'windows'
tests += {
  'timeloop' : {},
- 'iochannel-test' : {},
+   }
+-- 
+2.34.1
+
diff --git 
a/meta/recipes-core/glib-2.0/glib-2.0/0001-gio-tests-codegen.py-bump-timeout-to-100-seconds.patch
 
b/meta/recipes-core/glib-2.0/glib-2.0/0001-gio-tests-codegen.py-bump-timeout-to-100-seconds.patch
deleted file mode 100644
index 7690bb37a6be..
--- 
a/meta/recipes-core/glib-2.0/glib-2.0/0001-gio-tests-codegen.py-bump-timeout-to-100-seconds.patch
+++ /dev/null
@@ -1,26 +0,0 @@
-From b833254bcc9fcf4cdc2572027b1154d799535ca4 Mon Sep 17 00:00:00 2001
-From: Alexander Kanavin 
-Date: Sun, 20 Dec 2020 22:01:43 +0100
-Subject: [PATCH] gio/tests/codegen.py: bump timeout to 100 seconds
-
-This may be necessary on overloaded CI systems.
-
-Upstream-Status: Submitted 
[https://gitlab.gnome.org/GNOME/glib/-/merge_requests/2348]
-S

Re: [OE-core] [PATCH] binutils-cross-canadian: enable gold for mingw

[Christian Eggers]

On Wednesday, 9 March 2022, 10:13:23 CET, Samuli Piippo wrote:
> This reverts commit b5a595a4be09756b88e91f3353e3b221b165ab44,
> gold linker can now be built correctly for mingw SDKs.

I am interested how to use the gold linker from the SDK (on Linux).
I already tried this in the past [1], using -fuse-ld=gold, but without
much success.

[1] 
https://www.mail-archive.com/openembedded-core@lists.openembedded.org/msg142777.html
> 
> Signed-off-by: Samuli Piippo 
> ---
>  meta/recipes-devtools/binutils/binutils-cross-canadian.inc | 2 --
>  1 file changed, 2 deletions(-)
> 
> diff --git a/meta/recipes-devtools/binutils/binutils-cross-canadian.inc 
> b/meta/recipes-devtools/binutils/binutils-cross-canadian.inc
> index f44f46602d..b3d591e658 100644
> --- a/meta/recipes-devtools/binutils/binutils-cross-canadian.inc
> +++ b/meta/recipes-devtools/binutils/binutils-cross-canadian.inc
> @@ -13,8 +13,6 @@ EXTRA_OECONF += 
> "--with-sysroot=${SDKPATH}/sysroots/${TUNE_PKGARCH}${TARGET_VEND
>  # e.g. we switch between different machines with different tunes.
>  EXTRA_OECONF[vardepsexclude] = "TUNE_PKGARCH"
>  
> -LDGOLD:sdkmingw32 = ""
> -
>  do_install () {
>   autotools_do_install
>  
> 





-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#162970): 
https://lists.openembedded.org/g/openembedded-core/message/162970
Mute This Topic: https://lists.openembedded.org/mt/89658339/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [hardknott][PATCH] patch.py: Prevent git repo reinitialization

[Christian Eggers]

From: Pavel Zhukov 

There were few bugs in the _isInitialized() function which might trigger
git repo to be reinitialized and patches failing to apply.

Signed-off-by: Pavel Zhukov 
Signed-off-by: Richard Purdie 
---
 .../recipes-test/gitrepotest/gitrepotest.bb| 16 
 .../gitrepotest/0001-testpatch.patch   |  9 +
 meta/lib/oe/patch.py   | 11 ---
 meta/lib/oeqa/selftest/cases/bbtests.py| 18 --
 4 files changed, 49 insertions(+), 5 deletions(-)
 create mode 100644 meta-selftest/recipes-test/gitrepotest/gitrepotest.bb
 create mode 100644 
meta-selftest/recipes-test/gitrepotest/gitrepotest/0001-testpatch.patch

diff --git a/meta-selftest/recipes-test/gitrepotest/gitrepotest.bb 
b/meta-selftest/recipes-test/gitrepotest/gitrepotest.bb
new file mode 100644
index ..f1b6c55833bf
--- /dev/null
+++ b/meta-selftest/recipes-test/gitrepotest/gitrepotest.bb
@@ -0,0 +1,16 @@
+SUMMARY = "Test recipe for git repo initialization"
+HOMEPAGE = "https://git.yoctoproject.org/git/matchbox-panel-2;
+LICENSE = "GPL-2.0-or-later"
+LIC_FILES_CHKSUM = "file://COPYING;md5=94d55d512a9ba36caa9b7df079bae19f"
+
+INHIBIT_DEFAULT_DEPS = "1"
+
+PATCHTOOL="git"
+
+SRC_URI = 
"git://git.yoctoproject.org/git/matchbox-panel-2;branch=master;protocol=https \
+   file://0001-testpatch.patch \
+  "
+
+SRCREV = "f82ca3f42510fb3ef10f598b393eb373a2c34ca7"
+
+S = "${WORKDIR}/git"
diff --git 
a/meta-selftest/recipes-test/gitrepotest/gitrepotest/0001-testpatch.patch 
b/meta-selftest/recipes-test/gitrepotest/gitrepotest/0001-testpatch.patch
new file mode 100644
index ..bccda17ee9c6
--- /dev/null
+++ b/meta-selftest/recipes-test/gitrepotest/gitrepotest/0001-testpatch.patch
@@ -0,0 +1,9 @@
+diff --git a/Makefile.am b/Makefile.am
+index 432a9b4..bbf7c74 100644
+--- a/Makefile.am
 b/Makefile.am
+@@ -1,3 +1,4 @@
++## This is useless comment to test if patch works
+ ACLOCAL_AMFLAGS = -I m4
+ 
+ SUBDIRS = matchbox-panel applets data po
diff --git a/meta/lib/oe/patch.py b/meta/lib/oe/patch.py
index 950fe723dcfd..9034fcae0348 100644
--- a/meta/lib/oe/patch.py
+++ b/meta/lib/oe/patch.py
@@ -304,14 +304,19 @@ class GitApplyTree(PatchTree):
 
 def _isInitialized(self):
 cmd = "git rev-parse --show-toplevel"
-(status, output) = subprocess.getstatusoutput(cmd.split())
+try:
+output = runcmd(cmd.split(), self.dir).strip()
+except CmdError as err:
+## runcmd returned non-zero which most likely means 128
+## Not a git directory
+return False
 ## Make sure repo is in builddir to not break top-level git repos
-return status == 0 and os.path.samedir(output, self.dir)
+return os.path.samefile(output, self.dir)
 
 def _initRepo(self):
 runcmd("git init".split(), self.dir)
 runcmd("git add .".split(), self.dir)
-runcmd("git commit -a --allow-empty -m Patching_started".split(), 
self.dir)
+runcmd("git commit -a --allow-empty -m 
bitbake_patching_started".split(), self.dir)
 
 @staticmethod
 def extractPatchHeader(patchfile):
diff --git a/meta/lib/oeqa/selftest/cases/bbtests.py 
b/meta/lib/oeqa/selftest/cases/bbtests.py
index 0a618bb9a6c9..4187cb840a38 100644
--- a/meta/lib/oeqa/selftest/cases/bbtests.py
+++ b/meta/lib/oeqa/selftest/cases/bbtests.py
@@ -310,8 +310,22 @@ INHERIT_remove = \"report-error\"
 src = get_bb_var("SRC_URI",test_recipe)
 gitscm = re.search("git://", src)
 self.assertFalse(gitscm, "test_git_patchtool pre-condition failed: {} 
test recipe contains git repo!".format(test_recipe))
-result = bitbake('man-db -c patch', ignore_status=False)
+result = bitbake('{} -c patch'.format(test_recipe), 
ignore_status=False)
 fatal = re.search("fatal: not a git repository (or any of the parent 
directories)", result.output)
 self.assertFalse(fatal, "Failed to patch using PATCHTOOL=\"git\"")
 self.delete_recipeinc(test_recipe)
-bitbake('-cclean man-db')
+bitbake('-cclean {}'.format(test_recipe))
+
+def test_git_patchtool2(self):
+""" Test if PATCHTOOL=git works with git repo and doesn't reinitialize 
it
+"""
+test_recipe = "gitrepotest"
+src = get_bb_var("SRC_URI",test_recipe)
+gitscm = re.search("git://", src)
+self.assertTrue(gitscm, "test_git_patchtool pre-condition failed: {} 
test recipe doesn't contains git repo!".format(test_recipe))
+result = bitbake('{} -c patch'.format(test_recipe), 
ignore_status=False)
+srcdir = get_bb_var('S', test_recipe)
+result = runCmd("git log", cwd = srcdir)
+self.assertFalse("bitbake_patching_started" in result.output, msg = 
"Repository has been reinitialized. {}".format(srcdir))
+self.delete_recipeinc(test_recipe)
+bitbake('-cclean {}'.format(test_recipe))
-- 
2.34.1

[OE-core] [honister][PATCH] patch.py: Prevent git repo reinitialization

[Christian Eggers]

From: Pavel Zhukov 

There were few bugs in the _isInitialized() function which might trigger
git repo to be reinitialized and patches failing to apply.

Signed-off-by: Pavel Zhukov 
Signed-off-by: Richard Purdie 
---
 .../recipes-test/gitrepotest/gitrepotest.bb| 16 
 .../gitrepotest/0001-testpatch.patch   |  9 +
 meta/lib/oe/patch.py   | 11 ---
 meta/lib/oeqa/selftest/cases/bbtests.py| 18 --
 4 files changed, 49 insertions(+), 5 deletions(-)
 create mode 100644 meta-selftest/recipes-test/gitrepotest/gitrepotest.bb
 create mode 100644 
meta-selftest/recipes-test/gitrepotest/gitrepotest/0001-testpatch.patch

diff --git a/meta-selftest/recipes-test/gitrepotest/gitrepotest.bb 
b/meta-selftest/recipes-test/gitrepotest/gitrepotest.bb
new file mode 100644
index ..f1b6c55833bf
--- /dev/null
+++ b/meta-selftest/recipes-test/gitrepotest/gitrepotest.bb
@@ -0,0 +1,16 @@
+SUMMARY = "Test recipe for git repo initialization"
+HOMEPAGE = "https://git.yoctoproject.org/git/matchbox-panel-2;
+LICENSE = "GPL-2.0-or-later"
+LIC_FILES_CHKSUM = "file://COPYING;md5=94d55d512a9ba36caa9b7df079bae19f"
+
+INHIBIT_DEFAULT_DEPS = "1"
+
+PATCHTOOL="git"
+
+SRC_URI = 
"git://git.yoctoproject.org/git/matchbox-panel-2;branch=master;protocol=https \
+   file://0001-testpatch.patch \
+  "
+
+SRCREV = "f82ca3f42510fb3ef10f598b393eb373a2c34ca7"
+
+S = "${WORKDIR}/git"
diff --git 
a/meta-selftest/recipes-test/gitrepotest/gitrepotest/0001-testpatch.patch 
b/meta-selftest/recipes-test/gitrepotest/gitrepotest/0001-testpatch.patch
new file mode 100644
index ..bccda17ee9c6
--- /dev/null
+++ b/meta-selftest/recipes-test/gitrepotest/gitrepotest/0001-testpatch.patch
@@ -0,0 +1,9 @@
+diff --git a/Makefile.am b/Makefile.am
+index 432a9b4..bbf7c74 100644
+--- a/Makefile.am
 b/Makefile.am
+@@ -1,3 +1,4 @@
++## This is useless comment to test if patch works
+ ACLOCAL_AMFLAGS = -I m4
+ 
+ SUBDIRS = matchbox-panel applets data po
diff --git a/meta/lib/oe/patch.py b/meta/lib/oe/patch.py
index 950fe723dcfd..9034fcae0348 100644
--- a/meta/lib/oe/patch.py
+++ b/meta/lib/oe/patch.py
@@ -304,14 +304,19 @@ class GitApplyTree(PatchTree):
 
 def _isInitialized(self):
 cmd = "git rev-parse --show-toplevel"
-(status, output) = subprocess.getstatusoutput(cmd.split())
+try:
+output = runcmd(cmd.split(), self.dir).strip()
+except CmdError as err:
+## runcmd returned non-zero which most likely means 128
+## Not a git directory
+return False
 ## Make sure repo is in builddir to not break top-level git repos
-return status == 0 and os.path.samedir(output, self.dir)
+return os.path.samefile(output, self.dir)
 
 def _initRepo(self):
 runcmd("git init".split(), self.dir)
 runcmd("git add .".split(), self.dir)
-runcmd("git commit -a --allow-empty -m Patching_started".split(), 
self.dir)
+runcmd("git commit -a --allow-empty -m 
bitbake_patching_started".split(), self.dir)
 
 @staticmethod
 def extractPatchHeader(patchfile):
diff --git a/meta/lib/oeqa/selftest/cases/bbtests.py 
b/meta/lib/oeqa/selftest/cases/bbtests.py
index 31962b92d79e..a08001a173dc 100644
--- a/meta/lib/oeqa/selftest/cases/bbtests.py
+++ b/meta/lib/oeqa/selftest/cases/bbtests.py
@@ -310,8 +310,22 @@ INHERIT:remove = \"report-error\"
 src = get_bb_var("SRC_URI",test_recipe)
 gitscm = re.search("git://", src)
 self.assertFalse(gitscm, "test_git_patchtool pre-condition failed: {} 
test recipe contains git repo!".format(test_recipe))
-result = bitbake('man-db -c patch', ignore_status=False)
+result = bitbake('{} -c patch'.format(test_recipe), 
ignore_status=False)
 fatal = re.search("fatal: not a git repository (or any of the parent 
directories)", result.output)
 self.assertFalse(fatal, "Failed to patch using PATCHTOOL=\"git\"")
 self.delete_recipeinc(test_recipe)
-bitbake('-cclean man-db')
+bitbake('-cclean {}'.format(test_recipe))
+
+def test_git_patchtool2(self):
+""" Test if PATCHTOOL=git works with git repo and doesn't reinitialize 
it
+"""
+test_recipe = "gitrepotest"
+src = get_bb_var("SRC_URI",test_recipe)
+gitscm = re.search("git://", src)
+self.assertTrue(gitscm, "test_git_patchtool pre-condition failed: {} 
test recipe doesn't contains git repo!".format(test_recipe))
+result = bitbake('{} -c patch'.format(test_recipe), 
ignore_status=False)
+srcdir = get_bb_var('S', test_recipe)
+result = runCmd("git log", cwd = srcdir)
+self.assertFalse("bitbake_patching_started" in result.output, msg = 
"Repository has been reinitialized. {}".format(srcdir))
+self.delete_recipeinc(test_recipe)
+bitbake('-cclean {}'.format(test_recipe))
-- 
2.34.1

[OE-core] [PATCH] license: expand_wildcard_licenses: add AGPL-3.0* wildcard

[Christian Eggers]

The Yocto reference manual suggests also settings AGPL-3.0 in
INCOMPATIBLE_LICENSE.

https://www.yoctoproject.org/docs/latest/ref-manual/ref-manual.html#var-INCOMPATIBLE_LICENSE

Fixes: 724fc8047cae ("license: Rework INCOMPATIBLE_LICENSE wildcard handling")
Signed-off-by: Christian Eggers 
---
 meta/classes/license.bbclass | 1 +
 1 file changed, 1 insertion(+)

diff --git a/meta/classes/license.bbclass b/meta/classes/license.bbclass
index cb1f46983aca..94338be90a4f 100644
--- a/meta/classes/license.bbclass
+++ b/meta/classes/license.bbclass
@@ -282,6 +282,7 @@ def expand_wildcard_licenses(d, wildcard_licenses):
 """
 licenses = set(wildcard_licenses)
 mapping = {
+"AGPL-3.0*" : ["AGPL-3.0-only", "AGPL-3.0-or-later"],
 "GPL-3.0*" : ["GPL-3.0-only", "GPL-3.0-or-later"],
 "LGPL-3.0*" : ["LGPL-3.0-only", "LGPL-3.0-or-later"],
 }
-- 
2.34.1


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#162804): 
https://lists.openembedded.org/g/openembedded-core/message/162804
Mute This Topic: https://lists.openembedded.org/mt/89608892/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] [PATCH v3] patch.py: Prevent git repo reinitialization

[Christian Eggers]

esult.output)
>  self.assertFalse(fatal, "Failed to patch using PATCHTOOL=\"git\"")
>  self.delete_recipeinc(test_recipe)
> -bitbake('-cclean man-db')
> +bitbake('-cclean {}'.format(test_recipe))
> +
> +def test_git_patchtool2(self):
> +""" Test if PATCHTOOL=git works with git repo and doesn't 
> reinitialize it
> +"""
> +test_recipe = "gitrepotest"
> +src = get_bb_var("SRC_URI",test_recipe)
> +gitscm = re.search("git://", src)
> +self.assertTrue(gitscm, "test_git_patchtool pre-condition failed: {} 
> test recipe doesn't contains git repo!".format(test_recipe))
> +result = bitbake('{} -c patch'.format(test_recipe), 
> ignore_status=False)
> +srcdir = get_bb_var('S', test_recipe)
> +result = runCmd("git log", cwd = srcdir)
> +self.assertFalse("bitbake_patching_started" in result.output, msg = 
> "Repository has been reinitialized. {}".format(srcdir))
> +self.delete_recipeinc(test_recipe)
> +bitbake('-cclean {}'.format(test_recipe))
> +
>  
>  def test_git_unpack_nonetwork(self):
>  """
> 
Tested-by: Christian Eggers 







-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#162184): 
https://lists.openembedded.org/g/openembedded-core/message/162184
Mute This Topic: https://lists.openembedded.org/mt/89301375/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] [PATCH v2] patch.py: Prevent git repo reinitialization

[Christian Eggers]

On Monday, 21 February 2022, 15:20:07 CET, Pavel Zhukov wrote:
> There were few bugs in the _isInitialized() function which might trigger
> git repo to be reinitialized and patches failing to apply.
> 
> Signed-off-by: Pavel Zhukov 
> ---
>  .../net-tools/Add_missing_headers.patch   |  15 +++
>  .../net-tools/net-tools/net-tools-config.h|  75 +++
>  .../net-tools/net-tools/net-tools-config.make |  36 ++
>  .../recipes-test/net-tools/net-tools_2.10.bb  | 121 ++
>  meta/lib/oe/patch.py  |   6 +-
>  meta/lib/oeqa/selftest/cases/bbtests.py   |  20 ++-
>  6 files changed, 268 insertions(+), 5 deletions(-)
>  create mode 100644 
> meta-selftest/recipes-test/net-tools/net-tools/Add_missing_headers.patch
>  create mode 100644 
> meta-selftest/recipes-test/net-tools/net-tools/net-tools-config.h
>  create mode 100644 
> meta-selftest/recipes-test/net-tools/net-tools/net-tools-config.make
>  create mode 100644 meta-selftest/recipes-test/net-tools/net-tools_2.10.bb
> 
> [...]
>
> diff --git a/meta/lib/oe/patch.py b/meta/lib/oe/patch.py
> index 950fe723dc..2163e336c2 100644
> --- a/meta/lib/oe/patch.py
> +++ b/meta/lib/oe/patch.py
> @@ -304,14 +304,14 @@ class GitApplyTree(PatchTree):
>  
>  def _isInitialized(self):
>  cmd = "git rev-parse --show-toplevel"
> -(status, output) = subprocess.getstatusoutput(cmd.split())
> +(status, output) = subprocess.getstatusoutput(cmd)
>  ## Make sure repo is in builddir to not break top-level git repos
> -return status == 0 and os.path.samedir(output, self.dir)
> +return status == 0 and os.path.samefile(output, self.dir)

I have the following variable values here:
- status: 0
- output: /home/eggers/<...>/yocto
- self.dir: 
/home/eggers/<...>/build/devel-kirkstone/tmp/work/orbiter-poky-linux-musleabi/linux-stable-rt/5.10.100-rt62-orbiter+gitAUTOINC+ca44e6dff2-r0/git

So os.path.samefile() returns false and the existing Git repository
is still reinitialized.

regards
Christian




-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#162061): 
https://lists.openembedded.org/g/openembedded-core/message/162061
Mute This Topic: https://lists.openembedded.org/mt/89294279/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] [PATCH] patch.py: Prevent git repo reinitialization

[Christian Eggers]

On Monday, 21 February 2022, 14:57:46 CET, Pavel Zhukov wrote:
> There were few bugs in the _isInitialized() function which might trigger
> git repo to be reinitialized and patches failing to apply.
> 
> Signed-off-by: Pavel Zhukov 
> ---
>  .../net-tools/Add_missing_headers.patch   |  15 +++
>  .../net-tools/net-tools/net-tools-config.h|  75 +++
>  .../net-tools/net-tools/net-tools-config.make |  36 ++
>  .../recipes-test/net-tools/net-tools_2.10.bb  | 121 ++
I guess that these files have been added by accident.

>  meta/lib/oe/patch.py  |   6 +-
>  meta/lib/oeqa/selftest/cases/bbtests.py   |  20 ++-
>  6 files changed, 268 insertions(+), 5 deletions(-)
>  create mode 100644 
> meta-selftest/recipes-test/net-tools/net-tools/Add_missing_headers.patch
>  create mode 100644 
> meta-selftest/recipes-test/net-tools/net-tools/net-tools-config.h
>  create mode 100644 
> meta-selftest/recipes-test/net-tools/net-tools/net-tools-config.make
>  create mode 100644 meta-selftest/recipes-test/net-tools/net-tools_2.10.bb
> 
> ...
> 
> diff --git a/meta/lib/oe/patch.py b/meta/lib/oe/patch.py
> index 950fe723dc..b6fe9b2bdd 100644
> --- a/meta/lib/oe/patch.py
> +++ b/meta/lib/oe/patch.py
> @@ -304,14 +304,14 @@ class GitApplyTree(PatchTree):
>  
>  def _isInitialized(self):
>  cmd = "git rev-parse --show-toplevel"
> -(status, output) = subprocess.getstatusoutput(cmd.split())
> +(status, output) = subprocess.getstatusoutput(cmd)
>  ## Make sure repo is in builddir to not break top-level git repos
> -return status == 0 and os.path.samedir(output, self.dir)
> +return status == 0 and os.path.samefile(oиutput, self.dir)
extra character ('и')

regards
Christian

>  
>  def _initRepo(self):
>  runcmd("git init".split(), self.dir)
>  runcmd("git add .".split(), self.dir)
> -runcmd("git commit -a --allow-empty -m Patching_started".split(), 
> self.dir)
> +runcmd("git commit -a --allow-empty -m 
> bitbake_patching_started".split(), self.dir)
>  
>  @staticmethod
>  def extractPatchHeader(patchfile):
> diff --git a/meta/lib/oeqa/selftest/cases/bbtests.py 
> b/meta/lib/oeqa/selftest/cases/bbtests.py
> index ce72c4bcc6..ae3f230038 100644
> --- a/meta/lib/oeqa/selftest/cases/bbtests.py
> +++ b/meta/lib/oeqa/selftest/cases/bbtests.py
> @@ -307,11 +307,27 @@ INHERIT:remove = \"report-error\"
>  src = get_bb_var("SRC_URI",test_recipe)
>  gitscm = re.search("git://", src)
>  self.assertFalse(gitscm, "test_git_patchtool pre-condition failed: 
> {} test recipe contains git repo!".format(test_recipe))
> -result = bitbake('man-db -c patch', ignore_status=False)
> +result = bitbake('{} -c patch'.format(test_recipe), 
> ignore_status=False)
>  fatal = re.search("fatal: not a git repository (or any of the parent 
> directories)", result.output)
>  self.assertFalse(fatal, "Failed to patch using PATCHTOOL=\"git\"")
>  self.delete_recipeinc(test_recipe)
> -bitbake('-cclean man-db')
> +bitbake('-cclean {}'.format(test_recipe))
> +
> +def test_git_patchtool2(self):
> +""" Test if PATCHTOOL=git works with git repo and doesn't 
> reinitialize it
> +"""
> +test_recipe = "net-tools"
> +self.write_recipeinc(test_recipe, 'PATCHTOOL=\"git\"')
> +src = get_bb_var("SRC_URI",test_recipe)
> +gitscm = re.search("git://", src)
> +self.assertTrue(gitscm, "test_git_patchtool pre-condition failed: {} 
> test recipe doesn't contains git repo!".format(test_recipe))
> +result = bitbake('{} -c patch'.format(test_recipe), 
> ignore_status=False)
> +srcdir = get_bb_var('S', test_recipe)
> +result = runCmd("git log", cwd = srcdir)
> +self.assertFalse("bitbake_patching_started" in result.output, msg = 
> "Repository has been reinitialized. {}".format(srcdir))
> +self.delete_recipeinc(test_recipe)
> +bitbake('-cclean {}'.format(test_recipe))
> +
>  
>  def test_git_unpack_nonetwork(self):
>  """
> 





-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#162049): 
https://lists.openembedded.org/g/openembedded-core/message/162049
Mute This Topic: https://lists.openembedded.org/mt/89293670/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH] Revert "patch.py: Initialize git repo before patching"

[Christian Eggers]

This reverts commit 6184b56a7a0fc6f5d19fdfb81e7453667f7da940.

The above commit breaks building of recipes where sources are checked out from 
Git and PATCHTOOL = "git". The do_patch tasks fails before applying the
first patch without any helpful error messages.

GitApplyTree::_isInitialized():
+ cmd = "git rev-parse --show-toplevel"
+ (status, output) = subprocess.getstatusoutput(cmd.split())
The cmd.split() expression effectively runs GIT without any arguments.
Using `subprocess.getstatusoutput(cmd)` seems to work.

+ ## Make sure repo is in builddir to not break top-level git repos
+ return status == 0 and os.path.samedir(output, self.dir)
Exception: AttributeError: module 'posixpath' has no attribute 'samedir'

GitApplyTree::_initRepo():
+ runcmd("git init".split(), self.dir)
+ runcmd("git add .".split(), self.dir)
+ runcmd("git commit -a --allow-empty -m Patching_started".split(), 
self.dir)
If _isInitialized() returned false although there is actually an initialized
GIT repo, the last command fails with 128.

Signed-off-by: Christian Eggers 
---
The "git commit" command fails only on my CI machine but not on my dev system
(couldn't figure out why). This means that on both systems the existing GIT
respository is erroneously reinitialized, but on the CI this breaks the build
while on the dev machine the error is not noticeable.

 meta/lib/oe/patch.py| 16 +---
 meta/lib/oeqa/selftest/cases/bbtests.py | 15 ---
 2 files changed, 1 insertion(+), 30 deletions(-)

diff --git a/meta/lib/oe/patch.py b/meta/lib/oe/patch.py
index 950fe723dcfd..fccbedb519d7 100644
--- a/meta/lib/oe/patch.py
+++ b/meta/lib/oe/patch.py
@@ -4,7 +4,6 @@
 
 import oe.path
 import oe.types
-import subprocess
 
 class NotFoundError(bb.BBHandledException):
 def __init__(self, path):
@@ -26,6 +25,7 @@ class CmdError(bb.BBHandledException):
 
 def runcmd(args, dir = None):
 import pipes
+import subprocess
 
 if dir:
 olddir = os.path.abspath(os.curdir)
@@ -56,7 +56,6 @@ def runcmd(args, dir = None):
 if dir:
 os.chdir(olddir)
 
-
 class PatchError(Exception):
 def __init__(self, msg):
 self.msg = msg
@@ -299,19 +298,6 @@ class GitApplyTree(PatchTree):
 PatchTree.__init__(self, dir, d)
 self.commituser = d.getVar('PATCH_GIT_USER_NAME')
 self.commitemail = d.getVar('PATCH_GIT_USER_EMAIL')
-if not self._isInitialized():
-self._initRepo()
-
-def _isInitialized(self):
-cmd = "git rev-parse --show-toplevel"
-(status, output) = subprocess.getstatusoutput(cmd.split())
-## Make sure repo is in builddir to not break top-level git repos
-return status == 0 and os.path.samedir(output, self.dir)
-
-def _initRepo(self):
-runcmd("git init".split(), self.dir)
-runcmd("git add .".split(), self.dir)
-runcmd("git commit -a --allow-empty -m Patching_started".split(), 
self.dir)
 
 @staticmethod
 def extractPatchHeader(patchfile):
diff --git a/meta/lib/oeqa/selftest/cases/bbtests.py 
b/meta/lib/oeqa/selftest/cases/bbtests.py
index e2b9290200ac..487568c907e5 100644
--- a/meta/lib/oeqa/selftest/cases/bbtests.py
+++ b/meta/lib/oeqa/selftest/cases/bbtests.py
@@ -298,21 +298,6 @@ INHERIT:remove = \"report-error\"
 test_recipe_summary_after = get_bb_var('SUMMARY', test_recipe)
 self.assertEqual(expected_recipe_summary, test_recipe_summary_after)
 
-def test_git_patchtool(self):
-""" PATCHTOOL=git should work with non-git sources like tarballs
-test recipe for the test must NOT containt git:// repository in 
SRC_URI
-"""
-test_recipe = "man-db"
-self.write_recipeinc(test_recipe, 'PATCHTOOL=\"git\"')
-src = get_bb_var("SRC_URI",test_recipe)
-gitscm = re.search("git://", src)
-self.assertFalse(gitscm, "test_git_patchtool pre-condition failed: {} 
test recipe contains git repo!".format(test_recipe))
-result = bitbake('man-db -c patch', ignore_status=False)
-fatal = re.search("fatal: not a git repository (or any of the parent 
directories)", result.output)
-self.assertFalse(fatal, "Failed to patch using PATCHTOOL=\"git\"")
-self.delete_recipeinc(test_recipe)
-bitbake('-cclean man-db')
-
 def test_git_unpack_nonetwork(self):
 """
 Test that a recipe with a floating tag that needs to be resolved 
upstream doesn't
-- 
2.34.1


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#161948): 
https://lists.openembedded.org/g/openembedded-core/message/161948
Mute This Topic: https://lists.openembedded.org/mt/89238057/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] Building nativesdk package with Rust

[Christian Eggers]

In oe-core commit 67dd505c42 ("librsvg: do not enable nativesdk"),
nativesdk support for librsvg has been disabled:

> rust is not currently available for nativesdk builds:

> WARNING: Nothing PROVIDES 'nativesdk-libstd-rs' (but 
> virtual:nativesdk:/home/pokybuild/yocto-worker/a-full/build/meta/recipes-gnome/librsvg/librsvg_2.52.0.bb
>  DEPENDS on or otherwise requires it). Close matches:
> WARNING: Nothing PROVIDES 'virtual/nativesdk-x86_64-pokysdk-linux-rust' (but 
> virtual:nativesdk:/home/pokybuild/yocto-worker/a-full/build/meta/recipes-gnome/librsvg/librsvg_2.52.0.bb
>  DEPENDS on or otherwise requires it). Close matches:

After retrying today, only the 2nd warning was shown.

librsvg is used by graphviz which in turn is used by doxygen. I would like
to put this in my SDK again.

Is it possible to solve this problem in a foreseeable future?

regards
Christian




_______
Christian
Eggers
Software Engineer
​
ARRI
Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Arriweg 17,
83071
Stephanskirchen
www.arri.com

+49 8036 3009-3118
cegg...@arri.de

​
Get all the latest information from www.arri.com, Facebook, Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

​

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#161887): 
https://lists.openembedded.org/g/openembedded-core/message/161887
Mute This Topic: https://lists.openembedded.org/mt/89228760/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [meta-oe][hardknott][PATCH] graphviz: native: create /usr/lib/graphviz/config6 in populate_sysroot

[Christian Eggers]

The `dot` tool requires to be run once after installation in order to
create its configuration file.

The do_prepare_recipe_sysroot task uses do_populate_sysroot in order to
prepare the recipe-sysroot-native. Package postinstall scripts are not
executed for -native packages, but files under ${BINDIR}/postinst-* are.

This is quite the same as graphviz-setup.sh does for nativesdk. The
general idea has been taken from
OECORE/meta/classes/pixbufcache.bbclass.

Signed-off-by: Christian Eggers 
Signed-off-by: Khem Raj 
---
 meta-oe/recipes-graphics/graphviz/graphviz_2.44.1.bb | 11 +++
 1 file changed, 11 insertions(+)

diff --git a/meta-oe/recipes-graphics/graphviz/graphviz_2.44.1.bb 
b/meta-oe/recipes-graphics/graphviz/graphviz_2.44.1.bb
index 8fe837b75ceb..cc57761bc4fe 100644
--- a/meta-oe/recipes-graphics/graphviz/graphviz_2.44.1.bb
+++ b/meta-oe/recipes-graphics/graphviz/graphviz_2.44.1.bb
@@ -74,6 +74,17 @@ do_install_append_class-nativesdk() {
 }
 FILES_${PN}_class-nativesdk += "${SDKPATHNATIVE}"
 
+# create /usr/lib/graphviz/config6
+graphviz_sstate_postinst() {
+mkdir -p ${SYSROOT_DESTDIR}${bindir}
+dest=${SYSROOT_DESTDIR}${bindir}/postinst-${PN}
+echo '#!/bin/sh' > $dest
+echo '' >> $dest
+echo 'dot -c' >> $dest
+chmod 0755 $dest
+}
+SYSROOT_PREPROCESS_FUNCS_append_class-native = " graphviz_sstate_postinst"
+
 PACKAGES =+ "${PN}-python ${PN}-perl ${PN}-demo"
 
 FILES_${PN}-python += "${libdir}/python*/site-packages/ 
${libdir}/graphviz/python/"
-- 
2.34.1


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#161778): 
https://lists.openembedded.org/g/openembedded-core/message/161778
Mute This Topic: https://lists.openembedded.org/mt/89185407/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [meta-oe][honister][PATCH] graphviz: native: create /usr/lib/graphviz/config6 in populate_sysroot

[Christian Eggers]

The `dot` tool requires to be run once after installation in order to
create its configuration file.

The do_prepare_recipe_sysroot task uses do_populate_sysroot in order to
prepare the recipe-sysroot-native. Package postinstall scripts are not
executed for -native packages, but files under ${BINDIR}/postinst-* are.

This is quite the same as graphviz-setup.sh does for nativesdk. The
general idea has been taken from
OECORE/meta/classes/pixbufcache.bbclass.

Signed-off-by: Christian Eggers 
Signed-off-by: Khem Raj 
---
 meta-oe/recipes-graphics/graphviz/graphviz_2.44.1.bb | 11 +++
 1 file changed, 11 insertions(+)

diff --git a/meta-oe/recipes-graphics/graphviz/graphviz_2.44.1.bb 
b/meta-oe/recipes-graphics/graphviz/graphviz_2.44.1.bb
index 7d2515e07549..203e0206fbfe 100644
--- a/meta-oe/recipes-graphics/graphviz/graphviz_2.44.1.bb
+++ b/meta-oe/recipes-graphics/graphviz/graphviz_2.44.1.bb
@@ -74,6 +74,17 @@ do_install:append:class-nativesdk() {
 }
 FILES:${PN}:class-nativesdk += "${SDKPATHNATIVE}"
 
+# create /usr/lib/graphviz/config6
+graphviz_sstate_postinst() {
+mkdir -p ${SYSROOT_DESTDIR}${bindir}
+dest=${SYSROOT_DESTDIR}${bindir}/postinst-${PN}
+echo '#!/bin/sh' > $dest
+echo '' >> $dest
+echo 'dot -c' >> $dest
+chmod 0755 $dest
+}
+SYSROOT_PREPROCESS_FUNCS:append:class-native = " graphviz_sstate_postinst"
+
 PACKAGES =+ "${PN}-python ${PN}-perl ${PN}-demo"
 
 FILES:${PN}-python += "${libdir}/python*/site-packages/ 
${libdir}/graphviz/python/"
-- 
2.34.1


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#161776): 
https://lists.openembedded.org/g/openembedded-core/message/161776
Mute This Topic: https://lists.openembedded.org/mt/89185197/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [hardknott][PATCH] mc: fix build if ncurses have been configured without wide characters

[Christian Eggers]

Some distros like poky-tiny set ENABLE_WIDEC=false, which disables wide
character support for ncurses. The new patch fixes the build of mc for
this case.

Since 9000f8033662, NCURSES_WIDECHAR is set explicitly to 1 for musl.
This doesn't work for ENABLE_WIDEC==false. In this case,
NCURSES_WIDECHAR must be set explicitly to 0, as curses.h does not
record whether the ncurses library has actually been built with or
without wide characters.

Fixes: 9000f8033662 ("mc: upgrade 4.8.25 -> 4.8.26")
Signed-off-by: Christian Eggers 
Signed-off-by: Richard Purdie 
---
 ...FTBFS-with-ncurses-build-with-disabl.patch | 87 +++
 meta/recipes-extended/mc/mc_4.8.26.bb |  5 +-
 2 files changed, 91 insertions(+), 1 deletion(-)
 create mode 100644 
meta/recipes-extended/mc/files/0001-Ticket-4200-fix-FTBFS-with-ncurses-build-with-disabl.patch

diff --git 
a/meta/recipes-extended/mc/files/0001-Ticket-4200-fix-FTBFS-with-ncurses-build-with-disabl.patch
 
b/meta/recipes-extended/mc/files/0001-Ticket-4200-fix-FTBFS-with-ncurses-build-with-disabl.patch
new file mode 100644
index ..408473664f5d
--- /dev/null
+++ 
b/meta/recipes-extended/mc/files/0001-Ticket-4200-fix-FTBFS-with-ncurses-build-with-disabl.patch
@@ -0,0 +1,87 @@
+From e7bbf72544ab62db9c92bfe7bd1155227e78c621 Mon Sep 17 00:00:00 2001
+From: Andrew Borodin 
+Date: Sat, 28 Aug 2021 11:46:53 +0300
+Subject: [PATCH] Ticket #4200: fix FTBFS with ncurses build with
+ --disable-widec.
+
+Upstream-Status: Accepted 
[https://github.com/MidnightCommander/mc/commit/e7bbf72544]
+Signed-off-by: Andrew Borodin 
+---
+ lib/tty/tty-ncurses.c   | 8 
+ lib/tty/tty-ncurses.h   | 5 +
+ lib/tty/tty-slang.h | 2 ++
+ src/filemanager/boxes.c | 2 ++
+ 4 files changed, 17 insertions(+)
+
+diff --git a/lib/tty/tty-ncurses.c b/lib/tty/tty-ncurses.c
+index f619c0a7bf31..13058a624208 100644
+--- a/lib/tty/tty-ncurses.c
 b/lib/tty/tty-ncurses.c
+@@ -560,6 +560,7 @@ tty_fill_region (int y, int x, int rows, int cols, 
unsigned char ch)
+ void
+ tty_colorize_area (int y, int x, int rows, int cols, int color)
+ {
++#ifdef ENABLE_SHADOWS
+ cchar_t *ctext;
+ wchar_t wch[10];/* TODO not sure if the length is correct */
+ attr_t attrs;
+@@ -585,6 +586,13 @@ tty_colorize_area (int y, int x, int rows, int cols, int 
color)
+ }
+ 
+ g_free (ctext);
++#else
++(void) y;
++(void) x;
++(void) rows;
++(void) cols;
++(void) color;
++#endif /* ENABLE_SHADOWS */
+ }
+ 
+ /* 
-
 */
+diff --git a/lib/tty/tty-ncurses.h b/lib/tty/tty-ncurses.h
+index d75df9533ab9..8feb17ccd045 100644
+--- a/lib/tty/tty-ncurses.h
 b/lib/tty/tty-ncurses.h
+@@ -30,6 +30,11 @@
+ #define NCURSES_CONST const
+ #endif
+ 
++/* do not draw shadows if NCurses is built with --disable-widec */
++#if defined(NCURSES_WIDECHAR) && NCURSES_WIDECHAR
++#define ENABLE_SHADOWS 1
++#endif
++
+ /*** typedefs(not structures) and defined constants 
**/
+ 
+ /*** enums 
***/
+diff --git a/lib/tty/tty-slang.h b/lib/tty/tty-slang.h
+index 5b12c6512853..eeaade388af4 100644
+--- a/lib/tty/tty-slang.h
 b/lib/tty/tty-slang.h
+@@ -23,6 +23,8 @@
+ #define COLS  SLtt_Screen_Cols
+ #define LINES SLtt_Screen_Rows
+ 
++#define ENABLE_SHADOWS 1
++
+ /*** enums 
***/
+ 
+ enum
+diff --git a/src/filemanager/boxes.c b/src/filemanager/boxes.c
+index 3eb525be4a9b..98df5ff2ed9a 100644
+--- a/src/filemanager/boxes.c
 b/src/filemanager/boxes.c
+@@ -280,7 +280,9 @@ appearance_box_callback (Widget * w, Widget * sender, 
widget_msg_t msg, int parm
+ switch (msg)
+ {
+ case MSG_INIT:
++#ifdef ENABLE_SHADOWS
+ if (!tty_use_colors ())
++#endif
+ {
+ Widget *shadow;
+ 
+-- 
+2.34.1
+
diff --git a/meta/recipes-extended/mc/mc_4.8.26.bb 
b/meta/recipes-extended/mc/mc_4.8.26.bb
index 6bc7e6e8e159..906778400ec8 100644
--- a/meta/recipes-extended/mc/mc_4.8.26.bb
+++ b/meta/recipes-extended/mc/mc_4.8.26.bb
@@ -12,6 +12,7 @@ SRC_URI = 
"http://www.midnight-commander.org/downloads/${BPN}-${PV}.tar.bz2 \
file://0001-mc-replace-perl-w-with-use-warnings.patch \
file://nomandate.patch \
file://CVE-2021-36370.patch \
+   
file://0001-Ticket-4200-fix-FTBFS-with-ncurses-build-with-disabl.patch \
"
 SRC_URI[sha256sum] = 
"9d6358d0a351a455a1410aab57f33b6b48b0fcf31344b9a10b0ff497595979d1"
 
@@ -24,7 +25,9 @@ PACKAGECONFIG ??= ""
 PACKAGECONFIG[smb] = "--enable-vfs-smb,--disable-vfs-smb,samba,"
 PACKAGECONFIG[sftp] = "--enable-vfs-sftp,--disable-vfs-sftp,libssh2,"
 
-CFLAGS_append_libc-musl = ' -DNCURSES_WIDECHAR=1 '
+# enable NC

[OE-core] [honister][PATCH] mc: fix build if ncurses have been configured without wide characters

[Christian Eggers]

Some distros like poky-tiny set ENABLE_WIDEC=false, which disables wide
character support for ncurses. The new patch fixes the build of mc for
this case.

Since 9000f8033662, NCURSES_WIDECHAR is set explicitly to 1 for musl.
This doesn't work for ENABLE_WIDEC==false. In this case,
NCURSES_WIDECHAR must be set explicitly to 0, as curses.h does not
record whether the ncurses library has actually been built with or
without wide characters.

Fixes: 9000f8033662 ("mc: upgrade 4.8.25 -> 4.8.26")
Signed-off-by: Christian Eggers 
Signed-off-by: Richard Purdie 
---
 ...FTBFS-with-ncurses-build-with-disabl.patch | 87 +++
 meta/recipes-extended/mc/mc_4.8.27.bb |  5 +-
 2 files changed, 91 insertions(+), 1 deletion(-)
 create mode 100644 
meta/recipes-extended/mc/files/0001-Ticket-4200-fix-FTBFS-with-ncurses-build-with-disabl.patch

diff --git 
a/meta/recipes-extended/mc/files/0001-Ticket-4200-fix-FTBFS-with-ncurses-build-with-disabl.patch
 
b/meta/recipes-extended/mc/files/0001-Ticket-4200-fix-FTBFS-with-ncurses-build-with-disabl.patch
new file mode 100644
index ..408473664f5d
--- /dev/null
+++ 
b/meta/recipes-extended/mc/files/0001-Ticket-4200-fix-FTBFS-with-ncurses-build-with-disabl.patch
@@ -0,0 +1,87 @@
+From e7bbf72544ab62db9c92bfe7bd1155227e78c621 Mon Sep 17 00:00:00 2001
+From: Andrew Borodin 
+Date: Sat, 28 Aug 2021 11:46:53 +0300
+Subject: [PATCH] Ticket #4200: fix FTBFS with ncurses build with
+ --disable-widec.
+
+Upstream-Status: Accepted 
[https://github.com/MidnightCommander/mc/commit/e7bbf72544]
+Signed-off-by: Andrew Borodin 
+---
+ lib/tty/tty-ncurses.c   | 8 
+ lib/tty/tty-ncurses.h   | 5 +
+ lib/tty/tty-slang.h | 2 ++
+ src/filemanager/boxes.c | 2 ++
+ 4 files changed, 17 insertions(+)
+
+diff --git a/lib/tty/tty-ncurses.c b/lib/tty/tty-ncurses.c
+index f619c0a7bf31..13058a624208 100644
+--- a/lib/tty/tty-ncurses.c
 b/lib/tty/tty-ncurses.c
+@@ -560,6 +560,7 @@ tty_fill_region (int y, int x, int rows, int cols, 
unsigned char ch)
+ void
+ tty_colorize_area (int y, int x, int rows, int cols, int color)
+ {
++#ifdef ENABLE_SHADOWS
+ cchar_t *ctext;
+ wchar_t wch[10];/* TODO not sure if the length is correct */
+ attr_t attrs;
+@@ -585,6 +586,13 @@ tty_colorize_area (int y, int x, int rows, int cols, int 
color)
+ }
+ 
+ g_free (ctext);
++#else
++(void) y;
++(void) x;
++(void) rows;
++(void) cols;
++(void) color;
++#endif /* ENABLE_SHADOWS */
+ }
+ 
+ /* 
-
 */
+diff --git a/lib/tty/tty-ncurses.h b/lib/tty/tty-ncurses.h
+index d75df9533ab9..8feb17ccd045 100644
+--- a/lib/tty/tty-ncurses.h
 b/lib/tty/tty-ncurses.h
+@@ -30,6 +30,11 @@
+ #define NCURSES_CONST const
+ #endif
+ 
++/* do not draw shadows if NCurses is built with --disable-widec */
++#if defined(NCURSES_WIDECHAR) && NCURSES_WIDECHAR
++#define ENABLE_SHADOWS 1
++#endif
++
+ /*** typedefs(not structures) and defined constants 
**/
+ 
+ /*** enums 
***/
+diff --git a/lib/tty/tty-slang.h b/lib/tty/tty-slang.h
+index 5b12c6512853..eeaade388af4 100644
+--- a/lib/tty/tty-slang.h
 b/lib/tty/tty-slang.h
+@@ -23,6 +23,8 @@
+ #define COLS  SLtt_Screen_Cols
+ #define LINES SLtt_Screen_Rows
+ 
++#define ENABLE_SHADOWS 1
++
+ /*** enums 
***/
+ 
+ enum
+diff --git a/src/filemanager/boxes.c b/src/filemanager/boxes.c
+index 3eb525be4a9b..98df5ff2ed9a 100644
+--- a/src/filemanager/boxes.c
 b/src/filemanager/boxes.c
+@@ -280,7 +280,9 @@ appearance_box_callback (Widget * w, Widget * sender, 
widget_msg_t msg, int parm
+ switch (msg)
+ {
+ case MSG_INIT:
++#ifdef ENABLE_SHADOWS
+ if (!tty_use_colors ())
++#endif
+ {
+ Widget *shadow;
+ 
+-- 
+2.34.1
+
diff --git a/meta/recipes-extended/mc/mc_4.8.27.bb 
b/meta/recipes-extended/mc/mc_4.8.27.bb
index 546e615d1d1c..e877780ea051 100644
--- a/meta/recipes-extended/mc/mc_4.8.27.bb
+++ b/meta/recipes-extended/mc/mc_4.8.27.bb
@@ -11,6 +11,7 @@ RRECOMMENDS:${PN} = "ncurses-terminfo"
 SRC_URI = "http://www.midnight-commander.org/downloads/${BPN}-${PV}.tar.bz2 \
file://0001-mc-replace-perl-w-with-use-warnings.patch \
file://nomandate.patch \
+   
file://0001-Ticket-4200-fix-FTBFS-with-ncurses-build-with-disabl.patch \
"
 SRC_URI[sha256sum] = 
"2f52dd9c75c20d8eac7701bd3a8c6c125aaf8cdd9cf12b78ca50a0102b543407"
 
@@ -27,7 +28,9 @@ PACKAGECONFIG ??= ""
 PACKAGECONFIG[smb] = "--enable-vfs-smb,--disable-vfs-smb,samba,"
 PACKAGECONFIG[sftp] = "--enable-vfs-sftp,--disable-vfs-sftp,libssh2,"
 
-CFLAGS:append:libc-musl = ' -DNCURSES_WIDECHAR=1 '
+# enab

[OE-core] [PATCH] curl: configure with '--without-ssl' if ssl is disabled

[Christian Eggers]

curl requires that at least one SSL implementation or explicitly
'--without-ssl' is specified. This is particularly the case if
PACKAGECONFIG is empty.

| configure: error: select TLS backend(s) or disable TLS with
--without-ssl.
|
| Select from these:
|
|   --with-amissl
|   --with-bearssl
|   --with-gnutls
|   --with-mbedtls
|   --with-mesalink
|   --with-nss
|   --with-openssl (also works for BoringSSL and libressl)
|   --with-rustls
|   --with-schannel
|   --with-secure-transport
|   --with-wolfssl
|

Fixes: eef6c45fc6ec ("curl: Rework openssl and random PACKAGECONFIGs")
Signed-off-by: Christian Eggers 
---
 meta/recipes-support/curl/curl_7.81.0.bb | 1 +
 1 file changed, 1 insertion(+)

diff --git a/meta/recipes-support/curl/curl_7.81.0.bb 
b/meta/recipes-support/curl/curl_7.81.0.bb
index c22a13cfea3f..e7219def18d8 100644
--- a/meta/recipes-support/curl/curl_7.81.0.bb
+++ b/meta/recipes-support/curl/curl_7.81.0.bb
@@ -67,6 +67,7 @@ EXTRA_OECONF = " \
 --enable-debug \
 --enable-optimize \
 --disable-curldebug \
+${@'--without-ssl' if (bb.utils.filter('PACKAGECONFIG', 'gnutls mbedtls 
nss openssl', d) == '') else ''} \
 "
 
 do_install:append:class-target() {
-- 
2.34.1


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#161747): 
https://lists.openembedded.org/g/openembedded-core/message/161747
Mute This Topic: https://lists.openembedded.org/mt/89159635/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH] dev-manual: update example from kernel.bbclass

[Christian Eggers]

Since f7485646534e ("kernel: provide virtual/kernel for all kernels"),
all recipes which inherit kernel.bbclass provide "virtual/kernel",
independent of the value of `KERNEL_PACKAGE_NAME`.

Signed-off-by: Christian Eggers 
---
 documentation/dev-manual/common-tasks.rst | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/documentation/dev-manual/common-tasks.rst 
b/documentation/dev-manual/common-tasks.rst
index d35b590ee185..1856d4c3ad53 100644
--- a/documentation/dev-manual/common-tasks.rst
+++ b/documentation/dev-manual/common-tasks.rst
@@ -2114,7 +2114,7 @@ statement that essentially identifies itself as being 
able to provide
 ``virtual/kernel``. Here is one way through the
 :ref:`kernel ` class::
 
-   PROVIDES += "${@ "virtual/kernel" if (d.getVar("KERNEL_PACKAGE_NAME") == 
"kernel") else "" }"
+   PROVIDES += "virtual/kernel"
 
 Any recipe that inherits the :ref:`kernel ` class is
 going to utilize a :term:`PROVIDES` statement that identifies that recipe as
-- 
2.34.1


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#161720): 
https://lists.openembedded.org/g/openembedded-core/message/161720
Mute This Topic: https://lists.openembedded.org/mt/89139800/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [dunfell][PATCH] sdk: fix search for dynamic loader

[Christian Eggers]

if the package "nativesdk-glibc-dbg" is installed as part of the SDK,
the existing search expression finds two files:

$OECORE_NATIVE_SYSROOT/lib/.debug/ld-linux-x86-64.so.2
$OECORE_NATIVE_SYSROOT/lib/ld-linux-x86-64.so.2

The generated relocate_sdk.sh shell script contains then an extra
newline and segfaults during SDK relocation.

Limit the search depth to 1, to avoid finding the file in the .debug
directory.

Signed-off-by: Christian Eggers 
Signed-off-by: Richard Purdie 
---
 meta/files/toolchain-shar-relocate.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/files/toolchain-shar-relocate.sh 
b/meta/files/toolchain-shar-relocate.sh
index 3ece04db0a4a..cee9adbf399c 100644
--- a/meta/files/toolchain-shar-relocate.sh
+++ b/meta/files/toolchain-shar-relocate.sh
@@ -5,7 +5,7 @@ fi
 
 # fix dynamic loader paths in all ELF SDK binaries
 native_sysroot=$($SUDO_EXEC cat $env_setup_script |grep 
'OECORE_NATIVE_SYSROOT='|cut -d'=' -f2|tr -d '"')
-dl_path=$($SUDO_EXEC find $native_sysroot/lib -name "ld-linux*")
+dl_path=$($SUDO_EXEC find $native_sysroot/lib -maxdepth 1 -name "ld-linux*")
 if [ "$dl_path" = "" ] ; then
echo "SDK could not be set up. Relocate script unable to find 
ld-linux.so. Abort!"
exit 1
-- 
2.34.1


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#161702): 
https://lists.openembedded.org/g/openembedded-core/message/161702
Mute This Topic: https://lists.openembedded.org/mt/89133465/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [hardknott][PATCH] sdk: fix search for dynamic loader

[Christian Eggers]

if the package "nativesdk-glibc-dbg" is installed as part of the SDK,
the existing search expression finds two files:

$OECORE_NATIVE_SYSROOT/lib/.debug/ld-linux-x86-64.so.2
$OECORE_NATIVE_SYSROOT/lib/ld-linux-x86-64.so.2

The generated relocate_sdk.sh shell script contains then an extra
newline and segfaults during SDK relocation.

Limit the search depth to 1, to avoid finding the file in the .debug
directory.

Signed-off-by: Christian Eggers 
Signed-off-by: Richard Purdie 
---
 meta/files/toolchain-shar-relocate.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/files/toolchain-shar-relocate.sh 
b/meta/files/toolchain-shar-relocate.sh
index 3ece04db0a4a..cee9adbf399c 100644
--- a/meta/files/toolchain-shar-relocate.sh
+++ b/meta/files/toolchain-shar-relocate.sh
@@ -5,7 +5,7 @@ fi
 
 # fix dynamic loader paths in all ELF SDK binaries
 native_sysroot=$($SUDO_EXEC cat $env_setup_script |grep 
'OECORE_NATIVE_SYSROOT='|cut -d'=' -f2|tr -d '"')
-dl_path=$($SUDO_EXEC find $native_sysroot/lib -name "ld-linux*")
+dl_path=$($SUDO_EXEC find $native_sysroot/lib -maxdepth 1 -name "ld-linux*")
 if [ "$dl_path" = "" ] ; then
echo "SDK could not be set up. Relocate script unable to find 
ld-linux.so. Abort!"
exit 1
-- 
2.34.1


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#161701): 
https://lists.openembedded.org/g/openembedded-core/message/161701
Mute This Topic: https://lists.openembedded.org/mt/89133459/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [honister][PATCH] sdk: fix search for dynamic loader

[Christian Eggers]

if the package "nativesdk-glibc-dbg" is installed as part of the SDK,
the existing search expression finds two files:

$OECORE_NATIVE_SYSROOT/lib/.debug/ld-linux-x86-64.so.2
$OECORE_NATIVE_SYSROOT/lib/ld-linux-x86-64.so.2

The generated relocate_sdk.sh shell script contains then an extra
newline and segfaults during SDK relocation.

Limit the search depth to 1, to avoid finding the file in the .debug
directory.

Signed-off-by: Christian Eggers 
Signed-off-by: Richard Purdie 
---
 meta/files/toolchain-shar-relocate.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/files/toolchain-shar-relocate.sh 
b/meta/files/toolchain-shar-relocate.sh
index 3ece04db0a4a..cee9adbf399c 100644
--- a/meta/files/toolchain-shar-relocate.sh
+++ b/meta/files/toolchain-shar-relocate.sh
@@ -5,7 +5,7 @@ fi
 
 # fix dynamic loader paths in all ELF SDK binaries
 native_sysroot=$($SUDO_EXEC cat $env_setup_script |grep 
'OECORE_NATIVE_SYSROOT='|cut -d'=' -f2|tr -d '"')
-dl_path=$($SUDO_EXEC find $native_sysroot/lib -name "ld-linux*")
+dl_path=$($SUDO_EXEC find $native_sysroot/lib -maxdepth 1 -name "ld-linux*")
 if [ "$dl_path" = "" ] ; then
echo "SDK could not be set up. Relocate script unable to find 
ld-linux.so. Abort!"
exit 1
-- 
2.34.1


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#161700): 
https://lists.openembedded.org/g/openembedded-core/message/161700
Mute This Topic: https://lists.openembedded.org/mt/89133453/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH] mc: fix build if ncurses have been configured without wide characters

[Christian Eggers]

Some distros like poky-tiny set ENABLE_WIDEC=false, which disables wide
character support for ncurses. The new patch fixes the build of mc for
this case.

Since 9000f8033662, NCURSES_WIDECHAR is set explicitly to 1 for musl.
This doesn't work for ENABLE_WIDEC==false. In this case,
NCURSES_WIDECHAR must be set explicitly to 0, as curses.h does not
record whether the ncurses library has actually been built with or
without wide characters.

Fixes: 9000f8033662 ("mc: upgrade 4.8.25 -> 4.8.26")
Signed-off-by: Christian Eggers 
---
 ...FTBFS-with-ncurses-build-with-disabl.patch | 87 +++
 meta/recipes-extended/mc/mc_4.8.27.bb |  5 +-
 2 files changed, 91 insertions(+), 1 deletion(-)
 create mode 100644 
meta/recipes-extended/mc/files/0001-Ticket-4200-fix-FTBFS-with-ncurses-build-with-disabl.patch

diff --git 
a/meta/recipes-extended/mc/files/0001-Ticket-4200-fix-FTBFS-with-ncurses-build-with-disabl.patch
 
b/meta/recipes-extended/mc/files/0001-Ticket-4200-fix-FTBFS-with-ncurses-build-with-disabl.patch
new file mode 100644
index ..408473664f5d
--- /dev/null
+++ 
b/meta/recipes-extended/mc/files/0001-Ticket-4200-fix-FTBFS-with-ncurses-build-with-disabl.patch
@@ -0,0 +1,87 @@
+From e7bbf72544ab62db9c92bfe7bd1155227e78c621 Mon Sep 17 00:00:00 2001
+From: Andrew Borodin 
+Date: Sat, 28 Aug 2021 11:46:53 +0300
+Subject: [PATCH] Ticket #4200: fix FTBFS with ncurses build with
+ --disable-widec.
+
+Upstream-Status: Accepted 
[https://github.com/MidnightCommander/mc/commit/e7bbf72544]
+Signed-off-by: Andrew Borodin 
+---
+ lib/tty/tty-ncurses.c   | 8 
+ lib/tty/tty-ncurses.h   | 5 +
+ lib/tty/tty-slang.h | 2 ++
+ src/filemanager/boxes.c | 2 ++
+ 4 files changed, 17 insertions(+)
+
+diff --git a/lib/tty/tty-ncurses.c b/lib/tty/tty-ncurses.c
+index f619c0a7bf31..13058a624208 100644
+--- a/lib/tty/tty-ncurses.c
 b/lib/tty/tty-ncurses.c
+@@ -560,6 +560,7 @@ tty_fill_region (int y, int x, int rows, int cols, 
unsigned char ch)
+ void
+ tty_colorize_area (int y, int x, int rows, int cols, int color)
+ {
++#ifdef ENABLE_SHADOWS
+ cchar_t *ctext;
+ wchar_t wch[10];/* TODO not sure if the length is correct */
+ attr_t attrs;
+@@ -585,6 +586,13 @@ tty_colorize_area (int y, int x, int rows, int cols, int 
color)
+ }
+ 
+ g_free (ctext);
++#else
++(void) y;
++(void) x;
++(void) rows;
++(void) cols;
++(void) color;
++#endif /* ENABLE_SHADOWS */
+ }
+ 
+ /* 
-
 */
+diff --git a/lib/tty/tty-ncurses.h b/lib/tty/tty-ncurses.h
+index d75df9533ab9..8feb17ccd045 100644
+--- a/lib/tty/tty-ncurses.h
 b/lib/tty/tty-ncurses.h
+@@ -30,6 +30,11 @@
+ #define NCURSES_CONST const
+ #endif
+ 
++/* do not draw shadows if NCurses is built with --disable-widec */
++#if defined(NCURSES_WIDECHAR) && NCURSES_WIDECHAR
++#define ENABLE_SHADOWS 1
++#endif
++
+ /*** typedefs(not structures) and defined constants 
**/
+ 
+ /*** enums 
***/
+diff --git a/lib/tty/tty-slang.h b/lib/tty/tty-slang.h
+index 5b12c6512853..eeaade388af4 100644
+--- a/lib/tty/tty-slang.h
 b/lib/tty/tty-slang.h
+@@ -23,6 +23,8 @@
+ #define COLS  SLtt_Screen_Cols
+ #define LINES SLtt_Screen_Rows
+ 
++#define ENABLE_SHADOWS 1
++
+ /*** enums 
***/
+ 
+ enum
+diff --git a/src/filemanager/boxes.c b/src/filemanager/boxes.c
+index 3eb525be4a9b..98df5ff2ed9a 100644
+--- a/src/filemanager/boxes.c
 b/src/filemanager/boxes.c
+@@ -280,7 +280,9 @@ appearance_box_callback (Widget * w, Widget * sender, 
widget_msg_t msg, int parm
+ switch (msg)
+ {
+ case MSG_INIT:
++#ifdef ENABLE_SHADOWS
+ if (!tty_use_colors ())
++#endif
+ {
+ Widget *shadow;
+ 
+-- 
+2.34.1
+
diff --git a/meta/recipes-extended/mc/mc_4.8.27.bb 
b/meta/recipes-extended/mc/mc_4.8.27.bb
index 546e615d1d1c..e877780ea051 100644
--- a/meta/recipes-extended/mc/mc_4.8.27.bb
+++ b/meta/recipes-extended/mc/mc_4.8.27.bb
@@ -11,6 +11,7 @@ RRECOMMENDS:${PN} = "ncurses-terminfo"
 SRC_URI = "http://www.midnight-commander.org/downloads/${BPN}-${PV}.tar.bz2 \
file://0001-mc-replace-perl-w-with-use-warnings.patch \
file://nomandate.patch \
+   
file://0001-Ticket-4200-fix-FTBFS-with-ncurses-build-with-disabl.patch \
"
 SRC_URI[sha256sum] = 
"2f52dd9c75c20d8eac7701bd3a8c6c125aaf8cdd9cf12b78ca50a0102b543407"
 
@@ -27,7 +28,9 @@ PACKAGECONFIG ??= ""
 PACKAGECONFIG[smb] = "--enable-vfs-smb,--disable-vfs-smb,samba,"
 PACKAGECONFIG[sftp] = "--enable-vfs-sftp,--disable-vfs-sftp,libssh2,"
 
-CFLAGS:append:libc-musl = ' -DNCURSES_WIDECHAR=1 '
+# enable NCURSES_WIDECHAR=1 only if EN

[OE-core] [hardknott][PATCH] package_manager: ipk: Fix host manifest generation

[Christian Eggers]

From: Oleksiy Obitotskyy 

Since honister host manifest stopped to generate, i.e.
manifest file is empty but all ipks/files into sdk is
ok.

Signed-off-by: Oleksiy Obitotskyy 
Signed-off-by: Richard Purdie 
(cherry picked from commit 79a2392f5d2a4cb6509a83afb40bca01bac59914)
Signed-off-by: Anuj Mittal 
---
 meta/lib/oe/sdk.py | 4 
 1 file changed, 4 insertions(+)

diff --git a/meta/lib/oe/sdk.py b/meta/lib/oe/sdk.py
index 37b59afd1a99..27347667e846 100644
--- a/meta/lib/oe/sdk.py
+++ b/meta/lib/oe/sdk.py
@@ -115,6 +115,10 @@ def sdk_list_installed_packages(d, target, 
rootfs_dir=None):
 
 rootfs_dir = [sdk_output, os.path.join(sdk_output, 
target_path)][target is True]
 
+if target is False:
+ipkgconf_sdk_target = d.getVar("IPKGCONF_SDK")
+d.setVar("IPKGCONF_TARGET", ipkgconf_sdk_target)
+
 img_type = d.getVar('IMAGE_PKGTYPE')
 import importlib
 cls = importlib.import_module('oe.package_manager.' + img_type)
-- 
2.34.1


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#161653): 
https://lists.openembedded.org/g/openembedded-core/message/161653
Mute This Topic: https://lists.openembedded.org/mt/89068016/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [hardknott][PATCH] package_manager: ipk: Fix host manifest generation

[Christian Eggers]

From: Oleksiy Obitotskyy 

Since honister host manifest stopped to generate, i.e.
manifest file is empty but all ipks/files into sdk is
ok.

Signed-off-by: Oleksiy Obitotskyy 
Signed-off-by: Richard Purdie 
(cherry picked from commit 79a2392f5d2a4cb6509a83afb40bca01bac59914)
Signed-off-by: Anuj Mittal 
---
 meta/lib/oe/sdk.py | 4 
 1 file changed, 4 insertions(+)

diff --git a/meta/lib/oe/sdk.py b/meta/lib/oe/sdk.py
index 37b59afd1a99..27347667e846 100644
--- a/meta/lib/oe/sdk.py
+++ b/meta/lib/oe/sdk.py
@@ -115,6 +115,10 @@ def sdk_list_installed_packages(d, target, 
rootfs_dir=None):

 rootfs_dir = [sdk_output, os.path.join(sdk_output, 
target_path)][target is True]

+if target is False:
+ipkgconf_sdk_target = d.getVar("IPKGCONF_SDK")
+d.setVar("IPKGCONF_TARGET", ipkgconf_sdk_target)
+
 img_type = d.getVar('IMAGE_PKGTYPE')
 import importlib
 cls = importlib.import_module('oe.package_manager.' + img_type)
--
2.34.1


_______
Christian
Eggers
Software Engineer
​
ARRI
Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Arriweg 17,
83071
Stephanskirchen
www.arri.com

+49 8036 3009-3118
cegg...@arri.de

​
Get all the latest information from www.arri.com, Facebook, Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

​

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#161638): 
https://lists.openembedded.org/g/openembedded-core/message/161638
Mute This Topic: https://lists.openembedded.org/mt/89068016/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH] sdk: fix search for dynamic loader

[Christian Eggers]

if the package "nativesdk-glibc-dbg" is installed as part of the SDK,
the existing search expression finds two files:

$OECORE_NATIVE_SYSROOT/lib/.debug/ld-linux-x86-64.so.2
$OECORE_NATIVE_SYSROOT/lib/ld-linux-x86-64.so.2

The generated relocate_sdk.sh shell script contains then an extra
newline and segfaults during SDK relocation.

Limit the search depth to 1, to avoid finding the file in the .debug
directory.

Signed-off-by: Christian Eggers 
---
I ran into this problem at first on honister (as previous releases seem
not to include nativesdk-glibc-dbg automatically, e.g. due to "dbg-pkgs"
in SDKIMAGE_FEATURES). But I think that this patch may also be relevant
hardknott and dunfell.

 meta/files/toolchain-shar-relocate.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/files/toolchain-shar-relocate.sh 
b/meta/files/toolchain-shar-relocate.sh
index 3ece04db0a4a..cee9adbf399c 100644
--- a/meta/files/toolchain-shar-relocate.sh
+++ b/meta/files/toolchain-shar-relocate.sh
@@ -5,7 +5,7 @@ fi

 # fix dynamic loader paths in all ELF SDK binaries
 native_sysroot=$($SUDO_EXEC cat $env_setup_script |grep 
'OECORE_NATIVE_SYSROOT='|cut -d'=' -f2|tr -d '"')
-dl_path=$($SUDO_EXEC find $native_sysroot/lib -name "ld-linux*")
+dl_path=$($SUDO_EXEC find $native_sysroot/lib -maxdepth 1 -name "ld-linux*")
 if [ "$dl_path" = "" ] ; then
echo "SDK could not be set up. Relocate script unable to find 
ld-linux.so. Abort!"
exit 1
--
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler


___
Christian
Eggers
Software Engineer
​
ARRI
Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Arriweg 17,
83071
Stephanskirchen
www.arri.com

+49 8036 3009-3118
cegg...@arri.de

​
Get all the latest information from www.arri.com, Facebook, Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

​

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#161604): 
https://lists.openembedded.org/g/openembedded-core/message/161604
Mute This Topic: https://lists.openembedded.org/mt/89042978/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] [PATCH meta-oe] rtc-tool: Add a recipe

[Christian Eggers]

Hi Fabio, hi Alexandre,

On Monday, 24 January 2022, 20:18:34 CET, Fabio Estevam wrote:
> Hi Christian,
>

___
Christian
Eggers
Software Engineer
​
ARRI
Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Arriweg 17,
83071
Stephanskirchen
www.arri.com

+49 8036 3009-3118
cegg...@arri.de

​
Get all the latest information from www.arri.com, Facebook, Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

​
> On Mon, Jan 24, 2022 at 6:57 AM Christian Eggers  wrote:
> >
> > Hi Fabio,
> >
> > from my experience, custom build systems (including "bare" Makefiles)
> > are often hard to integrate/maintain for distributors.
>
> rtc-tools is a straightforward package and I have sent the Makefile
> patch upstream.
>
> Not sure why it can be hard to integrate or maintain it.
ok, maybe my statement was a little bit opinion based. It is likely
that the OE/Yocto maintainers have a different position to this.

I remember that I've read an article (but cannot find it again) that
custom build systems regularly causes headaches for distributors
(because they usually use custom/no ways for setting compiler switches
and installation directories which normally the distributor want to
control).

Well known systems like autotools (I don't fully understand them) or
cmake (I use it) have a standardized interface for setting
important switches. Additionally they provide common features like
cross-compiling, platform specific stuff, removing build artifacts
(clean), installation of generated files or out-of-tree builds.

Doing the same in bare Makefiles is possible, but requires much hand-crafted
code. Often Makefiles consist of a lot of repetitive code. You can use
macros for this. After putting everything into macros, you will have
reinvented either cmake or kbuild.

>
> > What about using CMake instead?
>
> I can do that if needed, but it is up to the rtc-tools maintainer, Alexandre.
I am not involved into rtc-tools and also not much into OE/Yocto.
So don't count too much on my opinion and choose the best solution
yourself.

regards
Christia
(having created hand-written Makefiles for 13 years)




-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#160896): 
https://lists.openembedded.org/g/openembedded-core/message/160896
Mute This Topic: https://lists.openembedded.org/mt/88625470/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] [PATCH meta-oe] rtc-tool: Add a recipe

[Christian Eggers]

Hi Fabio,

from my experience, custom build systems (including "bare" Makefiles)
are often hard to integrate/maintain for distributors.

What about using CMake instead?

regards
Christian

On Sunday, 23 January 2022, 16:03:58 CET, Fabio Estevam wrote:
> From: Fabio Estevam 
>
> rtc-tools is a useful utility developed by Alexandre Belloni
> for testing RTC kernel drivers, such as y2038 support.
>
> Based on the initial recipe from Heiko Schocher .
>
> Signed-off-by: Fabio Estevam 
> ---
>  .../0001-rtc-tools-Add-a-Makefile.patch   | 52 +++
>  .../rtc-tools/rtc-tools_1.0.0.bb  | 19 +++
>  2 files changed, 71 insertions(+)
>  create mode 100644 
> meta-oe/recipes-support/rtc-tools/rtc-tools/0001-rtc-tools-Add-a-Makefile.patch
>  create mode 100644 meta-oe/recipes-support/rtc-tools/rtc-tools_1.0.0.bb
>
> diff --git 
> a/meta-oe/recipes-support/rtc-tools/rtc-tools/0001-rtc-tools-Add-a-Makefile.patch
>  
> b/meta-oe/recipes-support/rtc-tools/rtc-tools/0001-rtc-tools-Add-a-Makefile.patch
> new file mode 100644
> index ..a372e390efb9
> --- /dev/null
> +++ 
> b/meta-oe/recipes-support/rtc-tools/rtc-tools/0001-rtc-tools-Add-a-Makefile.patch
> @@ -0,0 +1,52 @@
> +From faa6dd2e4d063afcaa19f16740a3fc413a399e60 Mon Sep 17 00:00:00 2001
> +
___
Christian
Eggers
Software Engineer
​
ARRI
Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Arriweg 17,
83071
Stephanskirchen
www.arri.com

+49 8036 3009-3118
cegg...@arri.de

​
Get all the latest information from www.arri.com, Facebook, Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

​
From: Fabio Estevam 
> +Date: Sun, 23 Jan 2022 09:22:19 -0300
> +Subject: [PATCH rtc-tools] rtc-tools: Add a Makefile
> +
> +Add a Makefile to make installation and uninstallation
> +process easier.
> +
> +Upstream-Status: Submitted 
> [https://marc.info/?l=linux-rtc=164294073129803=2]
> +Signed-off-by: Fabio Estevam 
> +---
> + Makefile | 29 +
> + 1 file changed, 29 insertions(+)
> + create mode 100644 Makefile
> +
> +diff --git a/Makefile b/Makefile
> +new file mode 100644
> +index ..4f244d9289f1
> +--- /dev/null
>  b/Makefile
> +@@ -0,0 +1,29 @@
> ++prefix ?= /usr
> ++bindir ?= $(prefix)/bin
> ++
> ++EXEC = rtc-range rtc rtc-sync
> ++
> ++all: $(EXEC)
> ++
> ++rtc-range: rtc-range.c
> ++$(CC) $(CFLAGS) -o $@ rtc-range.c
> ++
> ++rtc: rtc.c
> ++$(CC) $(CFLAGS) -o $@ rtc.c
> ++
> ++rtc-sync: rtc-sync.c
> ++$(CC) $(CFLAGS) -o $@ rtc-sync.c
> ++
> ++clean:
> ++rm -f $(EXEC)
> ++
> ++install: rtc-range rtc rtc-sync
> ++mkdir -p '$(DESTDIR)$(bindir)'
> ++install -m755 rtc-range '$(DESTDIR)$(bindir)/rtc-range'
> ++install -m755 rtc '$(DESTDIR)$(bindir)/rtc'
> ++install -m755 rtc-sync '$(DESTDIR)$(bindir)/rtc-sync'
> ++
> ++uninstall:
> ++rm -rf '$(DESTDIR)$(bindir)/rtc-range'
> ++rm -rf '$(DESTDIR)$(bindir)/rtc'
> ++rm -rf '$(DESTDIR)$(bindir)/rtc-sync'
> +--
> +2.25.1
> diff --git a/meta-oe/recipes-support/rtc-tools/rtc-tools_1.0.0.bb 
> b/meta-oe/recipes-support/rtc-tools/rtc-tools_1.0.0.bb
> new file mode 100644
> index ..8f0326829d8c
> --- /dev/null
> +++ b/meta-oe/recipes-support/rtc-tools/rtc-tools_1.0.0.bb
> @@ -0,0 +1,19 @@
> +DESCRIPTION = "Useful programs to test rtc drivers"
> +LICENSE = "GPLv2"
> +LIC_FILES_CHKSUM = "file://COPYING;md5=74274e8a218423e49eefdea80bc55038"
> +
> +SRC_URI = 
> "git://git.kernel.org/pub/scm/linux/kernel/git/abelloni/rtc-tools.git;protocol=https;branch=master
>  \
> +   file://0001-rtc-tools-Add-a-Makefile.patch \
> +   "
> +SRCREV ?= "acc442e7af4e1e783432a43d37f1a7938c692659"
> +
> +S = "${WORKDIR}/git"
> +
> +TARGET_CC_ARCH += "${LDFLAGS}"
> +EXTRA_OEMAKE = " \
> + DESTDIR=${D} \
> + all \
> + "
> +do_install() {
> + oe_runmake install
> +}
>





-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#160873): 
https://lists.openembedded.org/g/openembedded-core/message/160873
Mute This Topic: https://lists.openembedded.org/mt/88625470/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-Core][dunfell][PATCH] binutils: Fix a missing break in case statement

[Christian Eggers]

This was missed during patch forward porting
its only effective when printing options

Signed-off-by: Khem Raj 
Signed-off-by: Richard Purdie 
---
 ...-system-directories-when-cross-linki.patch | 26 +--
 1 file changed, 13 insertions(+), 13 deletions(-)

diff --git 
a/meta/recipes-devtools/binutils/binutils/0009-warn-for-uses-of-system-directories-when-cross-linki.patch
 
b/meta/recipes-devtools/binutils/binutils/0009-warn-for-uses-of-system-directories-when-cross-linki.patch
index 11a8110d40..88cce49e46 100644
--- 
a/meta/recipes-devtools/binutils/binutils/0009-warn-for-uses-of-system-directories-when-cross-linki.patch
+++ 
b/meta/recipes-devtools/binutils/binutils/0009-warn-for-uses-of-system-directories-when-cross-linki.patch
@@ -1,4 +1,4 @@
-From 7b24f81e04c9d00d96de7dbd250beade6d2c6e44 Mon Sep 17 00:00:00 2001
+From 12b658c0fe5771d16067baef933b7f34ed455def Mon Sep 17 00:00:00 2001
 From: Khem Raj 
 Date: Fri, 15 Jan 2016 06:31:09 +
 Subject: [PATCH] warn for uses of system directories when cross linking
@@ -59,8 +59,8 @@ Signed-off-by: Khem Raj 
  ld/ldfile.c | 17 +
  ld/ldlex.h  |  2 ++
  ld/ldmain.c |  2 ++
- ld/lexsup.c | 15 +++
- 9 files changed, 85 insertions(+)
+ ld/lexsup.c | 16 
+ 9 files changed, 86 insertions(+)
 
 diff --git a/ld/config.in b/ld/config.in
 index d93c9b0830..5da2742bea 100644
@@ -77,10 +77,10 @@ index d93c9b0830..5da2742bea 100644
  #undef EXTRA_SHLIB_EXTENSION
  
 diff --git a/ld/configure b/ld/configure
-index 811134a503..f8c17c19ae 100755
+index f432f4637d..a9da3c115e 100755
 --- a/ld/configure
 +++ b/ld/configure
-@@ -826,6 +826,7 @@ with_lib_path
+@@ -830,6 +830,7 @@ with_lib_path
  enable_targets
  enable_64_bit_bfd
  with_sysroot
@@ -88,7 +88,7 @@ index 811134a503..f8c17c19ae 100755
  enable_gold
  enable_got
  enable_compressed_debug_sections
-@@ -1491,6 +1492,8 @@ Optional Features:
+@@ -1495,6 +1496,8 @@ Optional Features:
--disable-largefile omit support for large files
--enable-targetsalternative target configurations
--enable-64-bit-bfd 64-bit support (on hosts with narrower word sizes)
@@ -97,7 +97,7 @@ index 811134a503..f8c17c19ae 100755
--enable-gold[=ARG] build gold [ARG={default,yes,no}]
--enable-got= GOT handling scheme (target, single, negative,
multigot)
-@@ -15788,6 +15791,19 @@ fi
+@@ -16624,6 +16627,19 @@ fi
  
  
  
@@ -222,10 +222,10 @@ index 5287f19a7f..55096e4fc9 100644
  
  /* The initial parser states.  */
 diff --git a/ld/ldmain.c b/ld/ldmain.c
-index da1ad17763..12d0b07d8a 100644
+index c4af10f4e9..95b56b2d2d 100644
 --- a/ld/ldmain.c
 +++ b/ld/ldmain.c
-@@ -274,6 +274,8 @@ main (int argc, char **argv)
+@@ -273,6 +273,8 @@ main (int argc, char **argv)
command_line.warn_mismatch = TRUE;
command_line.warn_search_mismatch = TRUE;
command_line.check_section_addresses = -1;
@@ -235,7 +235,7 @@ index da1ad17763..12d0b07d8a 100644
/* We initialize DEMANGLING based on the environment variable
   COLLECT_NO_DEMANGLE.  The gcc collect2 program will demangle the
 diff --git a/ld/lexsup.c b/ld/lexsup.c
-index 3d15cc491d..0e8b4f2b7a 100644
+index 3d15cc491d..6478821443 100644
 --- a/ld/lexsup.c
 +++ b/ld/lexsup.c
 @@ -550,6 +550,14 @@ static const struct ld_option ld_options[] =
@@ -253,10 +253,10 @@ index 3d15cc491d..0e8b4f2b7a 100644
  };
  
  #define OPTION_COUNT ARRAY_SIZE (ld_options)
-@@ -1603,6 +1611,13 @@ parse_args (unsigned argc, char **argv)
- 
+@@ -1604,6 +1612,14 @@ parse_args (unsigned argc, char **argv)
case OPTION_PRINT_MAP_DISCARDED:
  config.print_map_discarded = TRUE;
+ break;
 +
 +  case OPTION_NO_POISON_SYSTEM_DIRECTORIES:
 +command_line.poison_system_directories = FALSE;
@@ -264,6 +264,6 @@ index 3d15cc491d..0e8b4f2b7a 100644
 +
 +  case OPTION_ERROR_POISON_SYSTEM_DIRECTORIES:
 +command_line.error_poison_system_directories = TRUE;
- break;
++break;
}
  }
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#156784): 
https://lists.openembedded.org/g/openembedded-core/message/156784
Mute This Topic: https://lists.openembedded.org/mt/86180878/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] [SDK] How to use the gold linker

[Christian Eggers]

Hi Khem, hi Christopher,

On Tuesday, 15 December 2020, 18:15:52 CET, Khem Raj wrote:
>
> On 12/15/20 8:50 AM, Christopher Larson wrote:
> > This can be fixed via two changes, first removing real-ld, then adding
> > the missing links to fix -fuse-ld=. I have this in one of my
> > not-yet-submitted bbappends:
> >
> >  # Remove real-ld, as it prevents checking for ld or gold in the
> > search path
> >  rm -f ${dest}real-ld$suffix
> >  # Then add links to fix builds with -fuse-ld=
> >  ln -sf ${BINRELPATH}/${TARGET_PREFIX}ld.gold$suffix
> > ${dest}ld.gold$suffix
> >  ln -sf ${BINRELPATH}/${TARGET_PREFIX}ld.bfd$suffix ${dest}ld.bfd$suffix
> >
>
> right this looks ok. Can you submit this for upstreaming please

while this change works "fine for me" (I can now use ld.gold which is about
3 times faster for my project), there are side effects:

> ${CC} -fuse-ld=bfd -print-prog-name=ld
/opt/poky-tiny/3.1.4/sysroots/x86_64-pokysdk-linux/usr/libexec/arm-poky-linux-gnueabi/gcc/arm-poky-linux-gnueabi/9.3.0/ld.bfd

> ${CC} -fuse-ld=bfd -o test test.c
works fine

> ${CC} -fuse-ld=gold -print-prog-name=ld
/opt/poky-tiny/3.1.4/sysroots/x86_64-pokysdk-linux/usr/libexec/arm-poky-linux-gnueabi/gcc/arm-poky-linux-gnueabi/9.3.0/ld.gold

> ${CC} -fuse-ld=gold -o test test.c
also works fine

> ${CC} print-prog-name=ld
/opt/poky-tiny/3.1.4/sysroots/x86_64-pokysdk-linux/usr/libexec/arm-poky-linux-gnueabi/gcc/arm-poky-linux-gnueabi/9.3.0/ld

> ${CC} -o test test.c
collect2: fatal error: cannot find ‘ld’
compilation terminated.

> ls -l 
> /opt/poky-tiny/3.1.4/sysroots/x86_64-pokysdk-linux/usr/libexec/arm-poky-linux-gnueabi/gcc/arm-poky-linux-gnueabi/9.3.0/ld
/opt/poky-tiny/3.1.4/sysroots/x86_64-pokysdk-linux/usr/libexec/arm-poky-linux-gnueabi/gcc/arm-poky-linux-gnueabi/9.3.0/ld
 -> ../../../../../bin/arm-poky-linux-gnueabi/arm-poky-linux-gnueabi-ld



regards
Christian





 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 


Get all the latest information from www.arri.com, 
Facebook, 
Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#145858): 
https://lists.openembedded.org/g/openembedded-core/message/145858
Mute This Topic: https://lists.openembedded.org/mt/78948256/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [SDK] How to use the gold linker

[Christian Eggers]

Branch: dunfell
Arch: arm

When I set -fuse-ld=gold with the Yocto SDK, the bfd linker is used anyway:

> . 
> /opt/poky-tiny/3.1.4/environment-setup-cortexa7t2hf-neon-vfpv4-poky-linux-musleabi
> ${CC} -fuse-ld=gold -Wl,--threads -o test test.c
/opt/poky-tiny/3.1.4/sysroots/x86_64-pokysdk-linux/usr/libexec/arm-poky-linux-gnueabi/gcc/arm-poky-linux-gnueabi/9.3.0/real-ld:
 unrecognized option '--threads'
/opt/poky-tiny/3.1.4/sysroots/x86_64-pokysdk-linux/usr/libexec/arm-poky-linux-gnueabi/gcc/arm-poky-linux-gnueabi/9.3.0/real-ld:
 use the --help option for usage information

When I call the "real-ld" with -h, I can see that it's the BFD linker.

> ${CC} -print-prog-name=ld
/opt/poky-tiny/3.1.4/sysroots/x86_64-pokysdk-linux/usr/libexec/arm-poky-linux-gnueabi/gcc/arm-poky-linux-gnueabi/9.3.0/ld

> ${CC} -fuse-ld=bfd -print-prog-name=ld
ld.bfd

> ${CC} -fuse-ld=gold -print-prog-name=ld
ld.gold

Compared to other cross toolchains I use, the output of the previous two 
commands
miss the full path to the linker. The gold linker is part of my SDK:

> arm-poky-linux-musleabi-ld.gold --help
Usage: arm-poky-linux-musleabi-ld.gold [options] file...
Options:
  --help  Report usage information
...

How can I use the gold linker from gcc (without making it default)?

regards
Christian




 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 


Get all the latest information from www.arri.com, 
Facebook, 
Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#145548): 
https://lists.openembedded.org/g/openembedded-core/message/145548
Mute This Topic: https://lists.openembedded.org/mt/78948256/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH v2] packagegroup: rrecommend perf also for musl on ARM

[Christian Eggers]

The perf recipe builds fine for musl on ARM.

Fixes: d758a4445a ("packagegroup: Disable packages not available on musl")
Signed-off-by: Christian Eggers 
---
On Wednesday, 23 September 2020, 18:38:06 CEST, Khem Raj wrote:
> On Wed, Sep 23, 2020 at 1:46 AM Christian Eggers  wrote:
> > The perf recipe builds fine for musl.
> 
> which all architectures did you build on musl ?
only on ARM.

> Regardless of that we should still use PERF var so non-core
> architectures can still override it

 .../packagegroups/packagegroup-core-tools-profile.bb | 1 +
 1 file changed, 1 insertion(+)

diff --git a/meta/recipes-core/packagegroups/packagegroup-core-tools-profile.bb 
b/meta/recipes-core/packagegroups/packagegroup-core-tools-profile.bb
index 1c94653d72..d437e28311 100644
--- a/meta/recipes-core/packagegroups/packagegroup-core-tools-profile.bb
+++ b/meta/recipes-core/packagegroups/packagegroup-core-tools-profile.bb
@@ -28,6 +28,7 @@ PROFILETOOLS = "\
 "
 PERF = "perf"
 PERF_libc-musl = ""
+PERF_libc-musl_arm = "perf"
 
 # systemtap needs elfutils which is not fully buildable on some arches/libcs
 SYSTEMTAP = "systemtap"
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#142755): 
https://lists.openembedded.org/g/openembedded-core/message/142755
Mute This Topic: https://lists.openembedded.org/mt/77053925/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH] packagegroup: rrecommend perf also for musl

[Christian Eggers]

The perf recipe builds fine for musl.

Fixes: d758a4445a ("packagegroup: Disable packages not available on musl")
Signed-off-by: Christian Eggers 
---
 .../packagegroups/packagegroup-core-tools-profile.bb  | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/meta/recipes-core/packagegroups/packagegroup-core-tools-profile.bb 
b/meta/recipes-core/packagegroups/packagegroup-core-tools-profile.bb
index 1c94653d72..a5dca95c2a 100644
--- a/meta/recipes-core/packagegroups/packagegroup-core-tools-profile.bb
+++ b/meta/recipes-core/packagegroups/packagegroup-core-tools-profile.bb
@@ -17,7 +17,7 @@ PROFILE_TOOLS_X_nios2 = ""
 PROFILE_TOOLS_SYSTEMD = "${@bb.utils.contains('DISTRO_FEATURES', 'systemd', 
'systemd-analyze', '', d)}"
 
 RRECOMMENDS_${PN} = "\
-${PERF} \
+perf \
 blktrace \
 ${PROFILE_TOOLS_X} \
 ${PROFILE_TOOLS_SYSTEMD} \
@@ -26,8 +26,6 @@ RRECOMMENDS_${PN} = "\
 PROFILETOOLS = "\
 powertop \
 "
-PERF = "perf"
-PERF_libc-musl = ""
 
 # systemtap needs elfutils which is not fully buildable on some arches/libcs
 SYSTEMTAP = "systemtap"
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#142729): 
https://lists.openembedded.org/g/openembedded-core/message/142729
Mute This Topic: https://lists.openembedded.org/mt/77030965/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] gcc-runtime: Debug sources not packaged

[Christian Eggers]

It seems that there are no debug sources available for libstdc++. When
debugging with GDB, I get the following message:

> Can't find a source file at "/opt/orbiter-tiny/3.1.1/sysroots/cortexa7t2hf-
> neon-vfpv4-poky-linux-musleabi/usr/src/debug/gcc-runtime/9.3.0-r0/arm-poky-
> linux-musleabi/libstdc++-v3/libsupc++/../../../../../../../../../work-shared/
> gcc-9.3.0-r0/gcc-9.3.0/libstdc++-v3/libsupc++/eh_alloc.cc"
> Locate the file or edit the source lookup path to include its location.

Debug sources for all other libraries are available.

The directory work/.../gcc-runtime/.../packages-split/gcc-runtime-src
is empty and so there is no gcc-runtime-src ipk generated. It looks like
packaging debug sources does not work as the gcc sources are extracted
to the work-shared/ directory instead of work/

The gcc-runtime recipe creates special -fdebug-prefix-map options. It
looks like these settings make it impossible copy the debug sources
during packaging.

Any ideas how to generate a gcc-runtime-src package?

Best regards
Christian




 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 


Get all the latest information from www.arri.com, 
Facebook, 
Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#141285): 
https://lists.openembedded.org/g/openembedded-core/message/141285
Mute This Topic: https://lists.openembedded.org/mt/76044807/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] archiver.bbclass and GIT repositories

[Christian Eggers]

On Wednesday, 29 July 2020, 11:42:25 CEST, Paul Barker wrote:
> On Tue, 28 Jul 2020 at 13:56, Christian Eggers  wrote:
> > ...
> > So I have three questions:
> > 1. Is it possible to archive the git project in a way, that the GIT
> > history
> > can be examined later?
> > 2. If this is not possible, I would have to put my changes in an
> > "external" patch file alongside the bitbake recipe, right?
> > 3. Is there any value archiving the .git directory at all?
>
> What configuration are you using to enable the archiver? There may be
> ways to fix this by changing the configuration.
I use the archive as described here:
https://www.yoctoproject.org/docs/3.1.1/dev-manual/dev-manual.html#providing-the-source-code

>From my local.conf:
...
INHERIT += "archiver"
ARCHIVER_MODE[src] = "original"
...


>
> Also, which Yocto Project version/branch are you using?
I use Yocto 3.1.1.

regards
Christian




 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 
<https://microsites.arri.com/stellar/>

Get all the latest information from www.arri.com<https://www.arri.com/>, 
Facebook<https://www.facebook.com/TeamARRI>, 
Twitter<https://twitter.com/ARRIChannel>, Instagram<https://instagram.com/arri> 
and YouTube<https://www.youtube.com/user/ARRIChannel>.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#141083): 
https://lists.openembedded.org/g/openembedded-core/message/141083
Mute This Topic: https://lists.openembedded.org/mt/75843071/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] archiver.bbclass and GIT repositories

[Christian Eggers]

In my project, the linux kernel (together with dozens of our changes) is kept
in a company internal GIT repository. The GPLv2 requires the following:

> 2a) You must cause the modified files to carry prominent notices stating
> that you changed the files and the date of any change.

My idea is, that the GIT history (and diff) should be sufficient for documenting
our changes to the linux source code.

But when I unpack the .tar.gz file which was deploy by archiver.bbclass,
I get the following:

# tar -xf my-linux-kernel.tar.gz
# cd /git
# git log

> error: object directory 
> /srv/gitlab-runner/downloads/[..]/yocto/git2/[xxx].git/objects does not 
> exist; check .git/objects/info/alternates
> fatal: bad object HEAD

Note: The project has been built on another machine.

>From the bitbake user manual, 4.3.5, Git Fetcher:
> "This bare clone is then cloned into the work directory during the unpack
> stage when a specific tree is checked out. This is done using alternates and
> by reference to minimize the amount of duplicate data on the disk and make
> the unpack process fast."

It looks like the .git directory within the tar archive doesn't contain
the necessary information required for using the "git" command on
the extracted "working copy". If I understood the bitbake doc correctly,
the working copy is cloned "by reference" from a "bare" clone which
resided outside the archived directory (in the downloads dir).

So I have three questions:
1. Is it possible to archive the git project in a way, that the GIT history
can be examined later?
2. If this is not possible, I would have to put my changes in an
"external" patch file alongside the bitbake recipe, right?
3. Is there any value archiving the .git directory at all?

regards
Christian




 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 


Get all the latest information from www.arri.com, 
Facebook, 
Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#141061): 
https://lists.openembedded.org/g/openembedded-core/message/141061
Mute This Topic: https://lists.openembedded.org/mt/75843071/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] How to install individual -lic packages

[Christian Eggers]

In the meantime I could answer the question myself:

Am Freitag, 10. Juli 2020, 23:16:39 CEST schrieb Christian Eggers:
> [...]
> I think this is not what I require. How can I install -lic packages
> directly?

In order to known by other (image) recipes, -lic packages must be
added to PACKAGES_DYNAMIC. This is done in my distro.conf but should also work
from local.conf:

# Create separate ${PN}-lic packages for all recipes
LICENSE_CREATE_PACKAGE = "1"
# -lic packages are created dynamically. Satisfy dependencies to it (required
# if a -lic package needs to be installed directly (e.g. via IMAGE_INSTALL)
# instead of being installed due to RRECOMMENDS_${PN}).
PACKAGES_DYNAMIC += "^${PN}-lic"

After adding ${PN}-lic to PACKAGES_DYNAMIC, arbitrary -lic packages
can be added to an image recipe by using IMAGE_INSTALL. As an alternative,
other packages can RDEPEND on them.

regards
Christian




 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 
<https://microsites.arri.com/stellar/>

Get all the latest information from www.arri.com<https://www.arri.com/>, 
Facebook<https://www.facebook.com/TeamARRI>, 
Twitter<https://twitter.com/ARRIChannel>, Instagram<https://instagram.com/arri> 
and YouTube<https://www.youtube.com/user/ARRIChannel>.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#140616): 
https://lists.openembedded.org/g/openembedded-core/message/140616
Mute This Topic: https://lists.openembedded.org/mt/75394824/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] How to install individual -lic packages

[Christian Eggers]

Hi Christopher,

Am Donnerstag, 9. Juli 2020, 18:04:00 CEST schrieb Christopher Larson:
> COMPLEMENTARY_GLOB[lic-pkgs] = '*-lic'
> IMAGE_FEATURES += "lic-pkgs"
>
> This is how the dbg-pkgs, dev-pkgs, doc-pkgs, src-pkgs, ptest-pkgs,
> staticdev-pkgs are done.

I think this is not what I require. How can I install -lic packages directly?

I would like to have all necessary -lic packages in my rootfs, so that I
can publish the licenses via a webserver. I need some extra -lic packages
which are not installed automatically because the main package is not
part of the rootfs.

Example 1: My initramfs used dash from klibc. This code is licensed under
BSD-3-Clause, so I need the copyright disclaimer of this package. How
can I add the license of this package to my rootfs?

Example 2: My application depends on libuuid from util-linux. It looks
like the -lic package is not added to RRECOMMENDS_libuuid because
package splitting is done dynamically for util-linux. So I need a way to
install util-linux-lic manually.

regards
Christian





 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 


Get all the latest information from www.arri.com, 
Facebook, 
Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#140545): 
https://lists.openembedded.org/g/openembedded-core/message/140545
Mute This Topic: https://lists.openembedded.org/mt/75394824/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] [PATCH] libgcc: Set LICENSE for -lic package

[Christian Eggers]

Hi Quentin,

Am Donnerstag, 9. Juli 2020, 15:25:00 CEST schrieb Quentin Schulz:
> Not really directing the question to you but... does this actually makes
> sense? Is the license licensed? Can't we just empty LICENSE_${PN}-lic or
> not check the license for this package? That might be something that can
> be done in the license.bbclass for example.

I already asked myself what license the -lic package should have. Is this a
license of it's own (or none)?

>From the GPLv2:

"Copyright (C) 1989, 1991 Free Software Foundation, Inc.
51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA

Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed."

Other licenses like MIT have no such restrictions, so that they
are often forked. But setting an individual license for each -lic paket may
also complicate things.

The patch I sent, "works for me". But I'll be also happy with another
solution...

regards
Christian




 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 


Get all the latest information from www.arri.com, 
Facebook, 
Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#140483): 
https://lists.openembedded.org/g/openembedded-core/message/140483
Mute This Topic: https://lists.openembedded.org/mt/75396771/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH] libgcc: Set LICENSE for -lic package

[Christian Eggers]

The default license (GPL-3.0-with-GCC-exception & GPLv3) is not suitable
for installing libgcc-lic on an image where INCOMPATIBLE_LICENSE is set
to "GPL-3.0".

Signed-off-by: Christian Eggers 
---
 meta/recipes-devtools/gcc/libgcc.inc | 1 +
 1 file changed, 1 insertion(+)

diff --git a/meta/recipes-devtools/gcc/libgcc.inc 
b/meta/recipes-devtools/gcc/libgcc.inc
index 6d48ec9809..372da46459 100644
--- a/meta/recipes-devtools/gcc/libgcc.inc
+++ b/meta/recipes-devtools/gcc/libgcc.inc
@@ -35,6 +35,7 @@ PACKAGES = "\
 LICENSE_${PN} = "GPL-3.0-with-GCC-exception"
 LICENSE_${PN}-dev = "GPL-3.0-with-GCC-exception"
 LICENSE_${PN}-dbg = "GPL-3.0-with-GCC-exception"
+LICENSE_${PN}-lic = "GPL-3.0-with-GCC-exception"
 
 
 FILES_${PN}-dev = "\
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#140481): 
https://lists.openembedded.org/g/openembedded-core/message/140481
Mute This Topic: https://lists.openembedded.org/mt/75396771/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] How to install individual -lic packages

[Christian Eggers]

I have set LICENSE_CREATE_PACKAGE to "1" in order to create additional -lic
packages per recipe. As NO_RECOMMENDATIONS is set to "1" in my image recipe,
the -lic packages are not installed automatically.

How can I install specific -lic packages "manually". My idea was to create a
new recipe which rdepends on all -lic packages which I would like to install.

# myrecipe.bb:
RDEPENDS_${PN} += "curl-lic"

After that I get the following error message:
ERROR: Nothing RPROVIDES 'curl-lic' (but ... RDEPENDS on or otherwise requires
it)

The same happened when I tried to add the -lic packages to IMAGE_INSTALL in my
image recipe.

I also tried adding one line in image.bbclass, but this changed nothing:

def add_package_and_files(d):
...
d.setVar('PACKAGES', "%s %s" % (pn_lic, packages))
d.setVar('FILES_' + pn_lic, files)
+   d.setVar('RPROVIDES_' + pn_lic, pn_lic)


How can I install -lic packages in my rootfs without setting
NO_RECOMMENDATIONS to 0?

regards
Christian





 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 


Get all the latest information from www.arri.com, 
Facebook, 
Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#140476): 
https://lists.openembedded.org/g/openembedded-core/message/140476
Mute This Topic: https://lists.openembedded.org/mt/75394824/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] license.manifest misses header-only libraries

[Christian Eggers]

Am Mittwoch, 8. Juli 2020, 18:19:24 CEST schrieb Richard Purdie:
> On Wed, 2020-07-08 at 17:48 +0200, Christian Eggers wrote:
> > [...]
> > Is there a reliable way to get the header-only/static libraries into
> > the manifest file?
>
> The image manifest is a list of the things actually shipping in the
> image and their corresponding licenses. Most images don't ship
> development headers so they'll not be included there.
I'm more focusing on the (empty) main packages, not the -dev
packages with the header files.

> You can certainly get a list of everything used during the build
> instead. I'm not sure how you'd decide that a set of headers were used
> or not, compared to say the cross toolchain binaries or a native tool
> though so anything in between these two is harder.
I suppose that the "archiver" class already tracks everything which has been
used for building. But I'm only looking for licenses which are distributed
with my binary embedded software

> There is some work I did a while back which traces all SPDX identifiers
> referenced by debug symbols in the code to better audit license
> information but the lack of SPDX headers in much of the code we use
> makes that less useful right now.
>
> http://git.yoctoproject.org/cgit.cgi/poky-contrib/log/?h=rpurdie/license-exp
> eriments-osls (and a presentation/slides at OSLS 2019)
Thanks for this hint. But I think that this is not exactly what I'm locking
for. For now I'll stay at manually rdepending on the empty main packages
of header only libraries for getting them into the license manifest.
For the case of the boost library, I'll probably "invent" some additional
empty sub packages for tracking the license of the boost components
I use.

At the end I'll anyway publish everything collected by the archiver class on
our download site. But for the printed documentation of our product I would
like to focus on licenses which apply for the "binary distribution".

regards
Christian




 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 
<https://microsites.arri.com/stellar/>

Get all the latest information from www.arri.com<https://www.arri.com/>, 
Facebook<https://www.facebook.com/TeamARRI>, 
Twitter<https://twitter.com/ARRIChannel>, Instagram<https://instagram.com/arri> 
and YouTube<https://www.youtube.com/user/ARRIChannel>.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#140473): 
https://lists.openembedded.org/g/openembedded-core/message/140473
Mute This Topic: https://lists.openembedded.org/mt/75379155/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] license.manifest misses header-only libraries

[Christian Eggers]

According to [1], a license manifest is generated during image creation:

"The Yocto Project generates a license manifest during image creation that is 
located in ${DEPLOY_DIR}/licenses/image_name-datestamp to assist with any 
audits."

This manifest lists all packages installed on the target (either being installed
directly via IMAGE_INSTALL or due to other packages RDEPENDS).

The manifest seems to miss header-only libraries (like tclap or big parts
of boost), which were used during compilation (headers) but do not
contain dynamic libraries required at runtime. I guess that the same
may be true for libraries linked statically.

Setting RDEPENDS_${PN} += "tclap" in my own applications recipe
seems to improve the situation as now the (empty) package tclap
is included in my image and in the license.manifest. But this requires
manual work and cannot do this for all recipes shipped by OE.
Additionally this doesn't work for boost, as rdepending on boost
will pull in ALL boost library packages.

Is there a reliable way to get the header-only/static libraries into
the manifest file?

regards
Christian

[1] 
https://www.yoctoproject.org/docs/3.1.1/dev-manual/dev-manual.html#maintaining-open-source-license-compliance-during-your-products-lifecycle



 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 


Get all the latest information from www.arri.com, 
Facebook, 
Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#140424): 
https://lists.openembedded.org/g/openembedded-core/message/140424
Mute This Topic: https://lists.openembedded.org/mt/75379155/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] Patches are stuck in master-next

[Christian Eggers]

Am Mittwoch, 8. Juli 2020, 16:20:03 CEST schrieb Adrian Bunk:
> I would rather ask why you aren't running automatic unit tests
> built for the target inside qemu.
>
> cu
> Adrian

This may be an option. Currently the platform (x86 or arm)
decides whether real peripheral routines or mock objects
will be used. Of course this can easily be changed. But
running the tests within a (slow) emulator brings not much
benefit compared to building and running the tests directly
on x86.

Instead of using NATIVE, also the NATIVESDK build could
be executed within qemu on the build machine... (without
qemu, nativesdk executables cannot be executed on the
build machine due to non existing loader in .interp section).

>From Richard I got no response to the question whether
potentially every package may be extended for nativesdk/native,
or these platforms a reserved for packages required for "building".

regards
Christian



 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 


Get all the latest information from www.arri.com, 
Facebook, 
Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#140422): 
https://lists.openembedded.org/g/openembedded-core/message/140422
Mute This Topic: https://lists.openembedded.org/mt/75371450/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH] util-linux: Set license for library sub packages

[Christian Eggers]

- libblkid is exclusively licensed under LGPL-2.1+ (libblkid/COPYING)
- libfdisk is exclusively licensed under LGPL-2.1+ (libfdisk/COPYING)
- libmount is exclusively licensed under LGPL-2.1+ (libmount/COPYING)
- libsmartcols is exclusively licensed under LGPL-2.1+
(libsmartcols/COPYING)
- libuuid is exclusively licensed under BSD-3-Clause (libuuid/COPYING)

Signed-off-by: Christian Eggers 
---
 meta/recipes-core/util-linux/util-linux.inc | 10 +-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/meta/recipes-core/util-linux/util-linux.inc 
b/meta/recipes-core/util-linux/util-linux.inc
index 532cceb935..06fbf7a229 100644
--- a/meta/recipes-core/util-linux/util-linux.inc
+++ b/meta/recipes-core/util-linux/util-linux.inc
@@ -7,6 +7,11 @@ disk partitioning, kernel message management, filesystem 
creation, and system lo
 SECTION = "base"
 
 LICENSE = "GPLv2+ & LGPLv2.1+ & BSD-3-Clause & BSD-4-Clause"
+LICENSE_${PN}-libblkid = "LGPLv2.1+"
+LICENSE_${PN}-libfdisk = "LGPLv2.1+"
+LICENSE_${PN}-libmount = "LGPLv2.1+"
+LICENSE_${PN}-libsmartcols = "LGPLv2.1+"
+LICENSE_${PN}-libuuid = "BSD-3-Clause"
 
 LIC_FILES_CHKSUM = 
"file://README.licensing;md5=0fd5c050c6187d2bf0a4492b7f4e33da \
 file://COPYING;md5=b234ee4d69f5fce4486a80fdaf4a4263 \
@@ -16,7 +21,10 @@ LIC_FILES_CHKSUM = 
"file://README.licensing;md5=0fd5c050c6187d2bf0a4492b7f4e33da
 
file://Documentation/licenses/COPYING.BSD-4-Clause-UC;md5=263860f8968d8bafa5392cab74285262
 \
 
file://libuuid/COPYING;md5=6d2cafc999feb2c2de84d4d24b23290c \
 
file://libmount/COPYING;md5=7c7e39fb7d70ffe5d693a643e29987c2 \
-
file://libblkid/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04"
+
file://libblkid/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \
+
file://libfdisk/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \
+
file://libsmartcols/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \
+"
 
 #gtk-doc is not enabled as it requires xmlto which requires util-linux
 inherit autotools gettext manpages pkgconfig systemd update-alternatives 
python3-dir bash-completion ptest
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#140420): 
https://lists.openembedded.org/g/openembedded-core/message/140420
Mute This Topic: https://lists.openembedded.org/mt/75375643/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH] avahi: Fix typo in recipe

[Christian Eggers]

According to the PACKAGES variable, LICENSE_avahi-client is misspelled.
Additionally, the libavahi-client package actually only includes
LGPLv2.1+ software (as opposed to the global LICENSE variable).


Signed-off-by: Christian Eggers 
---
 meta/recipes-connectivity/avahi/avahi_0.8.bb | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/recipes-connectivity/avahi/avahi_0.8.bb 
b/meta/recipes-connectivity/avahi/avahi_0.8.bb
index 35d0c3ea75..c8a3f876aa 100644
--- a/meta/recipes-connectivity/avahi/avahi_0.8.bb
+++ b/meta/recipes-connectivity/avahi/avahi_0.8.bb
@@ -108,7 +108,7 @@ LICENSE_libavahi-gobject = "LGPLv2.1+"
 LICENSE_avahi-daemon = "LGPLv2.1+"
 LICENSE_libavahi-common = "LGPLv2.1+"
 LICENSE_libavahi-core = "LGPLv2.1+"
-LICENSE_avahi-client = "LGPLv2.1+"
+LICENSE_libavahi-client = "LGPLv2.1+"
 LICENSE_avahi-dnsconfd = "LGPLv2.1+"
 LICENSE_libavahi-glib = "LGPLv2.1+"
 LICENSE_avahi-autoipd = "LGPLv2.1+"
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#140416): 
https://lists.openembedded.org/g/openembedded-core/message/140416
Mute This Topic: https://lists.openembedded.org/mt/75373512/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] Patches are stuck in master-next

[Christian Eggers]

Hi Richard,

Am Mittwoch, 8. Juli 2020, 11:20:40 CEST schrieb Richard Purdie:
> The issue is I still don't really like the eudev one in particular. A
> native eudev really doesn't make sense and I'm reluctant to encourage
> people to build such a thing. I suspect you can't convince
> networkmanager-native to build without it (although I'm not convinced
> a native libnm is that useful either).

For NATIVESDK, I need the libs because I deploy a x86 version
of our embedded application for our internal integration test team. Sure, I
could built two cross SDKs (one for ARM, one for x86), but (mis?)using
NATIVESDK for getting the x86 version of our applications works
very fine for me.

For NATIVE, I need the libs for running automatic unit tests on the
build machine. One could probably ask whether unit tests should
depend on libraries like libudev or libnm.

Please let me know, which of the additions (NATIVE / NATIVESDK)
are welcome. I'm only offering to share my work, I'm also fine
keeping this in my private layer.

regards
Christian




 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 


Get all the latest information from www.arri.com, 
Facebook, 
Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#140413): 
https://lists.openembedded.org/g/openembedded-core/message/140413
Mute This Topic: https://lists.openembedded.org/mt/75371450/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] Patches are stuck in master-next

[Christian Eggers]

Two of my patches are in master-next since some time. Is there any reason why
they don't appear in master?

https://patchwork.openembedded.org/series/24760/#
https://patchwork.openembedded.org/series/24775/#

I would like to send further patches for meta-openembedded which
depend on the above ones.

regards
Christian




 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 


Get all the latest information from www.arri.com, 
Facebook, 
Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#140403): 
https://lists.openembedded.org/g/openembedded-core/message/140403
Mute This Topic: https://lists.openembedded.org/mt/75371450/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] [Patch v2 0/2] Extend dependencies of networkmanager for native/nativesdk

[Christian Eggers]

Am Freitag, 26. Juni 2020, 14:06:02 CEST schrieb Ross Burton:
> I'm curious what you're actually building for native here? (ie what
> bit of network manager is meaningful as a native build).
>
> Ross
>
Our own application (embedded product) links against libnm.
This application can either be built for the embedded system
or for PC platform.

I plan to perform unit tests of this application on the build machine
(native). This tests can also include testing the interfaces to
NetworkManager.

I'm still unsure whether I will built and run my own application
tests within the Yocto build environment (native Yocto package)
or doing this "externally" under the (native part of the) SDK. The
compiler/libraries from the build machine itself are not suitable
as they are usually too old.

>From your question it sounds like not every type of software
may be suitable for -native/-nativesdk? Shall these packages be
limited to software required for "building" (compiler, code generators,
documentation, ...)?

Sure, I could create separate SDKs for both machines, but having
the "target" SDK and the "PC" SDK together works very fine for us.

regards
Christian

> On Mon, 22 Jun 2020 at 13:54, Christian Eggers  wrote:
> > On Friday, 2020-06-19, Richard Pudie wrote:
> > > A native eudev is a really bad idea, a nativesdk one isn't much better.
> > > If there are libs we need from here, then the recipe should only be
> > > generating those but a "native eudev" makes no sense to be needed by a
> > > build system.
> >
> > Ack. I already expected that there might be a (hidden?) rule, what type of
> > software is (NOT) allowed for -native/-nativesdk.
> >
> > > A native or nativesdk consolekit does not make sense.
> > >
> > > If you need libs from networkmanager, you'll probably have to patch it
> > > just to generate the libs you need. We don't want builds polluted with
> > > confusing and potentially quite problematic binaries.
> >
> > Networkmanager can be built fine without consolekit. Probably this will
> > not make any difference for libnm.
> >
> >




 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 
<https://microsites.arri.com/stellar/>

Get all the latest information from www.arri.com<https://www.arri.com/>, 
Facebook<https://www.facebook.com/TeamARRI>, 
Twitter<https://twitter.com/ARRIChannel>, Instagram<https://instagram.com/arri> 
and YouTube<https://www.youtube.com/user/ARRIChannel>.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#140016): 
https://lists.openembedded.org/g/openembedded-core/message/140016
Mute This Topic: https://lists.openembedded.org/mt/75037411/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [OE-core] [PATCH 2/4] eudev: Extend for native/nativesdk

[Christian Eggers]

Dear Richard,

Am Freitag, 19. Juni 2020, 23:27:38 CEST schrieb Richard Purdie:
> A native eudev is a really bad idea, a nativesdk one isn't much better.
> If there are libs we need from here, then the recipe should only be
> generating those but a "native eudev" makes no sense to be needed by a
> build system.
Does v2 of my patch meet your requirements? In order to keep the recipe
simple, the full eudev is built, but only libudev is packaged at the end.

> In case I'm not being clear enough, is bitbake going to
> replace the udev/systemd from your host system at run time?
- libraries (libudev, libnm): yes
- tools (e.g. udevadm or nmcli): maybe
- daemons (udevd, networkmanager): no

"Replace" due to PATH / PKG_CONFIG_PATH variables, not really overwriting
them.

regards
Christian




 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 


Get all the latest information from www.arri.com, 
Facebook, 
Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#139877): 
https://lists.openembedded.org/g/openembedded-core/message/139877
Mute This Topic: https://lists.openembedded.org/mt/74987445/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH v2 1/2] libnl: Extend for native/nativesdk

[Christian Eggers]

libnl is required by networkmanager.

Signed-off-by: Christian Eggers 
---
 meta/recipes-support/libnl/libnl_3.5.0.bb | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/meta/recipes-support/libnl/libnl_3.5.0.bb 
b/meta/recipes-support/libnl/libnl_3.5.0.bb
index c0609c4d90..9d0e1441a9 100644
--- a/meta/recipes-support/libnl/libnl_3.5.0.bb
+++ b/meta/recipes-support/libnl/libnl_3.5.0.bb
@@ -40,3 +40,5 @@ FILES_${PN}-route = "${libdir}/libnl-route-3.so.*"
 FILES_${PN}-xfrm  = "${libdir}/libnl-xfrm-3.so.*"
 RREPLACES_${PN}-genl = "libnl-genl2"
 RCONFLICTS_${PN}-genl = "libnl-genl2"
+
+BBCLASSEXTEND = "native nativesdk"
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#139734): 
https://lists.openembedded.org/g/openembedded-core/message/139734
Mute This Topic: https://lists.openembedded.org/mt/75037413/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [Patch v2 0/2] Extend dependencies of networkmanager for native/nativesdk

[Christian Eggers]

On Friday, 2020-06-19, Richard Pudie wrote:
> A native eudev is a really bad idea, a nativesdk one isn't much better.
> If there are libs we need from here, then the recipe should only be
> generating those but a "native eudev" makes no sense to be needed by a
> build system.
Ack. I already expected that there might be a (hidden?) rule, what type of
software is (NOT) allowed for -native/-nativesdk.

> A native or nativesdk consolekit does not make sense.

> If you need libs from networkmanager, you'll probably have to patch it
> just to generate the libs you need. We don't want builds polluted with
> confusing and potentially quite problematic binaries.
Networkmanager can be built fine without consolekit. Probably this will
not make any difference for libnm.


Changelog v2:
--
- eudev: Don't build main package for -native(sdk), only libudev is
  required
- dbus/dbus-glib: Patch removed
- consolekit: Patch removed





 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 


Get all the latest information from www.arri.com, 
Facebook, 
Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#139733): 
https://lists.openembedded.org/g/openembedded-core/message/139733
Mute This Topic: https://lists.openembedded.org/mt/75037411/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH v2 2/2] eudev: Extend for native/nativesdk

[Christian Eggers]

libudev is required by networkmanager.

Signed-off-by: Christian Eggers 
---
 meta/recipes-core/udev/eudev_3.2.9.bb | 13 +
 1 file changed, 13 insertions(+)

diff --git a/meta/recipes-core/udev/eudev_3.2.9.bb 
b/meta/recipes-core/udev/eudev_3.2.9.bb
index f96f8cbe78..0009896ab9 100644
--- a/meta/recipes-core/udev/eudev_3.2.9.bb
+++ b/meta/recipes-core/udev/eudev_3.2.9.bb
@@ -61,12 +61,22 @@ do_install_prepend_class-target () {
sed -i -e 's:${RECIPE_SYSROOT_NATIVE}::g' \
${B}/src/udev/keyboard-keys-from-name.h
 }
+do_install_append_class-native() {
+   # for -native, we only want to install libudev
+   rm -r ${D}${sysconfdir} ${D}${nonarch_base_libdir}/udev 
${D}${base_sbindir} ${D}${bindir} ${D}${libexecdir}
+}
+do_install_append_class-nativesdk() {
+   # for -nativesdk, we only want to install libudev
+   rm -r ${D}${sysconfdir} ${D}${nonarch_base_libdir}/udev 
${D}${base_sbindir} ${D}${bindir} ${D}${libexecdir}
+}
 
 INITSCRIPT_NAME = "udev"
 INITSCRIPT_PARAMS = "start 04 S ."
 
 PACKAGES =+ "libudev"
 PACKAGES =+ "eudev-hwdb"
+# For nativesdk, we only want to package libudev
+PACKAGES_remove_class-nativesdk = "${PN} eudev-hwdb"
 
 
 FILES_${PN} += "${libexecdir} ${nonarch_base_libdir}/udev ${bindir}/udevadm"
@@ -81,6 +91,7 @@ RDEPENDS_eudev-hwdb += "eudev"
 
 RPROVIDES_${PN} = "hotplug udev"
 RPROVIDES_eudev-hwdb += "udev-hwdb"
+RPROVIDES_libudev_class-nativesdk = "${PN}"
 
 PACKAGE_WRITE_DEPS += "qemu-native"
 pkg_postinst_eudev-hwdb () {
@@ -94,3 +105,5 @@ pkg_postinst_eudev-hwdb () {
 pkg_prerm_eudev-hwdb () {
 rm -f $D${sysconfdir}/udev/hwdb.bin
 }
+
+BBCLASSEXTEND = "native nativesdk"
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#139735): 
https://lists.openembedded.org/g/openembedded-core/message/139735
Mute This Topic: https://lists.openembedded.org/mt/75037414/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH 3/4] dbus/dbus-glib: Extend for nativesdk

[Christian Eggers]

dbus-glib is required by consolekit which in turn is required by
networkmanager.

Signed-off-by: Christian Eggers 
---
 meta/recipes-core/dbus/dbus-glib_0.110.bb | 2 +-
 meta/recipes-core/dbus/dbus_1.12.18.bb| 1 -
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/meta/recipes-core/dbus/dbus-glib_0.110.bb 
b/meta/recipes-core/dbus/dbus-glib_0.110.bb
index 0b45805cf4..f3f20fc7e1 100644
--- a/meta/recipes-core/dbus/dbus-glib_0.110.bb
+++ b/meta/recipes-core/dbus/dbus-glib_0.110.bb
@@ -36,4 +36,4 @@ FILES_${PN}-dev += "${bindir}/dbus-binding-tool"
 RDEPENDS_${PN}-tests += "dbus-x11"
 FILES_${PN}-tests = "${datadir}/${BPN}/tests"
 
-BBCLASSEXTEND = "native"
+BBCLASSEXTEND = "native nativesdk"
diff --git a/meta/recipes-core/dbus/dbus_1.12.18.bb 
b/meta/recipes-core/dbus/dbus_1.12.18.bb
index 2fcb3079ad..5bbdd8dd50 100644
--- a/meta/recipes-core/dbus/dbus_1.12.18.bb
+++ b/meta/recipes-core/dbus/dbus_1.12.18.bb
@@ -43,7 +43,6 @@ DEBIANNAME_${PN} = "dbus-1"
 PACKAGES =+ "${PN}-lib"
 
 OLDPKGNAME = "dbus-x11"
-OLDPKGNAME_class-nativesdk = ""
 
 # for compatibility
 RPROVIDES_${PN} = "${OLDPKGNAME}"
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#139704): 
https://lists.openembedded.org/g/openembedded-core/message/139704
Mute This Topic: https://lists.openembedded.org/mt/74987452/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH 1/4] libnl: Extend for native/nativesdk

[Christian Eggers]

libnl is required by networkmanager.

Signed-off-by: Christian Eggers 
---
 meta/recipes-support/libnl/libnl_3.5.0.bb | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/meta/recipes-support/libnl/libnl_3.5.0.bb 
b/meta/recipes-support/libnl/libnl_3.5.0.bb
index c0609c4d90..9d0e1441a9 100644
--- a/meta/recipes-support/libnl/libnl_3.5.0.bb
+++ b/meta/recipes-support/libnl/libnl_3.5.0.bb
@@ -40,3 +40,5 @@ FILES_${PN}-route = "${libdir}/libnl-route-3.so.*"
 FILES_${PN}-xfrm  = "${libdir}/libnl-xfrm-3.so.*"
 RREPLACES_${PN}-genl = "libnl-genl2"
 RCONFLICTS_${PN}-genl = "libnl-genl2"
+
+BBCLASSEXTEND = "native nativesdk"
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#139702): 
https://lists.openembedded.org/g/openembedded-core/message/139702
Mute This Topic: https://lists.openembedded.org/mt/74987444/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH 4/4] consolekit: Extend for native/nativesdk

[Christian Eggers]

consolekit is required by networkmanager.

Signed-off-by: Christian Eggers 
---
 meta/recipes-support/consolekit/consolekit_0.4.6.bb | 5 -
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/meta/recipes-support/consolekit/consolekit_0.4.6.bb 
b/meta/recipes-support/consolekit/consolekit_0.4.6.bb
index 89f2d77b66..ad35e8c6d6 100644
--- a/meta/recipes-support/consolekit/consolekit_0.4.6.bb
+++ b/meta/recipes-support/consolekit/consolekit_0.4.6.bb
@@ -7,7 +7,8 @@ LIC_FILES_CHKSUM = 
"file://COPYING;md5=59530bdf33659b29e73d4adb9f9f6552 \
 
file://src/main.c;endline=21;md5=0a994e09769780220163255d8f9071c3"
 
 DEPENDS = "glib-2.0 glib-2.0-native dbus dbus-glib virtual/libx11"
-RDEPENDS_${PN} += "base-files"
+# base-files is only available for target
+RDEPENDS_${PN}_class-target += "base-files"
 
 inherit autotools pkgconfig features_check
 # depends on virtual/libx11
@@ -47,3 +48,5 @@ do_install_append() {
# Remove /var/ directories as the daemon creates them as required
rm -rf ${D}${localstatedir}
 }
+
+BBCLASSEXTEND = "native nativesdk"
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#139705): 
https://lists.openembedded.org/g/openembedded-core/message/139705
Mute This Topic: https://lists.openembedded.org/mt/74987454/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH 2/4] eudev: Extend for native/nativesdk

[Christian Eggers]

eudev is required by networkmanager.

Signed-off-by: Christian Eggers 
---
 meta/recipes-core/udev/eudev_3.2.9.bb | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/meta/recipes-core/udev/eudev_3.2.9.bb 
b/meta/recipes-core/udev/eudev_3.2.9.bb
index f96f8cbe78..a4cec456d9 100644
--- a/meta/recipes-core/udev/eudev_3.2.9.bb
+++ b/meta/recipes-core/udev/eudev_3.2.9.bb
@@ -94,3 +94,5 @@ pkg_postinst_eudev-hwdb () {
 pkg_prerm_eudev-hwdb () {
 rm -f $D${sysconfdir}/udev/hwdb.bin
 }
+
+BBCLASSEXTEND = "native nativesdk"
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#139703): 
https://lists.openembedded.org/g/openembedded-core/message/139703
Mute This Topic: https://lists.openembedded.org/mt/74987445/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [Patch 0/4] Extend dependencies of networkmanager for native/nativesdk

[Christian Eggers]

My project needs libnm for native/nativesdk. Patches for
meta-oe/meta-networking will follow after this has been applied.



 
[http://assets.arri.com/media/sign/2020-04-03-E-mail-signature-Stellar2_V1.jpg] 


Get all the latest information from www.arri.com, 
Facebook, 
Twitter, Instagram 
and YouTube.

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRA 57918
Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: 
HRB 54477
Geschäftsführer: Dr. Michael Neuhäuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#139701): 
https://lists.openembedded.org/g/openembedded-core/message/139701
Mute This Topic: https://lists.openembedded.org/mt/74987443/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH] tiff: Extend for nativesdk

[Christian Eggers]

Doxygen in meta-oe has recently been extended for nativesdk. Doxygen is
often used together with mscgen which in turn depends indirectly on
tiff (via gd library).

Signed-off-by: Christian Eggers 
---
 meta/recipes-multimedia/libtiff/tiff_4.1.0.bb | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/recipes-multimedia/libtiff/tiff_4.1.0.bb 
b/meta/recipes-multimedia/libtiff/tiff_4.1.0.bb
index 1f92c18513..9b907fbc24 100644
--- a/meta/recipes-multimedia/libtiff/tiff_4.1.0.bb
+++ b/meta/recipes-multimedia/libtiff/tiff_4.1.0.bb
@@ -47,4 +47,4 @@ do_install_append() {
 oe_multilib_header tiffconf.h
 }
 
-BBCLASSEXTEND = "native"
+BBCLASSEXTEND = "native nativesdk"
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#139101): 
https://lists.openembedded.org/g/openembedded-core/message/139101
Mute This Topic: https://lists.openembedded.org/mt/74627695/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[OE-core] [PATCH] librsvg: Extend for nativesdk

[Christian Eggers]

Doxygen in meta-oe has recently been extended for nativesdk. Doxygen is
often used together with dot (graphviz) which in turn depends on
librsvg.

Signed-off-by: Christian Eggers 
---
 meta/recipes-gnome/librsvg/librsvg_2.40.20.bb | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/recipes-gnome/librsvg/librsvg_2.40.20.bb 
b/meta/recipes-gnome/librsvg/librsvg_2.40.20.bb
index 6dd0533a5d..337299ff69 100644
--- a/meta/recipes-gnome/librsvg/librsvg_2.40.20.bb
+++ b/meta/recipes-gnome/librsvg/librsvg_2.40.20.bb
@@ -14,7 +14,7 @@ LIC_FILES_CHKSUM = 
"file://COPYING;md5=94d55d512a9ba36caa9b7df079bae19f \
 
 SECTION = "x11/utils"
 DEPENDS = "cairo gdk-pixbuf glib-2.0 libcroco libxml2 pango"
-BBCLASSEXTEND = "native"
+BBCLASSEXTEND = "native nativesdk"
 
 inherit gnomebase gtk-doc pixbufcache upstream-version-is-even 
gobject-introspection
 
-- 
Christian Eggers
Embedded software developer

Arnold & Richter Cine Technik GmbH & Co. Betriebs KG
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRA 57918
Persoenlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH
Sitz: Muenchen - Registergericht: Amtsgericht Muenchen - Handelsregisternummer: 
HRB 54477
Geschaeftsfuehrer: Dr. Michael Neuhaeuser; Stephan Schenk; Walter Trauninger; 
Markus Zeiler

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#139100): 
https://lists.openembedded.org/g/openembedded-core/message/139100
Mute This Topic: https://lists.openembedded.org/mt/74627673/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-