Re: [oe][meta-oe][PATCH 0/2] glmark2: fix an issue with eglconfig scoring
The QA check in oe-core would tell you that (as well as that the 'S' in Status needs to be uppercase). On Tue, Mar 5, 2024 at 10:47 PM Randolph Sapp wrote: > > On 3/5/24 15:42, Martin Jansa wrote: > > Invalid format of Upstream-Status and please squash those 2 commits as > > they don't make sense separately. > > > > On Tue, Mar 5, 2024 at 10:26 PM Randolph Sapp via > > lists.openembedded.org wrote: > >> > >> From: Randolph Sapp > >> > >> Backport a patch from the upstream glmark2 package to fix 2 primary issues: > >> > >> 1. Configs with stencil components were ranked unnecessarily low. > >> 2. If the target config was below 0 an optimal config may not be properly > >> selected. > >> > >> Randolph Sapp (2): > >>glmark2: add upstream patch to not care about stencil config > >>glmark2: add the stencil config patch to the package > >> > >> ...y-default-don-t-care-about-the-stenc.patch | 103 ++ > >> .../recipes-benchmark/glmark2/glmark2_git.bb | 1 + > >> 2 files changed, 104 insertions(+) > >> create mode 100644 > >> meta-oe/recipes-benchmark/glmark2/files/0003-GLVisualConfig-By-default-don-t-care-about-the-stenc.patch > >> > >> -- > >> 2.44.0 > > Roger. Would the correct status be: > > > Upstream-Status: Backport [5f23d540342ba69e12afeb6a1ac4f6fd36747975] > > (Reading the docs I was not aware the brackets were supposed to be > included...) -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#109139): https://lists.openembedded.org/g/openembedded-devel/message/109139 Mute This Topic: https://lists.openembedded.org/mt/104754322/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[oe][meta-oe][PATCHv2 0/1] glmark2: fix an issue with eglconfig scoring
From: Randolph Sapp Backport a patch from the upstream glmark2 package to fix 2 primary issues: 1. Configs with stencil components were ranked unnecessarily low. 2. If the target config was below 0 an optimal config may not be properly selected. v2: Fix upstream status and squash commits Randolph Sapp (1): glmark2: add upstream patch to not care about stencil config ...y-default-don-t-care-about-the-stenc.patch | 103 ++ .../recipes-benchmark/glmark2/glmark2_git.bb | 1 + 2 files changed, 104 insertions(+) create mode 100644 meta-oe/recipes-benchmark/glmark2/files/0003-GLVisualConfig-By-default-don-t-care-about-the-stenc.patch -- 2.44.0 -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#109138): https://lists.openembedded.org/g/openembedded-devel/message/109138 Mute This Topic: https://lists.openembedded.org/mt/104755766/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[oe][meta-oe][PATCHv2 1/1] glmark2: add upstream patch to not care about stencil config
From: Randolph Sapp
The upstream project already patched this, but considering this package
is currently locked to the latest tag, I'll add it here.
This fixes an issue with stencil configs being scored unnecessarily low.
It also addresses a problem when the target config is below 0.
Signed-off-by: Randolph Sapp
---
...y-default-don-t-care-about-the-stenc.patch | 103 ++
.../recipes-benchmark/glmark2/glmark2_git.bb | 1 +
2 files changed, 104 insertions(+)
create mode 100644
meta-oe/recipes-benchmark/glmark2/files/0003-GLVisualConfig-By-default-don-t-care-about-the-stenc.patch
diff --git
a/meta-oe/recipes-benchmark/glmark2/files/0003-GLVisualConfig-By-default-don-t-care-about-the-stenc.patch
b/meta-oe/recipes-benchmark/glmark2/files/0003-GLVisualConfig-By-default-don-t-care-about-the-stenc.patch
new file mode 100644
index 00..dc47af83e7
--- /dev/null
+++
b/meta-oe/recipes-benchmark/glmark2/files/0003-GLVisualConfig-By-default-don-t-care-about-the-stenc.patch
@@ -0,0 +1,103 @@
+From b59ec731c8c7e084b289e4ea92938faaebcc389d Mon Sep 17 00:00:00 2001
+From: Alexandros Frantzis
+Date: Wed, 1 Nov 2023 13:43:00 +0200
+Subject: [PATCH] GLVisualConfig: By default don't care about the stencil
+ config component
+
+Our benchmarks don't use a stencil buffer, but its presence doesn't hurt
+either, so don't mark configs that have one as unacceptable. Our scoring
+still favors configs without one, unless the user explicitly specifies
+otherwise with --visual-config.
+
+Upstream-Status: Backport
[https://github.com/glmark2/glmark2/commit/5f23d540342ba69e12afeb6a1ac4f6fd36747975]
+---
+ doc/glmark2.1.in | 2 +-
+ src/gl-visual-config.cpp | 9 ++---
+ src/gl-visual-config.h | 2 +-
+ src/options.cpp | 6 +++---
+ 4 files changed, 11 insertions(+), 8 deletions(-)
+
+diff --git a/doc/glmark2.1.in b/doc/glmark2.1.in
+index 015ba20..1f4a047 100644
+--- a/doc/glmark2.1.in
b/doc/glmark2.1.in
+@@ -38,7 +38,7 @@ Render to an off-screen surface
+ The visual configuration to use for the rendering target:
+
\'id=ID:red=R:green=G:blue=B:alpha=A:buffer=BUF:stencil=STENCIL:samples=SAMPLES'.
+ The parameters may be defined in any order, and any omitted parameters assume
a
+-default value of '0' (id, stencil, samples) or '1' (red, green, blue, alpha,
buffer).
++default value of '0' (id, samples), -1 (stencil) or '1' (red, green, blue,
alpha, buffer).
+ If 'id' is set to a non-zero value, all other parameters are ignored
+ .TP
+ \fB\-\-reuse\-context\fR
+diff --git a/src/gl-visual-config.cpp b/src/gl-visual-config.cpp
+index de92f93..665f53a 100644
+--- a/src/gl-visual-config.cpp
b/src/gl-visual-config.cpp
+@@ -26,7 +26,7 @@
+ #include
+
+ GLVisualConfig::GLVisualConfig(const std::string ) :
+-id(0), red(1), green(1), blue(1), alpha(1), depth(1), stencil(0),
buffer(1), samples(0)
++GLVisualConfig()
+ {
+ std::vector elems;
+
+@@ -85,7 +85,7 @@ GLVisualConfig::match_score(const GLVisualConfig )
const
+ score += score_component(blue, target.blue, 4);
+ score += score_component(alpha, target.alpha, 4);
+ score += score_component(depth, target.depth, 1);
+-score += score_component(stencil, target.stencil, 0);
++score += score_component(stencil, target.stencil, 1);
+ score += score_component(buffer, target.buffer, 1);
+ score += score_component(samples, target.samples, -1);
+
+@@ -135,11 +135,14 @@ GLVisualConfig::score_component(int component, int
target, int scale) const
+ * score for all components ranges from [0,MAXIMUM_COMPONENT_SCORE).
+ * If scale > 0, we reward the largest positive difference from
target,
+ * otherwise the smallest positive difference from target.
++ * We also reward the smallest positive difference from the target,
++ * if the target < 0, i.e., we don't care about this value.
+ */
+ int diff = std::abs(scale) * (component - target);
+ if (diff > 0)
+ {
+-score = scale < 0 ? MAXIMUM_COMPONENT_SCORE - diff : diff;
++score = (scale < 0 || target < 0) ?
++MAXIMUM_COMPONENT_SCORE - diff : diff;
+ score = std::min(MAXIMUM_COMPONENT_SCORE, score);
+ score = std::max(0, score);
+ }
+diff --git a/src/gl-visual-config.h b/src/gl-visual-config.h
+index b28473f..013ce14 100644
+--- a/src/gl-visual-config.h
b/src/gl-visual-config.h
+@@ -31,7 +31,7 @@ class GLVisualConfig
+ {
+ public:
+ GLVisualConfig():
+-id(0), red(1), green(1), blue(1), alpha(1), depth(1), stencil(0),
buffer(1), samples(0) {}
++id(0), red(1), green(1), blue(1), alpha(1), depth(1), stencil(-1),
buffer(1), samples(0) {}
+ GLVisualConfig(const std::string );
+
+ /**
+diff --git a/src/options.cpp b/src/options.cpp
+index 8d1ec16..3a31d3d 100644
+--- a/src/options.cpp
b/src/options.cpp
+@@ -213,9 +213,9 @@ Options::print_help()
+"
Re: [oe][meta-oe][PATCH 0/2] glmark2: fix an issue with eglconfig scoring
On 3/5/24 15:42, Martin Jansa wrote: Invalid format of Upstream-Status and please squash those 2 commits as they don't make sense separately. On Tue, Mar 5, 2024 at 10:26 PM Randolph Sapp via lists.openembedded.org wrote: From: Randolph Sapp Backport a patch from the upstream glmark2 package to fix 2 primary issues: 1. Configs with stencil components were ranked unnecessarily low. 2. If the target config was below 0 an optimal config may not be properly selected. Randolph Sapp (2): glmark2: add upstream patch to not care about stencil config glmark2: add the stencil config patch to the package ...y-default-don-t-care-about-the-stenc.patch | 103 ++ .../recipes-benchmark/glmark2/glmark2_git.bb | 1 + 2 files changed, 104 insertions(+) create mode 100644 meta-oe/recipes-benchmark/glmark2/files/0003-GLVisualConfig-By-default-don-t-care-about-the-stenc.patch -- 2.44.0 Roger. Would the correct status be: > Upstream-Status: Backport [5f23d540342ba69e12afeb6a1ac4f6fd36747975] (Reading the docs I was not aware the brackets were supposed to be included...) -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#109136): https://lists.openembedded.org/g/openembedded-devel/message/109136 Mute This Topic: https://lists.openembedded.org/mt/104754322/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [oe][meta-oe][PATCH 0/2] glmark2: fix an issue with eglconfig scoring
Invalid format of Upstream-Status and please squash those 2 commits as they don't make sense separately. On Tue, Mar 5, 2024 at 10:26 PM Randolph Sapp via lists.openembedded.org wrote: > > From: Randolph Sapp > > Backport a patch from the upstream glmark2 package to fix 2 primary issues: > > 1. Configs with stencil components were ranked unnecessarily low. > 2. If the target config was below 0 an optimal config may not be properly > selected. > > Randolph Sapp (2): > glmark2: add upstream patch to not care about stencil config > glmark2: add the stencil config patch to the package > > ...y-default-don-t-care-about-the-stenc.patch | 103 ++ > .../recipes-benchmark/glmark2/glmark2_git.bb | 1 + > 2 files changed, 104 insertions(+) > create mode 100644 > meta-oe/recipes-benchmark/glmark2/files/0003-GLVisualConfig-By-default-don-t-care-about-the-stenc.patch > > -- > 2.44.0 > > > > -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#109135): https://lists.openembedded.org/g/openembedded-devel/message/109135 Mute This Topic: https://lists.openembedded.org/mt/104754322/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[oe][meta-oe][PATCH 0/2] glmark2: fix an issue with eglconfig scoring
From: Randolph Sapp Backport a patch from the upstream glmark2 package to fix 2 primary issues: 1. Configs with stencil components were ranked unnecessarily low. 2. If the target config was below 0 an optimal config may not be properly selected. Randolph Sapp (2): glmark2: add upstream patch to not care about stencil config glmark2: add the stencil config patch to the package ...y-default-don-t-care-about-the-stenc.patch | 103 ++ .../recipes-benchmark/glmark2/glmark2_git.bb | 1 + 2 files changed, 104 insertions(+) create mode 100644 meta-oe/recipes-benchmark/glmark2/files/0003-GLVisualConfig-By-default-don-t-care-about-the-stenc.patch -- 2.44.0 -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#109133): https://lists.openembedded.org/g/openembedded-devel/message/109133 Mute This Topic: https://lists.openembedded.org/mt/104754322/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[oe][meta-oe][PATCH 2/2] glmark2: add the stencil config patch to the package
From: Randolph Sapp Add the previous upstream stencil config patch to the glmark2 package to address issues with EGL configs being ranked unnecessarily low, and another issue when the target config is below zero. Signed-off-by: Randolph Sapp --- meta-oe/recipes-benchmark/glmark2/glmark2_git.bb | 1 + 1 file changed, 1 insertion(+) diff --git a/meta-oe/recipes-benchmark/glmark2/glmark2_git.bb b/meta-oe/recipes-benchmark/glmark2/glmark2_git.bb index a2d5b152f2..0b61d6ed62 100644 --- a/meta-oe/recipes-benchmark/glmark2/glmark2_git.bb +++ b/meta-oe/recipes-benchmark/glmark2/glmark2_git.bb @@ -17,6 +17,7 @@ SRC_URI = " \ git://github.com/glmark2/glmark2.git;protocol=https;branch=master \ file://0001-fix-dispmanx-build.patch \ file://0002-run-dispmanx-fullscreen.patch \ +file://0003-GLVisualConfig-By-default-don-t-care-about-the-stenc.patch \ " SRCREV = "42e3d8fe3aa88743ef90348138f643f7b04a9237" -- 2.44.0 -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#109134): https://lists.openembedded.org/g/openembedded-devel/message/109134 Mute This Topic: https://lists.openembedded.org/mt/104754323/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[oe][meta-oe][PATCH 1/2] glmark2: add upstream patch to not care about stencil config
From: Randolph Sapp
The upstream project already patched this, but considering this package
is currently locked to the latest tag, I'll add it here.
This fixes an issue with stencil configs being scored unnecessarily low.
It also addresses a problem when the target config is below 0.
Signed-off-by: Randolph Sapp
---
...y-default-don-t-care-about-the-stenc.patch | 103 ++
1 file changed, 103 insertions(+)
create mode 100644
meta-oe/recipes-benchmark/glmark2/files/0003-GLVisualConfig-By-default-don-t-care-about-the-stenc.patch
diff --git
a/meta-oe/recipes-benchmark/glmark2/files/0003-GLVisualConfig-By-default-don-t-care-about-the-stenc.patch
b/meta-oe/recipes-benchmark/glmark2/files/0003-GLVisualConfig-By-default-don-t-care-about-the-stenc.patch
new file mode 100644
index 00..a9dcc8ef1a
--- /dev/null
+++
b/meta-oe/recipes-benchmark/glmark2/files/0003-GLVisualConfig-By-default-don-t-care-about-the-stenc.patch
@@ -0,0 +1,103 @@
+From b59ec731c8c7e084b289e4ea92938faaebcc389d Mon Sep 17 00:00:00 2001
+From: Alexandros Frantzis
+Date: Wed, 1 Nov 2023 13:43:00 +0200
+Subject: [PATCH] GLVisualConfig: By default don't care about the stencil
+ config component
+
+Our benchmarks don't use a stencil buffer, but its presence doesn't hurt
+either, so don't mark configs that have one as unacceptable. Our scoring
+still favors configs without one, unless the user explicitly specifies
+otherwise with --visual-config.
+
+Upstream-status: Backport 5f23d540342ba69e12afeb6a1ac4f6fd36747975
+---
+ doc/glmark2.1.in | 2 +-
+ src/gl-visual-config.cpp | 9 ++---
+ src/gl-visual-config.h | 2 +-
+ src/options.cpp | 6 +++---
+ 4 files changed, 11 insertions(+), 8 deletions(-)
+
+diff --git a/doc/glmark2.1.in b/doc/glmark2.1.in
+index 015ba20..1f4a047 100644
+--- a/doc/glmark2.1.in
b/doc/glmark2.1.in
+@@ -38,7 +38,7 @@ Render to an off-screen surface
+ The visual configuration to use for the rendering target:
+
\'id=ID:red=R:green=G:blue=B:alpha=A:buffer=BUF:stencil=STENCIL:samples=SAMPLES'.
+ The parameters may be defined in any order, and any omitted parameters assume
a
+-default value of '0' (id, stencil, samples) or '1' (red, green, blue, alpha,
buffer).
++default value of '0' (id, samples), -1 (stencil) or '1' (red, green, blue,
alpha, buffer).
+ If 'id' is set to a non-zero value, all other parameters are ignored
+ .TP
+ \fB\-\-reuse\-context\fR
+diff --git a/src/gl-visual-config.cpp b/src/gl-visual-config.cpp
+index de92f93..665f53a 100644
+--- a/src/gl-visual-config.cpp
b/src/gl-visual-config.cpp
+@@ -26,7 +26,7 @@
+ #include
+
+ GLVisualConfig::GLVisualConfig(const std::string ) :
+-id(0), red(1), green(1), blue(1), alpha(1), depth(1), stencil(0),
buffer(1), samples(0)
++GLVisualConfig()
+ {
+ std::vector elems;
+
+@@ -85,7 +85,7 @@ GLVisualConfig::match_score(const GLVisualConfig )
const
+ score += score_component(blue, target.blue, 4);
+ score += score_component(alpha, target.alpha, 4);
+ score += score_component(depth, target.depth, 1);
+-score += score_component(stencil, target.stencil, 0);
++score += score_component(stencil, target.stencil, 1);
+ score += score_component(buffer, target.buffer, 1);
+ score += score_component(samples, target.samples, -1);
+
+@@ -135,11 +135,14 @@ GLVisualConfig::score_component(int component, int
target, int scale) const
+ * score for all components ranges from [0,MAXIMUM_COMPONENT_SCORE).
+ * If scale > 0, we reward the largest positive difference from
target,
+ * otherwise the smallest positive difference from target.
++ * We also reward the smallest positive difference from the target,
++ * if the target < 0, i.e., we don't care about this value.
+ */
+ int diff = std::abs(scale) * (component - target);
+ if (diff > 0)
+ {
+-score = scale < 0 ? MAXIMUM_COMPONENT_SCORE - diff : diff;
++score = (scale < 0 || target < 0) ?
++MAXIMUM_COMPONENT_SCORE - diff : diff;
+ score = std::min(MAXIMUM_COMPONENT_SCORE, score);
+ score = std::max(0, score);
+ }
+diff --git a/src/gl-visual-config.h b/src/gl-visual-config.h
+index b28473f..013ce14 100644
+--- a/src/gl-visual-config.h
b/src/gl-visual-config.h
+@@ -31,7 +31,7 @@ class GLVisualConfig
+ {
+ public:
+ GLVisualConfig():
+-id(0), red(1), green(1), blue(1), alpha(1), depth(1), stencil(0),
buffer(1), samples(0) {}
++id(0), red(1), green(1), blue(1), alpha(1), depth(1), stencil(-1),
buffer(1), samples(0) {}
+ GLVisualConfig(const std::string );
+
+ /**
+diff --git a/src/options.cpp b/src/options.cpp
+index 8d1ec16..3a31d3d 100644
+--- a/src/options.cpp
b/src/options.cpp
+@@ -213,9 +213,9 @@ Options::print_help()
+" target:
'id=ID:red=R:green=G:blue=B:alpha=A:buffer=BUF:\n"
+"
[oe] [meta-oe][PATCH] signing.bbclass: fix typos
Signed-off-by: Ulrich Ölmann --- meta-oe/classes/signing.bbclass | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/meta-oe/classes/signing.bbclass b/meta-oe/classes/signing.bbclass index ec076d96851a..f52d861b763c 100644 --- a/meta-oe/classes/signing.bbclass +++ b/meta-oe/classes/signing.bbclass @@ -6,9 +6,9 @@ # This class provides a common workflow to use asymmetric (i.e. RSA) keys to # sign artifacts. Usually, the keys are either stored as simple files in the -# file system or on a HSM (Hardware Security Module). While files are easy to -# use, it's hard to verify that no copies of the private have been made and -# only authorized persons are able to use the key. Use of an HSM addresses +# file system or on an HSM (Hardware Security Module). While files are easy to +# use, it's hard to verify that no copies of the private key have been made +# and only authorized persons are able to use the key. Use of an HSM addresses # these risks by only allowing use of the key via an API (often PKCS #11). The # standard way of referring to a specific key in an HSM are PKCS #11 URIs (RFC # 7512). -- 2.39.2 -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#109131): https://lists.openembedded.org/g/openembedded-devel/message/109131 Mute This Topic: https://lists.openembedded.org/mt/104752838/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] [meta-oe][PATCH 1/2] mozjs-115: update 115.6.0 -> 115.8.0
Seeing the QA warning
https://autobuilder.yoctoproject.org/typhoon/#/builders/88/builds/3684/steps/14/logs/warnings
On Tue, Mar 5, 2024 at 3:56 AM Markus Volk wrote:
>
> - python 3.12 support was added, remove the patch
>
> Signed-off-by: Markus Volk
> ---
> .../mozjs/mozjs-115/py3.12.patch | 2496 -
> ...js-115_115.6.0.bb => mozjs-115_115.8.0.bb} |3 +-
> 2 files changed, 1 insertion(+), 2498 deletions(-)
> delete mode 100644 meta-oe/recipes-extended/mozjs/mozjs-115/py3.12.patch
> rename meta-oe/recipes-extended/mozjs/{mozjs-115_115.6.0.bb =>
> mozjs-115_115.8.0.bb} (96%)
>
> diff --git a/meta-oe/recipes-extended/mozjs/mozjs-115/py3.12.patch
> b/meta-oe/recipes-extended/mozjs/mozjs-115/py3.12.patch
> deleted file mode 100644
> index 732c8ff1c..0
> --- a/meta-oe/recipes-extended/mozjs/mozjs-115/py3.12.patch
> +++ /dev/null
> @@ -1,2496 +0,0 @@
> -From 7293cfae4fd68004901825ad1cabb83424d8729a Mon Sep 17 00:00:00 2001
> -From: serge-sans-paille
> -Date: Mon, 16 Oct 2023 13:35:24 +
> -Subject: [PATCH] Bug 1857492 - Upgrade vendored version of six and urllib3
> - r=saschanaz
> -
> -six -> 1.16
> -urllib3 -> 1.26.17
> -
> -Differential Revision: https://phabricator.services.mozilla.com/D190288
> -Upstream-Status: Backport
> [https://github.com/mozilla/gecko-dev/commit/7293cfae4fd68004901825ad1cabb83424d8729a]
> -Signed-off-by: Alexander Kanavin
>
> - third_party/python/poetry.lock| 22 +--
> - third_party/python/requirements.in| 4 +-
> - third_party/python/requirements.txt | 12 +-
> - .../python/six/six-1.13.0.dist-info/RECORD| 6 -
> - .../LICENSE | 2 +-
> - .../METADATA | 9 +-
> - .../python/six/six-1.16.0.dist-info/RECORD| 6 +
> - .../six-1.16.0.dist-info}/WHEEL | 2 +-
> - .../top_level.txt | 0
> - third_party/python/six/six.py | 91 ++---
> - .../urllib3/urllib3-1.26.0.dist-info/RECORD | 44 -
> - .../LICENSE.txt | 0
> - .../METADATA | 177 --
> - .../urllib3/urllib3-1.26.17.dist-info/RECORD | 44 +
> - .../urllib3-1.26.17.dist-info}/WHEEL | 2 +-
> - .../top_level.txt | 0
> - .../python/urllib3/urllib3/__init__.py| 17 ++
> - .../python/urllib3/urllib3/_version.py| 2 +-
> - .../python/urllib3/urllib3/connection.py | 62 --
> - .../python/urllib3/urllib3/connectionpool.py | 97 --
> - .../contrib/_securetransport/bindings.py | 2 +-
> - .../contrib/_securetransport/low_level.py | 1 +
> - .../urllib3/urllib3/contrib/appengine.py | 4 +-
> - .../urllib3/urllib3/contrib/ntlmpool.py | 13 +-
> - .../urllib3/urllib3/contrib/pyopenssl.py | 19 +-
> - .../urllib3/contrib/securetransport.py| 5 +-
> - .../python/urllib3/urllib3/contrib/socks.py | 2 +-
> - .../python/urllib3/urllib3/exceptions.py | 12 +-
> - .../urllib3/urllib3/packages/__init__.py | 5 -
> - .../packages/backports/weakref_finalize.py| 155 +++
> - .../python/urllib3/urllib3/packages/six.py| 125 +
> - .../packages/ssl_match_hostname/__init__.py | 22 ---
> - .../python/urllib3/urllib3/poolmanager.py | 3 +-
> - third_party/python/urllib3/urllib3/request.py | 21 +++
> - .../python/urllib3/urllib3/response.py| 72 ++-
> - .../python/urllib3/urllib3/util/connection.py | 5 +-
> - .../python/urllib3/urllib3/util/proxy.py | 1 +
> - .../python/urllib3/urllib3/util/request.py| 5 +-
> - .../python/urllib3/urllib3/util/retry.py | 37 +++-
> - .../python/urllib3/urllib3/util/ssl_.py | 53 --
> - .../ssl_match_hostname.py}| 15 +-
> - .../urllib3/urllib3/util/ssltransport.py | 6 +-
> - .../python/urllib3/urllib3/util/timeout.py| 9 +-
> - .../python/urllib3/urllib3/util/url.py| 17 +-
> - .../python/urllib3/urllib3/util/wait.py | 1 -
> - 45 files changed, 934 insertions(+), 275 deletions(-)
> - delete mode 100644 third_party/python/six/six-1.13.0.dist-info/RECORD
> - rename third_party/python/six/{six-1.13.0.dist-info =>
> six-1.16.0.dist-info}/LICENSE (96%)
> - rename third_party/python/six/{six-1.13.0.dist-info =>
> six-1.16.0.dist-info}/METADATA (85%)
> - create mode 100644 third_party/python/six/six-1.16.0.dist-info/RECORD
> - rename third_party/python/{urllib3/urllib3-1.26.0.dist-info =>
> six/six-1.16.0.dist-info}/WHEEL (70%)
> - rename third_party/python/six/{six-1.13.0.dist-info =>
> six-1.16.0.dist-info}/top_level.txt (100%)
> - delete mode 100644
> third_party/python/urllib3/urllib3-1.26.0.dist-info/RECORD
> - rename third_party/python/urllib3/{urllib3-1.26.0.dist-info =>
> urllib3-1.26.17.dist-info}/LICENSE.txt (100%)
> - rename
[oe] [meta-oe][PATCH 3/3] emacs: Add packageconfig for selinux support
Disable it by default, since selinux may not be present on build hosts and it can result in non-deterministic behavior during build. Signed-off-by: Khem Raj --- meta-oe/recipes-support/emacs/emacs_29.1.bb | 1 + 1 file changed, 1 insertion(+) diff --git a/meta-oe/recipes-support/emacs/emacs_29.1.bb b/meta-oe/recipes-support/emacs/emacs_29.1.bb index ae35367ac4..5cbe4551c0 100644 --- a/meta-oe/recipes-support/emacs/emacs_29.1.bb +++ b/meta-oe/recipes-support/emacs/emacs_29.1.bb @@ -18,6 +18,7 @@ CVE_STATUS[CVE-2007-6109] = "fixed-version: The CPE in the NVD database doesn't PACKAGECONFIG[gnutls] = "--with-gnutls=yes,--with-gnutls=no,gnutls" PACKAGECONFIG[kerberos] = "--with-kerberos=yes,--with-kerberos=no,krb5" PACKAGECONFIG[libgmp] = "--with-libgmp=yes,--with-libgmp=no,gmp" +PACKAGECONFIG[selinux] = "--with-selinux=yes,--with-selinux=no,libselinux" PACKAGECONFIG ??= "gnutls kerberos libgmp" -- 2.44.0 -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#109129): https://lists.openembedded.org/g/openembedded-devel/message/109129 Mute This Topic: https://lists.openembedded.org/mt/104751680/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[oe] [meta-networking][PATCH 1/3] pgpool2: Upgrade to 4.5.1
Signed-off-by: Khem Raj
---
...own-type-name-fd_set-error-with-musl.patch | 33 ---
.../{pgpool2_4.4.1.bb => pgpool2_4.5.1.bb}| 8 ++---
2 files changed, 2 insertions(+), 39 deletions(-)
delete mode 100644
meta-networking/recipes-support/pgpool2/pgpool2/0001-pgpool2-Fix-unknown-type-name-fd_set-error-with-musl.patch
rename meta-networking/recipes-support/pgpool2/{pgpool2_4.4.1.bb =>
pgpool2_4.5.1.bb} (87%)
diff --git
a/meta-networking/recipes-support/pgpool2/pgpool2/0001-pgpool2-Fix-unknown-type-name-fd_set-error-with-musl.patch
b/meta-networking/recipes-support/pgpool2/pgpool2/0001-pgpool2-Fix-unknown-type-name-fd_set-error-with-musl.patch
deleted file mode 100644
index 26f8afe4d6..00
---
a/meta-networking/recipes-support/pgpool2/pgpool2/0001-pgpool2-Fix-unknown-type-name-fd_set-error-with-musl.patch
+++ /dev/null
@@ -1,33 +0,0 @@
-From 5d335b5ddd052f047759cae0e9b94ec49f9f3884 Mon Sep 17 00:00:00 2001
-From: Lei Maohui
-Date: Mon, 20 Mar 2023 02:03:04 +
-Subject: [PATCH] pgpool2: Fix unknown type name 'fd_set' error with musl.
-
-| read.c: In function 'read_until_ready_for_query':
-| read.c:53:24: error: storage size of 'timeoutval' isn't known
-|53 | struct timeval timeoutval;
-| |^~
-| read.c:54:9: error: unknown type name 'fd_set'
-|54 | fd_set readmask;
-| | ^~
-
-Upstream-Status: Inappropriate [embedded specific]
-Signed-off-by: Lei Maohui
- src/tools/pgproto/read.c | 1 +
- 1 file changed, 1 insertion(+)
-
-diff --git a/src/tools/pgproto/read.c b/src/tools/pgproto/read.c
-index c7fb548..4efe641 100644
a/src/tools/pgproto/read.c
-+++ b/src/tools/pgproto/read.c
-@@ -19,6 +19,7 @@
- #include "../../include/config.h"
- #include "pgproto/pgproto.h"
- #include
-+#include
- #include
- #include
- #include
---
-2.34.1
diff --git a/meta-networking/recipes-support/pgpool2/pgpool2_4.4.1.bb
b/meta-networking/recipes-support/pgpool2/pgpool2_4.5.1.bb
similarity index 87%
rename from meta-networking/recipes-support/pgpool2/pgpool2_4.4.1.bb
rename to meta-networking/recipes-support/pgpool2/pgpool2_4.5.1.bb
index 5e6e00caeb..b127d69625 100644
--- a/meta-networking/recipes-support/pgpool2/pgpool2_4.4.1.bb
+++ b/meta-networking/recipes-support/pgpool2/pgpool2_4.5.1.bb
@@ -1,4 +1,4 @@
-SUMMARY = "a language independent connection pool server for PostgreSQL."
+SUMMARY = "A language independent connection pool server for PostgreSQL."
DESCRIPTION = "Pgpool-II is a middleware that works between PostgreSQL \
servers and a PostgreSQL database client. It is distributed \
@@ -12,11 +12,10 @@ LIC_FILES_CHKSUM =
"file://COPYING;md5=e4b38de086d73e0521de0bbdbaa4a1a9"
SRC_URI = "http://www.pgpool.net/mediawiki/images/pgpool-II-${PV}.tar.gz \
file://0001-Fix-build-error-when-build-this-file.patch \
-
file://0001-pgpool2-Fix-unknown-type-name-fd_set-error-with-musl.patch \
file://pgpool.sysconfig \
file://pgpool.service \
"
-SRC_URI[sha256sum] =
"4b379bbba8e178128a1cee4a5bd1ae116dedb3da6121b728c18f0f54c881f328"
+SRC_URI[sha256sum] =
"8e14b0558a15dae8767c8e1acee3f2f6c7c08ebfffda66a359367eaaa56c3936"
S = "${WORKDIR}/pgpool-II-${PV}"
@@ -34,9 +33,6 @@ PACKAGECONFIG[postgresql] =
"--with-pgsql-includedir=${STAGING_INCDIR}/postgresq
EXTRA_OECONF += "--disable-static \
--disable-rpath \
"
-
-CFLAGS:append = " -fcommon "
-
FILES:${PN} += "${datadir}/pgpool-II/ "
do_configure:append() {
--
2.44.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#109127):
https://lists.openembedded.org/g/openembedded-devel/message/109127
Mute This Topic: https://lists.openembedded.org/mt/104751677/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[oe] [meta-networking][PATCH 2/3] pgpool2: Fix build with postgresql 16+
Signed-off-by: Khem Raj
---
.../pgpool2/pgpool2/define_SIGNAL_ARGS.patch | 25 +++
.../recipes-support/pgpool2/pgpool2_4.5.1.bb | 1 +
2 files changed, 26 insertions(+)
create mode 100644
meta-networking/recipes-support/pgpool2/pgpool2/define_SIGNAL_ARGS.patch
diff --git
a/meta-networking/recipes-support/pgpool2/pgpool2/define_SIGNAL_ARGS.patch
b/meta-networking/recipes-support/pgpool2/pgpool2/define_SIGNAL_ARGS.patch
new file mode 100644
index 00..3ae9b1bd9a
--- /dev/null
+++ b/meta-networking/recipes-support/pgpool2/pgpool2/define_SIGNAL_ARGS.patch
@@ -0,0 +1,25 @@
+Define SIGNAL_ARGS if its not defined, this is flagged with
+newer C compilers e.g. clang18, where if it is not defined
+then it ends up with compiler errors
+
+/usr/include/postgresql/server/port.h:488:28: error: a parameter l
+ist without types is only allowed in a function definition
+| 488 | typedef void (*pqsigfunc) (SIGNAL_ARGS);
+| |^
+
+Upstream-Status: Pending
+Signed-off-by: Khem Raj
+
+--- a/src/include/pool_type.h
b/src/include/pool_type.h
+@@ -125,7 +125,9 @@ extern void ExceptionalCondition(const c
+ #define LONGALIGN(LEN)TYPEALIGN(ALIGNOF_LONG, (LEN))
+ #define DOUBLEALIGN(LEN) TYPEALIGN(ALIGNOF_DOUBLE, (LEN))
+ #define MAXALIGN(LEN) TYPEALIGN(MAXIMUM_ALIGNOF, (LEN))
+-
++#ifndef SIGNAL_ARGS
++#define SIGNAL_ARGS int postgres_signal_arg
++#endif
+ /*
+ * It seems that sockaddr_storage is now commonly used in place of sockaddr.
+ * So, define it if it is not define yet, and create new SockAddr structure
diff --git a/meta-networking/recipes-support/pgpool2/pgpool2_4.5.1.bb
b/meta-networking/recipes-support/pgpool2/pgpool2_4.5.1.bb
index b127d69625..ffb879952c 100644
--- a/meta-networking/recipes-support/pgpool2/pgpool2_4.5.1.bb
+++ b/meta-networking/recipes-support/pgpool2/pgpool2_4.5.1.bb
@@ -12,6 +12,7 @@ LIC_FILES_CHKSUM =
"file://COPYING;md5=e4b38de086d73e0521de0bbdbaa4a1a9"
SRC_URI = "http://www.pgpool.net/mediawiki/images/pgpool-II-${PV}.tar.gz \
file://0001-Fix-build-error-when-build-this-file.patch \
+ file://define_SIGNAL_ARGS.patch \
file://pgpool.sysconfig \
file://pgpool.service \
"
--
2.44.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#109128):
https://lists.openembedded.org/g/openembedded-devel/message/109128
Mute This Topic: https://lists.openembedded.org/mt/104751678/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] [meta-oe][PATCH 1/1] googletest: allow for shared libraries
On Tue, Mar 5, 2024 at 8:31 AM Slater, Joseph wrote:
>
>
>
> > -Original Message-
> > From: Khem Raj
> > Sent: Monday, March 4, 2024 3:50 PM
> > To: Slater, Joseph
> > Cc: openembedded-devel@lists.openembedded.org; MacLeod, Randy
> >
> > Subject: Re: [oe] [meta-oe][PATCH 1/1] googletest: allow for shared
> > libraries
> >
> > On Mon, Mar 4, 2024 at 1:50 PM Joe Slater via lists.openembedded.org
> > wrote:
> > >
> > > From: Joe Slater
> > >
> > > Add a PACKAGECONFIG option to produce shared libraries.
> > > Allow staticdev to be empty if no static libraries are built.
> > >
> >
> > why would be not build static libs ? and if we do not then why do we need to
> > create empty package ?
>
> If we set -DBUILD_SHARED_LIBS=ON static libs will not be built and -staticdev
> will not be created, so -dev will produce an error if we try to, for example,
> populate an sdk. I guess we could put static libs in -dev and get rid of
> -staticdev, but I didn't want to mess with the packaging.
>
maybe you can unbuckle the rdep for dev on staticdev package
> Joe
>
> >
> > > Signed-off-by: Joe Slater
> > > ---
> > > meta-oe/recipes-test/googletest/googletest_1.14.0.bb | 8 +++-
> > > 1 file changed, 7 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/meta-oe/recipes-test/googletest/googletest_1.14.0.bb
> > > b/meta-oe/recipes-test/googletest/googletest_1.14.0.bb
> > > index 8c75f936de..1b05f3d336 100644
> > > --- a/meta-oe/recipes-test/googletest/googletest_1.14.0.bb
> > > +++ b/meta-oe/recipes-test/googletest/googletest_1.14.0.bb
> > > @@ -10,11 +10,17 @@ S = "${WORKDIR}/git"
> > > SRCREV = "f8d7d77c06936315286eb55f8de22cd23c188571"
> > > SRC_URI =
> > "git://github.com/google/googletest.git;branch=main;protocol=https"
> > >
> > > -inherit cmake
> > > +inherit cmake pkgconfig
> > > +
> > > +# allow for shared libraries, but do not default to them #
> > > +PACKAGECONFIG[shared] = "-DBUILD_SHARED_LIBS=ON,-
> > DBUILD_SHARED_LIBS=OFF,,"
> > > +
> > >
> > > ALLOW_EMPTY:${PN} = "1"
> > > ALLOW_EMPTY:${PN}-dbg = "1"
> > >
> > > +ALLOW_EMPTY:${PN}-staticdev = "1"
> > > RDEPENDS:${PN}-dev += "${PN}-staticdev"
> > >
> > > BBCLASSEXTEND = "native nativesdk"
> > > --
> > > 2.25.1
> > >
> > >
> > >
> > >
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#109126):
https://lists.openembedded.org/g/openembedded-devel/message/109126
Mute This Topic: https://lists.openembedded.org/mt/104732413/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] [meta-oe][PATCH 1/1] googletest: allow for shared libraries
> -Original Message-
> From: Khem Raj
> Sent: Monday, March 4, 2024 3:50 PM
> To: Slater, Joseph
> Cc: openembedded-devel@lists.openembedded.org; MacLeod, Randy
>
> Subject: Re: [oe] [meta-oe][PATCH 1/1] googletest: allow for shared libraries
>
> On Mon, Mar 4, 2024 at 1:50 PM Joe Slater via lists.openembedded.org
> wrote:
> >
> > From: Joe Slater
> >
> > Add a PACKAGECONFIG option to produce shared libraries.
> > Allow staticdev to be empty if no static libraries are built.
> >
>
> why would be not build static libs ? and if we do not then why do we need to
> create empty package ?
If we set -DBUILD_SHARED_LIBS=ON static libs will not be built and -staticdev
will not be created, so -dev will produce an error if we try to, for example,
populate an sdk. I guess we could put static libs in -dev and get rid of
-staticdev, but I didn't want to mess with the packaging.
Joe
>
> > Signed-off-by: Joe Slater
> > ---
> > meta-oe/recipes-test/googletest/googletest_1.14.0.bb | 8 +++-
> > 1 file changed, 7 insertions(+), 1 deletion(-)
> >
> > diff --git a/meta-oe/recipes-test/googletest/googletest_1.14.0.bb
> > b/meta-oe/recipes-test/googletest/googletest_1.14.0.bb
> > index 8c75f936de..1b05f3d336 100644
> > --- a/meta-oe/recipes-test/googletest/googletest_1.14.0.bb
> > +++ b/meta-oe/recipes-test/googletest/googletest_1.14.0.bb
> > @@ -10,11 +10,17 @@ S = "${WORKDIR}/git"
> > SRCREV = "f8d7d77c06936315286eb55f8de22cd23c188571"
> > SRC_URI =
> "git://github.com/google/googletest.git;branch=main;protocol=https"
> >
> > -inherit cmake
> > +inherit cmake pkgconfig
> > +
> > +# allow for shared libraries, but do not default to them #
> > +PACKAGECONFIG[shared] = "-DBUILD_SHARED_LIBS=ON,-
> DBUILD_SHARED_LIBS=OFF,,"
> > +
> >
> > ALLOW_EMPTY:${PN} = "1"
> > ALLOW_EMPTY:${PN}-dbg = "1"
> >
> > +ALLOW_EMPTY:${PN}-staticdev = "1"
> > RDEPENDS:${PN}-dev += "${PN}-staticdev"
> >
> > BBCLASSEXTEND = "native nativesdk"
> > --
> > 2.25.1
> >
> >
> >
> >
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#109125):
https://lists.openembedded.org/g/openembedded-devel/message/109125
Mute This Topic: https://lists.openembedded.org/mt/104732413/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [oe] [meta-oe][PATCH] postgresql: upgrade 15.5 -> 16.2
this seems to be causing below failure with pgpool2
https://errors.yoctoproject.org/Errors/Details/755777/
On Mon, Mar 4, 2024 at 10:34 PM Changqing Li
wrote:
>
> From: Changqing Li
>
> License-Update: Update lincense year to 2024
>
> In version 16.2, ICU support is enabled by default, add PACKAGECONFIG
> icu to align with upstream, enable icu by default. And fix buildpaths
> QA warning.
>
> Signed-off-by: Changqing Li
> ---
> .../files/0001-Add-support-for-RISC-V.patch | 31 +--
> ...tch => 0002-Improve-reproducibility.patch} | 8 ++---
> ...-bypass-autoconf-2.69-version-check.patch} | 10 +++---
> ...config_info.c-not-expose-build-info.patch} | 20 ++--
> ...resql-fix-ptest-failure-of-sysviews.patch} | 15 ++---
> meta-oe/recipes-dbs/postgresql/postgresql.inc | 13 +++-
> .../recipes-dbs/postgresql/postgresql_15.5.bb | 16 --
> .../recipes-dbs/postgresql/postgresql_16.2.bb | 16 ++
> 8 files changed, 72 insertions(+), 57 deletions(-)
> rename
> meta-oe/recipes-dbs/postgresql/files/{0001-Improve-reproducibility.patch =>
> 0002-Improve-reproducibility.patch} (89%)
> rename
> meta-oe/recipes-dbs/postgresql/files/{0001-configure.ac-bypass-autoconf-2.69-version-check.patch
> => 0003-configure.ac-bypass-autoconf-2.69-version-check.patch} (75%)
> rename
> meta-oe/recipes-dbs/postgresql/files/{0001-config_info.c-not-expose-build-info.patch
> => 0004-config_info.c-not-expose-build-info.patch} (83%)
> rename
> meta-oe/recipes-dbs/postgresql/files/{0001-postgresql-fix-ptest-failure-of-sysviews.patch
> => 0005-postgresql-fix-ptest-failure-of-sysviews.patch} (71%)
> delete mode 100644 meta-oe/recipes-dbs/postgresql/postgresql_15.5.bb
> create mode 100644 meta-oe/recipes-dbs/postgresql/postgresql_16.2.bb
>
> diff --git
> a/meta-oe/recipes-dbs/postgresql/files/0001-Add-support-for-RISC-V.patch
> b/meta-oe/recipes-dbs/postgresql/files/0001-Add-support-for-RISC-V.patch
> index 7a4ba9897..34d34ecad 100644
> --- a/meta-oe/recipes-dbs/postgresql/files/0001-Add-support-for-RISC-V.patch
> +++ b/meta-oe/recipes-dbs/postgresql/files/0001-Add-support-for-RISC-V.patch
> @@ -1,21 +1,19 @@
> -From 780fd27ea6f7f2c446c46a7a5e26d94106c67efd Mon Sep 17 00:00:00 2001
> +From ba079b8d6a50796db41bb0ddf4c22bfe022ef898 Mon Sep 17 00:00:00 2001
> From: "Richard W.M. Jones"
> Date: Sun, 20 Nov 2016 15:04:52 +
> -Subject: [PATCH] Add support for RISC-V.
> +Subject: [PATCH 1/5] Add support for RISC-V.
>
> The architecture is sufficiently similar to aarch64 that simply
> extending the existing aarch64 macro works.
> ---
> -Upstream-Status: Pending
> -
> - src/include/storage/s_lock.h | 6 +++---
> - 1 file changed, 3 insertions(+), 3 deletions(-)
> + src/include/storage/s_lock.h | 5 +++--
> + 1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/src/include/storage/s_lock.h b/src/include/storage/s_lock.h
> -index 4d3ffc7..22e27bf 100644
> +index c9fa84c..9b491e8 100644
> --- a/src/include/storage/s_lock.h
> +++ b/src/include/storage/s_lock.h
> -@@ -317,11 +317,12 @@ tas(volatile slock_t *lock)
> +@@ -252,11 +252,12 @@ spin_delay(void)
>
> /*
>* On ARM and ARM64, we use __sync_lock_test_and_set(int *, int) if
> available.
> @@ -24,21 +22,20 @@ index 4d3ffc7..22e27bf 100644
>* We use the int-width variant of the builtin because it works on more
> chips
>* than other widths.
>*/
> --#if defined(__arm__) || defined(__arm) || defined(__aarch64__) ||
> defined(__aarch64)
> -+#if defined(__arm__) || defined(__arm) || defined(__aarch64__) ||
> defined(__aarch64) || defined(__riscv)
> +-#if defined(__arm__) || defined(__arm) || defined(__aarch64__)
> ++#if defined(__arm__) || defined(__arm) || defined(__aarch64__) ||
> defined(__riscv)
> #ifdef HAVE_GCC__SYNC_INT32_TAS
> #define HAS_TEST_AND_SET
>
> -@@ -355,8 +356,7 @@ spin_delay(void)
> +@@ -290,7 +291,7 @@ spin_delay(void)
>
> - #endif /* __aarch64__ || __aarch64 */
> + #endif /* __aarch64__ */
> #endif /* HAVE_GCC__SYNC_INT32_TAS */
> --#endif /* __arm__ || __arm || __aarch64__ || __aarch64 */
> --
> -+#endif /* __arm__ || __arm || __aarch64__ || __aarch64 || __riscv */
> +-#endif /* __arm__ || __arm || __aarch64__ */
> ++#endif /* __arm__ || __arm || __aarch64__ || __riscv */
> +
>
> /* S/390 and S/390x Linux (32- and 64-bit zSeries) */
> - #if defined(__s390__) || defined(__s390x__)
> --
> -2.34.1
> +2.25.1
>
> diff --git
> a/meta-oe/recipes-dbs/postgresql/files/0001-Improve-reproducibility.patch
> b/meta-oe/recipes-dbs/postgresql/files/0002-Improve-reproducibility.patch
> similarity index 89%
> rename from
> meta-oe/recipes-dbs/postgresql/files/0001-Improve-reproducibility.patch
> rename to
> meta-oe/recipes-dbs/postgresql/files/0002-Improve-reproducibility.patch
> index 02f4c9e51..b3e87cbc4 100644
> --- a/meta-oe/recipes-dbs/postgresql/files/0001-Improve-reproducibility.patch
> +++
[oe] [meta-oe][PATCH] toybox-inittab: Fix serial getty reproducibility
The enable_getty test uses a bash-only syntax "[[ ... ]]".
On /bin/sh==bash hosts, it enables the serial getty.
On /bin/sh!=bash hosts, it does not enable it.
Fix this non-reproducibility by using the standard shell "[ ... ]"
syntax which should work in any POSIX shell[0].
[0]: https://www.shellcheck.net/wiki/SC3010
Signed-off-by: Yoann Congal
---
meta-oe/recipes-core/toybox/toybox-inittab_0.8.2.bb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta-oe/recipes-core/toybox/toybox-inittab_0.8.2.bb
b/meta-oe/recipes-core/toybox/toybox-inittab_0.8.2.bb
index 603f365cf3..4cad7dce11 100644
--- a/meta-oe/recipes-core/toybox/toybox-inittab_0.8.2.bb
+++ b/meta-oe/recipes-core/toybox/toybox-inittab_0.8.2.bb
@@ -34,7 +34,7 @@ do_configure() {
cp orig/* .
for config in ${PACKAGECONFIG_CONFARGS}; do
- if [[ ${config} == "enable_getty" ]]; then
+ if [ "${config}" = "enable_getty" ]; then
echo "# generated by bitbake recipe ${PN}" >>
${S}/inittab
for console in "${SERIAL_CONSOLES}"; do
param=$(echo ${console} | sed s/\;/\ /g)
--
2.39.2
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#109121):
https://lists.openembedded.org/g/openembedded-devel/message/109121
Mute This Topic: https://lists.openembedded.org/mt/104746462/21656
Group Owner: openembedded-devel+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[oe] [meta-oe][PATCH 1/2] mozjs-115: update 115.6.0 -> 115.8.0
- python 3.12 support was added, remove the patch
Signed-off-by: Markus Volk
---
.../mozjs/mozjs-115/py3.12.patch | 2496 -
...js-115_115.6.0.bb => mozjs-115_115.8.0.bb} |3 +-
2 files changed, 1 insertion(+), 2498 deletions(-)
delete mode 100644 meta-oe/recipes-extended/mozjs/mozjs-115/py3.12.patch
rename meta-oe/recipes-extended/mozjs/{mozjs-115_115.6.0.bb =>
mozjs-115_115.8.0.bb} (96%)
diff --git a/meta-oe/recipes-extended/mozjs/mozjs-115/py3.12.patch
b/meta-oe/recipes-extended/mozjs/mozjs-115/py3.12.patch
deleted file mode 100644
index 732c8ff1c..0
--- a/meta-oe/recipes-extended/mozjs/mozjs-115/py3.12.patch
+++ /dev/null
@@ -1,2496 +0,0 @@
-From 7293cfae4fd68004901825ad1cabb83424d8729a Mon Sep 17 00:00:00 2001
-From: serge-sans-paille
-Date: Mon, 16 Oct 2023 13:35:24 +
-Subject: [PATCH] Bug 1857492 - Upgrade vendored version of six and urllib3
- r=saschanaz
-
-six -> 1.16
-urllib3 -> 1.26.17
-
-Differential Revision: https://phabricator.services.mozilla.com/D190288
-Upstream-Status: Backport
[https://github.com/mozilla/gecko-dev/commit/7293cfae4fd68004901825ad1cabb83424d8729a]
-Signed-off-by: Alexander Kanavin
- third_party/python/poetry.lock| 22 +--
- third_party/python/requirements.in| 4 +-
- third_party/python/requirements.txt | 12 +-
- .../python/six/six-1.13.0.dist-info/RECORD| 6 -
- .../LICENSE | 2 +-
- .../METADATA | 9 +-
- .../python/six/six-1.16.0.dist-info/RECORD| 6 +
- .../six-1.16.0.dist-info}/WHEEL | 2 +-
- .../top_level.txt | 0
- third_party/python/six/six.py | 91 ++---
- .../urllib3/urllib3-1.26.0.dist-info/RECORD | 44 -
- .../LICENSE.txt | 0
- .../METADATA | 177 --
- .../urllib3/urllib3-1.26.17.dist-info/RECORD | 44 +
- .../urllib3-1.26.17.dist-info}/WHEEL | 2 +-
- .../top_level.txt | 0
- .../python/urllib3/urllib3/__init__.py| 17 ++
- .../python/urllib3/urllib3/_version.py| 2 +-
- .../python/urllib3/urllib3/connection.py | 62 --
- .../python/urllib3/urllib3/connectionpool.py | 97 --
- .../contrib/_securetransport/bindings.py | 2 +-
- .../contrib/_securetransport/low_level.py | 1 +
- .../urllib3/urllib3/contrib/appengine.py | 4 +-
- .../urllib3/urllib3/contrib/ntlmpool.py | 13 +-
- .../urllib3/urllib3/contrib/pyopenssl.py | 19 +-
- .../urllib3/contrib/securetransport.py| 5 +-
- .../python/urllib3/urllib3/contrib/socks.py | 2 +-
- .../python/urllib3/urllib3/exceptions.py | 12 +-
- .../urllib3/urllib3/packages/__init__.py | 5 -
- .../packages/backports/weakref_finalize.py| 155 +++
- .../python/urllib3/urllib3/packages/six.py| 125 +
- .../packages/ssl_match_hostname/__init__.py | 22 ---
- .../python/urllib3/urllib3/poolmanager.py | 3 +-
- third_party/python/urllib3/urllib3/request.py | 21 +++
- .../python/urllib3/urllib3/response.py| 72 ++-
- .../python/urllib3/urllib3/util/connection.py | 5 +-
- .../python/urllib3/urllib3/util/proxy.py | 1 +
- .../python/urllib3/urllib3/util/request.py| 5 +-
- .../python/urllib3/urllib3/util/retry.py | 37 +++-
- .../python/urllib3/urllib3/util/ssl_.py | 53 --
- .../ssl_match_hostname.py}| 15 +-
- .../urllib3/urllib3/util/ssltransport.py | 6 +-
- .../python/urllib3/urllib3/util/timeout.py| 9 +-
- .../python/urllib3/urllib3/util/url.py| 17 +-
- .../python/urllib3/urllib3/util/wait.py | 1 -
- 45 files changed, 934 insertions(+), 275 deletions(-)
- delete mode 100644 third_party/python/six/six-1.13.0.dist-info/RECORD
- rename third_party/python/six/{six-1.13.0.dist-info =>
six-1.16.0.dist-info}/LICENSE (96%)
- rename third_party/python/six/{six-1.13.0.dist-info =>
six-1.16.0.dist-info}/METADATA (85%)
- create mode 100644 third_party/python/six/six-1.16.0.dist-info/RECORD
- rename third_party/python/{urllib3/urllib3-1.26.0.dist-info =>
six/six-1.16.0.dist-info}/WHEEL (70%)
- rename third_party/python/six/{six-1.13.0.dist-info =>
six-1.16.0.dist-info}/top_level.txt (100%)
- delete mode 100644 third_party/python/urllib3/urllib3-1.26.0.dist-info/RECORD
- rename third_party/python/urllib3/{urllib3-1.26.0.dist-info =>
urllib3-1.26.17.dist-info}/LICENSE.txt (100%)
- rename third_party/python/urllib3/{urllib3-1.26.0.dist-info =>
urllib3-1.26.17.dist-info}/METADATA (86%)
- create mode 100644 third_party/python/urllib3/urllib3-1.26.17.dist-info/RECORD
- rename third_party/python/{six/six-1.13.0.dist-info =>
urllib3/urllib3-1.26.17.dist-info}/WHEEL (70%)
- rename third_party/python/urllib3/{urllib3-1.26.0.dist-info =>
[oe] [meta-oe][PATCH 2/2] polkit: update 123 -> 124
- support for mozjs-115 was added, remove the patch
- update 0001-polkit.service.in-disable-MemoryDenyWriteExecute.patch
- include missing files to avoid:
ERROR: polkit-124-r0 do_package: QA Issue: polkit: Files/directories were
installed but not shipped in any package:
/usr/lib/pam.d
/usr/lib/sysusers.d
/usr/lib/pam.d/polkit-1
/usr/lib/sysusers.d/polkit.conf
Please set FILES such that these items are packaged. Alternatively if they are
unneeded, avoid installing them or delete them within do_install.
---
.../0001-jsauthority-Bump-mozjs-to-115.patch | 26 ---
...ce.in-disable-MemoryDenyWriteExecute.patch | 11 +++-
.../polkit/{polkit_123.bb => polkit_124.bb} | 13 +++---
3 files changed, 14 insertions(+), 36 deletions(-)
delete mode 100644
meta-oe/recipes-extended/polkit/polkit/0001-jsauthority-Bump-mozjs-to-115.patch
rename meta-oe/recipes-extended/polkit/{polkit_123.bb => polkit_124.bb} (89%)
diff --git
a/meta-oe/recipes-extended/polkit/polkit/0001-jsauthority-Bump-mozjs-to-115.patch
b/meta-oe/recipes-extended/polkit/polkit/0001-jsauthority-Bump-mozjs-to-115.patch
deleted file mode 100644
index 163a03cfc..0
---
a/meta-oe/recipes-extended/polkit/polkit/0001-jsauthority-Bump-mozjs-to-115.patch
+++ /dev/null
@@ -1,26 +0,0 @@
-From 2f0de2a831ab106fce210c1d65baef041256bc18 Mon Sep 17 00:00:00 2001
-From: Xi Ruoyao
-Date: Mon, 18 Sep 2023 01:53:04 +0800
-Subject: [PATCH] jsauthority: Bump mozjs to 115
-
-No code change is needed!
-
-Upstream-Status: Backport
[https://gitlab.freedesktop.org/polkit/polkit/-/commit/b340f50b7bb963863ede7c63f9a0b5c50c80c1e1]
-Signed-off-by: Alexander Kanavin
- meson.build | 2 +-
- 1 file changed, 1 insertion(+), 1 deletion(-)
-
-diff --git a/meson.build b/meson.build
-index 3b96562..92b68fd 100644
a/meson.build
-+++ b/meson.build
-@@ -153,7 +153,7 @@ if js_engine == 'duktape'
- func = 'pthread_condattr_setclock'
- config_h.set('HAVE_' + func.to_upper(), cc.has_function(func, prefix :
'#include '))
- elif js_engine == 'mozjs'
-- js_dep = dependency('mozjs-102')
-+ js_dep = dependency('mozjs-115')
-
- _system = host_machine.system().to_lower()
- if _system.contains('freebsd')
diff --git
a/meta-oe/recipes-extended/polkit/polkit/0001-polkit.service.in-disable-MemoryDenyWriteExecute.patch
b/meta-oe/recipes-extended/polkit/polkit/0001-polkit.service.in-disable-MemoryDenyWriteExecute.patch
index 9a097274a..4f008f7a9 100644
---
a/meta-oe/recipes-extended/polkit/polkit/0001-polkit.service.in-disable-MemoryDenyWriteExecute.patch
+++
b/meta-oe/recipes-extended/polkit/polkit/0001-polkit.service.in-disable-MemoryDenyWriteExecute.patch
@@ -1,4 +1,4 @@
-From 046d853818f18bac5df4dfc007151e06fd64a5b3 Mon Sep 17 00:00:00 2001
+From 95148a804be66092564f81306a02f625d5b8a5d0 Mon Sep 17 00:00:00 2001
From: Markus Volk
Date: Sun, 17 Sep 2023 23:26:59 +0200
Subject: [PATCH] polkit.service.in: disable MemoryDenyWriteExecute
@@ -16,11 +16,11 @@ Signed-off-by: Markus Volk
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/data/polkit.service.in b/data/polkit.service.in
-index 2113ff7..42dfd90 100644
+index e6db351..4390cce 100644
--- a/data/polkit.service.in
+++ b/data/polkit.service.in
-@@ -14,7 +14,7 @@ Group=@polkitd_user@
- IPAddressDeny=any
+@@ -12,7 +12,7 @@ ExecStart=@libprivdir@/polkitd --no-debug
+ User=@polkitd_user@
LimitMEMLOCK=0
LockPersonality=yes
-MemoryDenyWriteExecute=yes
@@ -28,6 +28,3 @@ index 2113ff7..42dfd90 100644
NoNewPrivileges=yes
PrivateDevices=yes
PrivateNetwork=yes
---
-2.41.0
-
diff --git a/meta-oe/recipes-extended/polkit/polkit_123.bb
b/meta-oe/recipes-extended/polkit/polkit_124.bb
similarity index 89%
rename from meta-oe/recipes-extended/polkit/polkit_123.bb
rename to meta-oe/recipes-extended/polkit/polkit_124.bb
index df9d25e9f..3eb0d5280 100644
--- a/meta-oe/recipes-extended/polkit/polkit_123.bb
+++ b/meta-oe/recipes-extended/polkit/polkit_124.bb
@@ -6,11 +6,10 @@ LIC_FILES_CHKSUM =
"file://COPYING;md5=155db86cdbafa7532b41f390409283eb"
SRC_URI =
"git://gitlab.freedesktop.org/polkit/polkit.git;protocol=https;branch=master \
file://0001-polkit.service.in-disable-MemoryDenyWriteExecute.patch \
- file://0001-jsauthority-Bump-mozjs-to-115.patch \
"
S = "${WORKDIR}/git"
-SRCREV = "fc8b07e71d99f88a29258cde99b913b44da1846d"
+SRCREV = "82f0924dc0eb23b9df68e88dbaf9e07c81940a5a"
DEPENDS = "expat glib-2.0"
@@ -51,4 +50,12 @@ do_install:append() {
chown polkitd:root ${D}/${sysconfdir}/polkit-1/rules.d
}
-FILES:${PN} += "${libdir}/polkit-1 ${nonarch_libdir}/polkit-1 ${datadir}"
+FILES:${PN} += " \
+ ${libdir}/pam.d/polkit-1 \
+ ${libdir}/sysusers.d \
+ ${libdir}/polkit-1 \
+ ${nonarch_libdir}/pam.d/polkit-1 \
+ ${nonarch_libdir}/sysusers.d \
+ ${nonarch_libdir}/polkit-1 \
+ ${datadir} \
+"
--
2.43.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this
Re: [oe] [meta-networking][PATCH] drbd-utils: Fix a udev rule reproducibility
Le mar. 5 mars 2024 à 07:49, Yoann CONGAL a écrit :
> Le mar. 5 mars 2024 à 00:48, Khem Raj a écrit :
>
>> which branch is it meant for ?
>>
>
> That was for the master branch but I missed the 9.27.0 update, sorry.
>
> I'll rebase and send a v2.
>
v2 sent!
Thanks!
>
> On Mon, Mar 4, 2024 at 12:13 PM Yoann Congal
>> wrote:
>> >
>> > drbd-utils configure step check the build host udev version to enable
>> > or disable the target udev rule. This leads to a clear
>> > non-reproducibility.
>> >
>> > This patch fixes this by adding a configure option to the configure step
>> > which allows to skip the udev version checks and unconditionally enable
>> > the udev rule.
>> >
>> > Signed-off-by: Yoann Congal
>> > ---
>> > ...-an-option-to-disable-host-udev-vers.patch | 55 +++
>> > .../recipes-support/drbd/drbd-utils_9.22.0.bb | 2 +
>> > 2 files changed, 57 insertions(+)
>> > create mode 100644
>> meta-networking/recipes-support/drbd/drbd-utils/0001-configure.ac-Add-an-option-to-disable-host-udev-vers.patch
>> >
>> > diff --git
>> a/meta-networking/recipes-support/drbd/drbd-utils/0001-configure.ac-Add-an-option-to-disable-host-udev-vers.patch
>> b/meta-networking/recipes-support/drbd/drbd-utils/0001-configure.ac-Add-an-option-to-disable-host-udev-vers.patch
>> > new file mode 100644
>> > index 00..5e03ff7b32
>> > --- /dev/null
>> > +++
>> b/meta-networking/recipes-support/drbd/drbd-utils/0001-configure.ac-Add-an-option-to-disable-host-udev-vers.patch
>> > @@ -0,0 +1,55 @@
>> > +From 50749a5b0eb0becb39a4edb5fe844b13bbeb5836 Mon Sep 17 00:00:00 2001
>> > +From: Yoann Congal
>> > +Date: Sat, 2 Mar 2024 16:23:07 +0100
>> > +Subject: [PATCH] configure.ac: Add an option to disable host udev
>> version
>> > + checks
>> > +
>> > +In cross-compilation environment, the build host might have an outdated
>> > +udev or no udev at all. But the user may still want to build with the
>> > +enabled udev rule (for its udev-enabled target).
>> > +
>> > +This patch adds a "--disable-udevchecks" option the disable build host
>> > +udev version check at configure-time and unconditionally install the
>> > +enabled udev rule. Without this new option, the behavior stays the same
>> > +(checks enabled).
>> > +
>> > +Signed-off-by: Yoann Congal
>> > +Upstream-Status: Submitted [
>> https://lists.linbit.com/pipermail/drbd-dev/2024-March/007201.html]
>> > +---
>> > + configure.ac | 9 +++--
>> > + 1 file changed, 7 insertions(+), 2 deletions(-)
>> > +
>> > +diff --git a/configure.ac b/configure.ac
>> > +index d5a0e733..da083240 100644
>> > +--- a/configure.ac
>> > b/configure.ac
>> > +@@ -71,6 +71,11 @@ AC_ARG_WITH([udev],
>> > + [AS_HELP_STRING([--with-udev],
>> > + [Enable udev integration])],
>> > + [WITH_UDEV=$withval])
>> > ++AC_ARG_ENABLE([udevchecks],
>> > ++ [AS_HELP_STRING([--disable-udevchecks],
>> > ++ [Disable host udev version checks])],
>> > ++ [],
>> > ++ [enable_udevchecks=yes])
>> > + AC_ARG_WITH([xen],
>> > + [AS_HELP_STRING([--with-xen],
>> > + [Enable Xen integration])],
>> > +@@ -309,7 +314,7 @@ if test -z $GIT; then
>> > +AC_MSG_WARN(Cannot update buildtag without git. You may safely
>> ignore this warning when building from a tarball.)
>> > + fi
>> > +
>> > +-if test $UDEVADM = false && test $UDEVINFO = false; then
>> > ++if test "x$enable_udevchecks" != "xno" && test $UDEVADM = false &&
>> test $UDEVINFO = false; then
>> > +if test "$WITH_UDEV" = "yes"; then
>> > + AC_MSG_WARN([udev support enabled, but neither udevadm nor
>> udevinfo found on this system.])
>> > +fi
>> > +@@ -413,7 +418,7 @@ dnl default.
>> > + test -z $INITDIR && INITDIR="$sysconfdir/init.d"
>> > +
>> > + dnl Our udev rules file is known to work only with udev >= 85
>> > +-if test "$WITH_UDEV" = "yes"; then
>> > ++if test "x$enable_udevchecks" != "xno" && test "$WITH_UDEV" = "yes";
>> then
>> > +udev_version=$( set -- $($UDEVADM version); echo $1 )
>> > +if test -z "$udev_version"; then
>> > + udev_version=$( set -- $($UDEVINFO -V); echo $3 )
>> > diff --git a/meta-networking/recipes-support/drbd/drbd-utils_9.22.0.bb
>> b/meta-networking/recipes-support/drbd/drbd-utils_9.22.0.bb
>> > index 960260fd39..65a43d9b26 100644
>> > --- a/meta-networking/recipes-support/drbd/drbd-utils_9.22.0.bb
>> > +++ b/meta-networking/recipes-support/drbd/drbd-utils_9.22.0.bb
>> > @@ -12,6 +12,7 @@ SRC_URI = "git://
>> github.com/LINBIT/drbd-utils;name=drbd-utils;branch=master;prot
>> > file://0001-drbdmon-add-LDFLAGS-when-linking.patch \
>> > file://0001-replace-off64_t-with-off_t.patch \
>> >
>>
>> ${@bb.utils.contains('DISTRO_FEATURES','usrmerge','file://0001-drbd-utils-support-usrmerge.patch','',d)}
>> \
>> > +
>> file://0001-configure.ac-Add-an-option-to-disable-host-udev-vers.patch \
>> > "
>> > SRC_URI:append:libc-musl = "
>>
[oe] [meta-networking][PATCH v2 2/2] drbd-utils: Fix a udev rule reproducibility
drbd-utils configure step check the build host udev version to enable
or disable the target udev rule. This leads to a clear
non-reproducibility.
This patch fixes this by adding a configure option to the configure step
which allows to skip the udev version checks and unconditionally enable
the udev rule.
Signed-off-by: Yoann Congal
---
v1->v2:
* Rebased on master (which updated to 9.27.0)
* Update patch upstream status as it was merged in the meantime
---
...-an-option-to-disable-host-udev-vers.patch | 55 +++
.../recipes-support/drbd/drbd-utils_9.27.0.bb | 2 +
2 files changed, 57 insertions(+)
create mode 100644
meta-networking/recipes-support/drbd/drbd-utils/0001-configure.ac-Add-an-option-to-disable-host-udev-vers.patch
diff --git
a/meta-networking/recipes-support/drbd/drbd-utils/0001-configure.ac-Add-an-option-to-disable-host-udev-vers.patch
b/meta-networking/recipes-support/drbd/drbd-utils/0001-configure.ac-Add-an-option-to-disable-host-udev-vers.patch
new file mode 100644
index 0..b98e1e70c
--- /dev/null
+++
b/meta-networking/recipes-support/drbd/drbd-utils/0001-configure.ac-Add-an-option-to-disable-host-udev-vers.patch
@@ -0,0 +1,55 @@
+From 61991db099f66348dddbc3408e7ee8c05bda85cb Mon Sep 17 00:00:00 2001
+From: Yoann Congal
+Date: Sat, 2 Mar 2024 16:23:07 +0100
+Subject: [PATCH] configure.ac: Add an option to disable host udev version
+ checks
+
+In cross-compilation environment, the build host might have an outdated
+udev or no udev at all. But the user may still want to build with the
+enabled udev rule (for its udev-enabled target).
+
+This patch adds a "--disable-udevchecks" option the disable build host
+udev version check at configure-time and unconditionally install the
+enabled udev rule. Without this new option, the behavior stays the same
+(checks enabled).
+
+Signed-off-by: Yoann Congal
+Upstream-Status: Backport
[https://github.com/LINBIT/drbd-utils/commit/f2b5a7335ae52109dfb95a9d99ae20519c43d59a]
+---
+ configure.ac | 9 +++--
+ 1 file changed, 7 insertions(+), 2 deletions(-)
+
+diff --git a/configure.ac b/configure.ac
+index f1d69ea3..4c56d33b 100644
+--- a/configure.ac
b/configure.ac
+@@ -71,6 +71,11 @@ AC_ARG_WITH([udev],
+ [AS_HELP_STRING([--with-udev],
+ [Enable udev integration])],
+ [WITH_UDEV=$withval])
++AC_ARG_ENABLE([udevchecks],
++ [AS_HELP_STRING([--disable-udevchecks],
++ [Disable host udev version checks])],
++ [],
++ [enable_udevchecks=yes])
+ AC_ARG_WITH([xen],
+ [AS_HELP_STRING([--with-xen],
+ [Enable Xen integration])],
+@@ -315,7 +320,7 @@ if test -z $GIT; then
+AC_MSG_WARN(Cannot update buildtag without git. You may safely ignore this
warning when building from a tarball.)
+ fi
+
+-if test $UDEVADM = false && test $UDEVINFO = false; then
++if test "x$enable_udevchecks" != "xno" && test $UDEVADM = false && test
$UDEVINFO = false; then
+if test "$WITH_UDEV" = "yes"; then
+ AC_MSG_WARN([udev support enabled, but neither udevadm nor udevinfo
found on this system.])
+fi
+@@ -423,7 +428,7 @@ else
+ test -z $INITDIR && INITDIR="$sysconfdir/init.d"
+
+ dnl Our udev rules file is known to work only with udev >= 85
+-if test "$WITH_UDEV" = "yes"; then
++if test "x$enable_udevchecks" != "xno" && test "$WITH_UDEV" = "yes"; then
+udev_version=$( set -- $($UDEVADM version); echo $1 )
+if test -z "$udev_version"; then
+ udev_version=$( set -- $($UDEVINFO -V); echo $3 )
diff --git a/meta-networking/recipes-support/drbd/drbd-utils_9.27.0.bb
b/meta-networking/recipes-support/drbd/drbd-utils_9.27.0.bb
index c99717839..63f0fd18c 100644
--- a/meta-networking/recipes-support/drbd/drbd-utils_9.27.0.bb
+++ b/meta-networking/recipes-support/drbd/drbd-utils_9.27.0.bb
@@ -11,6 +11,7 @@ SRC_URI =
"git://github.com/LINBIT/drbd-utils;name=drbd-utils;branch=master;prot
git://github.com/LINBIT/drbd-headers;name=drbd-headers;destsuffix=git/drbd-headers;branch=master;protocol=https
\
file://0001-drbdmon-add-LDFLAGS-when-linking.patch \
${@bb.utils.contains('DISTRO_FEATURES','usrmerge','file://0001-drbd-utils-support-usrmerge.patch','',d)}
\
+
file://0001-configure.ac-Add-an-option-to-disable-host-udev-vers.patch \
"
SRCREV_drbd-utils = "fdd9a4d603a9dc99d110d8bd0e288d7c0b6f586e"
SRCREV_drbd-headers = "0349f00825b4198d4ec3248f43884114a187676a"
@@ -40,6 +41,7 @@ EXTRA_OECONF = " \
--with-initscripttype=both\
--with-systemdunitdir=${systemd_unitdir}/system \
--without-manual \
+--disable-udevchecks \
"
# If we have inherited reproducible_build, we want to use it.
--
2.39.2
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#109115):
[oe] [meta-networking][PATCH v2 1/2] drbd-utils: Drop a duplicated line in DESCRIPTION
Line "DRBD mirrors a block device over the network to another machine" is written twice in DESCRIPTION. Signed-off-by: Yoann Congal --- meta-networking/recipes-support/drbd/drbd-utils_9.27.0.bb | 1 - 1 file changed, 1 deletion(-) diff --git a/meta-networking/recipes-support/drbd/drbd-utils_9.27.0.bb b/meta-networking/recipes-support/drbd/drbd-utils_9.27.0.bb index e1fd8e448..c99717839 100644 --- a/meta-networking/recipes-support/drbd/drbd-utils_9.27.0.bb +++ b/meta-networking/recipes-support/drbd/drbd-utils_9.27.0.bb @@ -1,6 +1,5 @@ SUMMARY = "Distributed block device driver for Linux" DESCRIPTION = "DRBD mirrors a block device over the network to another machine.\ -DRBD mirrors a block device over the network to another machine.\ Think of it as networked raid 1. It is a building block for\ setting up high availability (HA) clusters." HOMEPAGE = "http://www.drbd.org/; -- 2.39.2 -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#109114): https://lists.openembedded.org/g/openembedded-devel/message/109114 Mute This Topic: https://lists.openembedded.org/mt/104741300/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
