From: Jermain Horsman
get_layer_revisions() uses the os.path module but it is not included
in the imports.
Signed-off-by: Jermain Horsman
---
meta/lib/oe/buildcfg.py | 1 +
1 file changed, 1 insertion(+)
diff --git a/meta/lib/oe/buildcfg.py b/meta/lib/oe/buildcfg.py
index 90f5e05715
From: Jermain Horsman
Several functions included the 'd' parameter but never used it,
additionally the value passed is always None.
Signed-off-by: Jermain Horsman
---
meta/lib/oe/buildcfg.py | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/meta/lib/oe/buildcfg.py
From: Jermain Horsman
During development it can be useful to use a branch instead of a
revision, e.g. when rebasing often or just for easy upgrading of
the setup.
This is a first iteration, it updates the generation of the
setup-layers.json file as well as the json schema for it.
This allows
From: Jermain Horsman
Having a default remote present allows for the checkout of a
branch in the case multiple remotes are available.
This requires an update to the json-schema and associated checks.
Signed-off-by: Jermain Horsman
---
meta/files/layers.example.json| 5
From: Jermain Horsman
This allows for modifications of an existing configuration file
using current revisions.
Signed-off-by: Jermain Horsman
---
meta/lib/bblayers/updatesetup.py | 78
1 file changed, 78 insertions(+)
create mode 100644 meta/lib/bblayers
From: Jermain Horsman
The test creates a layer setup, modifies it to use a branch,
attempts a checkout using the modified file, it modifies the
layers setup again to use a revision restoring it to the
original state.
A final checkout is skipped as the revision might not be
present, additionally
From: Jermain Horsman
These changes allow for situations where one or more layers are checked out
using a branch instead of a revision, care is taken to make sure this works
when using multiple remotes.
All changes made are backwards compatible with older setup-layer json files.
Signed-off
From: Jermain Horsman
Includes a git describe and git toplevel function and functions
to get info for git remotes.
Signed-off-by: Jermain Horsman
---
meta/lib/oe/buildcfg.py | 29 +
1 file changed, 29 insertions(+)
diff --git a/meta/lib/oe/buildcfg.py b/meta/lib
From: Jermain Horsman
Refactor to use git functions from library.
Signed-off-by: Jermain Horsman
---
meta/lib/bblayers/makesetup.py | 32 ++--
1 file changed, 10 insertions(+), 22 deletions(-)
diff --git a/meta/lib/bblayers/makesetup.py b/meta/lib/bblayers
From: Jermain Horsman
Includes a function to determine the default remote.
Signed-off-by: Jermain Horsman
---
meta/lib/oe/buildcfg.py | 16
1 file changed, 16 insertions(+)
diff --git a/meta/lib/oe/buildcfg.py b/meta/lib/oe/buildcfg.py
index f9cb0cc74c..2ca654e56e 100644
From: Jermain Horsman
Previously _is_repo_git_repo() would return a result containing b'true\n' or
b'false\n' if 'git rev-parse' ran successfully, instead of True of False.
While this can be solved using e.g. result.strip().decode("utf-8") == "true",
there are some oth
From: Jermain Horsman
These changes allow for situations where one or more layers are checked out
using a branch instead of a revision, care is taken to make sure this works
when using multiple remotes.
All changes made are backwards compatible with older setup-layer json files.
Signed-off
I was mostly looking for some quick feedback,
As a proof of concept, this was easily enough testable on my end.
However before spending too much time on the actual implementation,
I wanted to know if this was roughly the right direction.
Sincerely,
Jermain Horsman
-Original Message
From: Jermain Horsman
The writer always printed the script was written even if this
was not the case when specififying '--json-only'.
Signed-off-by: Jermain Horsman
---
meta/lib/bblayers/setupwriters/oe-setup-layers.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta
mportant to make sure that it is.
Sincerely,
Jermain Horsman
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#190353):
https://lists.openembedded.org/g/openembedded-core/message/190353
Mute This Topic: https://lists.openembedded.org/mt/102444605/21656
Group Owner:
be used if we decide to
use remote/branch.
Sincerely,
Jermain Horsman
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#190352):
https://lists.openembedded.org/g/openembedded-core/message/190352
Mute This Topic: https://lists.openembedded.org/mt/
From: Jermain Horsman
When building an image cve_check_write_rootfs_manifest() would sometimes fail
with a FileNotFoundError when writing the manifest.cve due to the parent
directory (DEPLOY_DIR_IMAGE) not (yet) existing.
The image task will provide the manifest in the deploy directory
From: Jermain Horsman
When building an image cve_check_write_rootfs_manifest() would sometimes fail
with a FileNotFoundError when writing the manifest.cve due to the parent
directory (DEPLOY_DIR_IMAGE) not (yet) existing.
The image task will provide the manifest in the deploy directory
included as it is?
Sincerely,
Jermain Horsman
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#178340):
https://lists.openembedded.org/g/openembedded-core/message/178340
Mute This Topic: https://lists.openembedded.org/mt/97522170/21656
Group Owner: openembedd
Would it be appropriate to backport this to dunfell too?
If so, should I send in a patch? I'm not entirely sure what the general process
is in these cases.
Sincerely,
Jermain Horsman
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#178481
that disable the systemd pager
when busybox `less` is detected.
Signed-off-by: Alban Bedel
Signed-off-by: Richard Purdie
(cherry picked from commit de7e36a7858ebca4615975967fcad1c399eacdb0)
Signed-off-by: Jermain Horsman
---
meta/recipes-core/systemd/systemd/systemd-pager.sh | 7 +++
meta/recipes
on dunfell (using its version of systemd) and as far
as I can tell it works there too.
Would it be alright to send this patch in for dunfell, since there does not
seem to be any conflict, but it normally does not apply to it either?
Sincerely,
Jermain Horsman
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive
the "After=" discussion, "After" should only affect the ordering of jobs
that are scheduled, but not actually start any by itself.
Therefore it should be fairly save to include this patch.
Sincerely,
Jermain Horsman
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to th
helpful and isn't what most
applications expect.
Signed-off-by: Richard Purdie
(cherry picked from commit e730d005fa8aec07f9ae25c58d4566eaa92a6997)
Signed-off-by: Jermain Horsman
---
meta/recipes-core/base-files/base-files/hosts | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
On Thu, May 4, 2023 at 02:12 AM, Khem Raj wrote:
>
> should it depend on ldconfig being in distro features ? it may not be
> there on all systems.
>
Do you mean we should add it conditionally based upon the value of distro
features?
Or if it is required to have ldconfig included in distro
From: Jermain Horsman
The first time logrotate runs it reports an error:
error: state file /var/lib/logrotate.status is
world-readable and thus can be locked from other
unprivileged users. Skipping lock acquisition...
This check was added with
https://github.com/logrotate/logrotate
af38a39b4bed9e43c5075008be47ca72191a489e)
Signed-off-by: Jermain Horsman
---
meta/recipes-connectivity/openssh/openssh/sshd.socket | 1 +
meta/recipes-connectivity/openssh/openssh/sshd@.service | 2 --
2 files changed, 1 insertion(+), 2 deletions(-)
diff --git a/meta/recipes-connectivity/openssh/openssh/sshd.socket
b/meta
From: Jermain Horsman
The first time logrotate runs it reports an error:
error: state file /var/lib/logrotate.status is
world-readable and thus can be locked from other
unprivileged users. Skipping lock acquisition...
This check was added with
https://github.com/logrotate/logrotate
at the same time.
Signed-off-by: Arturo Buzarra
Signed-off-by: Richard Purdie
(cherry picked from commit 4e9d812e127dc6743f52f4881e509e8e2e833afe)
Signed-off-by: Jermain Horsman
---
.../run-postinsts/run-postinsts/run-postinsts.service | 2 +-
1 file changed, 1 insertion(+), 1 deletion
at the same time.
Signed-off-by: Arturo Buzarra
Signed-off-by: Richard Purdie
(cherry picked from commit 4e9d812e127dc6743f52f4881e509e8e2e833afe)
Signed-off-by: Jermain Horsman
---
.../run-postinsts/run-postinsts/run-postinsts.service | 2 +-
1 file changed, 1 insertion(+), 1 deletion
a bit of a
different direction,
as a result I did not change the file that used these functions, and therefore
did not include this patch.
I'd be more than happy to send this as a standalone though.
Sincerely,
Jermain Horsman
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View
,
so not too much progress since, however, I have reserved some
time this month and was planning to work on this.
Sincerely,
Jermain Horsman
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#191885):
https://lists.openembedded.org/g/openembedded-
tion like '--use-custom-reference-all' seems undesirable to me.
Sincerely,
Jermain Horsman
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193349):
https://lists.openembedded.org/g/openembedded-core/message/193349
Mute This Top
From: Jermain Horsman
This includes a simple test which creates a layer setup using
custom references, and subsequently modifies the resulting layers
setup using a different custom reference.
Signed-off-by: Jermain Horsman
---
meta/lib/oeqa/selftest/cases/bblayers.py | 53
From: Jermain Horsman
This uses an existing setup-layers configuration and modifies one
or more repositories using a reference provided by the user.
This is a very minimal implementation, no validation of this reference
is done and it is left to the user to provide a valid value.
Signed-off
From: Jermain Horsman
During development it can be useful to use a branch instead of a
revision, e.g. when rebasing often or just for easy upgrading of
the setup.
This iteration has a very minimal implementation, which covers many
use cases, but certainly not all.
As previously discussed
> Is comma (,) as a separator okay? Maybe : is better?
In my opinion a comma suggest a list, whereas a semicolon
suggests a relation, so I've opted to use ':'.
I did implement a new version, it is available on the mailing list,
it should include all other feedback as well.
Sincerely,
Jerm
From: Jermain Horsman
This includes a simple test which creates a layer setup using
custom references, and subsequently modifies the resulting layers
setup using a different custom reference.
Signed-off-by: Jermain Horsman
---
meta/lib/oeqa/selftest/cases/bblayers.py | 53
From: Jermain Horsman
This creates a new layers setup with, or, modifies an existing layers
setup using, one or more repositories where the references are provided
by the user.
This is a very minimal implementation, no validation of any reference
is done and it is left to the user to provide
From: Jermain Horsman
Signed-off-by: Jermain Horsman
---
meta/lib/bblayers/makesetup.py | 2 --
1 file changed, 2 deletions(-)
diff --git a/meta/lib/bblayers/makesetup.py b/meta/lib/bblayers/makesetup.py
index 5fb6f1469e..4f27c565ee 100644
--- a/meta/lib/bblayers/makesetup.py
+++ b/meta/lib
From: Jermain Horsman
Signed-off-by: Jermain Horsman
---
meta/lib/bblayers/buildconf.py | 7 ---
1 file changed, 7 deletions(-)
diff --git a/meta/lib/bblayers/buildconf.py b/meta/lib/bblayers/buildconf.py
index ccab332adf..87a5e5a55b 100644
--- a/meta/lib/bblayers/buildconf.py
+++ b/meta
From: Jermain Horsman
This includes a simple test which creates a layer setup,
modifies it using a test string and checks if the resulting
layer setup is written correctly.
Signed-off-by: Jermain Horsman
---
meta/lib/oeqa/selftest/cases/bblayers.py | 26
1 file
From: Jermain Horsman
This uses an existing setup-layers configuration and modifies one
or more repositories using a reference provided by the user.
This is a very minimal implementation, no validation of this reference
is done and it is left to the user to provide a valid value.
Signed-off
From: Jermain Horsman
During development it can be useful to use a branch instead of a
revision, e.g. when rebasing often or just for easy upgrading of
the setup.
This iteration has a very minimal implementation, which covers many
use cases, but certainly not all.
As previously discussed
y' as I would like to use '--update' with
'--repository', without '--reference'
at some point, in which case '--use-custom-reference' does not seem appropriate.
I'm open to suggestions and I'd like to know what people think about this.
Sincerely,
Jermain Horsman
-=-=-=-=-=-=-=-=-=-=-=-
Links: You rece
From: Jermain Horsman
This uses an existing setup-layers configuration and modifies one
or more repositories using a reference provided by the user.
This is a very minimal implementation, no validation of this reference
is done and it is left to the user to provide a valid value.
Signed-off
From: Jermain Horsman
This includes a simple test which creates a layer setup,
modifies it using a test string and checks if the resulting
layer setup is written correctly.
Signed-off-by: Jermain Horsman
---
meta/lib/oeqa/selftest/cases/bblayers.py | 26
1 file
From: Jermain Horsman
During development it can be useful to use a branch instead of a
revision, e.g. when rebasing often or just for easy upgrading of
the setup.
This iteration has a very minimal implementation, which covers many
use cases, but certainly not all.
As previously discussed
From: Jermain Horsman
I ran into the following when trying to compile rust for arm (on kirkstone):
Exception: TypeError: int() argument must be a string, a bytes-like object or a
number, not 'NoneType'
After some investigation I notices some patches on master which were not
available
4aa4c1257c)
Signed-off-by: Jermain Horsman
---
meta/recipes-devtools/rust/rust-common.inc | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/meta/recipes-devtools/rust/rust-common.inc
b/meta/recipes-devtools/rust/rust-common.inc
index db0bd8fc1b..ac00fd3d43 100644
--- a/me
1912c4e9e0ecf9655f3b3a41588b54d7956f5899)
Signed-off-by: Jermain Horsman
---
meta/recipes-devtools/rust/rust-cross-canadian-common.inc | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/meta/recipes-devtools/rust/rust-cross-canadian-common.inc
b/meta/recipes-devtools/rust/rust-cross-canadian
Purdie
(cherry picked from commit ff3c3dbbd2bf1bb7bb70b55cca203e9eedcf14a8)
Signed-off-by: Jermain Horsman
---
meta/recipes-devtools/rust/rust-common.inc | 6 ++
meta/recipes-devtools/rust/rust-cross.inc | 7 +--
2 files changed, 7 insertions(+), 6 deletions(-)
diff --git a/meta/recipes
From: Richard Purdie
Match the code in rust-cross-canadian so that further simplifications
can be considered in future.
Signed-off-by: Richard Purdie
(cherry picked from commit 5cb62d0a33777cb9afb1eea5f736a2580ce50dc7)
Signed-off-by: Jermain Horsman
---
meta/recipes-devtools/rust/rust
from commit f8edeead263708889d31a7ff578ef8274cb678b4)
Signed-off-by: Jermain Horsman
---
meta/classes/native.bbclass | 2 ++
1 file changed, 2 insertions(+)
diff --git a/meta/classes/native.bbclass b/meta/classes/native.bbclass
index 4de96cd59b..56726301bd 100644
--- a/meta/classes/native.bbclass
> If recipe A requires the useradd actions of recipe B we need to
> ensure that recipe B is part of the recipe A dependancy chain. In
> order to do that, we introduce USERADD_DEPENDS. This makes sure
> that the do_populate_sysroot_setscene of recipe B exists for
> recipe A in case of a missing
Purdie
(cherry picked from commit ff3c3dbbd2bf1bb7bb70b55cca203e9eedcf14a8)
Signed-off-by: Jermain Horsman
---
meta/recipes-devtools/rust/rust-common.inc | 6 ++
meta/recipes-devtools/rust/rust-cross.inc | 7 +--
2 files changed, 7 insertions(+), 6 deletions(-)
diff --git a/meta/recipes
From: Richard Purdie
Match the code in rust-cross-canadian so that further simplifications
can be considered in future.
Signed-off-by: Richard Purdie
(cherry picked from commit 5cb62d0a33777cb9afb1eea5f736a2580ce50dc7)
Signed-off-by: Jermain Horsman
---
meta/recipes-devtools/rust/rust
From: Jermain Horsman
I ran into the following when trying to compile rust for arm (on kirkstone):
Exception: TypeError: int() argument must be a string, a bytes-like object or a
number, not 'NoneType'
After some investigation I notices some patches on master which were not
available
-Original Message-
From: Alexander Kanavin
Sent: Tuesday, November 14, 2023 2:28 PM
> I'm fine with that. Can you tweak the tools?
Yeah I will, just need some time to work on this,
so it might take a bit of time.
Sincerely,
Jermain Horsman
-=-=-=-=-=-=-=-=-=-=-=-
Links: You rece
From: Jermain Horsman
This allows other classes to make use of these as well.
Includes a git describe and git toplevel function and functions
to get info for git remotes.
Signed-off-by: Jermain Horsman
---
Cc: Martin Jansa
meta/lib/bblayers/makesetup.py | 33
ot;; then
> + if test "x$1" = "xgroupmems"; then
As these are quotes variables the added 'x' doesn't really add anything here.
> + if test "x$1" = "x"; then
Same here, but this could just test for a zero length variable instead using
'test -z
61 matches
Mail list logo
