From: Jan Kiszka <jan.kiszka@siemens.com>
To: nobuhiro1.iwamatsu@toshiba.co.jp, cip-dev@lists.cip-project.org
Subject: Re: [isar-cip-core][PATCH 1/2] kas/opt: Add option for not install kernel
Date: Wed, 5 Mar 2025 06:38:08 +0100 [thread overview]
Message-ID: <143de2ca-60aa-472a-ac99-e10e24a7e882@siemens.com> (raw)
In-Reply-To: <TY7PR01MB14818077A700A2DEE3DB42BDC92CB2@TY7PR01MB14818.jpnprd01.prod.outlook.com>
On 05.03.25 05:45, nobuhiro1.iwamatsu@toshiba.co.jp wrote:
> Hi Jan,
>
>> -----Original Message-----
>> From: Jan Kiszka <jan.kiszka@siemens.com>
>> Sent: Saturday, March 1, 2025 12:40 AM
>> To: iwamatsu nobuhiro(岩松 信洋 ○DITC□DIT○OST)
>> <nobuhiro1.iwamatsu@toshiba.co.jp>; cip-dev@lists.cip-project.org
>> Subject: Re: [isar-cip-core][PATCH 1/2] kas/opt: Add option for not install
>> kernel
>>
>> On 28.02.25 13:46, Nobuhiro Iwamatsu wrote:
>>> This adds an option to create an image that does not contain the kernel.
>>>
>>
>> What is the use case? How does this relate to kas/opt/kernelci.yml?
>>
>
> This is not related kas/opt/kernelci.yml. This will be used to create rootfs for CIP/LAVA testing.
> The current recipe creates an image containing the kernel binary. Kernel binaries are not needed in
> rootfs for kernel testing. Therefore, we have added an option to remove it.
> This is also useful because some users may need rootfs without kernel binaries.
For my understanding: When is kernelci.yml used, when the kernel-free
image? For kernelci, we also deploy (not right now as upstream's API is
changing), but that was always with the kernel - why there?
>
>>> Signed-off-by: Nobuhiro Iwamatsu <nobuhiro1.iwamatsu@toshiba.co.jp>
>>> ---
>>> .gitlab-ci.yml | 28 ++++++++++++++++++++++-
>>> kas/opt/no_kernel.yml | 17 ++++++++++++++
>>> scripts/deploy-cip-core.sh | 47
>>> +++++++++++++++++++++++++++-----------
>>> 3 files changed, 78 insertions(+), 14 deletions(-) create mode
>>> 100644 kas/opt/no_kernel.yml
>>>
>>> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml index 4df521b..a09eedd
>>> 100644
>>> --- a/.gitlab-ci.yml
>>> +++ b/.gitlab-ci.yml
>>> @@ -27,6 +27,7 @@ variables:
>>> build_swu_v2: disable
>>> swupdate_version: default
>>> test_function: swupdate
>>> + no_kernel: disable
>>>
>>> stages:
>>> - build
>>> @@ -67,9 +68,10 @@ default:
>>> - if [ "${watchdog}" = "disable" ]; then
>> base_yaml="${base_yaml}:kas/opt/disable-watchdog.yml"; fi
>>> - if [ "${security_test}" = "enable" ]; then
>> base_yaml="${base_yaml}:kas/opt/security_test.yml"; fi
>>> - if [ "${swupdate_version}" = "2022.12" ]; then
>>> base_yaml="${base_yaml}:kas/opt/swupdate-2022.12.yaml"; fi
>>> + - if [ "${no_kernel}" = "enable" ]; then
>>> + base_yaml="${base_yaml}:kas/opt/no_kernel.yml"; fi
>>> - echo "Building ${base_yaml}"
>>> - kas build ${base_yaml}
>>> - - if [ "${deploy}" = "enable" ]; then scripts/deploy-cip-core.sh
>> ${release} ${target} ${extension} ${dtb} ${CI_COMMIT_REF_SLUG} wic; fi
>>> + - if [ "${deploy}" = "enable" ]; then scripts/deploy-cip-core.sh
>>> + ${release} ${target} ${extension} ${no_kernel} ${dtb}
>>> + ${CI_COMMIT_REF_SLUG} wic; fi
>>> - if [ "${build_swu_v2}" = "enable" ]; then
>>> mkdir build/previous-image;
>>> if [ "${extension}" = "security" ] || [ "${extension}" =
>>> "ebg-secure-boot-snakeoil" ]; then @@ -221,6 +223,30 @@
>> build:qemu-arm-test:
>>> target: qemu-arm
>>> extension: test
>>>
>>> +build:qemu-amd64-test-nokernel:
>>> + extends:
>>> + - .build_base
>>> + variables:
>>> + target: qemu-amd64
>>> + extension: test
>>> + no_kernel: enable
>>> +
>>> +build:qemu-arm64-test-nokernel:
>>> + extends:
>>> + - .build_base
>>> + variables:
>>> + target: qemu-arm64
>>> + extension: test
>>> + no_kernel: enable
>>> +
>>> +build:qemu-arm-test-nokernel:
>>> + extends:
>>> + - .build_base
>>> + variables:
>>> + target: qemu-arm
>>> + extension: test
>>> + no_kernel: enable
>>> +
>>
>> Do we always need to build and deploy those extra images?
>
> It will not be necessary every time.
> It will need to be built when LTP is updated or when a point release is had in Debian.
> The current gitlab pipeline is not that well thought out and I think it needs to be considered.
>
If we do not pick up the artifacts automatically from a pipeline build,
we can make these variants also manually triggered. But I still need to
understand the apparent overlap with the kernelci builds (which happen
unconditionally so far).
BTW, where is the riscv64 variant?
>>
>>> build:x86-uefi-secure-boot:
>>> extends:
>>> - .build_base
>>> diff --git a/kas/opt/no_kernel.yml b/kas/opt/no_kernel.yml new file
>>> mode 100644 index 0000000..2338177
>>> --- /dev/null
>>> +++ b/kas/opt/no_kernel.yml
>>> @@ -0,0 +1,17 @@
>>> +#
>>> +# CIP Core, generic profile
>>> +#
>>> +# Copyright (c) Toshiba Corporation 2025 # # Authors:
>>> +# Nobuhiro Iwamatsu <nobuhiro1.iwamatsu@toshiba.co.jp>
>>> +#
>>> +# SPDX-License-Identifier: MIT
>>> +#
>>> +
>>> +header:
>>> + version: 14
>>> +
>>> +local_conf_header:
>>> + kernel-version: |
>>> + IMAGE_INSTALL:remove = "linux-image-cip linux-image-cip-rt"
>>
>> KERNEL_NAME = "", that will also disable the now pointless build of the
>> kernel.
>
> Thank you.
> This is simpler and better.
>
> Best regards,
> Nobuhiro
--
Siemens AG, Foundational Technologies
Linux Expert Center
next prev parent reply other threads:[~2025-03-05 5:38 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-02-28 12:46 [isar-cip-core][PATCH 1/2] kas/opt: Add option for not install kernel Nobuhiro Iwamatsu
2025-02-28 12:46 ` [isar-cip-core][PATCH 2/2] Kconfig: Add config NO_KERNEL for kernel selection Nobuhiro Iwamatsu
2025-02-28 15:41 ` Jan Kiszka
2025-03-05 4:53 ` [cip-dev] " nobuhiro1.iwamatsu
2025-02-28 15:40 ` [isar-cip-core][PATCH 1/2] kas/opt: Add option for not install kernel Jan Kiszka
2025-03-05 4:45 ` nobuhiro1.iwamatsu
2025-03-05 5:38 ` Jan Kiszka [this message]
2025-03-06 5:54 ` [cip-dev] " nobuhiro1.iwamatsu
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=143de2ca-60aa-472a-ac99-e10e24a7e882@siemens.com \
--to=jan.kiszka@siemens.com \
--cc=cip-dev@lists.cip-project.org \
--cc=nobuhiro1.iwamatsu@toshiba.co.jp \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox