From: Jan Kiszka <jan.kiszka@siemens.com>
To: "Schmidt, Adriaan (T CED SES-DE)" <adriaan.schmidt@siemens.com>,
"Anton Mikanovich" <amikan@ilbers.de>,
"isar-users@googlegroups.com" <isar-users@googlegroups.com>
Subject: Re: [PATCH 2/2] u-boot-script: Split packages by kernel
Date: Wed, 12 Jan 2022 14:59:29 +0100 [thread overview]
Message-ID: <8860aaba-abd6-303b-6bf2-2d540f7fc6b3@siemens.com> (raw)
In-Reply-To: <AS4PR10MB5318B5B6AAA4015628169CC0ED529@AS4PR10MB5318.EURPRD10.PROD.OUTLOOK.COM>
On 12.01.22 14:51, Schmidt, Adriaan (T CED SES-DE) wrote:
> Anton Mikanovich, Mittwoch, 12. Januar 2022 14:35:
>> 11.01.2022 18:56, Jan Kiszka wrote:
>>> On 11.01.22 16:05, Anton Mikanovich wrote:
>>>> 11.01.2022 15:03, Jan Kiszka wrote:
>>>>> It makes no sense to build the same package with the same content
>>>>> multiple times. We likely rather need to make it depend on a generic
>>>>> kernel package / create a generic kernel package name.
>>>>>
>>>>> Jan
>>>>
>>>> I'm afraid it will not help - moving this dependency into another
>>>> package will also move the issue into this new package.
>>>>
>>>
>>> It will. We just have to make u-boot-script depend on a generic kernel
>>> so that this package is only installed after some kernel - no matter
>>> which one - is already installed on the target image. There is NO
>>> dependency on the concrete kernel in this packages, just study it.
>>>
>>> Jan
>>
>> Can you please provide any details about how it can be implemented?
>> We have linux-image-generic package in Ubuntu, but not in Debian. And we
>> can't just create it, because of the same deps conflict issue in Isar.
>> I understood u-boot-script depends on any kernel, but we need to declare
>> this 'any' somehow.
>
> OE has the concept of virtual providers:
> https://www.yoctoproject.org/docs/current/mega-manual/mega-manual.html#metadata-virtual-providers
> And it looks like the PREFERRED_PROVIDER_* variables are a bitbake feature, so it shouldn't be too hard to do something like this in Isar.
>
> @Jan: I assume this is what you had in mind?
Yes and no: Yes, we would need just a provider, no, his won't help us as
we need it as runtime dependency, and Debian kernel do not expose that.
See my other reply.
Jan
--
Siemens AG, Technology
Competence Center Embedded Linux
next prev parent reply other threads:[~2022-01-12 13:59 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-01-11 9:36 [PATCH 0/2] Fix u-boot-script packages conflict Anton Mikanovich
2022-01-11 9:36 ` [PATCH 1/2] dpkg: Allow parallel build of different package versions Anton Mikanovich
2022-01-11 9:48 ` Henning Schild
2022-01-11 9:36 ` [PATCH 2/2] u-boot-script: Split packages by kernel Anton Mikanovich
2022-01-11 9:50 ` Henning Schild
2022-01-11 12:03 ` Jan Kiszka
2022-01-11 15:05 ` Anton Mikanovich
2022-01-11 15:56 ` Jan Kiszka
2022-01-12 13:34 ` Anton Mikanovich
2022-01-12 13:51 ` Schmidt, Adriaan
2022-01-12 13:59 ` Jan Kiszka [this message]
2022-01-12 13:58 ` Jan Kiszka
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=8860aaba-abd6-303b-6bf2-2d540f7fc6b3@siemens.com \
--to=jan.kiszka@siemens.com \
--cc=adriaan.schmidt@siemens.com \
--cc=amikan@ilbers.de \
--cc=isar-users@googlegroups.com \
/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