public inbox for isar-users@googlegroups.com
 help / color / mirror / Atom feed
From: Jan Kiszka <jan.kiszka@siemens.com>
To: Alexander Smirnov <asmirnov@ilbers.de>,
	isar-users <isar-users@googlegroups.com>
Subject: Re: [PATCH] Factor out meta-examples
Date: Wed, 24 Jan 2018 13:15:07 +0100	[thread overview]
Message-ID: <3e2ad3a4-9631-80bd-7c75-315d628b8d5c@siemens.com> (raw)
In-Reply-To: <9fa572d2-8a59-b3b3-ab96-451975c72235@ilbers.de>

On 2018-01-24 11:04, Alexander Smirnov wrote:
> On 01/24/2018 10:56 AM, Jan Kiszka wrote:
>> On 2018-01-24 08:52, Jan Kiszka wrote:
>>> On 2018-01-24 08:49, Alexander Smirnov wrote:
>>>> Hi,
>>>>
>>>> On 01/24/2018 10:34 AM, Jan Kiszka wrote:
>>>>> From: Jan Kiszka <jan.kiszka@siemens.com>
>>>>>
>>>>> Move recipes that have exemplary character (currently everything under
>>>>> recipes-app) into a separate layer that can easily be dropped when
>>>>> building real images. This also helps to ensure that no assumptions
>>>>> about customizations and core recipes sharing the same layer sneak in
>>>>> again.
>>>>>
>>>>> The meta-isar layer is still not completely clean as we keep
>>>>> bblayers.conf.sample and local.conf.sample prepared for building
>>>>> example
>>>>> images.
>>>>>
>>>>
>>>> Isar core is in 'meta' layer. Current 'meta-isar' is the example
>>>> itself,
>>>> it's not intended to be used in real projects, it's a kind of template
>>>> for your 'meta-product'.
>>>
>>> This is the old view, it's no longer correct. meta-isar has become
>>> already a core layer that products build upon. We need to fix this now.
>>
>> BTW, we should eventually discuss the folding of meta and meta-isar or a
>> clarification what should go where. For now, I will be add the reusable
>> recipes for custom kernel builds into meta-isar, but it's not really
>> intuitive yet.
>>
>> We definitely must stop proposing users to fork isar in order to build
>> own images. That is a completely obsolete and counterproductive pattern.
>>
> 
> As I see this patch is just a small step to new direction, but I have no
> idea what this direction is itself, it wasn't discussed here.
> 
> Could you please describe the vision you assume to see finally? What is
> Isar core, what is example?

The direction is not really new: do not tough isar for own images,
create "product" layers - simply the OE model. That's the key point of
using bitbake and layers.

> 
> BTW, below is my assumption what could be reasonable regarding Isar
> future layout:
> 
> 1. 'meta' contains:
>  - Isar classes
>  - buildchroot
>  - caching/reproducibility core;
> 
> 2. meta-isar contains:
>  - distro: debian-wheezy, debian-jessie, debian-stretch
>  - machine: qemuarm, qemuamd64, qemui386
>  - images: isar-image-base (minimal multirstap image)
>  - No local.conf, bblayers.conf etc...
> 
> 3. meta-isar-example:
>  - distro: raspbian-jessie
>  - machine: rpi
>  - images: isar-image-example
>  - apps: hello, example-raw
>  - local.conf, bblayers.conf etc...

I specifically like the structure or meta[-isar]-example. Still, I have
problems with the separation of meta and meta-isar.

Could you describe a use case where one would only pull meta? If there
is one, we will also have to think about how to ensure that no
dependencies from meta on meta-isar prevent this (test strategy).

> 
> With this layout I assume that:
> 
> 1. Debian is the reference Isar distro and it's included to Isar core by
> default.
> 
> 2. Debian will be the most popular distro for customer projects.
> 
> 3. To create new product, the user needs to create meta-prodcut with:
>  - local.conf, bblayers.conf
>  - isar-image-base.bbappend with custom packages set

Right.

> 
> This layout will help to hide back-end magic like multistrap from user.
> So if, for example, we will switch from multistrap to debootstrap,
> 'meta-product' will work as it is.

Exactly. And it avoid the trap of forking isar and then running into
update/rebase problems with local changes.

Jan

-- 
Siemens AG, Corporate Technology, CT RDA IOT SES-DE
Corporate Competence Center Embedded Linux

  reply	other threads:[~2018-01-24 12:15 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-01-24  7:34 Jan Kiszka
2018-01-24  7:49 ` Alexander Smirnov
2018-01-24  7:52   ` Jan Kiszka
2018-01-24  7:56     ` Jan Kiszka
2018-01-24 10:04       ` Alexander Smirnov
2018-01-24 12:15         ` Jan Kiszka [this message]
2018-01-24 13:08           ` Baurzhan Ismagulov
2018-01-24 13:39             ` Jan Kiszka
2018-01-24 14:09               ` Baurzhan Ismagulov

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=3e2ad3a4-9631-80bd-7c75-315d628b8d5c@siemens.com \
    --to=jan.kiszka@siemens.com \
    --cc=asmirnov@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