Got it. I see your rational and I now understand why you proposed a bbclass.
I'll use this bbclass method and send out V3.

Regards,
Qi

-----Original Message-----
From: Alexander Kanavin <[email protected]> 
Sent: Thursday, May 7, 2026 7:33 PM
To: Chen, Qi <[email protected]>
Cc: [email protected]; MacLeod, Randy 
<[email protected]>
Subject: Re: [OE-core][PATCH V2 1/4] recipeutils: add optional stable_upgrade 
parameter to get_recipe_upgrade_status

On Thu, 7 May 2026 at 08:20, <[email protected]> wrote:
> If it's not set, we give
> it a reasonable default value, which means if a version consists of at 
> least three parts, the bump in last one is considered as a stable 
> upgrade.

I strongly disagree and think we should be a lot more careful. This default 
will result in a lot of point version updates where the LTS maintainer would 
have to go and check whether the updates include new features or not. Yoann 
already had to reject a bunch of such updates, produced in a similar way with 
custom AUH tweaks.

Rather this should be enabled on a case by case basis for components that are 
known to do stable maintenance (a good indicator is presence of stable branches 
with a major.minor version in branch name somewhere), by setting 
UPSTREAM_STABLE_RELEASE_REGEX for them (perhaps via class inherit to avoid 
repetition).

No explicitly set UPSTREAM_STABLE_RELEASE_REGEX means no stable updates exist, 
or the situation hasn't yet been checked.

Alex
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#236631): 
https://lists.openembedded.org/g/openembedded-core/message/236631
Mute This Topic: https://lists.openembedded.org/mt/119193131/21656
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

Reply via email to