Re: [PATCH RFC v2 2/2] default-registry: consolidate OE and Poky release configs

Alexander Kanavin <[email protected]> Fri, 19 Jun 2026 13:12:20 +0200
Newsgroups org.openembedded.lists.bitbake-devel
Message-ID <CANNYZj8e8gD0=7V_szu-paBWv8fdvLKf35ZOHFe6Tb=USBrvnA@mail.gmail.com>
On Thu, 18 Jun 2026 at 20:04, Ernest Van Hoecke
<[email protected]> wrote:
>  .../configurations/oe-nodistro-master.conf.json    |  47 ------
>  .../oe-nodistro-whinlatter.conf.json               |  56 -------
>  .../configurations/oe-nodistro-wrynose.conf.json   |  48 ------
>  .../configurations/oe-nodistro.conf.json           | 117 +++++++++++++++
>  .../configurations/poky-master.conf.json           |  73 ----------
>  .../configurations/poky-whinlatter.conf.json       |  80 ----------
>  .../configurations/poky-wrynose.conf.json          |  74 ----------
>  default-registry/configurations/poky.conf.json     | 161 +++++++++++++++++++++

Thanks! It isn't yet clear whether the configurations in the default
bitbake registry will be replaced like this, or the new
configutrations will co-exist (for a while at least). A potential
issue is that people have existing checkouts based on the existing
configurations, and they will not appreciate no longer being able to
update them. We'll probably be able to get away with it for master,
but the release configs (whinlatter, wrynose) should be kept. This can
be decided later, this commit is intended to give an example.

Anyway! One thing we should check is: how well does 'bitbake-setup
status' (and 'update') handle the situation when a configuration file
has been removed from the registry, http location, or on local disk?
Does it print an unhelpful traceback and fail horribly? Can you check?
It would be good to ensure this is handled gracefully: e.g. no
tracebacks, but a nice note to the user that the operation cannot
proceed as the latest version of the config cannot anymore be
obtained, and a pointer to where it's supposed to be.

Other than this and the comments I made elsewhere, I think you can
start working on tests and docs, and doing this will take the patchset
out of RFC and into a real submission.

Alex