From: Henning Schild <henning.schild@siemens.com>
To: Alexander Smirnov <asmirnov@ilbers.de>
Cc: <isar-users@googlegroups.com>
Subject: Re: Isar apt further steps
Date: Wed, 18 Oct 2017 16:08:46 +0200 [thread overview]
Message-ID: <20171018160846.4e12413f@md1em3qc> (raw)
In-Reply-To: <20171018160030.05b119f4@md1em3qc>
On Wed, 18 Oct 2017 16:00:30 +0200
"[ext] Henning Schild" <henning.schild@siemens.com> wrote:
> On Wed, 18 Oct 2017 14:05:01 +0300
> Alexander Smirnov <asmirnov@ilbers.de> wrote:
>
> > Hi Henning,
> >
> > I'd like to start separate discussion about further apt steps
> > because previous series was applied before your last comment:
>
> As i just wrote you in the original thread, please remove these
> patches, they are not ready to get merged and are on a rebasing
> branch.
With the "URGENT" kernel changes we now have a situation where next
needs to be rebased anyways, or do you plan to merge?
Henning
> > > - create one task in dpkg-base.bbclass that does the following
> > > - config and init reprepro if no other recipe did that
> > > before
> >
> > Hmm, in general repository generation should be performed only
> > once, moreover in most cases it should be done for dedicated Isar
> > image, otherwise no need to cache binaries to apt if it's not used.
> > So that's why I've put repo creation to image class.
>
> If there is a way to express "image step to run before any package
> recipe starts". I guess all recipes "do_populate_apt" need a [depends]
> "image:do_cache_config"
>
> > > - add package
> >
> > If task[lockfile] works for multiconfig, then I definitely like
> > this approach.
>
> I certainly hope so. If not, you know my opinion on multiconfig ;).
>
> > > - use bitkages task[lockfiles] to deal with races between
> > > recipes, put distro into lockfile-name so we have one lock per
> > > distro
> > > - drop do_populate
> >
> > That's already in todo, the blocking issue was locking mechanism
> > for multiconfig.
> >
> > > - call the new task instead of, or in do_deploy_deb
> >
> >
> >
> >
> > So, to summarize my vision:
> >
> > 1. image.bbclass: do_cache_config(). Use [lockfiles] = ${DISTRO} to
> > handle multiconfig and 'bitbake image-A image-B' parallel builds.
> >
> > 2. dpkg-base.bbclass: do_populate_apt(). Use [lockfiles] = ${DISTRO}
> > to avoid races between recipes and for multiconfig.
> >
> > 3. Drop image.bbclass: do_populate()
> >
> > 4. Remove ${DEPLOY_DIR_DEB} from Isar.
> >
> > What do you think?
>
> Sounds good!
>
> Henning
>
> > Alex
>
prev parent reply other threads:[~2017-10-18 14:08 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-10-18 11:05 Alexander Smirnov
2017-10-18 14:00 ` Henning Schild
2017-10-18 14:08 ` Henning Schild [this message]
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=20171018160846.4e12413f@md1em3qc \
--to=henning.schild@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