All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Jan-Simon Möller" <dl9pf@gmx.de>
To: Richard Purdie <richard.purdie@linuxfoundation.org>,
	yocto@lists.yoctoproject.org, yocto@lists.yoctoproject.org,
	duy.dang.yw@renesas.com
Subject: Re: [yocto] Any recommendation to make software layer Yocto Compatible?
Date: Sat, 13 Apr 2024 14:40:11 +0200	[thread overview]
Message-ID: <1944482.7Z3S40VBb9@arbeit> (raw)
In-Reply-To: <20226.1712579798480481849@lists.yoctoproject.org>

Hi Duy,

The recipe in question was done by the Instrument Cluster EG, we can work on 
this with the group (main contact Yamaguchi-san). They do meet every other 
monday. See: https://lists.automotivelinux.org/g/agl-dev-community/calendar 

We do exacly what Paul describes all the time in meta-agl to pass the yocto-
check-layer:

e.g.
https://git.automotivelinux.org/AGL/meta-agl/tree/meta-agl-core/recipes-core/
systemd/systemd_%25.bbappend
and
https://git.automotivelinux.org/AGL/meta-agl/tree/meta-agl-core/recipes-core/
systemd/systemd_aglcore.inc

The condition can be on DISTRO_FEATURES or other variables. We try not to 
overload the usage of DISTRO_FEATURES as a change there will trigger a reparse 
in a lot of locations and hence possibly rebuilds. Thus the use of 
AGL_FEATURES in above example for anything that is directly related to AGL and 
is not a DISTRO_FEATURE in yocto already. 

For dlt-daemon in your example, we have these options and recommendations:
a) use such a conditional include as shown above ... simple but it will not fix 
all the issues
b) use :append:<some_other_conditional>  only in .bbappend files
     (name says it all ;) )
    e.g. :append:aglcontainerguest
    or
    e.g.  :append:aglcontainerhost
    The 2nd condition / override is important here to pass the check.
    A simple :append  would triger
c) distill the recipe down to just the required changes & if possible upstream 
these (e.g. PACKAGECONFIG options) ... config file changes can also be added in 
a "dlt-daemon-conf" package and replace or amend the original files. This helps 
most.

a) will have immediate effect but only c) and b) will lower the workload mid-
term. 

A path forward would then be:
Upstream the PACKAGECONFIG bits . It might make sense to rework the package so 
there is a configuration sub-package that is replaceable by the user. Then most 
conf file changes are in a separate package and can be easily replaced by the 
user w/o even touching the main (binary) package. That is of importance for 
reproducible builds and binary feeds. 



Am Montag, 8. April 2024, 14:36:38 CEST schrieb Duy via 
lists.yoctoproject.org:
> Hi Richard,
> 
> Thanks for your response.
> Here is one of the recipe bbappend files I'm working on:
> meta-agl-ic-container/recipes-extended/dlt-daemon/dlt-daemon_%.bbappend ·
> master · Automotive Grade Linux / AGL / meta-agl-devel · GitLab (
> https://gitlab.com/automotivegradelinux/AGL/meta-agl-devel/-/blob/master/me
> ta-agl-ic-container/recipes-extended/dlt-daemon/dlt-daemon_%25.bbappend?ref_
> type=heads )
> meta-agl-ic-container/recipes-extended/dlt-daemon/dlt-daemon_%.bbappend ·
> master · Automotive Grade Linux / AGL / meta-agl-devel · GitLab (
> https://gitlab.com/automotivegradelinux/AGL/meta-agl-devel/-/blob/master/me
> ta-agl-ic-container/recipes-extended/dlt-daemon/dlt-daemon_%25.bbappend?ref_
> type=heads )
> 
> I think some configurations are better to be upstreaming, e.g: Adding new
> PACKAGES, adding new PACKAGECONFIG. It avoids adding too much to bbappend.
> If you have any idea, please share it.
> 
> Best Regards,
> Duy Dang






  parent reply	other threads:[~2024-04-13 12:40 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-08  9:25 Any recommendation to make software layer Yocto Compatible? Duy
2024-04-08  9:54 ` [yocto] " Richard Purdie
2024-04-08 12:36   ` Duy
2024-04-11 20:45     ` Paul Barker
2024-04-13 11:41     ` Richard Purdie
2024-04-13 12:40     ` Jan-Simon Möller [this message]
2024-04-15  2:56       ` Duy

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=1944482.7Z3S40VBb9@arbeit \
    --to=dl9pf@gmx.de \
    --cc=duy.dang.yw@renesas.com \
    --cc=richard.purdie@linuxfoundation.org \
    --cc=yocto@lists.yoctoproject.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.