Buildroot Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Arnout Vandecappelle <arnout@mind.be>
To: buildroot@busybox.net
Subject: [Buildroot] Recommended way of working on project
Date: Tue, 27 Oct 2015 22:59:43 +0100	[thread overview]
Message-ID: <562FF3CF.8090502@mind.be> (raw)
In-Reply-To: <CAHsRNLiVzUfg7YhRwRedC3oYpc9QTSSvxFaCduvcMDp3pFqehg@mail.gmail.com>

On 27-10-15 21:11, Gabe Evans wrote:
> Hi David,
> 
> I've used several strategies for different projects we've deployed in
> production and BR2_EXTERNAL is what I've settled on. It's by far the best
> option for the majority of projects. I would only recommend forking
> Buildroot when making changes you plan to merge upstream. Otherwise,
> you're likely to make upgrading to future releases difficult.
> 
> Is there a specific package you're trying to tweak that's resulted in
> you having to make changes in multiple packages depending on it?
> 
> Some of the ways I've added customizations to a package without necessarily
> copying the entirety into my own tree include:
> 
> 1. Making small tweaks to code by using patches in BR2_GLOBAL_PATCH_DIR.

 That's indeed another good way to make changes, but doesn't help if you want to
change some configure or install option.

> 
> 2. Using a different source by creating a custom package that only adjusts
>    the <PACKAGE>_SOURCE/SITE of the original package.

 Note that this is not possible for _VERSION, _INSTALL_STAGING/TARGET,
_SITE_METHOD and probably a few others.

> 3. Adding an additional build step to a package by doing the same as #2 to
>    add a pre/post hook to the original package.
> 
> This has made maintenance trivial and in cases where there's a conflict,
> I'm able to focus my attention on a couple lines instead of on the
> entirety of the package.

 But doesn't that make things more complicated? I mean, you don't have an
overview of what's really happening, and when you look at the package.mk file
you don't see those overrides so you don't understand why it's behaving differently.

 Also, when you merge with upstream and something relevant has changed, there
will _not_ be a merge conflict but instead things go wrong in unexpected ways.

 So I'm not in favour of this approach. But if it works for you, by all means
continue doing so!

 Regards,
 Arnout

> Hopefully others can chime in on any other
> pitfalls there could be with this strategy.
> 

[snip]
-- 
Arnout Vandecappelle                          arnout at mind be
Senior Embedded Software Architect            +32-16-286500
Essensium/Mind                                http://www.mind.be
G.Geenslaan 9, 3001 Leuven, Belgium           BE 872 984 063 RPR Leuven
LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle
GPG fingerprint:  7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF

  reply	other threads:[~2015-10-27 21:59 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-10-26 21:56 [Buildroot] Recommended way of working on project David Kosir
2015-10-27 20:11 ` Gabe Evans
2015-10-27 21:59   ` Arnout Vandecappelle [this message]
2015-10-27 21:38 ` Arnout Vandecappelle
2015-10-28 12:07 ` Carlos Santos

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=562FF3CF.8090502@mind.be \
    --to=arnout@mind.be \
    --cc=buildroot@busybox.net \
    /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