Buildroot Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
To: buildroot@busybox.net
Subject: [Buildroot] Bumping packages: some comments/suggestions
Date: Sun, 13 Oct 2013 15:42:29 +0200	[thread overview]
Message-ID: <20131013154229.36a438c9@skate> (raw)
In-Reply-To: <CAFRkauCECxjxqB5=ufBW5Sve+2aGyMBKSOY+9vpPe0gueV7S1Q@mail.gmail.com>

Dear Axel Lin,

On Sun, 13 Oct 2013 18:44:31 +0800, Axel Lin wrote:

> > Recently, both of you have worked on and contributed a number of
> > patches bumping a significant number of Buildroot packages. This is of
> > course really great, and I'd like to thank you for those contributions.
> >
> > That being said, I would have two suggestions:
> >
> >  *) It would be great if you could check that the reverse dependencies
> >     of the package you're bumping still continue to build. For example,
> >     Axel bumped 'ortp', but didn't realize bumping it would break
> >     the linphone and mediastreamer. While we certainly cannot expect
> This is my bad.
> I did try to compile ortp with various combination of build config.
> I didn't realize the reverse dependencies issue when
> I sent the patch bumping ortp version.
> A lesson learnt. I'll be more careful when bump version.
> 
> >     contributors to test package bumps in all possible configurations
> >     (especially for packages having a large number of
> >     reverse dependencies), checking at least a few of them is a good
> >     idea. Also, when bumping from one major release to another (such as
> >     berkeleydb 5.x to berkeleydb 6.x), even more care should be taken.
> Also my bad. Will be checking licenses as well when bump versions.

Thanks. Note that my comments were really meant as suggestions to
improve your contributions: these will continue to be very welcome.

> >  *) To make this "bumping" effort a bit more systematic, I believe it
> >     would be useful to introduce an infrastructure in Buildroot to
> >     automatically check if upstream has a new package. In many cases,
> >     the upstream site has a directory with all the different versions
> >     of the tarball, so checking if there's a newer one in an automated
> >     way would be possible. If we do this for many packages, then we can
> >     run a script every day, and check if there are new upstream
> >     releases available. Debian has such a mechanism with the 'watch'
> >     mechanism (see https://wiki.debian.org/debian/watch/). Gentoo has
> >     the euscan utility (see https://github.com/iksaif/euscan). It would
> >     be nice having something like this, that we could integrate in the
> >     Buildroot per-package stats at
> >     http://autobuild.buildroot.org/stats/ to get a clear vision of
> >     which packages need to be upgraded. If one of you is interested in
> >     doing this, it'd be great!

Any opinion about this? I believe it would make more sense to invest
time doing this than doing many many bumps on all packages.

Best regards,

Thomas
-- 
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

  reply	other threads:[~2013-10-13 13:42 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-10-13  9:42 [Buildroot] Bumping packages: some comments/suggestions Thomas Petazzoni
2013-10-13 10:44 ` Axel Lin
2013-10-13 13:42   ` Thomas Petazzoni [this message]
2013-10-13 15:01     ` Thomas De Schampheleire
2013-10-13 15:06       ` Thomas Petazzoni
2013-10-13 21:04 ` Jerzy Grzegorek
2013-10-14  7:09   ` Thomas Petazzoni
2013-10-14  7:40     ` Jeremy Rosen
2013-10-14  9:30       ` Thomas Petazzoni
2013-10-14  9:38         ` Jeremy Rosen
2013-10-14 10:02           ` Thomas De Schampheleire
2013-10-14 10:04             ` Thomas Petazzoni
2013-10-14 12:58               ` arnaud aujon
2013-10-14 13:56                 ` Thomas Petazzoni
2013-10-14 17:11                   ` arnaud aujon
2013-10-14 21:55             ` Arnout Vandecappelle
2013-10-15  7:34               ` Thomas Petazzoni
2013-10-15 19:55                 ` Arnout Vandecappelle
2013-10-14 13:45         ` rjbarnet at rockwellcollins.com

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=20131013154229.36a438c9@skate \
    --to=thomas.petazzoni@free-electrons.com \
    --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