From: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
To: buildroot@busybox.net
Subject: [Buildroot] [PATCH v7 3/8] core: implement per-package SDK and target
Date: Wed, 6 Nov 2019 08:57:33 +0100 [thread overview]
Message-ID: <20191106085733.3b339da7@windsurf> (raw)
In-Reply-To: <CAJ4jsafpnaBRuxtd-6RFAGVcFgkEgmqNbjDLgPEYQV5DXGMxMw@mail.gmail.com>
Hello,
On Tue, 5 Nov 2019 16:05:50 -0300
Carlos Santos <unixmania@gmail.com> wrote:
> > Or we could:
> >
> > (1) Move the /etc/resolv.conf symlink from system/skeleton/ to
> > package/skeleton-init-sysv/skeleton/etc.
> >
> > (2) Tweak the systemd.mk so that it always creates resolv.conf, either
> > pointing to ../run/systemd/resolve/resolv.conf when resolved is
> > enabled, or to /tmp/resolv.conf otherwise.
> >
> > This would avoid the overwriting entirely.
> >
> > What do you think ?
> >
> > Best regards,
> >
> This would prevent NetworkManager from updating resolv.conf. See
> https://bugs.busybox.net/show_bug.cgi?id=9881
Then I guess the TARGET_FINALIZE_HOOKS is the only solution here.
But thinking more about this issue, it really means that no package
should overwrite any file installed by one of its dependencies: this
means that the check-uniq-files stuff that Yann recently removed was in
fact useful.
Indeed, if we have the following scenario:
- Package A installs file /foo/bar
- Package B depends on A, and overwrites /foo/bar with some contents
- Package C depends on A
Then, at the end in the target root filesystem, we will have A's
version of /foo/bar, and not B's version. Just like Jan explained, the
final rsync will rsync the per-package target directories in this order
(due to alphabetic ordering):
A, B, C
So, in the end, it is the version of /foo/bar in the per-package
directory of C that wins, and because C depends on A, the version of
/foo/bar in C's per-package directory is the one coming from package A.
So, it is really important for packages *not* to overwrite any file
installed by any other package, even if said package is in its
dependencies.
Am I missing something here ?
Best regards,
Thomas
--
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
next prev parent reply other threads:[~2019-11-06 7:57 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-12-28 10:43 [Buildroot] [PATCH v7 0/8] Top-level parallel build support Thomas Petazzoni
2018-12-28 10:43 ` [Buildroot] [PATCH v7 1/8] support/scripts/check-host-rpath: document existing functions Thomas Petazzoni
2018-12-28 12:47 ` Yann E. MORIN
2019-01-17 21:33 ` Peter Korsgaard
2018-12-28 10:43 ` [Buildroot] [PATCH v7 2/8] Makefile: move definition of TARGET_DIR inside .config condition Thomas Petazzoni
2018-12-28 12:49 ` Yann E. MORIN
2019-01-17 21:35 ` Peter Korsgaard
2018-12-28 10:43 ` [Buildroot] [PATCH v7 3/8] core: implement per-package SDK and target Thomas Petazzoni
2018-12-30 21:52 ` Yann E. MORIN
2018-12-31 14:31 ` Thomas Petazzoni
2018-12-31 14:45 ` Yann E. MORIN
2019-01-08 18:02 ` Jan Kundrát
2019-11-05 16:38 ` Thomas Petazzoni
2019-11-05 19:05 ` Carlos Santos
2019-11-06 7:57 ` Thomas Petazzoni [this message]
2019-11-06 8:13 ` Jan Kundrát
2018-12-28 10:43 ` [Buildroot] [PATCH v7 4/8] Makefile: allow top-level parallel build with BR2_PER_PACKAGE_DIRECTORIES=y Thomas Petazzoni
2018-12-28 12:51 ` Yann E. MORIN
2018-12-28 10:43 ` [Buildroot] [PATCH v7 5/8] package/pkg-generic: make libtool .la files compatible with per-package directories Thomas Petazzoni
2018-12-31 8:44 ` Yann E. MORIN
2018-12-28 10:43 ` [Buildroot] [PATCH v7 6/8] package/pkg-kconfig: handle KCONFIG_DEPENDENCIES " Thomas Petazzoni
2018-12-28 10:43 ` [Buildroot] [PATCH v7 7/8] docs/manual: add details about top-level parallel build support Thomas Petazzoni
2018-12-28 13:03 ` Yann E. MORIN
2018-12-28 13:08 ` Thomas Petazzoni
2018-12-31 8:46 ` Yann E. MORIN
2018-12-28 10:43 ` [Buildroot] [PATCH v7 8/8] docs/manual: document the effect of per-package directory on variables Thomas Petazzoni
2018-12-28 17:21 ` [Buildroot] [PATCH v7 0/8] Top-level parallel build support Thomas Petazzoni
2019-02-22 16:18 ` Andreas Naumann
2019-02-22 18:07 ` Vadim Kochan
2019-02-22 20:29 ` Thomas Petazzoni
2019-02-25 1:10 ` Vadim Kochan
2019-02-25 8:05 ` Thomas Petazzoni
2019-02-25 8:33 ` Vadim Kochan
2019-03-01 14:50 ` Vadym Kochan
2019-03-01 17:18 ` Yann E. MORIN
2019-03-04 7:24 ` Arnout Vandecappelle
2019-03-04 10:22 ` Thomas Petazzoni
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=20191106085733.3b339da7@windsurf \
--to=thomas.petazzoni@bootlin.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