Buildroot Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Arnout Vandecappelle <arnout@mind.be>
To: buildroot@busybox.net
Subject: [Buildroot] busybox installation with --noclobber
Date: Thu, 03 Apr 2014 18:45:49 +0200	[thread overview]
Message-ID: <533D903D.6080906@mind.be> (raw)
In-Reply-To: <CANxTyt4JBk+nF2fWxxGvZOfGzYCuUAUaP7Y71fqXN+U89h4HZA@mail.gmail.com>

On 03/04/14 03:08, Danomi Manchego wrote:
> All,
> 
> A tangential work flow issue came up while helping some co-workers
> recently.  Basically, they have a busybox configuration that must be
> duplicating the "real" version of apps installed by other packages.
> When they rebuild and re-install busybox (say, after a "make
> busybox-menuconfig"), their rootfs becomes unusable, presumably
> because the re-installation of the busybox symlinks must be wiping out
> something important.  To recover functionality, they have to either
> delete all of the .stamp_target_installed files to re-install
> everything, or else rebuild from scratch.
> 
> As it turns out, the busybox install script has a --noclobber option,
> where symlinks will not be installed if something already exists at
> the destination (be it file or symlink).  This makes the busybox
> re-install work for my peers without subsequent head ache.  I provided
> a patch to them to force this option on.  So I'm now considering if
> such a change should be considered applicable more generally.
> 
> I was originally inclined against such a patch, to encourage the
> active trimming down of busybox to eliminate the redundancy (by not
> dulling the pain).  But the reality is that identifying the duplicates
> sometimes is not so easy for my co-workers, so I fear that the project
> maintainers simply won't bother.
> 
> There is also an argument that the --noclobber can be beneficial to
> the effort to trim down busybox.  For example:
> 1. Build project from a clean state
> 2. Delete busybox from that target directory:
>         rm -vf $(TARGET_DIR)/bin/busybox
> 3. Delete broken symlinks from target directory's bin, sbin, usr/bin, usr/sbin:
>         find -L $(TARGET_DIR)/bin $(TARGET_DIR)/sbin
> $(TARGET_DIR)/usr/bin $(TARGET_DIR)/usr/sbin -type l -print -delete
> 4. Re-install busybox:
>         make busybox-rebuild
> 5. During install, take note of which symlinks are reported as
> "blahblah already exists"
> 6. Remove the noted items from busybox configuration if appropriate
> 
> Opinions?  Would there be interest in such a patch?

 It would be a good idea, I think. Currently, packages that provide the
same functionality as busybox have to depend on it to make sure that
busybox is built first, and this indeed makes busybox-rebuild impossible.
It would be much simpler to use the noclobber option.

 Regards,
 Arnout

-- 
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:  7CB5 E4CC 6C2E EFD4 6E3D A754 F963 ECAB 2450 2F1F

      reply	other threads:[~2014-04-03 16:45 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-04-03  1:08 [Buildroot] busybox installation with --noclobber Danomi Manchego
2014-04-03 16:45 ` Arnout Vandecappelle [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=533D903D.6080906@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