From: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
To: buildroot@busybox.net
Subject: [Buildroot] [RFC] *clean targets
Date: Fri, 9 Oct 2009 09:51:52 +0200 [thread overview]
Message-ID: <20091009095152.555ceed2@surf> (raw)
In-Reply-To: <20091007162422.GI10261@mx.loc>
Hello,
Good to get a discussion on clean targets. They are currently a mess.
Thanks for raising the topic and starting a discussion on this topic.
Le Wed, 7 Oct 2009 18:24:22 +0200,
Bernhard Reutner-Fischer <rep.dot.nop@gmail.com> a ?crit :
> # wipe target
> clean:
I'm not a fan of a make target that only cleans the target. People
might think that by doing so, there are able to remove packages from
their system. However, previously selected packages will still be
present in the staging/ directory, which is not clean. Consider the
following scenario :
* We have package A that has an optional dependency on package B. I.e,
A does not depend on B, but when A ./configure detects that B is
present, then it uses B features, otherwise it just disables this
feature flawlessly.
* The user select A and B, and compiles the system. In the staging/
and target/ directories, we have both A and B
* The user removes B, and does "make clean" with the semantic of your
proposal. So target/ is empty, but staging/ still contains A and B
* The user runs "make" to regenerate the target/. This will reinstall
a version of A that thinks that B will be present on the target,
which is not longer the case.
Moreover, just removing the target/ directory doesn't trigger the
reinstallation of packages using the Makefile.autotools.in machinery,
since this machinery uses stamp files outside of the target/ directory.
Until we have a proven working way of cleanly removing packages, I'd
prefer not to let the user think that he can remove packages safely.
> # remove generated files, retain configs
> realclean:
Could you specify "remove generated files" ? And the semantic when O=
is used ?
> # remove generated files, including configs (but not DL_DIR!)
> distclean:
Agreed on realclean vs. distclean concerning the removal of .config,
and the fact that DL_DIR must be kept.
But still need to define what are the ? generated files ? and how you
plan to remove them.
Sincerly,
Thomas
--
Thomas Petazzoni, Free Electrons
Kernel, drivers and embedded Linux development,
consulting, training and support.
http://free-electrons.com
next prev parent reply other threads:[~2009-10-09 7:51 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-10-07 16:24 [Buildroot] [RFC] *clean targets Bernhard Reutner-Fischer
2009-10-07 16:32 ` Sven Neumann
2009-10-07 18:12 ` Peter Korsgaard
2009-10-07 18:25 ` Bernhard Reutner-Fischer
2009-10-07 18:26 ` Peter Korsgaard
2009-10-07 18:31 ` Bernhard Reutner-Fischer
2009-10-07 19:31 ` Peter Korsgaard
2009-10-09 8:01 ` Thomas Petazzoni
2009-10-07 18:27 ` Peter Korsgaard
2009-10-09 7:51 ` Thomas Petazzoni [this message]
2009-10-09 8:00 ` Thomas Petazzoni
2009-10-09 9:48 ` Bernhard Reutner-Fischer
2009-10-09 11:28 ` Peter Korsgaard
2009-11-20 13:04 ` Peter Korsgaard
2009-11-20 15:28 ` Bernhard Reutner-Fischer
2009-11-20 15:55 ` Peter Korsgaard
2009-11-20 16:19 ` Bernhard Reutner-Fischer
2009-11-20 16:20 ` Peter Korsgaard
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=20091009095152.555ceed2@surf \
--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