Linux LVM users
 help / color / mirror / Atom feed
From: "Kevin P. Fleming" <kpfleming@backtobasicsmgmt.com>
To: linux-lvm@redhat.com
Subject: Re: [linux-lvm] [RFC][PATCH] proper DESTDIR handling for device-mapper and LVM2
Date: Fri Mar 12 10:25:00 2004	[thread overview]
Message-ID: <4051D6EF.9070507@backtobasicsmgmt.com> (raw)
In-Reply-To: <20040312013822.C4622@homer.msp.redhat.com>

Alasdair G Kergon wrote:

> Thanks for the tidying: I'll work through this properly late next week.  
> (Office moves today.)  Quick comments: I'd prefer not to see DESTDIR in 
> Makefiles though: only in the template, and some (I thought most?) people 
> *do* want to pick up some include files etc. from their DESTDIR (they're not
> installed on their main system, or the ones on their system are different
> versions from the ones in DESTDIR which has the correct environment the
> software uses when it runs): perhaps that should be split into a separate
> configuration option (enabled by default)?

To keep DESTDIR only in the template, you'll have to replace the

bindir = @bindir@

lines (and similar) in make.tmpl with

bindir = $(DESTDIR)@bindir@

to avoid the problem of @bindir@ not including ${prefix}. Just don't 
bother putting $(DESTDIR) into prefix/exec_prefix at all, IMHO.

I can't speak to what everyone else does with DESTDIR, but most 
packaging systems point DESTDIR to an _empty_ subtree to do the install 
into, so they can package up all the resulting files. If someone wanted 
to build LVM2 on a system that had device-mapper _only_ installed into 
DESTDIR (i.e. /usr/lib did not contain libdevmapper.so, /usr/include did 
not contain libdevmapper.h, etc.) this would take a lot more work than 
what was already in the Makefiles, and even then it would tend to break 
because /usr/sbin/lvm would end up linked to

/foo/usr/lib/libdevmapper.so

instead of

/usr/lib/libdevmapper.so

and so /usr/sbin/lvm wouldn't run properly once the DESTDIR directory 
was used as a real environment.

So just to be completely redundant <G>, I'll reiterate the way I've 
always seen DESTDIR used: it's used _only_ at "make install" time to 
redirect the installation to a different path than the package was 
compiled to run from.

  reply	other threads:[~2004-03-12 15:27 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-03-11 23:10 [linux-lvm] [RFC][PATCH] proper DESTDIR handling for device-mapper and LVM2 Kevin P. Fleming
2004-03-12  2:36 ` Alasdair G Kergon
2004-03-12 10:25   ` Kevin P. Fleming [this message]
2004-03-16 21:27     ` Alasdair G Kergon
2004-03-16 21:38       ` Kevin P. Fleming

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=4051D6EF.9070507@backtobasicsmgmt.com \
    --to=kpfleming@backtobasicsmgmt.com \
    --cc=linux-lvm@redhat.com \
    /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