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] [PATCH 1/1] package/pkg-kconfig: define KCONFIG_EDITORS_MAKE_ENV
Date: Sat, 4 Jul 2015 14:30:14 +0200	[thread overview]
Message-ID: <20150704143014.21a3e1c1@free-electrons.com> (raw)
In-Reply-To: <20150702212031.GH3838@free.fr>

Dear Yann E. MORIN,

On Thu, 2 Jul 2015 23:20:32 +0200, Yann E. MORIN wrote:

> On 2015-07-02 17:59 +0200, Thomas Petazzoni spake thusly:
> > On Thu, 2 Jul 2015 17:41:40 +0300, Tal Zilcer wrote:
> > 
> > >  $$(addprefix $(1)-,$$($(2)_KCONFIG_EDITORS)): @D=$$($(2)_DIR)
> > >  $$(addprefix $(1)-,$$($(2)_KCONFIG_EDITORS)): $$($(2)_DIR)/.stamp_kconfig_fixup_done
> > > -	$$($(2)_MAKE_ENV) $$(MAKE) -C $$($(2)_DIR) \
> > > +	$$($(2)_KCONFIG_EDITORS_MAKE_ENV) $$(MAKE) -C $$($(2)_DIR) \
> > 
> > I actually don't think adding a KCONFIG_EDITORS_MAKE_ENV variable is
> > really useful here. Since it's a bunch of host tools, maybe we should
> > just:
> > 
> > 	$$(HOST_MAKE_ENV) $$(MAKE) -C ...
> 
> I do agree with Thomas here.

Except that it doesn't work: HOST_MAKE_ENV defines:

        PKG_CONFIG_SYSROOT_DIR="/" \
        PKG_CONFIG_LIBDIR="$(HOST_DIR)/usr/lib/pkgconfig"

So this is perfectly OK to force pkg-config to find libraries built for
the host instead of the target. However, since we're passing
PKG_CONFIG_LIBDIR, we completely override the default pkg-config search
path, so it does not even check for libraries installed system-wide, it
only searches for libraries we have installed in $(HOST_DIR). And since
Qt is not built by us, it doesn't work.

Using PKG_CONFIG_PATH would allow to search for the host installed
libraries. However, our pkg-config wrapper enforces the use of the
cross-PKG_CONFIG_LIBDIR if no PKG_CONFIG_LIBDIR value is passed.

Bottom line, using $(HOST_MAKE_ENV) is clearly not sufficient to solve
the problem.

So we have to decide whether we want to use the system-provided
pkg-config, or our own pkg-config. But since linux does not depend on
host-pkgconf, maybe we should simply just use the one available on the
system, since we anyway don't build ncurses/qt/gtk ourselves for the
configuration interfaces. If that's what we want to do, I would then
suggest to do pass any variable in the make environment.

Yann, what do you think?

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

      reply	other threads:[~2015-07-04 12:30 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-02 14:41 [Buildroot] [PATCH 1/1] package/pkg-kconfig: define KCONFIG_EDITORS_MAKE_ENV Tal Zilcer
2015-07-02 15:59 ` Thomas Petazzoni
2015-07-02 21:20   ` Yann E. MORIN
2015-07-04 12:30     ` Thomas Petazzoni [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=20150704143014.21a3e1c1@free-electrons.com \
    --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