From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mailout4.zoneedit.com (mailout4.zoneedit.com [64.68.198.17]) by mail.openembedded.org (Postfix) with ESMTP id B91E57199D for ; Wed, 21 Jun 2017 15:50:27 +0000 (UTC) Received: from localhost (localhost [127.0.0.1]) by mailout4.zoneedit.com (Postfix) with ESMTP id 65CC020ABA; Wed, 21 Jun 2017 15:50:28 +0000 (UTC) Received: from mailout4.zoneedit.com ([127.0.0.1]) by localhost (zmo03-pco.easydns.vpn [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id KIKPPmpltVs6; Wed, 21 Jun 2017 15:50:28 +0000 (UTC) Received: from mail.denix.org (pool-100-15-85-143.washdc.fios.verizon.net [100.15.85.143]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mailout4.zoneedit.com (Postfix) with ESMTPSA id 280DE204E0; Wed, 21 Jun 2017 15:50:24 +0000 (UTC) Received: by mail.denix.org (Postfix, from userid 1000) id AE86316250E; Wed, 21 Jun 2017 11:50:23 -0400 (EDT) Date: Wed, 21 Jun 2017 11:50:23 -0400 From: Denys Dmytriyenko To: Tom Rini Message-ID: <20170621155023.GQ28053@denix.org> References: <20170620221512.21860-1-otavio@ossystems.com.br> <20170620235544.GO28053@denix.org> <20170621120317.GG27196@bill-the-cat> MIME-Version: 1.0 In-Reply-To: <20170621120317.GG27196@bill-the-cat> User-Agent: Mutt/1.5.20 (2009-06-14) Cc: marex@denx.de, brad.mouring@ni.com, Otavio Salvador , OpenEmbedded Core Mailing List Subject: Re: [PATCH] u-boot-fw-utils: Allow replacement of fw_env.config X-BeenThere: openembedded-core@lists.openembedded.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Patches and discussions about the oe-core layer List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 21 Jun 2017 15:50:28 -0000 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline On Wed, Jun 21, 2017 at 08:03:17AM -0400, Tom Rini wrote: > On Tue, Jun 20, 2017 at 07:55:44PM -0400, Denys Dmytriyenko wrote: > > On Tue, Jun 20, 2017 at 07:15:12PM -0300, Otavio Salvador wrote: > > > If a fw_env.config file is found in workdir, this is preferred over > > > the U-Boot example. > > > > > > Signed-off-by: Otavio Salvador > > > --- > > > > > > meta/recipes-bsp/u-boot/u-boot-fw-utils_2017.05.bb | 9 ++++++++- > > > 1 file changed, 8 insertions(+), 1 deletion(-) > > > > > > diff --git a/meta/recipes-bsp/u-boot/u-boot-fw-utils_2017.05.bb b/meta/recipes-bsp/u-boot/u-boot-fw-utils_2017.05.bb > > > index c2e8f0fb84..0682f9274b 100644 > > > --- a/meta/recipes-bsp/u-boot/u-boot-fw-utils_2017.05.bb > > > +++ b/meta/recipes-bsp/u-boot/u-boot-fw-utils_2017.05.bb > > > @@ -19,7 +19,14 @@ do_install () { > > > install -d ${D}${sysconfdir} > > > install -m 755 ${S}/tools/env/fw_printenv ${D}${base_sbindir}/fw_printenv > > > install -m 755 ${S}/tools/env/fw_printenv ${D}${base_sbindir}/fw_setenv > > > - install -m 0644 ${S}/tools/env/fw_env.config ${D}${sysconfdir}/fw_env.config > > > + > > > + # If a specific file is added in a .bbappend, this is used instead > > > + # of the generic one > > > + if [ -e ${WORKDIR}/fw_env.config ]; then > > > + install -m 0644 ${WORKDIR}/fw_env.config ${D}${sysconfdir}/fw_env.config > > > > I don't get it - if it's expected that a .bbappend will be adding a more > > specific version of fw_env.config, why that .bbappend can't simply install > > it with do_install_append()? > > Well, this is a lot more user friendly, and it's quite likely that if a > platform intends to ship u-boot-fw-utils they intend to provide a > functional one as well. It would be a best-practices to provide one > that works for a machine.conf that supports U-Boot, even (and has env > stored somewhere, I mean it even supports env as a file). Why don't we go all the way there? If we say machine.conf is meant to define a machine-specific env config, let's do it generically. What I'm arguing against here is some partial solution - modifying the global recipe in oe-core, but still require .bbappend to provide a file. It's should either be "do everything in .bbappend, like we all do now" or "do everything in main oe-core recipe and let machine.conf define it". -- Denys