From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Petazzoni Date: Wed, 5 Apr 2017 18:14:24 +0200 Subject: [Buildroot] [RFC 4/4] board/acmesystems/aria-g25: set BR2_GENIMAGE_CFG_FILES In-Reply-To: <02a1a330-c947-2a64-29bb-88f6ac7d9c9f@mind.be> References: <20170329145120.11863-1-etienne.phelip@savoirfairelinux.com> <20170329145120.11863-5-etienne.phelip@savoirfairelinux.com> <20170401155116.66464885@free-electrons.com> <8760ilvozx.fsf@dell.be.48ers.dk> <2ce8812c-d95e-bd4a-9cde-f02ab8e3ca60@mind.be> <87wpazuaa1.fsf@dell.be.48ers.dk> <02a1a330-c947-2a64-29bb-88f6ac7d9c9f@mind.be> Message-ID: <20170405181424.1d69fee5@free-electrons.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net Hello, On Wed, 5 Apr 2017 17:02:36 +0200, Arnout Vandecappelle wrote: > Combined with the other discussion in this thread, long-term I see us evolving > towards: > > - deprecate some of the finer-grain fs tuning options we have now; > - always using genimage to generate the target filesystems; > - position our fs targets more as a 'quick fix' solution than for production; > - make host-e2fsprogs etc. blind options again. To be honest, I am not sure I share this long-term view. I like the way things are done today, very modular: we can generate just a filesystem image, optionally use genimage afterwards, etc. So the "always using genimage to generate the target filesystem" is not something that I see as an improvement, for example. Best regards, Thomas -- Thomas Petazzoni, CTO, Free Electrons Embedded Linux and Kernel engineering http://free-electrons.com