From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Petazzoni Date: Sun, 12 Oct 2014 11:02:13 +0200 Subject: [Buildroot] [PATCH 2/2] package/mono: new package In-Reply-To: References: <1413100208-12776-1-git-send-email-angelo.compagnucci@gmail.com> <1413100208-12776-3-git-send-email-angelo.compagnucci@gmail.com> <20141012102554.39aa340a@free-electrons.com> Message-ID: <20141012110213.5b7c6c8a@free-electrons.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net Dear Angelo Compagnucci, On Sun, 12 Oct 2014 10:44:21 +0200, Angelo Compagnucci wrote: > Yes, you are right, but changing runtime version means you have to > recompile a lot of things. This way, you compile at first and simply > removing .stap_target_installed, you can have free a new mono runtime > version! > If you think that it necessary to have them disabled at compile time, > I'll do it! Yes, please, that's how we do things for all packages: if a configure option is there to not build something that is actually disabled in the Buildroot configuration, then we use that. > >> +ifneq ($(ASSEMBLY_INCLUDED),) > >> +$(eval $(host-autotools-package)) > >> +endif > > > > Conditional not needed. I think we should ensure at least one runtime > > version is enabled. > > That's not always true. There is ways in mono to "freeze" the > assemblies your application requires, so only the mono vm is required. > Moreover, you can write an application that ships required assemblies > in your project folder and not using the system wide installed. > I think that having an option to only compile the vm is good! Then, the conditional shouldn't be here, but around the dependency of mono on host-mono. That being said, I would recommend you to keep things simple for now, and handle only this use case as a follow-up contribution. Thomas -- Thomas Petazzoni, CTO, Free Electrons Embedded Linux, Kernel and Android engineering http://free-electrons.com