From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Petazzoni Date: Sun, 30 Mar 2014 11:13:50 +0200 Subject: [Buildroot] [PATCH 00/91] Update mesa3d & xorg In-Reply-To: <20140329231259.GO3227@free.fr> References: <20140329235039.4bf4983e@skate> <20140329231259.GO3227@free.fr> Message-ID: <20140330111350.4782647c@skate> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net Dear Yann E. MORIN, On Sun, 30 Mar 2014 00:12:59 +0100, Yann E. MORIN wrote: > The problem seems related to libevas, where the dependency on openGL is > poorly handled. > > I'm working on it, but you are free to investigate too. ;-) > > So far, I managed to get rid of the circular hell, with this: > > diff --git a/package/efl/libevas/Config.in b/package/efl/libevas/Config.in > index 2afa888..8eea23d 100644 > --- a/package/efl/libevas/Config.in > +++ b/package/efl/libevas/Config.in > @@ -148,7 +148,7 @@ choice > config BR2_PACKAGE_LIBEVAS_GL > bool "generic OpenGL" > - select BR2_PACKAGE_MESA3D > + depends on BR2_PACKAGE_MESA3D > select BR2_PACKAGE_LIBEET But shouldn't BR2_PACKAGE_LIBEVAS_GL instead depend on a new virtual package for full OpenGL? We can imagine having in the future other implementations that Mesa3D for the full OpenGL, no? Like if you install the NVidia proprietary drivers, I believe they come with their own implementation of OpenGL, completely independent from Mesa3D, no? In this case, since we can "select" a virtual package because that doesn't make sense, the dependency would have to be a "depends on", and we're good, no? Thomas -- Thomas Petazzoni, CTO, Free Electrons Embedded Linux, Kernel and Android engineering http://free-electrons.com