Date: Sat, 16 Sep 2017 18:59:02 +0200 From: "Tobias C. Berner" <tcberner@freebsd.org> To: Jan Beich <jbeich@freebsd.org> Cc: Adriaan de Groot <adridg@freebsd.org>, ports-committers@freebsd.org, svn-ports-all@freebsd.org, svn-ports-head@freebsd.org Subject: Re: svn commit: r449853 - in head: . astro/libosmpbf/files audio/sayonara/files databases/kdb/files devel devel/cmake devel/cmake-doc devel/cmake-gui devel/cmake-gui/files devel/cmake-modules devel/cma... Message-ID: <CAOshKtdP1aw3oFrdqMQ9%2B70LZRXsFD=DfdkdFbir7tqqsryo1A@mail.gmail.com> In-Reply-To: <vaki-ziyw-wny@FreeBSD.org> References: <201709142102.v8EL2q5R011625@repo.freebsd.org> <vaki-ziyw-wny@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
The issue is not emulators/ppsspp but its slave emulators/ppsspp-qt5 which gets a dependency cycle: ==> Building for ppsspp-qt5-1.4.2 ninja: error: dependency cycle: cmake_object_order_depends_target_Core -> Core_autogen -> CMakeFiles/Core_autogen -> /wrkdirs/usr/ports/emulators/ppsspp-qt5/work/ppsspp-1.4.2/git-version.cpp -> cmake_object_order_depends_target_Core mfg Tobias On 16 September 2017 at 18:49, Jan Beich <jbeich@freebsd.org> wrote: > Adriaan de Groot <adridg@FreeBSD.org> writes: > > > Author: adridg > > Date: Thu Sep 14 21:02:51 2017 > > New Revision: 449853 > > URL: https://svnweb.freebsd.org/changeset/ports/449853 > > > > Log: > > Update CMake to 3.9, merge devel/cmake-modules into devel/cmake > [...] > > Index: files/patch-system-libpng16 > > =================================================================== > > --- files/patch-system-libpng16 (revision 449852) > > +++ files/patch-system-libpng16 (revision 449853) > > @@ -1,8 +1,10 @@ > > Make build glue accept libpng 1.6.x and don't hardcode suffix > > > > ---- CMakeLists.txt.orig 2015-11-19 15:07:48 UTC > > +One hunk in src/CMakeLists.txt was added to fix target generation with > ninja. > > Don't mix unrelated changes into topic patches. This file may go away once > graphics/png switches to 1.7.*. It's similar to a backport from upstream. > > What was broken? Without the change the port still builds fine using > CMake 3.9 (both -j1 and -j8). Otherwise, that one-line rationale isn't > sufficient for a "portmgr blanket". > >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAOshKtdP1aw3oFrdqMQ9%2B70LZRXsFD=DfdkdFbir7tqqsryo1A>