On Tue, Apr 22, 2014 at 10:15:14PM -0700, Warner Losh wrote: > > On Apr 22, 2014, at 10:12 PM, Glen Barber <g...@freebsd.org> wrote: > > > On Wed, Apr 23, 2014 at 01:09:23AM -0400, Glen Barber wrote: > >> On Tue, Apr 22, 2014 at 10:02:03PM -0700, Warner Losh wrote: > >>> Ditto. This should be backed out and you should make whoever broke this > >>> fix it in head. > >>> > >> > >> Well, this commit only moves where the definition is, from the script > >> itself to the configuration file. > >> > >> But, to be fair, I think you broke it in r264660. > >> > > > > To clarify further, if it was you that did break it in head/, I am not > > complaining about it - I fixed the use case I need. > > > > What I *want* to happen is for the WITH_CLANG_IS_CC and > > WITHOUT_CLANG_IS_CC to stop conflicting with WITH_GCC, and additional > > convoluted and unnecessary knobs to go away. > > > > But you already know that. :) > > Yea, see my recent posts to arm@ on that very topic. CLANG_IS_CC > is evil and must die. There’s too many stupid special cases for > clang in the build system, and it is really horking me off… But > there will be fewer in about a week or so when I push my next round > of fixes in... >
Feel free to run your changes by me for testing. I'm happy to complain^W test. :) Glen
pgpt_RYm8mO0D.pgp
Description: PGP signature
