>>> On 14.12.15 at 18:52, <car...@cardoe.com> wrote: > On 12/14/15 10:35 AM, Jan Beulich wrote: >>>>> On 10.12.15 at 17:48, <car...@cardoe.com> wrote: >>> --- /dev/null >>> +++ b/xen/arch/x86/Kconfig >>> @@ -0,0 +1,17 @@ >>> +config X86_64 >>> + def_bool y >>> + >>> +config X86 >>> + def_bool y >>> + >>> +config ARCH_DEFCONFIG >>> + string >>> + default "arch/x86/configs/x86_64_defconfig" >>> + >>> +menu "Architecture Features" >>> + >>> +endmenu >>> + >>> +source "common/Kconfig" >>> + >>> +source "drivers/Kconfig" >> >> I'm still missing "config 64BIT" in this file. > > You had wanted me to factor that out in earlier series. So now its only > in the arm side which is the only place its used.
>From what I recall, I've always been asking to make this a general setting, not an ARM specific one (irrespective of it only being used in ARM right now). >>> +# provide the host compiler >>> +HOSTCC := gcc >>> +HOSTCXX := g++ >> >> Didn't you mean to inherit these instead of forcing them here? > > No. I pointed out that the only real way to inherit them is to pull in > top-level Config.mk which is too heavy handed. If we have some agreement > to break up the pieces that truly belong in there then sure. > > Honestly I would prefer to do that break up after the fact. I've being > working on this series for 3 months now and a lot of the issues brought > up can happen after the fact because, like this one, it involves fixing > up existing Xen makefiles rather than issues with this series. That's fine as long as it doesn't get forgotten. Jan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
