On Tue, May 26, 2015 at 1:12 AM, Paolo Bonzini <pbonz...@redhat.com> wrote: > > > On 26/05/2015 08:38, Peter Crosthwaite wrote: >> Rather than an explicit inclusion of cpu.h. This maked it more >> consistent with other core code files, which either just rely on >> qemu-common.h inclusion or preceed cpu.h with qemu-common.h anyway. > > I'd rather keep the cpu.h inclusion. It would be nice to get rid of the > automagic inclusion of cpu.h from qemu-common.h. >
OK, but it is an odd one out, would you be willing to accept the double include that seems to be widespread? +#include "qemu_common.h" #include "cpu.h" -#include "exec/cpu-all.h" Regards, Peter > Paolo > >> Signed-off-by: Peter Crosthwaite <crosthwaite.pe...@gmail.com> >> --- >> Picked up by my multi arch WIP where target-multi/cpu.h cant handle >> random core code inclusion without preceeded qemu-common.h. I guess >> this is the only one in tree? >> --- >> memory_mapping.c | 3 +-- >> 1 file changed, 1 insertion(+), 2 deletions(-) >> >> diff --git a/memory_mapping.c b/memory_mapping.c >> index 7b69801..7cf5548 100644 >> --- a/memory_mapping.c >> +++ b/memory_mapping.c >> @@ -13,8 +13,7 @@ >> >> #include <glib.h> >> >> -#include "cpu.h" >> -#include "exec/cpu-all.h" >> +#include "qemu-common.h" >> #include "sysemu/memory_mapping.h" >> #include "exec/memory.h" >> #include "exec/address-spaces.h" >> -- 1.9.1 >
