On 08/24/2011 02:38 PM, Peter Maydell wrote:
On 24 August 2011 11:47, Avi Kivity<a...@redhat.com> wrote:
> On 08/24/2011 01:35 PM, Peter Maydell wrote:
>>
>> On 24 August 2011 11:11, Avi Kivity<a...@redhat.com> wrote:
>> > diff --git a/hw/mcf5208.c b/hw/mcf5208.c
>> > index f4d4fbd..73c6961 100644
>> > --- a/hw/mcf5208.c
>> > +++ b/hw/mcf5208.c
>> > @@ -5,6 +5,7 @@
>> > *
>> > * This code is licensed under the GPL
>> > */
>> > +#include<glib.h>
>> > #include "hw.h"
>> > #include "mcf.h"
>> > #include "qemu-timer.h"
>>
>> You shouldn't need to include glib.h here, I think? hw.h includes
>> qemu-common.h
>> which includes glib.h.
>>
>
> My preference is not to depend on indirect includes; but I can remove this
> include if it's disliked by many.
qemu-common.h says
/* we put basic includes here to avoid repeating them in device drivers */
so direct includes of system headers in hw/ files is going against
current practice. I don't feel strongly either way on the question
of direct vs indirect, but we should be consistent, and the choice
we've made is "rely on qemu-common.h".
Okay, I'll drop the include (here and elsewhere).
--
I have a truly marvellous patch that fixes the bug which this
signature is too narrow to contain.
