On Wed, Jun 21, 2017 at 09:23:11AM -0300, Eduardo Habkost wrote:
> On Wed, Jun 21, 2017 at 03:52:00PM +0800, Peter Xu wrote:
> > Introduce this new field for the accelerator classes so that each
> > specific accelerator in the future can register its own global
> > properties to be used further by the system. It works just like how the
> > old machine compatible properties do, but only tailored for
> > accelerators.
> >
> > Use the newly exported register_compat_prop() to pass the accelerator
> > global properties to the global_props list.
> >
> > Suggested-by: Eduardo Habkost <ehabk...@redhat.com>
> > Signed-off-by: Peter Xu <pet...@redhat.com>
> > ---
> > accel/accel.c | 10 ++++++++++
> > include/sysemu/accel.h | 10 ++++++++++
> > vl.c | 1 +
> > 3 files changed, 21 insertions(+)
> >
> > diff --git a/accel/accel.c b/accel/accel.c
> > index 7c079a5..212581c 100644
> > --- a/accel/accel.c
> > +++ b/accel/accel.c
> > @@ -120,6 +120,16 @@ void configure_accelerator(MachineState *ms)
> > }
> > }
> >
> > +void accel_register_compat_props(AccelState *accel)
> > +{
> > + AccelClass *class = ACCEL_GET_CLASS(accel);
> > + GlobalProperty *prop;
> > +
> > + for (prop = class->global_props; prop && prop->driver; prop++) {
> > + register_compat_prop(prop->driver, prop->property, prop->value);
> > + }
> > +}
>
> I suggest adding a generic register_compat_props_array(GlobalProperty *props)
> helper, so we can reuse it at machine_register_compat_props()
> later (once we fix global property ordering in
> qdev_prop_set_globals() and make the object_class_foreach() hack
> in machine_register_compat_props() unnecessary).
Sure.
>
> > +
> > static void register_accel_types(void)
> > {
> > type_register_static(&accel_type);
> > diff --git a/include/sysemu/accel.h b/include/sysemu/accel.h
> > index 15944c1..8a01e51 100644
> > --- a/include/sysemu/accel.h
> > +++ b/include/sysemu/accel.h
> > @@ -24,6 +24,7 @@
> > #define HW_ACCEL_H
> >
> > #include "qom/object.h"
> > +#include "hw/qdev-properties.h"
> >
> > typedef struct AccelState {
> > /*< private >*/
> > @@ -40,6 +41,13 @@ typedef struct AccelClass {
> > int (*available)(void);
> > int (*init_machine)(MachineState *ms);
> > bool *allowed;
> > + /*
> > + * Array of gobal properties that would be applied when specific
>
> "global"
Will fix.
>
> > + * accelerator is chosen. It works just like
> > + * MachineClass.compat_props but it's for accelerators not
> > + * machines.
> > + */
> > + GlobalProperty *global_props;
> > } AccelClass;
> >
> > #define TYPE_ACCEL "accel"
> > @@ -57,5 +65,7 @@ typedef struct AccelClass {
> > extern int tcg_tb_size;
> >
> > void configure_accelerator(MachineState *ms);
> > +/* Register accelerator specific global properties */
> > +void accel_register_compat_props(AccelState *accel);
> >
> > #endif
> > diff --git a/vl.c b/vl.c
> > index 59fea15..4452d7a 100644
> > --- a/vl.c
> > +++ b/vl.c
> > @@ -4571,6 +4571,7 @@ int main(int argc, char **argv, char **envp)
> > exit (i == 1 ? 1 : 0);
> > }
> >
>
> I suggest a comment here warning about global property
> registration ordering. e.g.:
>
> /*
> * Ordering of global property registration matters:
> * - machine compat_props should override accelerator-specific
> * globals.
> * - user-provided globals (-global, and global properties
> * derived from other command-line options like -cpu) should
> * override machine compat_props and accelerator-specific
> * globals.
> */
I have a standalone patch (next one) to unify these three places and
added some comment. Would that suite?
Thanks,
--
Peter Xu
