On Mon, 16 Jun 2014 10:14:16 -0600 Eric Blake <ebl...@redhat.com> wrote:
> On 06/05/2014 08:36 AM, Igor Mammedov wrote: > > ... using TYPE_ACPI_DEVICE_IF interface. > > Which provides status reporting of ACPI declared memory devices > > > > Signed-off-by: Igor Mammedov <imamm...@redhat.com> > > --- > > v2: > > - set slot-type to DIMM > > - take into account that 'device' field could be optional > > --- > > hw/acpi/ich9.c | 7 +++++++ > > hw/acpi/memory_hotplug.c | 31 +++++++++++++++++++++++++++++++ > > hw/acpi/piix4.c | 11 +++++++++++ > > hw/isa/lpc_ich9.c | 3 +++ > > include/hw/acpi/ich9.h | 3 +++ > > include/hw/acpi/memory_hotplug.h | 1 + > > 6 files changed, 56 insertions(+), 0 deletions(-) > > > > > +void acpi_memory_ospm_status(MemHotplugState *mem_st, ACPIOSTInfoList > > ***list); > > Triple indirection looks unusual, but that doesn't make it wrong :) I plan to move ACPI code inside of interface handlers and drop piix4/ich9 stubs in favor of using QOM casting directly. So it would be -1 layer of indirection and not only for ospm_status(). In addition interface allows to avoid adding global function + stub where functionality is not implemented. > > Reviewed-by: Eric Blake <ebl...@redhat.com> >
