On 31/05/2023 16:00, Markus Armbruster wrote:
Philippe Mathieu-Daudé <phi...@linaro.org> writes:
On 31/5/23 14:53, Mark Cave-Ayland wrote:
Also change the instantiation of the CPU to use object_initialize_child()
followed by a separate realisation.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayl...@ilande.co.uk>
---
hw/m68k/q800.c | 13 ++++++++-----
include/hw/m68k/q800.h | 2 ++
2 files changed, 10 insertions(+), 5 deletions(-)
diff --git a/hw/m68k/q800.c b/hw/m68k/q800.c
index 3730b30dd1..c34b2548ca 100644
--- a/hw/m68k/q800.c
+++ b/hw/m68k/q800.c
@@ -364,7 +364,7 @@ static uint8_t fake_mac_rom[] = {
static void q800_machine_init(MachineState *machine)
{
- M68kCPU *cpu = NULL;
+ Q800MachineState *m = Q800_MACHINE(machine);
int linux_boot;
int32_t kernel_size;
uint64_t elf_entry;
@@ -407,8 +407,10 @@ static void q800_machine_init(MachineState *machine)
}
/* init CPUs */
- cpu = M68K_CPU(cpu_create(machine->cpu_type));
- qemu_register_reset(main_cpu_reset, cpu);
+ object_initialize_child(OBJECT(machine), "cpu", &m->cpu,
+ M68K_CPU_TYPE_NAME("m68040"));
+ object_property_set_bool(OBJECT(&m->cpu), "realized", true, &error_fatal);
CPUs are QDev-based, shouldn't we use qdev_realize()?
Yes, we should.
[...]
Interesting. I remember thinking that CPUs were different, so I'm fairly sure I
borrowed this from some similar code in hw/arm :)
Shouldn't the above be directly equivalent to qdev_realize(dev, NULL, &error_fatal)
given that the CPU doesn't connect to a bus?
ATB,
Mark.
