Re: [RFC PATCH] target/openrisc: Call cpu_openrisc_clock_init() in cpu_realize()

[Richard Henderson]

On 1/14/25 15:13, Philippe Mathieu-Daudé wrote:
> OpenRISC timer is architecturally tied to the CPU.
>
> It doesn't belong to the machine init() code to
> instanciate it: move its creation when a vCPU is
> realized (after being created).
>
> Reported-by: Paolo Bonzini <pbonzini@redhat.com>
> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
> ---
>   hw/openrisc/openrisc_sim.c | 2 --
>   hw/openrisc/virt.c         | 2 --
>   target/openrisc/cpu.c      | 2 ++
>   3 files changed, 2 insertions(+), 4 deletions(-)

Acked-by: Richard Henderson <richard.henderson@linaro.org>

r~

> diff --git a/hw/openrisc/openrisc_sim.c b/hw/openrisc/openrisc_sim.c
> index e0da4067ba3..d7f97d19913 100644
> --- a/hw/openrisc/openrisc_sim.c
> +++ b/hw/openrisc/openrisc_sim.c
> @@ -306,8 +306,6 @@ static void openrisc_sim_init(MachineState *machine)
>               exit(1);
>           }
>   
> -        cpu_openrisc_clock_init(cpus[n]);
> -
>           qemu_register_reset(main_cpu_reset, cpus[n]);
>       }
>   
> diff --git a/hw/openrisc/virt.c b/hw/openrisc/virt.c
> index 7b60bf85094..676cf8d692d 100644
> --- a/hw/openrisc/virt.c
> +++ b/hw/openrisc/virt.c
> @@ -487,8 +487,6 @@ static void openrisc_virt_init(MachineState *machine)
>               exit(1);
>           }
>   
> -        cpu_openrisc_clock_init(cpus[n]);
> -
>           qemu_register_reset(main_cpu_reset, cpus[n]);
>       }
>   
> diff --git a/target/openrisc/cpu.c b/target/openrisc/cpu.c
> index b7bab0d7abf..291fec04fda 100644
> --- a/target/openrisc/cpu.c
> +++ b/target/openrisc/cpu.c
> @@ -165,6 +165,8 @@ static void openrisc_cpu_realizefn(DeviceState *dev, Error 
> **errp)
>       qemu_init_vcpu(cs);
>       cpu_reset(cs);
>   
> +    cpu_openrisc_clock_init(OPENRISC_CPU(dev));
> +
>       occ->parent_realize(dev, errp);
>   }
>