Re: [PATCH 2/5] hw/s390x/pv: Un-inline s390_pv_init()

[Philippe Mathieu-Daudé]

On 16/12/22 23:04, Philippe Mathieu-Daudé wrote:
> There is no point in having s390_pv_init() inlined.
>
> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
> ---
>   hw/s390x/pv.c         | 13 +++++++++++++
>   include/hw/s390x/pv.h | 14 +-------------
>   2 files changed, 14 insertions(+), 13 deletions(-)


> diff --git a/include/hw/s390x/pv.h b/include/hw/s390x/pv.h
> index 9360aa1091..5bca5bcaf1 100644
> --- a/include/hw/s390x/pv.h
> +++ b/include/hw/s390x/pv.h
> @@ -12,7 +12,6 @@
>   #ifndef HW_S390_PV_H
>   #define HW_S390_PV_H
>   
> -#include "qapi/error.h"
>   #include "sysemu/kvm.h"
>   
>   #ifdef CONFIG_KVM
> @@ -78,17 +77,6 @@ static inline int kvm_s390_dump_completion_data(void *buff) 
> { return 0; }
>   #endif /* CONFIG_KVM */
>   
>   int s390_pv_kvm_init(ConfidentialGuestSupport *cgs, Error **errp);
> -static inline int s390_pv_init(ConfidentialGuestSupport *cgs, Error **errp)
> -{
> -    if (!cgs) {
> -        return 0;
> -    }
> -    if (kvm_enabled()) {
> -        return s390_pv_kvm_init(cgs, errp);
> -    }
> -
> -    error_setg(errp, "Protected Virtualization requires KVM");
> -    return -1;
> -}
> +int s390_pv_init(ConfidentialGuestSupport *cgs, Error **errp);

OK I understood why this function is inlined. It is called from:

$ git grep s390_pv_init
hw/s390x/pv.c:29:int s390_pv_init(ConfidentialGuestSupport *cgs, Error 
**errp)
hw/s390x/s390-virtio-ccw.c:259:    s390_pv_init(machine->cgs, &error_fatal);
include/hw/s390x/pv.h:86:int s390_pv_init(ConfidentialGuestSupport *cgs, 
Error **errp);

Now note in meson.build:

s390x_ss.add(when: 'CONFIG_KVM', if_true: files(
  'tod-kvm.c',
  's390-skeys-kvm.c',
  's390-stattrib-kvm.c',
  'pv.c',
  's390-pci-kvm.c',
))

So when CONFIG_KVM=false s390-virtio-ccw.c doesn't have to link with
the kvm-specific pv.o.

I'll rework this patch.