| Message ID | 20240801090117.3841080-10-tabba@google.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | KVM: Restricted mapping of guest_memfd at the host and pKVM/arm64 support | expand |
Hi Fuad, On Thu, 2024-08-01 at 10:01 +0100, Fuad Tabba wrote: > Implement kvm_arch_has_private_mem() in arm64, making it > dependent on the configuration option. > > Signed-off-by: Fuad Tabba <tabba@google.com> > --- > arch/arm64/include/asm/kvm_host.h | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h > index 36b8e97bf49e..8f7d78ee9557 100644 > --- a/arch/arm64/include/asm/kvm_host.h > +++ b/arch/arm64/include/asm/kvm_host.h > @@ -1414,4 +1414,7 @@ bool kvm_arm_vcpu_stopped(struct kvm_vcpu *vcpu); > (pa + pi + pa3) == 1; \ > }) > > +#define kvm_arch_has_private_mem(kvm) \ > + (IS_ENABLED(CONFIG_KVM_PRIVATE_MEM) && is_protected_kvm_enabled()) > + Would it make sense to have some ARM equivalent of KVM_X86_SW_PROTECTED_VM here? Both for easier testing of guest_memfd on ARM, as well as for future non-coco usecases. > #endif /* __ARM64_KVM_HOST_H__ */ > -- > 2.46.0.rc1.232.g9752f9e123-goog > Best, Patrick
Hi Patrick, On Thu, 15 Aug 2024 at 07:27, Patrick Roy <roypat@amazon.co.uk> wrote: > > Hi Fuad, > > On Thu, 2024-08-01 at 10:01 +0100, Fuad Tabba wrote: > > Implement kvm_arch_has_private_mem() in arm64, making it > > dependent on the configuration option. > > > > Signed-off-by: Fuad Tabba <tabba@google.com> > > --- > > arch/arm64/include/asm/kvm_host.h | 3 +++ > > 1 file changed, 3 insertions(+) > > > > diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h > > index 36b8e97bf49e..8f7d78ee9557 100644 > > --- a/arch/arm64/include/asm/kvm_host.h > > +++ b/arch/arm64/include/asm/kvm_host.h > > @@ -1414,4 +1414,7 @@ bool kvm_arm_vcpu_stopped(struct kvm_vcpu *vcpu); > > (pa + pi + pa3) == 1; \ > > }) > > > > +#define kvm_arch_has_private_mem(kvm) \ > > + (IS_ENABLED(CONFIG_KVM_PRIVATE_MEM) && is_protected_kvm_enabled()) > > + > > Would it make sense to have some ARM equivalent of > KVM_X86_SW_PROTECTED_VM here? Both for easier testing of guest_memfd on > ARM, as well as for future non-coco usecases. I don't really have a strong opinion about this. I thought that if/when that were to happen, it would be trivial to modify this macro. Cheers, /fuad > > #endif /* __ARM64_KVM_HOST_H__ */ > > -- > > 2.46.0.rc1.232.g9752f9e123-goog > > > > Best, > Patrick
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h index 36b8e97bf49e..8f7d78ee9557 100644 --- a/arch/arm64/include/asm/kvm_host.h +++ b/arch/arm64/include/asm/kvm_host.h @@ -1414,4 +1414,7 @@ bool kvm_arm_vcpu_stopped(struct kvm_vcpu *vcpu); (pa + pi + pa3) == 1; \ }) +#define kvm_arch_has_private_mem(kvm) \ + (IS_ENABLED(CONFIG_KVM_PRIVATE_MEM) && is_protected_kvm_enabled()) + #endif /* __ARM64_KVM_HOST_H__ */
Implement kvm_arch_has_private_mem() in arm64, making it dependent on the configuration option. Signed-off-by: Fuad Tabba <tabba@google.com> --- arch/arm64/include/asm/kvm_host.h | 3 +++ 1 file changed, 3 insertions(+)