Re: [PATCH v12 10/31] LoongArch: KVM: Implement vcpu ENABLE_CAP ioctl interface

[Tianrui Zhao]

在 2023年06月05日 21:12, bibo, mao 写道:
> 在 2023/5/30 09:52, Tianrui Zhao 写道:
> > Implement LoongArch vcpu KVM_ENABLE_CAP ioctl interface.
> >
> > Signed-off-by: Tianrui Zhao <zhaotianrui@xxxxxxxxxxx>
> > ---
> >   arch/loongarch/kvm/vcpu.c | 26 ++++++++++++++++++++++++++
> >   1 file changed, 26 insertions(+)
> >
> > diff --git a/arch/loongarch/kvm/vcpu.c b/arch/loongarch/kvm/vcpu.c
> > index 278fbafc59b4..5a88f815c412 100644
> > --- a/arch/loongarch/kvm/vcpu.c
> > +++ b/arch/loongarch/kvm/vcpu.c
> > @@ -186,6 +186,23 @@ int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
> >   	return 0;
> >   }
> >   
> > +static int kvm_vcpu_ioctl_enable_cap(struct kvm_vcpu *vcpu,
> > +				     struct kvm_enable_cap *cap)
> > +{
> > +	int r = 0;
> > +
> > +	if (!kvm_vm_ioctl_check_extension(vcpu->kvm, cap->cap))
> > +		return -EINVAL;
> It is a little strange to check extension of the whole vm in enable vcap capability.
> can we change to usage like  general architectures?
Thanks, I look up this interface of other archs and re-consider it, and 
I think it should be removed.

Thanks
Tianrui Zhao
> > +	if (cap->flags)
> > +		return -EINVAL;
> > +	if (cap->args[0])
> > +		return -EINVAL;
> > +	if (cap->cap)
> > +		return -EINVAL;
> Do we need check args[0] and cap here ?
>
> Regards
> Bibo, Mao
No need, I will remove the two conditions.

Thanks
Tianrui Zhao
> > +
> > +	return r;
> > +}
> > +
> >   long kvm_arch_vcpu_ioctl(struct file *filp,
> >   			 unsigned int ioctl, unsigned long arg)
> >   {
> > @@ -209,6 +226,15 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
> >   			r = _kvm_get_reg(vcpu, &reg);
> >   		break;
> >   	}
> > +	case KVM_ENABLE_CAP: {
> > +		struct kvm_enable_cap cap;
> > +
> > +		r = -EFAULT;
> > +		if (copy_from_user(&cap, argp, sizeof(cap)))
> > +			break;
> > +		r = kvm_vcpu_ioctl_enable_cap(vcpu, &cap);
> > +		break;
> > +	}
> >   	default:
> >   		r = -ENOIOCTLCMD;
> >   		break;