Re: [PATCH v4 5/5] x86: KVM: SEV: implement kvm_lock_all_vcpus and use it

[Sean Christopherson <seanjc@google.com>]

On Wed, Apr 30, 2025, Maxim Levitsky wrote:
> diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
> index 834f08dfa24c..9211b07b0565 100644
> --- a/virt/kvm/kvm_main.c
> +++ b/virt/kvm/kvm_main.c
> @@ -1392,6 +1392,31 @@ int kvm_trylock_all_vcpus(struct kvm *kvm)
>  }
>  EXPORT_SYMBOL_GPL(kvm_trylock_all_vcpus);
>  
> +/*
> + * Lock all of the VM's vCPUs.
> + * Assumes that the kvm->lock is held.

Add a lockdep assertion instead of a comment.

> + * Returns -EINTR if the process is killed.
> + */
> +int kvm_lock_all_vcpus(struct kvm *kvm)
> +{
> +	struct kvm_vcpu *vcpu;
> +	unsigned long i, j;
> +
> +	kvm_for_each_vcpu(i, vcpu, kvm)

Needs curly braces.

> +		if (mutex_lock_killable_nest_lock(&vcpu->mutex, &kvm->lock))

I'd rather return mutex_lock_killable_nest_lock()'s error code verbatim.  Then
the function comment can go away, because the only thing remaining would be:

	/*
	 * Lock all of the VM's vCPUs.
	 /*

and that should be completely self-explanatory.  E.g.

int kvm_lock_all_vcpus(struct kvm *kvm)
{
	struct kvm_vcpu *vcpu;
	unsigned long i, j;
	int r;

	lockdep_assert_held(&kvm->lock);

	kvm_for_each_vcpu(i, vcpu, kvm) {
		r = mutex_lock_killable_nest_lock(&vcpu->mutex, &kvm->lock);
		if (r)
			goto out_unlock;
	}
	return 0;

out_unlock:
	kvm_for_each_vcpu(j, vcpu, kvm) {
		if (i == j)
			break;
		mutex_unlock(&vcpu->mutex);
	}
	return r;
}
EXPORT_SYMBOL_GPL(kvm_lock_all_vcpus);

-- 
kvm-riscv mailing list
kvm-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kvm-riscv