Hi james, Below is the check [1] I was referring to in today's LOD call and it looks you have already fixed it in your latest patches [2] posted
virtual_cpu_hotplug/rfc/v0.2
#Hunk
diff --git a/arch/arm64/kvm/hypercalls.c b/arch/arm64/kvm/hypercalls.c index fcab358de228..3c2136cd7a3f 100644 --- a/arch/arm64/kvm/hypercalls.c +++ b/arch/arm64/kvm/hypercalls.c @@ -126,7 +126,7 @@ int kvm_hvc_user(struct kvm_vcpu *vcpu) int i; struct kvm_run *run = vcpu->run;
if (test_bit(KVM_ARCH_FLAG_HVC_TO_USER,
&vcpu->kvm->arch.flags)) {
if (!test_bit(KVM_ARCH_FLAG_HVC_TO_USER,
&vcpu->kvm->arch.flags)) {
smccc_set_retval(vcpu, SMCCC_RET_NOT_SUPPORTED, 0, 0,
0);
return 1; }Makes sense - thanks! I couldn't test this without the Qemu changes.
[1] https://op-lists.linaro.org/archives/list/linaro-open-discussions@op-lists.l... [2] https://lore.kernel.org/linux-arm-kernel/20230203135043.409192-30-james.mors...