Avi Kivity wrote:
> Zhang, Xiantao wrote:
>> Hi, Avi
>> This is the key fix for 2.6.28 merge. Without this patch, guest
>> may hang once configured with more than 2 vcpus, it is because x86
>> side changed the halt handling's common logic, but it misses to
>> change ia64 side.
>>
>
>> @@ -398,8 +400,11 @@ static int handle_global_purge(struct kvm_vcpu
>> *vcpu, struct kvm_run *kvm_run)
>>
>> if (kvm->vcpus[i]->cpu != -1) {
>> call_data.vcpu = kvm->vcpus[i];
>> - smp_call_function_single(kvm->vcpus[i]->cpu,
>> + if (kvm->vcpus[i]->cpu != smp_processor_id())
>> +
>> smp_call_function_single(kvm->vcpus[i]->cpu,
>> vcpu_global_purge, &call_data,
>> 1);
>> + else
>> + vcpu_global_purge(&call_data);
>>
>
> smp_call_function_single() will call the function on the current cpu
> if instructed, so this change is unneeded.
Aha, Thanks for pointing this. Yes, it is unnecessary. Could you help to
remove it directly ? or need me have a resend.
Xiantao
--
To unsubscribe from this list: send the line "unsubscribe kvm-ia64" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
