Re: [PATCH v2: kvm 4/4] Fix hotplug of CPUs for KVM.

[Zachary Amsden <zamsden@redhat.com>]

On 09/28/2009 10:30 PM, Avi Kivity wrote:
> On 09/29/2009 06:04 AM, Zachary Amsden wrote:
>> Both VMX and SVM require per-cpu memory allocation, which is done at 
>> module
>> init time, for only online cpus.
>>
>> Backend was not allocating enough structure for all possible CPUs, so
>> new CPUs coming online could not be hardware enabled.
>>
>> diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
>> index e27b7a9..2cd8bc2 100644
>> --- a/virt/kvm/kvm_main.c
>> +++ b/virt/kvm/kvm_main.c
>> @@ -1716,9 +1716,6 @@ static int kvm_cpu_hotplug(struct 
>> notifier_block *notifier, unsigned long val,
>>   {
>>       int cpu = (long)v;
>>
>> -    if (!kvm_usage_count)
>> -        return NOTIFY_OK;
>> -
>>       val&= ~CPU_TASKS_FROZEN;
>>       switch (val) {
>>       case CPU_DYING:
>
> I still don't see how this bit can work.  Maybe if we move the 
> notification registration to the point where kvm_usage_count is bumped.

That was stray junk in the patch.  Let me rediff...