Re: [PATCH 16/38] KVM: PPC: BookE: Add check_requests helper function

[Scott Wood <scottwood@freescale.com>]

On 08/14/2012 07:13 PM, Alexander Graf wrote:
> 
> On 15.08.2012, at 02:10, Scott Wood wrote:
> 
>> On 08/14/2012 06:04 PM, Alexander Graf wrote:
>>> We need a central place to check for pending requests in. Add one that
>>> only does the timer check we already do in a different place.
>>>
>>> Later, this central function can be extended by more checks.
>>>
>>> Signed-off-by: Alexander Graf <agraf@suse.de>
>>> ---
>>> arch/powerpc/kvm/booke.c |   24 +++++++++++++++++-------
>>> 1 files changed, 17 insertions(+), 7 deletions(-)
>>>
>>> diff --git a/arch/powerpc/kvm/booke.c b/arch/powerpc/kvm/booke.c
>>> index 1d4ce9a..bcf87fe 100644
>>> --- a/arch/powerpc/kvm/booke.c
>>> +++ b/arch/powerpc/kvm/booke.c
>>> @@ -419,13 +419,6 @@ static void kvmppc_core_check_exceptions(struct kvm_vcpu *vcpu)
>>> 	unsigned long *pending = &vcpu->arch.pending_exceptions;
>>> 	unsigned int priority;
>>>
>>> -	if (vcpu->requests) {
>>> -		if (kvm_check_request(KVM_REQ_PENDING_TIMER, vcpu)) {
>>> -			smp_mb();
>>> -			update_timer_ints(vcpu);
>>> -		}
>>> -	}
>>> -
>>> 	priority = __ffs(*pending);
>>> 	while (priority < BOOKE_IRQPRIO_MAX) {
>>> 		if (kvmppc_booke_irqprio_deliver(vcpu, priority))
>>> @@ -461,6 +454,14 @@ int kvmppc_core_prepare_to_enter(struct kvm_vcpu *vcpu)
>>> 	return r;
>>> }
>>>
>>> +static void kvmppc_check_requests(struct kvm_vcpu *vcpu)
>>> +{
>>> +	if (vcpu->requests) {
>>> +		if (kvm_check_request(KVM_REQ_PENDING_TIMER, vcpu))
>>> +			update_timer_ints(vcpu);
>>> +	}
>>> +}
>>> +
>>> /*
>>>  * Common checks before entering the guest world.  Call with interrupts
>>>  * disabled.
>>> @@ -485,6 +486,15 @@ static int kvmppc_prepare_to_enter(struct kvm_vcpu *vcpu)
>>> 			break;
>>> 		}
>>>
>>> +		smp_mb();
>>> +		if (vcpu->requests) {
>>> +			/* Make sure we process requests preemptable */
>>> +			local_irq_enable();
>>> +			kvmppc_check_requests(vcpu);
>>> +			local_irq_disable();
>>> +			continue;
>>> +		}
>>
>> What previous memory access is the smp_mb() ordering against?
> 
> Against itself really (see the continue?). I might have just gotten this wrong though :).

Ordering is already ensured by the hardware in that case since it's the
same address -- and the stuff inside the if statement is more than
enough to ensure the compiler doesn't cache an old value of vcpu->requests.

That said, I think the problem is not that the smp_mb() shouldn't be
there, but that there should be a vcpu->mode setting before it.

-Scott