On 11/23/2009 04:06 PM, Gleb Natapov wrote:
If guest access swapped out memory do not swap it in from vcpu thread
context. Setup slow work to do swapping and send async page fault to
a guest.
Allow async page fault injection only when guest is in user mode since
otherwise guest may be in non-sleepable context and will not be able to
reschedule.
+
+void kvm_arch_inject_async_page_present(struct kvm_vcpu *vcpu,
+ struct kvm_async_pf *work)
+{
+ put_user(KVM_PV_REASON_PAGE_READY, vcpu->arch.apf_data);
+ kvm_inject_page_fault(vcpu, work->arch.token, 0);
+ trace_kvm_send_async_pf(work->arch.token, work->gva,
+ KVM_PV_REASON_PAGE_READY);
+}
What if the guest is now handling a previous asynv pf or ready
notification? We're clobbering the data structure.
+
+bool kvm_arch_can_inject_async_page_present(struct kvm_vcpu *vcpu)
+{
+ return !kvm_event_needs_reinjection(vcpu)&&
+ kvm_x86_ops->interrupt_allowed(vcpu);
+}
Okay, so this is only allowed with interrupts disabled. Need to make
sure the entire pf path up to async pf executes with interrupts disabled.
--
error compiling committee.c: too many arguments to function
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
