Yikes! Did I forget to include "kvm: nVMX: Set nested_run_pending
before prepare_vmcs02()"? Sorry. That obviates the need for
"from_vmentry," and is a little less awkward, I think. (The problem is
that kvm can exit to userspace with vmx->nested.nested_run_pending
set. If VMX state is saved at that time, then the restore code has to
behave as if "from_vmentry" is true. In any event, your version looks
fine, and I can always clean it up later (or not).
After commit 06750494993be18e ("kvm: nVMX: Refactor
nested_get_vmcs12_pages()"), vmx.c:9688 should read:
if (vmx->nested.virtual_apic_page) {
i.e., drop the '!'
On Wed, Feb 8, 2017 at 8:24 AM, Paolo Bonzini <pbonzini@xxxxxxxxxx> wrote:
>
>
> On 30/11/2016 21:03, Jim Mattson wrote:
>> This series of patches introduces checkpoint and restore operations
>> for the VMX state of a VM, so that VMX-capable VMs can be migrated.
>>
>> Two new ioctls are introduced: KVM_GET_VMX_STATE and
>> KVM_SET_VMX_STATE. The VMX state that is checkpointed/restored
>> consists of the VMXON region address (if any), the current VMCS
>> address (if any), and the cached current VMCS contents (if any). One
>> piece of implementation-specific state that is also
>> checkpointed/restored is the nested_run_pending bit.
>>
>> On live migration, our userspace process does not have access to guest
>> memory when the VM is set up on the target host, so some operations
>> that require a GPA->HPA mapping are deferred until the next vcpu_run
>> using a new vcpu request.
>
> I merged patches 1-6 into kvm/queue. Rebasing took a bit of work, so
> please take a look.
>
> Paolo
