Messages in this thread | | | Date | Thu, 23 Jun 2011 14:40:41 +0800 | From | Xiao Guangrong <> | Subject | Re: [PATCH v2 21/22] KVM: MMU: mmio page fault support |
| |
On 06/23/2011 11:19 AM, Xiao Guangrong wrote:
>> Otherwise you can move the mmio info from an mmio spte back to >> mmio_gva/mmio_gfn after a TLB flush, without rereading the guest >> pagetable. >> > > We do not read the guest page table when mmio page fault occurred, > we just do it as you say: > - Firstly, walk the shadow page table to get the mmio spte > - Then, cache the mmio spte info to mmio_gva/mmio_gfn > > I missed your meaning? >
Marcelo, sorry for my poor English, i missed your meaning, please ignore this. :(
| |