On 2017/5/22 20:26, Paolo Bonzini wrote: > > > On 20/05/2017 02:46, Yang Hongyang wrote: >>>> How about do O(1) for stopping stage of live migration? >>>> Because the cpu is stopped in this phase, it wouldn't cause >>>> side effects IMHO, right? >>> O(1) update is probably not quite possible, but it's possible to do one >>> O(n) update. We tried doing that, but something in virtio broke. >> How were you do that? did you put extra begin/commit before/after destination >> guest load, so that the destination guest only do one update(of n devs) >> throughout >> the whole migration? >> > > Right. However, this is a problem if post_load hooks do DMA. You need > to move that DMA to a VM RunState change notifier.
Right. How about force memory_region_commit before post_load if a device has post_load hook? > > Paolo > -- Thanks, Yang
