On 12/08/21 01:37, Jing Zhang wrote:
A per VM stat dirty_pages is added to record the number of dirtied pages in the life cycle of a VM. The growth rate of this stat is a good indicator during the process of live migrations. The exact number of dirty pages at the moment doesn't matter. That's why we define dirty_pages as a cumulative counter instead of an instantaneous one.
Why not make it a per-CPU stat? mark_page_dirty_in_slot can use kvm_get_running_vcpu() and skip the logging in the rare case it's NULL.
Paolo
