On Sat, 20 Oct 2018 at 11:18, Junaid Shahid <junaids@xxxxxxxxxx> wrote: > > This patch series implements a new dirty logging mode utilizing the Dirty > bits in the SPTEs, when available, instead of relying on write protection. > On hosts that do not support PML, this can result in significant > improvement in VM performance for some workloads during the brownout phase > of a live migration when compared to write protection based dirty logging, > because of the elimination of all the VM exits on memory writes. > > However, this new mode does incur additional time in the get_dirty_log > IOCTL because of having to scan all the SPTEs for dirtiness, which can > in some cases, result in slightly elongating the blackout time. Two > strategies can be used to mitigate that. The first, included in this > series, is to have the ability to dynamically switch between D Bit based > and write protection based dirty logging modes. Then the mode could be > switched to write protection just before entering blackout, thus incurring > less time in the get_dirty_log IOCTL during blackout, while keeping the > reduced overhead of the D Bit based mode during brownout. Secondly, the > scan in the get_dirty_log IOCTL can be parallelized, which results in that > IOCTL becoming faster for all modes. That patch series will be sent out > separately a little later. Interesting, I also implement a version of PTE D Bit based dirty logging 4 years ago w/o sending to upstream. The PTE D Bit based mode will switch to write protection mode each time before KVM_GET_DIRTY_LOG ioctl or just once before the xx_start? Regards, Wanpeng Li
