On Wed, Aug 27, 2025 at 12:24:39PM +0100, Daniel P. Berrangé wrote: > So every MSHV vCPU has a corresponding Linux thread, similar > to the model with KVM. In libvirt we rely on the vCPU thread > being controllable with all the normal Linux process related > APIs. For example, setting thread CPU affinity, setting NUMA > memory policy, setting scheduler priorities, putting threads > into cgroups and applying a wide variety of cgroup controls. > > Will there be any significant "gotchas" with the threads for > MSHV vCPUs, that would mean the above libvirt controls would > either raise errors, or silently not have any effect ? >
Hi Daniel, I am not aware of any such gotchas. The MSHV vCPU threads should be regular threads that spend most of their time blocked in ioctl(MSHV_RUN) calls, and as such they should be controllable by the facilities you mentioned. I know that that folks who tested this code have been using numactl for reliable performance assessments without running into issues. best, magnus
