On 08/03/2012 12:36 AM, Hu Tao wrote:
From: Wen Congyang <wency(a)cn.fujitsu.com>
Create a new cgroup and move all hypervisor threads to the new cgroup.
And then we can do the other things:
1. limit only vcpu usage rather than the whole qemu
2. limit for hypervisor threads(include vhost-net threads)
Signed-off-by: Wen Congyang <wency(a)cn.fujitsu.com>
Signed-off-by: Tang Chen <tangchen(a)cn.fujitsu.com>
Signed-off-by: Hu Tao <hutao(a)cn.fujitsu.com>
---
+int qemuSetupCgroupForHypervisor(struct qemud_driver *driver,
+ virDomainObjPtr vm)
+{
+
+ for (i = 0; i < VIR_CGROUP_CONTROLLER_LAST; i++) {
+ if (!qemuCgroupControllerActive(driver, i)) {
+ VIR_WARN("cgroup %d is not active", i);
+ continue;
+ }
Do we need to do this for every controller, or only for the cpu and
cpuacct controllers?
Again, looks reasonable in isolation.
--
Eric Blake eblake(a)redhat.com +1-919-301-3266
Libvirt virtualization library
http://libvirt.org