[Xen-devel] How to migrate vCPUs based on Credit Scheduler

[甘清甜 <qingtiangan@xxxxxxxxx>]

Hi,

 I'm now designing new vCPU scheduler in Xen, and trying to implement 

the scheduler based on the Credit scheduler in Xen-4.5.1. But I encountered

 come problems when debuging the code.

Most of the code modification is done in function csched_schedule() in 

file: xen/common/csched_schedule.c . And the core code is as followed:

if( vcpu_runnable(current) )

{

        if( match the migration contition )

        {

            cpu_affinity = pick_pcpu_runq();  // this function is defined by myself

            

            pcpulock = pcpu_schedule_lock_irqsave(cpu_affinity, 

                   &pcpulock_flag);

            TRACE_3D(TRC_CSCHED_STOLEN_VCPU, cpu_affinity  , domain_id,

                   vcpu_id);

            SCHED_VCPU_STAT_CRANK(scurr, migrate_q);

            SCHED_STAT_CRANK(migrate_queued);

            WARN_ON(scurr->vcpu->is_urgent);

        scurr->vcpu->processor = cpu_affinity;

             __runq_insert(cpu_affinity, scurr);

            pcpu_schedule_unlock_irqrestore(pcpulock, pcpulock_flag, cpu_affinity  );

        }

        else 

            __runq_insert(cpu, scurr);

}

else 

        BUG_ON( is_idle_vcpu(current) || list_empty(runq) );

I try to run the modified Xen. But according to the log I found that, 

although I insert the vCPU into the runqueue  of another pCPU, the 

vCPU still appears at the old pCPU in the following scheduling period. 

Now I have a few questions:

1. Does the Xen scheduler framework support changing the pCPU of a 

vCPU after using out the scheduling time slice, but not just to steal one 

vCPU from runqueue of other pCPU in load_balance period?

2. If yes, what status of the vCPU should be changed before inserting 

the vCPU into the destination pCPU?

Thank you very much!

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel