Re: [Xen-devel] [PATCH v3 4/4] sched: credit2: consider per-vcpu soft affinity

2015-04-22 Thread George Dunlap
On Thu, Mar 26, 2015 at 9:48 AM, Justin T. Weaver wrote: [snip] > Added a function to determine the number of soft cpus gained (or lost) by a > given vcpu if it is migrated from a given source run queue to a given > destination run queue. > > Modified algorithm in balance_load and consider... > *

Re: [Xen-devel] [PATCH v3 4/4] sched: credit2: consider per-vcpu soft affinity

2015-04-20 Thread George Dunlap
On Thu, Mar 26, 2015 at 9:48 AM, Justin T. Weaver wrote: > * choose_cpu > > choose_cpu now tries to find the run queue with the most cpus in the given > vcpu's soft affinity. It uses minimum run queue load as a tie breaker. [snip] > * choose_cpu: added balance loop to find cpu for given vcpu tha

Re: [Xen-devel] [PATCH v3 4/4] sched: credit2: consider per-vcpu soft affinity

2015-03-31 Thread George Dunlap
On 03/26/2015 09:48 AM, Justin T. Weaver wrote: > when making decisions for vcpus (run queue assignment, run queue migration, > cpu assignment, and cpu tickling). > > Added soft affinity balancing loops to... > * get_fallback_cpu > * runq_tickle (one for idle, but not tickled; one for non-idle,

[Xen-devel] [PATCH v3 4/4] sched: credit2: consider per-vcpu soft affinity

2015-03-26 Thread Justin T. Weaver
when making decisions for vcpus (run queue assignment, run queue migration, cpu assignment, and cpu tickling). Added soft affinity balancing loops to... * get_fallback_cpu * runq_tickle (one for idle, but not tickled; one for non-idle, and not tickled) * choose_cpu choose_cpu now tries to f