Re: [PATCH RFC 5/5] sched/fair: remove impossible condition from find_idlest_group_cpu

2017-10-30 Thread Vincent Guittot
On 30 October 2017 at 17:19, Joel Fernandes wrote: > Hi Vincent, > > On Mon, Oct 30, 2017 at 9:00 AM, Vincent Guittot > wrote: >> On 28 October 2017 at 11:59, Joel Fernandes wrote: >>> find_idlest_group_cpu goes through CPUs of a group previous selected by >>> find_idlest_group. find_idlest_grou

Re: [PATCH RFC 5/5] sched/fair: remove impossible condition from find_idlest_group_cpu

2017-10-30 Thread Joel Fernandes
Hi Vincent, On Mon, Oct 30, 2017 at 9:00 AM, Vincent Guittot wrote: > On 28 October 2017 at 11:59, Joel Fernandes wrote: >> find_idlest_group_cpu goes through CPUs of a group previous selected by >> find_idlest_group. find_idlest_group returns NULL if the local group is the >> selected one and d

Re: [PATCH RFC 5/5] sched/fair: remove impossible condition from find_idlest_group_cpu

2017-10-30 Thread Vincent Guittot
On 28 October 2017 at 11:59, Joel Fernandes wrote: > find_idlest_group_cpu goes through CPUs of a group previous selected by > find_idlest_group. find_idlest_group returns NULL if the local group is the > selected one and doesn't execute find_idlest_group_cpu if the group to which > 'cpu' belongs

Re: [PATCH RFC 5/5] sched/fair: remove impossible condition from find_idlest_group_cpu

2017-10-30 Thread Brendan Jackman
On Sat, Oct 28 2017 at 09:59, Joel Fernandes wrote: > find_idlest_group_cpu goes through CPUs of a group previous selected by > find_idlest_group. find_idlest_group returns NULL if the local group is the > selected one and doesn't execute find_idlest_group_cpu if the group to which > 'cpu' belongs

[PATCH RFC 5/5] sched/fair: remove impossible condition from find_idlest_group_cpu

2017-10-28 Thread Joel Fernandes
find_idlest_group_cpu goes through CPUs of a group previous selected by find_idlest_group. find_idlest_group returns NULL if the local group is the selected one and doesn't execute find_idlest_group_cpu if the group to which 'cpu' belongs to is chosen. So we're always guaranteed to call find_idlest