On 9 October 2014 17:18, Peter Zijlstra [off-list ref] wrote:

On Thu, Oct 09, 2014 at 04:18:02PM +0200, Vincent Guittot wrote:

quoted

On 9 October 2014 14:16, Peter Zijlstra [off-list ref] wrote:

quoted

On Tue, Oct 07, 2014 at 02:13:36PM +0200, Vincent Guittot wrote:

quoted

+static inline bool
+group_has_capacity(struct lb_env *env, struct sg_lb_stats *sgs)
 {
+     if ((sgs->group_capacity * 100) >
+                     (sgs->group_usage * env->sd->imbalance_pct))
+             return true;

Why the imb_pct there? We're looking for 100% utilization, not 130 or
whatnot, right?

Having exactly 100% is quite difficult because of various rounding.
So i have added a margin/threshold to prevent any excessive change of the state.
I have just to use the same margin/threshold than in other place in
load balance.

Yet you failed to mention this anywhere. Also does it really matter?

yes i think it latter because it give a more stable view of the
"overload state" and "have  free capacity state" of the CPU.
One additional point is that the imbalance_pct will ensure that a
cpu/group will not been seen as having capacity if its available
capacity is only 1-5% which will generate spurious task migration

I will add these details in the commit log and in a comment in the code