Re: [RFC v1] sched/fair: search a task from the tail of the queue

[Peter Zijlstra]

On Fri, Aug 25, 2017 at 12:11:31AM +0200, Uladzislau Rezki (Sony) wrote:
> From: Uladzislau Rezki <urezki@xxxxxxxxx>
> 
> As a first step this patch makes cfs_tasks list as MRU one.
> It means, that when a next task is picked to run on physical
> CPU it is moved to the front of the list.
> 
> Thefore, the cfs_tasks list is more or less sorted (except woken
> tasks) starting from recently given CPU time tasks toward tasks
> with max wait time in a run-queue, i.e. MRU list.
> 
> Second, as part of the load balance operation, this approach
> starts detach_tasks()/detach_one_task() from the tail of the
> queue instead of the head, giving some advantages:
> 
> - tends to pick a task with highest wait time;
> - tasks located in the tail are less likely cache-hot,
>   therefore the can_migrate_task() decision is higher.
> 
> hackbench illustrates slightly better performance. For example
> doing 1000 samples and 40 groups on i5-3320M CPU, it shows below
> figures:
> 
> default: 0.644 avg
> patched: 0.637 avg
> 
> Signed-off-by: Uladzislau Rezki (Sony) <urezki@xxxxxxxxx>
> ---
>  kernel/sched/fair.c | 19 ++++++++++++++-----
>  1 file changed, 14 insertions(+), 5 deletions(-)
> 
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index c77e4b1d51c0..cda281c6bb29 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -6357,7 +6357,7 @@ pick_next_task_fair(struct rq *rq, struct task_struct *prev, struct rq_flags *rf
>  	if (hrtick_enabled(rq))
>  		hrtick_start_fair(rq, p);
>  
> -	return p;
> +	goto done;
>  simple:
>  	cfs_rq = &rq->cfs;
>  #endif
> @@ -6378,6 +6378,14 @@ pick_next_task_fair(struct rq *rq, struct task_struct *prev, struct rq_flags *rf
>  	if (hrtick_enabled(rq))
>  		hrtick_start_fair(rq, p);
>  
> +done: __maybe_unused
> +	/*
> +	 * Move the next running task to the front of
> +	 * the list, so our cfs_tasks list becomes MRU
> +	 * one.
> +	 */
> +	list_move(&se->group_node, &rq->cfs_tasks);
> +
>  	return p;
>  
>  idle:

Could you also run something like:

$ taskset 1 perf bench sched pipe

to make sure the added list_move() doesn't hurt, I'm not sure group_node
and cfs_tasks are in cachelines we already touch for that operation.

And if you can see that list_move() hurt in "perf annotate", try moving
those members around to lines that we already need anyway.