Re: [PATCH] sched/fair: Refactor cpu_util_without()

From: Vincent Guittot
Date: Wed Mar 02 2022 - 04:10:04 EST


On Tue, 1 Mar 2022 at 18:17, Dietmar Eggemann <dietmar.eggemann@xxxxxxx> wrote:
>
> Except the 'task has no contribution or is new' condition at the
> beginning of cpu_util_without(), a cpu_util_next(..., dst_cpu = -1)
> call can replace the rest of this function.
>
> The UTIL_EST specific check for a race between select_task_rq_fair()
> and detach_task() in case of an enqueued or running WF_EXEC task has
> to be moved to cpu_util_next().
> This was initially introduced by commit c469933e7721
> ("sched/fair: Fix cpu_util_wake() for 'execl' type workloads").
> UnixBench's `execl` throughput tests were run on the dual socket 40
> CPUs Intel E5-2690 v2 machine to make sure the regression doesn't
> occur again.
>
> Signed-off-by: Dietmar Eggemann <dietmar.eggemann@xxxxxxx>

I have only minor comment

> ---
>
> There is still a lot of CPU utilization related code. cpu_util_without()
> and cpu_util_next() are very similar. In fact the former can be
> refactored to use a call to the latter to be able to remove some
> redundancy.
>
> kernel/sched/fair.c | 143 ++++++++++++++++++--------------------------
> 1 file changed, 59 insertions(+), 84 deletions(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 16874e112fe6..c084c2e29e40 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -6511,6 +6511,64 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target)
> return target;
> }
>
> +/*
> + * Predicts what cpu_util(@cpu) would return if @p was migrated (and enqueued)
> + * to @dst_cpu.
> + */
> +static unsigned long cpu_util_next(int cpu, struct task_struct *p, int dst_cpu)
> +{
> + struct cfs_rq *cfs_rq = &cpu_rq(cpu)->cfs;
> + unsigned long util_est, util = READ_ONCE(cfs_rq->avg.util_avg);
> +
> + /*
> + * If @p migrates from @cpu to another, remove its contribution. Or,
> + * if @p migrates from another CPU to @cpu, add its contribution. In
> + * the other cases, @cpu is not impacted by the migration, so the
> + * util_avg should already be correct.
> + */
> + if (task_cpu(p) == cpu && dst_cpu != cpu)
> + lsub_positive(&util, task_util(p));
> + else if (task_cpu(p) != cpu && dst_cpu == cpu)
> + util += task_util(p);
> +
> + if (sched_feat(UTIL_EST)) {
> + util_est = READ_ONCE(cfs_rq->avg.util_est.enqueued);
> +
> + /*
> + * During wake-up, the task isn't enqueued yet and doesn't
> + * appear in the cfs_rq->avg.util_est.enqueued of any rq,
> + * so just add it (if needed) to "simulate" what will be
> + * cpu_util after the task has been enqueued.
> + */
> + if (dst_cpu == cpu)
> + util_est += _task_util_est(p);
> +

Could you add a comment that explains why the addition above will not
be removed below by the lsub_positive below so it isn't worth trying
to optimize such a case?

> + /*
> + * Despite the following checks we still have a small window
> + * for a possible race, when an execl's select_task_rq_fair()
> + * races with LB's detach_task():
> + *
> + * detach_task()
> + * p->on_rq = TASK_ON_RQ_MIGRATING;
> + * ---------------------------------- A
> + * deactivate_task() \
> + * dequeue_task() + RaceTime
> + * util_est_dequeue() /
> + * ---------------------------------- B
> + *
> + * The additional check on "current == p" it's required to
> + * properly fix the execl regression and it helps in further
> + * reducing the chances for the above race.
> + */
> + if (unlikely(task_on_rq_queued(p) || current == p))
> + lsub_positive(&util_est, _task_util_est(p));
> +
> + util = max(util, util_est);
> + }
> +
> + return min(util, capacity_orig_of(cpu));
> +}
> +
> /*
> * cpu_util_without: compute cpu utilization without any contributions from *p
> * @cpu: the CPU which utilization is requested
> @@ -6526,19 +6584,10 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target)
> */
> static unsigned long cpu_util_without(int cpu, struct task_struct *p)
> {
> - struct cfs_rq *cfs_rq;
> - unsigned int util;
> -
> /* Task has no contribution or is new */
> if (cpu != task_cpu(p) || !READ_ONCE(p->se.avg.last_update_time))
> return cpu_util_cfs(cpu);
>
> - cfs_rq = &cpu_rq(cpu)->cfs;
> - util = READ_ONCE(cfs_rq->avg.util_avg);
> -
> - /* Discount task's util from CPU's util */
> - lsub_positive(&util, task_util(p));
> -
> /*
> * Covered cases:
> *
> @@ -6560,82 +6609,8 @@ static unsigned long cpu_util_without(int cpu, struct task_struct *p)
> * estimation of the spare capacity on that CPU, by just
> * considering the expected utilization of tasks already
> * runnable on that CPU.

The comment about the covered cases above should be moved in
cpu_util_next() which is where the cases are covered now

> - *
> - * Cases a) and b) are covered by the above code, while case c) is
> - * covered by the following code when estimated utilization is
> - * enabled.
> */
> - if (sched_feat(UTIL_EST)) {
> - unsigned int estimated =
> - READ_ONCE(cfs_rq->avg.util_est.enqueued);
> -
> - /*
> - * Despite the following checks we still have a small window
> - * for a possible race, when an execl's select_task_rq_fair()
> - * races with LB's detach_task():
> - *
> - * detach_task()
> - * p->on_rq = TASK_ON_RQ_MIGRATING;
> - * ---------------------------------- A
> - * deactivate_task() \
> - * dequeue_task() + RaceTime
> - * util_est_dequeue() /
> - * ---------------------------------- B
> - *
> - * The additional check on "current == p" it's required to
> - * properly fix the execl regression and it helps in further
> - * reducing the chances for the above race.
> - */
> - if (unlikely(task_on_rq_queued(p) || current == p))
> - lsub_positive(&estimated, _task_util_est(p));
> -
> - util = max(util, estimated);
> - }
> -
> - /*
> - * Utilization (estimated) can exceed the CPU capacity, thus let's
> - * clamp to the maximum CPU capacity to ensure consistency with
> - * cpu_util.
> - */
> - return min_t(unsigned long, util, capacity_orig_of(cpu));
> -}
> -
> -/*
> - * Predicts what cpu_util(@cpu) would return if @p was migrated (and enqueued)
> - * to @dst_cpu.
> - */
> -static unsigned long cpu_util_next(int cpu, struct task_struct *p, int dst_cpu)
> -{
> - struct cfs_rq *cfs_rq = &cpu_rq(cpu)->cfs;
> - unsigned long util_est, util = READ_ONCE(cfs_rq->avg.util_avg);
> -
> - /*
> - * If @p migrates from @cpu to another, remove its contribution. Or,
> - * if @p migrates from another CPU to @cpu, add its contribution. In
> - * the other cases, @cpu is not impacted by the migration, so the
> - * util_avg should already be correct.
> - */
> - if (task_cpu(p) == cpu && dst_cpu != cpu)
> - lsub_positive(&util, task_util(p));
> - else if (task_cpu(p) != cpu && dst_cpu == cpu)
> - util += task_util(p);
> -
> - if (sched_feat(UTIL_EST)) {
> - util_est = READ_ONCE(cfs_rq->avg.util_est.enqueued);
> -
> - /*
> - * During wake-up, the task isn't enqueued yet and doesn't
> - * appear in the cfs_rq->avg.util_est.enqueued of any rq,
> - * so just add it (if needed) to "simulate" what will be
> - * cpu_util after the task has been enqueued.
> - */
> - if (dst_cpu == cpu)
> - util_est += _task_util_est(p);
> -
> - util = max(util, util_est);
> - }
> -
> - return min(util, capacity_orig_of(cpu));
> + return cpu_util_next(cpu, p, -1);
> }
>
> /*
> --
> 2.25.1
>