Re: [PATCH] Revert "sched/core: Reduce cost of sched_move_task when config autogroup"
From: Vincent Guittot
Date: Fri Mar 14 2025 - 12:06:42 EST
On Fri, 14 Mar 2025 at 16:13, Dietmar Eggemann <dietmar.eggemann@xxxxxxx> wrote:
>
> This reverts commit eff6c8ce8d4d7faef75f66614dd20bb50595d261.
>
> Hazem reported a 30% drop in UnixBench spawn test with commit
> eff6c8ce8d4d ("sched/core: Reduce cost of sched_move_task when config
> autogroup") on a m6g.xlarge AWS EC2 instance with 4 vCPUs and 16 GiB RAM
> (aarch64) (single level MC sched domain) [1].
>
> There is an early bail from sched_move_task() if p->sched_task_group is
> equal to p's 'cpu cgroup' (sched_get_task_group()). E.g. both are
> pointing to taskgroup '/user.slice/user-1000.slice/session-1.scope'
> (Ubuntu '22.04.5 LTS').
>
> So in:
>
> do_exit()
>
> sched_autogroup_exit_task()
>
> sched_move_task()
>
> if sched_get_task_group(p) == p->sched_task_group
> return
>
> /* p is enqueued */
> dequeue_task() \
> sched_change_group() |
> task_change_group_fair() |
> detach_task_cfs_rq() | (1)
> set_task_rq() |
> attach_task_cfs_rq() |
> enqueue_task() /
>
> (1) isn't called for p anymore.
>
> Turns out that the regression is related to sgs->group_util in
> group_is_overloaded() and group_has_capacity(). If (1) isn't called for
> all the 'spawn' tasks then sgs->group_util is ~900 and
> sgs->group_capacity = 1024 (single CPU sched domain) and this leads to
> group_is_overloaded() returning true (2) and group_has_capacity() false
> (3) much more often compared to the case when (1) is called.
>
> I.e. there are much more cases of 'group_is_overloaded' and
> 'group_fully_busy' in WF_FORK wakeup sched_balance_find_dst_cpu() which
> then returns much more often a CPU != smp_processor_id() (5).
>
> This isn't good for these extremely short running tasks (FORK + EXIT)
> and also involves calling sched_balance_find_dst_group_cpu() unnecessary
> (single CPU sched domain).
>
> Instead if (1) is called for 'p->flags & PF_EXITING' then the path
> (4),(6) is taken much more often.
>
> select_task_rq_fair(..., wake_flags = WF_FORK)
>
> cpu = smp_processor_id()
>
> new_cpu = sched_balance_find_dst_cpu(..., cpu, ...)
>
> group = sched_balance_find_dst_group(..., cpu)
>
> do {
>
> update_sg_wakeup_stats()
>
> sgs->group_type = group_classify()
>
> if group_is_overloaded() (2)
> return group_overloaded
>
> if !group_has_capacity() (3)
> return group_fully_busy
>
> return group_has_spare (4)
>
> } while group
>
> if local_sgs.group_type > idlest_sgs.group_type
> return idlest (5)
>
> case group_has_spare:
>
> if local_sgs.idle_cpus >= idlest_sgs.idle_cpus
> return NULL (6)
>
> Unixbench Tests './Run -c 4 spawn' on:
>
> (a) VM AWS instance (m7gd.16xlarge) with v6.13 ('maxcpus=4 nr_cpus=4')
> and Ubuntu 22.04.5 LTS (aarch64).
>
> Shell & test run in '/user.slice/user-1000.slice/session-1.scope'.
>
> w/o patch w/ patch
> 21005 27120
>
> (b) i7-13700K with tip/sched/core ('nosmt maxcpus=8 nr_cpus=8') and
> Ubuntu 22.04.5 LTS (x86_64).
>
> Shell & test run in '/A'.
>
> w/o patch w/ patch
> 67675 88806
>
> CONFIG_SCHED_AUTOGROUP=y & /sys/proc/kernel/sched_autogroup_enabled equal
> 0 or 1.
>
> [1] https://lkml.kernel.org/r/20250205151026.13061-1-hagarhem@xxxxxxxxxx
>
> Reported-by: Hazem Mohamed Abuelfotoh <abuehaze@xxxxxxxxxx>
> Tested-by: Hagar Hemdan <hagarhem@xxxxxxxxxx>
> Signed-off-by: Dietmar Eggemann <dietmar.eggemann@xxxxxxx>
Reviewed-by: Vincent Guittot <vincent.guittot@xxxxxxxxxx>
> ---
>
> Previous discussion under:
> https://lkml.kernel.org/r/20250306162635.2614376-1-dietmar.eggemann@xxxxxxx
>
> kernel/sched/core.c | 21 +++------------------
> 1 file changed, 3 insertions(+), 18 deletions(-)
>
> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> index 621cfc731c5b..532119a3f72b 100644
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -9015,7 +9015,7 @@ void sched_release_group(struct task_group *tg)
> spin_unlock_irqrestore(&task_group_lock, flags);
> }
>
> -static struct task_group *sched_get_task_group(struct task_struct *tsk)
> +static void sched_change_group(struct task_struct *tsk)
> {
> struct task_group *tg;
>
> @@ -9027,13 +9027,7 @@ static struct task_group *sched_get_task_group(struct task_struct *tsk)
> tg = container_of(task_css_check(tsk, cpu_cgrp_id, true),
> struct task_group, css);
> tg = autogroup_task_group(tsk, tg);
> -
> - return tg;
> -}
> -
> -static void sched_change_group(struct task_struct *tsk, struct task_group *group)
> -{
> - tsk->sched_task_group = group;
> + tsk->sched_task_group = tg;
>
> #ifdef CONFIG_FAIR_GROUP_SCHED
> if (tsk->sched_class->task_change_group)
> @@ -9054,20 +9048,11 @@ void sched_move_task(struct task_struct *tsk, bool for_autogroup)
> {
> int queued, running, queue_flags =
> DEQUEUE_SAVE | DEQUEUE_MOVE | DEQUEUE_NOCLOCK;
> - struct task_group *group;
> struct rq *rq;
>
> CLASS(task_rq_lock, rq_guard)(tsk);
> rq = rq_guard.rq;
>
> - /*
> - * Esp. with SCHED_AUTOGROUP enabled it is possible to get superfluous
> - * group changes.
> - */
> - group = sched_get_task_group(tsk);
> - if (group == tsk->sched_task_group)
> - return;
> -
> update_rq_clock(rq);
>
> running = task_current_donor(rq, tsk);
> @@ -9078,7 +9063,7 @@ void sched_move_task(struct task_struct *tsk, bool for_autogroup)
> if (running)
> put_prev_task(rq, tsk);
>
> - sched_change_group(tsk, group);
> + sched_change_group(tsk);
> if (!for_autogroup)
> scx_cgroup_move_task(tsk);
>
> --
> 2.34.1
>