Re: [PATCH v7 26/40] sched/deadline: Make a rebuild count a task in the root domain it runs in

From: sashiko-bot

Date: Tue Nov 17 2026 - 10:31:23 EST


Thank you for your contribution! Sashiko AI review found 2 potential
issues to consider:
- [Medium] sched/deadline: dl_task_rebuild_cpu() uses recorded CPU
outside task cpuset
- [Medium] sched/deadline: dl_task_rebuild_cpu() fallback reads stale
HK_TYPE_DOMAIN
--

commit 5f53c9545c6d42e90c32f229213e080ced9acb3f
Author: Mathieu Desnoyers <mathieu.desnoyers@xxxxxxxxxxxx>

sched/deadline: Make a rebuild count a task in the root domain it runs in

This commit introduces dl_task_rebuild_cpu() to select an active CPU in
the root domain where a deadline task runs during root domain bandwidth
rebuilds, and leaves out tasks with no active CPU until offline migration
or a later rebuild places them.

> diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
> index a9b8bd019d4ba..9ff79141c6306 100644
> --- a/kernel/sched/deadline.c
> +++ b/kernel/sched/deadline.c

[ ... ]

> @@ -3820,6 +3826,60 @@ static void dl_get_task_effective_cpus(struct task_struct *p, struct cpumask *cp
> cpuset_cpus_allowed_locked(p, cpus);
> }
>
> +/*
> + * Return a CPU of the root domain where a rebuild of the accounting should
> + * count the bandwidth of @p, or nr_cpu_ids if there is none.
> + *
> + * Prefer the CPU @p is on, if it is active and in the mask of @p. The
> + * task runs, or wakes up, in the root domain of that CPU. A mask can hold
> + * CPUs of two root domains: housekeeping CPUs and CPUs isolated by
> + * isolcpus="domain". When every CPU of a mask goes offline,
> + * select_fallback_rq() can widen it to both. It moves no bandwidth.
> + * Prefer that CPU also if @p is queued there, out of its mask, and the
> + * CPU is the one recorded. Without admission control,
> + * dl_task_offline_migration() can put a throttled task and its bandwidth
> + * on any active CPU. The task runs there until it blocks.
> + *
> + * Else prefer the CPU recorded with the bandwidth, if it is active and in
> + * the mask. It is in the root domain that admitted the bandwidth, or that
> + * its last move put it in. A task moved asleep is still on a CPU of the
> + * root domain it left, out of its mask. Else take the first active CPU of
> + * the mask, a housekeeping one if there is one. A task on isolated CPUs
> + * that all went offline can have no active CPU at all.
> + */
> +static unsigned int dl_task_rebuild_cpu(struct task_struct *p)
> +{
> + struct cpumask *msk = this_cpu_cpumask_var_ptr(local_cpu_mask_dl);
> + int recorded = p->dl.dl_bw_cpu;
> + unsigned int cpu;
> +
> + /*
> + * Look at the mask of the task first. A change of the cpusets sets
> + * the mask of each task before its rebuild. The housekeeping mask
> + * that dl_get_task_effective_cpus() reads is updated after it. If no
> + * CPU of the mask is active, take the CPU recorded if it is active.
> + * dl_task_offline_migration() moved the task there, or an earlier
> + * rebuild counted it there. Else ask the cpuset.
> + */
> + cpumask_and(msk, &p->cpus_mask, cpu_active_mask);
> + cpu = task_cpu(p);
> + if (cpumask_test_cpu(cpu, msk) ||
> + (cpu == recorded && cpu_active(cpu) && task_on_rq_queued(p)))
> + return cpu;
> + if (cpumask_empty(msk)) {
> + if (recorded >= 0 && cpu_active(recorded))
> + return recorded;

[Severity: Medium]
Can returning recorded here without checking the task's cpuset charge the
bandwidth to a newly created partition that no longer contains the task?

Suppose a sleeping task in top_cpuset has user_cpus_ptr restricted via
sched_setaffinity() to CPUs 0 and 1, and both go offline while CPUs 2 and
3 remain online. The hotplug rebuild sees an empty msk, falls back to
dl_get_task_effective_cpus(), and records CPU 2 in p->dl.dl_bw_cpu.

If CPU 2 is then moved into a new cpuset partition root,
cpuset_update_tasks_cpumask() updates top_cpuset tasks with
possible_mask & ~subpartitions_cpus. Because __set_cpus_allowed_ptr()
intersects that mask with p->user_cpus_ptr ({0, 1}), p->cpus_mask stays
{0, 1} and msk remains empty. This check then sees CPU 2 is still active
and returns it, counting the top_cpuset task's bandwidth in the new
partition on CPU 2 instead of top_cpuset on CPU 3.

> + dl_get_task_effective_cpus(p, msk);
> + cpumask_and(msk, msk, cpu_active_mask);
> + } else if (recorded >= 0 && cpumask_test_cpu(recorded, msk)) {
> + return recorded;
> + }
> + cpu = cpumask_first_and(msk, housekeeping_cpumask(HK_TYPE_DOMAIN_BOOT));
> + if (cpu < nr_cpu_ids)
> + return cpu;
> + return cpumask_first(msk);
> +}

[Severity: Medium]
Does calling dl_get_task_effective_cpus() here read a stale
HK_TYPE_DOMAIN mask during cpuset partition transitions?

dl_get_task_effective_cpus() checks housekeeping_cpumask(HK_TYPE_DOMAIN):

kernel/sched/deadline.c:dl_get_task_effective_cpus() {
hk_msk = housekeeping_cpumask(HK_TYPE_DOMAIN);
if (housekeeping_enabled(HK_TYPE_DOMAIN)) {
if (!cpumask_intersects(p->cpus_ptr, hk_msk)) {
cpumask_andnot(cpus, cpu_active_mask, hk_msk);
return;
}
}
}

HK_TYPE_DOMAIN excludes both boot-time isolcpus=domain CPUs and runtime
isolated cpuset partitions, and cpuset_update_sd_hk_unlock() rebuilds
root domain accounting before housekeeping_update() refreshes
HK_TYPE_DOMAIN. If a task is confined to an offline isolcpus=domain CPU
(so msk is empty and recorded is inactive) when a runtime isolated
partition on another CPU transitions to root or member,
cpu_active_mask & ~hk_msk still includes that partition's CPU.
dl_add_task_root_domain() will then charge the task's bandwidth to that
root domain and record the CPU in p->dl.dl_bw_cpu.

Should dl_get_task_effective_cpus() use HK_TYPE_DOMAIN_BOOT to match
dl_task_rebuild_cpu()?

--
Sashiko AI review · https://sashiko.dev/#/patchset/cover.1794922146.git.mathieu.desnoyers@xxxxxxxxxxxx?part=26