From: Waiman Long <llong@redhat.com>
To: Frederic Weisbecker <frederic@kernel.org>,
LKML <linux-kernel@vger.kernel.org>
Cc: "Michal Koutný" <mkoutny@suse.com>,
"Andrew Morton" <akpm@linux-foundation.org>,
"Bjorn Helgaas" <bhelgaas@google.com>,
"Catalin Marinas" <catalin.marinas@arm.com>,
"Chen Ridong" <chenridong@huawei.com>,
"Danilo Krummrich" <dakr@kernel.org>,
"David S . Miller" <davem@davemloft.net>,
"Eric Dumazet" <edumazet@google.com>,
"Gabriele Monaco" <gmonaco@redhat.com>,
"Greg Kroah-Hartman" <gregkh@linuxfoundation.org>,
"Ingo Molnar" <mingo@redhat.com>,
"Jakub Kicinski" <kuba@kernel.org>,
"Jens Axboe" <axboe@kernel.dk>,
"Johannes Weiner" <hannes@cmpxchg.org>,
"Lai Jiangshan" <jiangshanlai@gmail.com>,
"Marco Crivellari" <marco.crivellari@suse.com>,
"Michal Hocko" <mhocko@suse.com>,
"Muchun Song" <muchun.song@linux.dev>,
"Paolo Abeni" <pabeni@redhat.com>,
"Peter Zijlstra" <peterz@infradead.org>,
"Phil Auld" <pauld@redhat.com>,
"Rafael J . Wysocki" <rafael@kernel.org>,
"Roman Gushchin" <roman.gushchin@linux.dev>,
"Shakeel Butt" <shakeel.butt@linux.dev>,
"Simon Horman" <horms@kernel.org>, "Tejun Heo" <tj@kernel.org>,
"Thomas Gleixner" <tglx@linutronix.de>,
"Vlastimil Babka" <vbabka@suse.cz>,
"Will Deacon" <will@kernel.org>,
cgroups@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
linux-block@vger.kernel.org, linux-mm@kvack.org,
linux-pci@vger.kernel.org, netdev@vger.kernel.org
Subject: Re: [PATCH 29/33] sched/arm64: Move fallback task cpumask to HK_TYPE_DOMAIN
Date: Fri, 26 Dec 2025 18:46:02 -0500 [thread overview]
Message-ID: <19b001a4-0696-4efd-aaa3-902ffa42ff9e@redhat.com> (raw)
In-Reply-To: <20251224134520.33231-30-frederic@kernel.org>
On 12/24/25 8:45 AM, Frederic Weisbecker wrote:
> When none of the allowed CPUs of a task are online, it gets migrated
> to the fallback cpumask which is all the non nohz_full CPUs.
>
> However just like nohz_full CPUs, domain isolated CPUs don't want to be
> disturbed by tasks that have lost their CPU affinities.
>
> And since nohz_full rely on domain isolation to work correctly, the
> housekeeping mask of domain isolated CPUs should always be a superset of
> the housekeeping mask of nohz_full CPUs (there can be CPUs that are
> domain isolated but not nohz_full, OTOH there shouldn't be nohz_full
> CPUs that are not domain isolated):
>
> HK_TYPE_DOMAIN | HK_TYPE_KERNEL_NOISE == HK_TYPE_DOMAIN
>
> Therefore use HK_TYPE_DOMAIN as the appropriate fallback target for
> tasks and since this cpumask can be modified at runtime, make sure
> that 32 bits support CPUs on ARM64 mismatched systems are not isolated
> by cpusets.
>
> Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
> ---
> arch/arm64/kernel/cpufeature.c | 18 +++++++++++++++---
> include/linux/cpu.h | 4 ++++
> kernel/cgroup/cpuset.c | 17 ++++++++++++++---
> 3 files changed, 33 insertions(+), 6 deletions(-)
>
> diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c
> index c840a93b9ef9..70b0e45e299a 100644
> --- a/arch/arm64/kernel/cpufeature.c
> +++ b/arch/arm64/kernel/cpufeature.c
> @@ -1656,6 +1656,18 @@ has_cpuid_feature(const struct arm64_cpu_capabilities *entry, int scope)
> return feature_matches(val, entry);
> }
>
> +/*
> + * 32 bits support CPUs can't be isolated because tasks may be
> + * arbitrarily affine to them, defeating the purpose of isolation.
> + */
> +bool arch_isolated_cpus_can_update(struct cpumask *new_cpus)
> +{
> + if (static_branch_unlikely(&arm64_mismatched_32bit_el0))
> + return !cpumask_intersects(cpu_32bit_el0_mask, new_cpus);
> + else
> + return true;
> +}
> +
> const struct cpumask *system_32bit_el0_cpumask(void)
> {
> if (!system_supports_32bit_el0())
> @@ -1669,7 +1681,7 @@ const struct cpumask *system_32bit_el0_cpumask(void)
>
> const struct cpumask *task_cpu_fallback_mask(struct task_struct *p)
> {
> - return __task_cpu_possible_mask(p, housekeeping_cpumask(HK_TYPE_TICK));
> + return __task_cpu_possible_mask(p, housekeeping_cpumask(HK_TYPE_DOMAIN));
> }
>
> static int __init parse_32bit_el0_param(char *str)
> @@ -3987,8 +3999,8 @@ static int enable_mismatched_32bit_el0(unsigned int cpu)
> bool cpu_32bit = false;
>
> if (id_aa64pfr0_32bit_el0(info->reg_id_aa64pfr0)) {
> - if (!housekeeping_cpu(cpu, HK_TYPE_TICK))
> - pr_info("Treating adaptive-ticks CPU %u as 64-bit only\n", cpu);
> + if (!housekeeping_cpu(cpu, HK_TYPE_DOMAIN))
> + pr_info("Treating domain isolated CPU %u as 64-bit only\n", cpu);
> else
> cpu_32bit = true;
> }
> diff --git a/include/linux/cpu.h b/include/linux/cpu.h
> index 487b3bf2e1ea..0b48af25ab5c 100644
> --- a/include/linux/cpu.h
> +++ b/include/linux/cpu.h
> @@ -229,4 +229,8 @@ static inline bool cpu_attack_vector_mitigated(enum cpu_attack_vectors v)
> #define smt_mitigations SMT_MITIGATIONS_OFF
> #endif
>
> +struct cpumask;
> +
> +bool arch_isolated_cpus_can_update(struct cpumask *new_cpus);
> +
> #endif /* _LINUX_CPU_H_ */
> diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
> index cd6119c02beb..1cc83a3c25f6 100644
> --- a/kernel/cgroup/cpuset.c
> +++ b/kernel/cgroup/cpuset.c
> @@ -1408,14 +1408,22 @@ static void partition_xcpus_del(int old_prs, struct cpuset *parent,
> cpumask_or(parent->effective_cpus, parent->effective_cpus, xcpus);
> }
>
> +bool __weak arch_isolated_cpus_can_update(struct cpumask *new_cpus)
> +{
> + return true;
> +}
> +
> /*
> - * isolated_cpus_can_update - check for isolated & nohz_full conflicts
> + * isolated_cpus_can_update - check for conflicts against housekeeping and
> + * CPUs capabilities.
> * @add_cpus: cpu mask for cpus that are going to be isolated
> * @del_cpus: cpu mask for cpus that are no longer isolated, can be NULL
> * Return: false if there is conflict, true otherwise
> *
> - * If nohz_full is enabled and we have isolated CPUs, their combination must
> - * still leave housekeeping CPUs.
> + * Check for conflicts:
> + * - If nohz_full is enabled and there are isolated CPUs, their combination must
> + * still leave housekeeping CPUs.
> + * - Architecture has CPU capabilities incompatible with being isolated
> *
> * TBD: Should consider merging this function into
> * prstate_housekeeping_conflict().
> @@ -1426,6 +1434,9 @@ static bool isolated_cpus_can_update(struct cpumask *add_cpus,
> cpumask_var_t full_hk_cpus;
> int res = true;
>
> + if (!arch_isolated_cpus_can_update(add_cpus))
> + return false;
> +
> if (!housekeeping_enabled(HK_TYPE_KERNEL_NOISE))
> return true;
>
Reviewed-by: Waiman Long <longman@redhat.com>
next prev parent reply other threads:[~2025-12-26 23:46 UTC|newest]
Thread overview: 68+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-12-24 13:44 [PATCH 00/33 v5] cpuset/isolation: Honour kthreads preferred affinity Frederic Weisbecker
2025-12-24 13:44 ` [PATCH 01/33] PCI: Prepare to protect against concurrent isolated cpuset change Frederic Weisbecker
2025-12-29 3:23 ` Zhang Qiao
2025-12-29 3:53 ` Waiman Long
2025-12-31 13:18 ` Frederic Weisbecker
2025-12-30 22:38 ` Frederic Weisbecker
2025-12-24 13:44 ` [PATCH 02/33] cpu: Revert "cpu/hotplug: Prevent self deadlock on CPU hot-unplug" Frederic Weisbecker
2025-12-24 13:44 ` [PATCH 03/33] memcg: Prepare to protect against concurrent isolated cpuset change Frederic Weisbecker
2025-12-26 23:56 ` Tejun Heo
2025-12-24 13:44 ` [PATCH 04/33] mm: vmstat: " Frederic Weisbecker
2025-12-24 13:44 ` [PATCH 05/33] sched/isolation: Save boot defined domain flags Frederic Weisbecker
2025-12-25 22:27 ` Waiman Long
2025-12-31 13:45 ` Frederic Weisbecker
2025-12-24 13:44 ` [PATCH 06/33] cpuset: Convert boot_hk_cpus to use HK_TYPE_DOMAIN_BOOT Frederic Weisbecker
2025-12-25 22:31 ` Waiman Long
2025-12-24 13:44 ` [PATCH 07/33] driver core: cpu: Convert /sys/devices/system/cpu/isolated " Frederic Weisbecker
2025-12-24 13:44 ` [PATCH 08/33] net: Keep ignoring isolated cpuset change Frederic Weisbecker
2025-12-24 13:44 ` [PATCH 09/33] block: Protect against concurrent " Frederic Weisbecker
2025-12-30 0:37 ` Jens Axboe
2025-12-31 14:02 ` Frederic Weisbecker
2025-12-31 15:30 ` Jens Axboe
2025-12-24 13:44 ` [PATCH 10/33] timers/migration: Prevent from lockdep false positive warning Frederic Weisbecker
2025-12-24 13:44 ` [PATCH 11/33] cpu: Provide lockdep check for CPU hotplug lock write-held Frederic Weisbecker
2025-12-24 13:44 ` [PATCH 12/33] cpuset: Provide lockdep check for cpuset lock held Frederic Weisbecker
2025-12-24 13:45 ` [PATCH 13/33] sched/isolation: Convert housekeeping cpumasks to rcu pointers Frederic Weisbecker
2025-12-24 13:45 ` [PATCH 14/33] cpuset: Update HK_TYPE_DOMAIN cpumask from cpuset Frederic Weisbecker
2025-12-26 2:24 ` Waiman Long
2025-12-26 3:20 ` Waiman Long
2025-12-26 8:08 ` Chen Ridong
2025-12-31 14:21 ` Frederic Weisbecker
2025-12-24 13:45 ` [PATCH 15/33] sched/isolation: Flush memcg workqueues on cpuset isolated partition change Frederic Weisbecker
2025-12-24 13:45 ` [PATCH 16/33] sched/isolation: Flush vmstat " Frederic Weisbecker
2025-12-24 13:45 ` [PATCH 17/33] PCI: Flush PCI probe workqueue " Frederic Weisbecker
2025-12-26 8:48 ` Chen Ridong
2025-12-31 14:37 ` Frederic Weisbecker
2025-12-24 13:45 ` [PATCH 18/33] cpuset: Propagate cpuset isolation update to workqueue through housekeeping Frederic Weisbecker
2025-12-26 20:31 ` Waiman Long
2025-12-27 0:18 ` Tejun Heo
2025-12-24 13:45 ` [PATCH 19/33] cpuset: Propagate cpuset isolation update to timers " Frederic Weisbecker
2025-12-26 20:40 ` Waiman Long
2025-12-31 15:53 ` Frederic Weisbecker
2025-12-24 13:45 ` [PATCH 20/33] timers/migration: Remove superfluous cpuset isolation test Frederic Weisbecker
2025-12-26 20:45 ` Waiman Long
2025-12-24 13:45 ` [PATCH 21/33] cpuset: Remove cpuset_cpu_is_isolated() Frederic Weisbecker
2025-12-26 20:48 ` Waiman Long
2025-12-24 13:45 ` [PATCH 22/33] sched/isolation: Remove HK_TYPE_TICK test from cpu_is_isolated() Frederic Weisbecker
2025-12-26 21:26 ` Waiman Long
2025-12-24 13:45 ` [PATCH 23/33] PCI: Remove superfluous HK_TYPE_WQ check Frederic Weisbecker
2025-12-24 13:45 ` [PATCH 24/33] kthread: Refine naming of affinity related fields Frederic Weisbecker
2025-12-26 21:37 ` Waiman Long
2025-12-24 13:45 ` [PATCH 25/33] kthread: Include unbound kthreads in the managed affinity list Frederic Weisbecker
2025-12-26 22:11 ` Waiman Long
2025-12-24 13:45 ` [PATCH 26/33] kthread: Include kthreadd to " Frederic Weisbecker
2025-12-26 22:13 ` Waiman Long
2025-12-24 13:45 ` [PATCH 27/33] kthread: Rely on HK_TYPE_DOMAIN for preferred affinity management Frederic Weisbecker
2025-12-26 22:16 ` Waiman Long
2025-12-24 13:45 ` [PATCH 28/33] sched: Switch the fallback task allowed cpumask to HK_TYPE_DOMAIN Frederic Weisbecker
2025-12-26 23:08 ` Waiman Long
2025-12-24 13:45 ` [PATCH 29/33] sched/arm64: Move fallback task " Frederic Weisbecker
2025-12-26 23:46 ` Waiman Long [this message]
2025-12-24 13:45 ` [PATCH 30/33] kthread: Honour kthreads preferred affinity after cpuset changes Frederic Weisbecker
2025-12-26 23:59 ` Waiman Long
2025-12-24 13:45 ` [PATCH 31/33] kthread: Comment on the purpose and placement of kthread_affine_node() call Frederic Weisbecker
2025-12-24 13:45 ` [PATCH 32/33] kthread: Document kthread_affine_preferred() Frederic Weisbecker
2025-12-24 13:45 ` [PATCH 33/33] doc: Add housekeeping documentation Frederic Weisbecker
2025-12-27 0:39 ` Waiman Long
2025-12-31 15:25 ` Frederic Weisbecker
2025-12-31 17:35 ` Waiman Long
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=19b001a4-0696-4efd-aaa3-902ffa42ff9e@redhat.com \
--to=llong@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=axboe@kernel.dk \
--cc=bhelgaas@google.com \
--cc=catalin.marinas@arm.com \
--cc=cgroups@vger.kernel.org \
--cc=chenridong@huawei.com \
--cc=dakr@kernel.org \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=frederic@kernel.org \
--cc=gmonaco@redhat.com \
--cc=gregkh@linuxfoundation.org \
--cc=hannes@cmpxchg.org \
--cc=horms@kernel.org \
--cc=jiangshanlai@gmail.com \
--cc=kuba@kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-pci@vger.kernel.org \
--cc=marco.crivellari@suse.com \
--cc=mhocko@suse.com \
--cc=mingo@redhat.com \
--cc=mkoutny@suse.com \
--cc=muchun.song@linux.dev \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=pauld@redhat.com \
--cc=peterz@infradead.org \
--cc=rafael@kernel.org \
--cc=roman.gushchin@linux.dev \
--cc=shakeel.butt@linux.dev \
--cc=tglx@linutronix.de \
--cc=tj@kernel.org \
--cc=vbabka@suse.cz \
--cc=will@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox