linux-kernel - Re: [PATCH RFC v2 11/22] cpuset: introduce local_partition

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <3569d5d9-1a5d-4e01-a24b-4d5817181899@huaweicloud.com>
Date: Thu, 13 Nov 2025 11:27:13 +0800
From: Chen Ridong <chenridong@...weicloud.com>
To: Waiman Long <llong@...hat.com>, tj@...nel.org, hannes@...xchg.org,
 mkoutny@...e.com
Cc: cgroups@...r.kernel.org, linux-kernel@...r.kernel.org,
 lujialin4@...wei.com, chenridong@...wei.com
Subject: Re: [PATCH RFC v2 11/22] cpuset: introduce local_partition_disable()



On 2025/11/13 6:10, Waiman Long wrote:
> On 10/25/25 2:48 AM, Chen Ridong wrote:
>> From: Chen Ridong <chenridong@...wei.com>
>>
>> The partition_disable() function introduced earlier can be extended to
>> handle local partition disablement.
>>
>> The local_partition_disable() functions is introduced, which extracts the
>> local partition disable logic from update_parent_effective_cpumask(). It
>> calls partition_disable() to complete the disablement process.
>>
>> This refactoring establishes a clear separation between local and remote
>> partition operations while promoting code reuse through the shared
>> partition_disable() infrastructure.
>>
>> Signed-off-by: Chen Ridong <chenridong@...wei.com>
>> ---
>>   kernel/cgroup/cpuset.c | 55 ++++++++++++++++++++++++++++++------------
>>   1 file changed, 39 insertions(+), 16 deletions(-)
>>
>> diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
>> index b308d9f80eef..f36d17a4d8cd 100644
>> --- a/kernel/cgroup/cpuset.c
>> +++ b/kernel/cgroup/cpuset.c
>> @@ -1877,6 +1877,43 @@ static int local_partition_enable(struct cpuset *cs,
>>       return 0;
>>   }
>>   +/**
>> + * local_partition_disable - Disable a local partition
>> + * @cs: Target cpuset (local partition root) to disable
>> + * @part_error: partition error when @cs is disabled
>> + * @tmp: Temporary masks for CPU calculations
>> + */
>> +static void local_partition_disable(struct cpuset *cs, enum prs_errcode part_error,
>> +                    struct tmpmasks *tmp)
>> +{
>> +    struct cpuset *parent = parent_cs(cs);
>> +    bool cpumask_updated = false;
>> +    int new_prs;
>> +
>> +    lockdep_assert_held(&cpuset_mutex);
>> +    WARN_ON_ONCE(is_remote_partition(cs));    /* For local partition only */
>> +
>> +    if (!is_partition_valid(cs))
>> +        return;
>> +
>> +    new_prs = part_error ? -cs->partition_root_state : 0;
>> +    /*
>> +     * May need to add cpus back to parent's effective_cpus
>> +     * (and maybe removed from subpartitions_cpus/isolated_cpus)
>> +     * for valid partition root. xcpus may contain CPUs that
>> +     * shouldn't be removed from the two global cpumasks.
>> +     */
>> +    if (is_partition_valid(parent))
>> +        cpumask_updated = !cpumask_empty(cs->effective_xcpus);
> 
> If cs is a valid local partition, parent must be a valid partition. So the
> is_partition_valid(parent) check is meaningless. Also the effective_xcpus must not be empty. IOW,
> cpumask_updated must be true.
> 
> Cheers,
> Longman
> 

I believe is_partition_valid(parent) is meaningless and can be removed. We do support valid
partitions with empty effective CPUs—meaning cs->effective_xcpus may be empty.

>> +
>> +    partition_disable(cs, parent, new_prs, part_error);
>> +
>> +    if (cpumask_updated) {
>> +        cpuset_update_tasks_cpumask(parent, tmp->addmask);
>> +        update_sibling_cpumasks(parent, cs, tmp);
>> +    }
>> +}
>> +
>>   /**
>>    * update_parent_effective_cpumask - update effective_cpus mask of parent cpuset
>>    * @cs:      The cpuset that requests change in partition root state
>> @@ -1967,19 +2004,7 @@ static int update_parent_effective_cpumask(struct cpuset *cs, int cmd,
>>         nocpu = tasks_nocpu_error(parent, cs, xcpus);
>>   -    if (cmd == partcmd_disable) {
>> -        /*
>> -         * May need to add cpus back to parent's effective_cpus
>> -         * (and maybe removed from subpartitions_cpus/isolated_cpus)
>> -         * for valid partition root. xcpus may contain CPUs that
>> -         * shouldn't be removed from the two global cpumasks.
>> -         */
>> -        if (is_partition_valid(cs)) {
>> -            cpumask_copy(tmp->addmask, cs->effective_xcpus);
>> -            adding = true;
>> -        }
>> -        new_prs = PRS_MEMBER;
>> -    } else if (newmask) {
>> +    if (newmask) {
>>           /*
>>            * Empty cpumask is not allowed
>>            */
>> @@ -3110,9 +3135,7 @@ static int update_prstate(struct cpuset *cs, int new_prs)
>>           if (is_remote_partition(cs))
>>               remote_partition_disable(cs, &tmpmask);
>>           else
>> -            update_parent_effective_cpumask(cs, partcmd_disable,
>> -                            NULL, &tmpmask);
>> -
>> +            local_partition_disable(cs, PERR_NONE, &tmpmask);
>>           /*
>>            * Invalidation of child partitions will be done in
>>            * update_cpumasks_hier().
> 

-- 
Best regards,
Ridong