[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <fd3d0d3b-a58c-890e-8a41-841a1d0a149d@redhat.com>
Date: Wed, 25 Oct 2023 14:46:01 -0400
From: Waiman Long <longman@...hat.com>
To: Tejun Heo <tj@...nel.org>
Cc: Zefan Li <lizefan.x@...edance.com>,
Johannes Weiner <hannes@...xchg.org>,
Jonathan Corbet <corbet@....net>,
Lai Jiangshan <jiangshanlai@...il.com>,
Shuah Khan <shuah@...nel.org>, cgroups@...r.kernel.org,
linux-doc@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-kselftest@...r.kernel.org
Subject: Re: [PATCH-cgroup 3/4] cgroup/cpuset: Keep track of CPUs in isolated
partitions
On 10/23/23 23:25, Tejun Heo wrote:
> Hello, Waiman.
>
> On Wed, Oct 18, 2023 at 02:24:00PM -0400, Waiman Long wrote:
>> If you mean saving the exclusion cpumask no matter who the caller is, we can
>> add another exclusion cpumask to save it and expose it to sysfs. This should
>> be done in the first workqueue patch, not as part of this patch. I expose
>> this isolated cpumask for testing purpose to be checked by the
>> test_cpuset_prs.sh script for correctness. As said, I can expose it without
>> cgroup_debug if you think the information is useful in general.
> I don't really care where the cpumask is in the source tree. I just want all
> the workqueue cpumasks presented to the userspace in a single place. Also, I
> think it makes sense to publish it to userspace in an easily accessible
> manner as what the eventual configuration ends up being can be confusing and
> the effect it has on the system subtle.
I have added 2 more read-only cpumask sysfs files in v2 to expose the
information.
Cheers,
Longman
Powered by blists - more mailing lists