________________________________________ 发件人: Thomas Gleixner <t...@linutronix.de> 发送时间: 2020年10月28日 17:23 收件人: Zhang, Qiang; pmla...@suse.com; t...@kernel.org 抄送: a...@linux-foundation.org; linux...@kvack.org; linux-kernel@vger.kernel.org 主题: Re: 回复: [PATCH v2] kthread_worker: re-set CPU affinities if CPU come online
> [Please note this e-mail is from an EXTERNAL e-mail address] > > On Wed, Oct 28 2020 at 15:30, qiang zhang wrote: > >>How is that addressing any of the comments I made on V1 of this? > > Do you mean the following problem: > > "The dynamic hotplug states run late. What's preventing work to be queued > on such a worker before it is bound to the CPU again?" > >This is one problem, but there are more and I explained them in great >length. If there is anything unclear, then please ask. Really, this patch is not considered that work may be put into the queue after the bound CPU is offline. in addition, when the bound CPU goes online again, before restoring the worker's CPU affinity, work may be put into the queue. Although int this (powerclamp) way，that's not a problem, that it is solved by destroying and creating tasks when the CPU hotplug, in addition, when CPU going down , this need call 'cancel_work_sync' func in offline callback, this may be blocked long time. these operation is expensive. this patch only just to recover the worker task's affinity when CPU go to online again that create by "kthread_create_worker_on_cpu" func , likely per-CPU worker method when CPU hotplug in "workqueue" and "io-wq". Thanks Qiang > >Thanks, > > tglx
