Received-SPF: pass (google.com: domain of linux-kernel+bounces-30-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Sender: Tejun Heo <htejun@gmail.com>
Date: Thu, 14 Dec 2023 09:47:31 -1000
From: Tejun Heo <tj@kernel.org>
To: Juri Lelli <juri.lelli@redhat.com>
Cc: Aaron Tomlin <atomlin@atomlin.com>, linux-kernel@vger.kernel.org,
	jiangshanlai@gmail.com, peterz@infradead.org
Subject: Re: [RFC PATCH 0/2] workqueue: Introduce PF_WQ_RESCUE_WORKER
Message-ID: <ZXtb066P-ZnjxfgK@slm.duckdns.org>
References: <20230729135334.566138-1-atomlin@atomlin.com>
 <um77hym4t6zyypfbhwbaeqxpfdzc657oa7vgowdfah7cuctjak@pexots3mfb24>
 <ZXdXdBzvbkI4Y4fL@slm.duckdns.org>
 <ZXguMgcKLCLn16T4@localhost.localdomain>
 <ZXiVCOKk90Fjpmhw@slm.duckdns.org>
 <ZXlyfjDsFGbYcMU6@localhost.localdomain>
 <ZXnPVtISKQ2JFDNn@slm.duckdns.org>
 <ZXn4qiMetd7zY1sb@localhost.localdomain>
 <ZXn6J5bN-dPC1WSk@slm.duckdns.org>
 <ZXrmJYvekzrLSaGo@localhost.localdomain>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <ZXrmJYvekzrLSaGo@localhost.localdomain>

Hello,

On Thu, Dec 14, 2023 at 12:25:25PM +0100, Juri Lelli wrote:
> > So, we have to use set_cpus_allowed_ptr() but we still don't want to change
> > the affinity of a rescuer which is already running a task for a pool.
> 
> But then, even today, a rescuer might keep handling work on a cpu
> outside its wq cpumask if the associated wq cpumask change can proceed
> w/o waiting for it to finish the iteration?

Yeah, that can happen and pool cpumasks naturally being subsets of the wq's
cpumask that they're serving, your original approach likely isn't broken
either.

> BTW, apologies for all the questions, but I'd like to make sure I can
> get the implications hopefully right. :)

I obviously haven't thought through it very well, so thanks for the
questions. So, yeah, I think we actually need to set the rescuer's cpumask
when wq's cpumask changes and doing it where you were suggesting should
probably work.

Thanks.

-- 
tejun