Message-ID: <bac7b124-9087-1712-c88a-fac54f3d2cc0@linux.alibaba.com>
Date:   Fri, 10 Dec 2021 07:31:29 +0800
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:91.0)
 Gecko/20100101 Thunderbird/91.2.1
Subject: Re: [PATCH 6/7] workqueue: Remove the cacheline_aligned for
 nr_running
Content-Language: en-US
To:     Tejun Heo <tj@kernel.org>, Lai Jiangshan <jiangshanlai@gmail.com>
Cc:     linux-kernel@vger.kernel.org
References: <20211207073543.61092-1-jiangshanlai@gmail.com>
 <20211207073543.61092-7-jiangshanlai@gmail.com>
 <YbJ+E+HB+UCYHxSi@slm.duckdns.org>
From:   Lai Jiangshan <laijs@linux.alibaba.com>
In-Reply-To: <YbJ+E+HB+UCYHxSi@slm.duckdns.org>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk


On 2021/12/10 06:07, Tejun Heo wrote:
> On Tue, Dec 07, 2021 at 03:35:42PM +0800, Lai Jiangshan wrote:
>> From: Lai Jiangshan <laijs@linux.alibaba.com>
>>
>> nr_running is never modified remotely after the schedule callback in
>> wakeup path is removed.
>>
>> Rather nr_running is often accessed with other fields in the pool
>> together, so the cacheline_aligned for nr_running isn't needed.
> 
> Does it even need to be atomic anymore?
> 

It doesn't need to be atomic, it is only modified in its associated CPU
in process context.