[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87bkjv1oq5.ffs@tglx>
Date: Mon, 10 Apr 2023 22:12:02 +0200
From: Thomas Gleixner <tglx@...utronix.de>
To: Ye Bin <yebin@...weicloud.com>, dennis@...nel.org, tj@...nel.org,
cl@...ux.com, linux-mm@...ck.org, yury.norov@...il.com,
andriy.shevchenko@...ux.intel.com, linux@...musvillemoes.dk
Cc: linux-kernel@...r.kernel.org, dchinner@...hat.com,
yebin10@...wei.com, yebin@...weicloud.com
Subject: Re: [PATCH v2 1/2] cpu/hotplug: introduce 'num_dying_cpus' to get
dying CPUs count
On Thu, Apr 06 2023 at 09:56, Ye Bin wrote:
> From: Ye Bin <yebin10@...wei.com>
>
> Introduce '__num_dying_cpus' variable to cache the number of dying CPUs
> in the core and just return the cached variable.
Why?
That atomic counter is racy too if read and acted upon w/o having CPUs
read locked.
All it does is making the race window smaller vs. the cpumask_weight()
based implementation. It's still racy and incorrect.
So no, this is not going to happen.
Thanks,
tglx
Powered by blists - more mailing lists