Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Tue, 6 Aug 2019 16:19:22 +0200
From:   Peter Zijlstra <peterz@infradead.org>
To:     Cheng Jian <cj.chengjian@huawei.com>
Cc:     mingo@redhat.com, xiexiuqi@huawei.com, huawei.libin@huawei.com,
        bobo.shaobowang@huawei.com, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] sched/core: decrease rq->nr_uninterruptible before
 set_task_cpu
Message-ID: <20190806141922.GT2332@hirez.programming.kicks-ass.net>
References: <1565100770-110193-1-git-send-email-cj.chengjian@huawei.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1565100770-110193-1-git-send-email-cj.chengjian@huawei.com>
User-Agent: Mutt/1.10.1 (2018-07-13)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue, Aug 06, 2019 at 10:12:50PM +0800, Cheng Jian wrote:
> Migration may occur when wake up a process, so we must update
> the rq->nr_uninterruptible before set_task_cpu, otherwise we
> will decrease the nr_interuptible of the incorrect rq. Over
> time, it cause some rq accounting according to be too large,
> but others are negative.
> 
> Also change the type of rq->nr_uninterruptible to atomic_t.

NAK. Also there is no actual problem description there.