Subject: Re: [PATCH v2 3/5] md: use md_thread api to wake up sync_thread
To:     Yu Kuai <yukuai1@huaweicloud.com>,
        kernel test robot <oliver.sang@intel.com>
Cc:     oe-lkp@lists.linux.dev, lkp@intel.com, linux-raid@vger.kernel.org,
        agk@redhat.com, snitzer@kernel.org, song@kernel.org,
        linux-kernel@vger.kernel.org, yi.zhang@huawei.com,
        yangerkun@huawei.com, logang@deltatee.com, guoqing.jiang@linux.dev,
        pmenzel@molgen.mpg.de, "yukuai (C)" <yukuai3@huawei.com>
References: <202303170904.640a8ece-oliver.sang@intel.com>
 <3e171048-060f-af7c-35ec-76fad4fb6f08@huaweicloud.com>
From:   Yu Kuai <yukuai1@huaweicloud.com>
Message-ID: <89b8377c-e953-f5c5-0ad9-d7b426f1a371@huaweicloud.com>
Date:   Sat, 18 Mar 2023 10:31:52 +0800
User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:60.0) Gecko/20100101
 Thunderbird/60.8.0
MIME-Version: 1.0
In-Reply-To: <3e171048-060f-af7c-35ec-76fad4fb6f08@huaweicloud.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk

Hi,

在 2023/03/17 11:30, Yu Kuai 写道:
> Hi,
> 
> 在 2023/03/17 10:30, kernel test robot 写道:
>>
>> Greeting,
>>
>> FYI, we noticed INFO:task_blocked_for_more_than#seconds due to commit 
>> (built with gcc-11):
>>
>> commit: af2203c7e88c00d3ce072f18c18a36e2936372fd ("[PATCH v2 3/5] md: 
>> use md_thread api to wake up sync_thread")
>> url: 
>> https://github.com/intel-lab-lkp/linux/commits/Yu-Kuai/md-pass-a-md_thread-pointer-to-md_register_thread/20230315-142018 
>>
>> base: git://git.kernel.org/cgit/linux/kernel/git/song/md.git md-next
>> patch link: 
>> https://lore.kernel.org/all/20230315061810.653263-4-yukuai1@huaweicloud.com/ 
>>
>> patch subject: [PATCH v2 3/5] md: use md_thread api to wake up 
>> sync_thread
>>
> 
> I don't expect there is any difference between:
> 
> if (mddev->sync_thread)
>      wake_up_process(mddev->sync_thread->tsk);
> 
> and:
> 
> md_wakeup_thread(mddev->sync_thread);

I understand that they are different now.

md_wakeup_thread() only wakeup wait_event() from md_thread(), it will
not wake up 'md_thread->tsk' if it's runing.

Hence this patch is wrong.

Thanks,
Kuai