Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Subject: Re: [PATCH v2 1/2] blk-iocost: add refcounting for iocg
To:     Tejun Heo <tj@kernel.org>, Yu Kuai <yukuai1@huaweicloud.com>
Cc:     hch@infradead.org, josef@toxicpanda.com, axboe@kernel.dk,
        cgroups@vger.kernel.org, linux-block@vger.kernel.org,
        linux-kernel@vger.kernel.org, yi.zhang@huawei.com,
        "yukuai (C)" <yukuai3@huawei.com>
References: <Y7iCId3pnEnLqY8G@slm.duckdns.org>
 <875eb43e-202d-5b81-0bff-ef0434358d99@huaweicloud.com>
 <Y7xbpidpq7+DqJan@slm.duckdns.org>
 <a71f997f-6cae-d57b-85dd-2fd499d238f6@huaweicloud.com>
 <Y72wF/b0/xNRmP7f@slm.duckdns.org>
 <53b30ac8-d608-ba0b-8b1b-7fe5cfed6d61@huaweicloud.com>
 <Y77s0f741mFfGlTO@slm.duckdns.org>
 <4aeef320-c6c8-d9b4-8826-d58f00ea6264@huaweicloud.com>
 <Y8CrloCDGhbU42OH@slm.duckdns.org>
 <efa1c73b-e94f-373f-e535-2cfc32ce2433@huaweicloud.com>
 <Y8CwwghZ0adMsHFC@slm.duckdns.org>
From:   Yu Kuai <yukuai1@huaweicloud.com>
Message-ID: <ac95dfb8-b1b6-8916-bde8-1edb573e7ca5@huaweicloud.com>
Date:   Fri, 13 Jan 2023 09:25:11 +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: <Y8CwwghZ0adMsHFC@slm.duckdns.org>
Content-Type: text/plain; charset=gbk; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk

Hi,

?? 2023/01/13 9:15, Tejun Heo д??:
> Hello,
> 
> On Fri, Jan 13, 2023 at 09:10:25AM +0800, Yu Kuai wrote:
>>> only activate the policy when the controller is actually enabled. So, idk.
>>> What's wrong with synchronizing the two removal paths? blkcg policies are
>>> combinations of cgroups and block device configurations, so having exit
>>> paths from both sides is kinda natural.
>>
>> I still can't figure out how to synchronizing them will a mutex. Maybe
>> I'm being foolish...
> 
> Hmm... can't you just use e.g. per-bdev mutex which is grabbed by both
> blkg_free_workfn() and blkcg_deactivate_policy()?
> 

I think hold the lock in blkg_free_workfn() is too late, pd_free_fn()
for parent from blkcg_deactivate_policy() can be called first.

t1: remove cgroup t1/t2
blkcg_destroy_blkgs
  blkg_destroy
   percpu_ref_kill(&blkg->refcnt)
    blkg_release
     blkg_free
      schedule_work(&blkg->free_work)
      // t1 is done

t2: handle t1 from removing device
blkcg_deactivate_policy
  pd_free_fn
  // free parent
				t3: from t1
				blkg_free_workfn
				 pd_free_fn
				 // free child

Thanks,
Kuai