Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Sender: Tejun Heo <htejun@gmail.com>
Date:   Mon, 11 Oct 2021 07:16:59 -1000
From:   Tejun Heo <tj@kernel.org>
To:     Yu Kuai <yukuai3@huawei.com>
Cc:     axboe@kernel.dk, cgroups@vger.kernel.org,
        linux-block@vger.kernel.org, linux-kernel@vger.kernel.org,
        yi.zhang@huawei.com
Subject: Re: [PATCH] blk-cgroup: check blkcg policy is enabled in
 blkg_create()
Message-ID: <YWRxi2OaIHhG9rOc@slm.duckdns.org>
References: <20211008072720.797814-1-yukuai3@huawei.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20211008072720.797814-1-yukuai3@huawei.com>
Precedence: bulk

On Fri, Oct 08, 2021 at 03:27:20PM +0800, Yu Kuai wrote:
> diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
> index eb48090eefce..00e1d97621ea 100644
> --- a/block/blk-cgroup.c
> +++ b/block/blk-cgroup.c
> @@ -226,6 +226,20 @@ struct blkcg_gq *blkg_lookup_slowpath(struct blkcg *blkcg,
>  }
>  EXPORT_SYMBOL_GPL(blkg_lookup_slowpath);
>  
> +static void blkg_check_pd(struct request_queue *q, struct blkcg_gq *blkg)
> +{
> +	int i;
> +
> +	for (i = 0; i < BLKCG_MAX_POLS; i++) {
> +		struct blkcg_policy *pol = blkcg_policy[i];
> +
> +		if (blkg->pd[i] && !blkcg_policy_enabled(q, pol)) {
> +			pol->pd_free_fn(blkg->pd[i]);
> +			blkg->pd[i] = NULL;
> +		}
> +	}
> +}
> +
>  /*
>   * If @new_blkg is %NULL, this function tries to allocate a new one as
>   * necessary using %GFP_NOWAIT.  @new_blkg is always consumed on return.
> @@ -252,6 +266,9 @@ static struct blkcg_gq *blkg_create(struct blkcg *blkcg,
>  		goto err_free_blkg;
>  	}
>  
> +	if (new_blkg)
> +		blkg_check_pd(q, new_blkg);
> +

Can't this happen the other way around too? ie. Linking a pd which doesn't
have an entry for a policy which got enabled inbetween? And what if an
existing policy was de-registered and another policy got the policy id
inbetween? I think the correct solution here would be synchronizing alloc -
create blocks against policy deactivation rather than trying to patch an
allocated blkg later. Deactivation being a really slow path, there are
plenty of options. The main challenge would making it difficult to make
mistakes with, I guess.

Thanks.

-- 
tejun