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;
Date:   Thu, 17 Feb 2022 07:48:33 -0800
From:   Christoph Hellwig <hch@infradead.org>
To:     Jens Axboe <axboe@kernel.dk>
Cc:     Christoph Hellwig <hch@infradead.org>,
        "Wang Jianchao (Kuaishou)" <jianchao.wan9@gmail.com>,
        Bart Van Assche <bvanassche@acm.org>,
        linux-block@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] blk: do rq_qos_exit in blk_cleanup_queue
Message-ID: <Yg5uUeMONPn4jnng@infradead.org>
References: <20220216113212.83000-1-jianchao.wan9@gmail.com>
 <Yg392sqFydj9p3My@infradead.org>
 <96deb063-f4ef-fadd-543b-e905495898f6@kernel.dk>
 <Yg5VnRtsrrQX4mFG@infradead.org>
 <eb8e2477-d8f6-b032-6177-8ce187de12ab@kernel.dk>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <eb8e2477-d8f6-b032-6177-8ce187de12ab@kernel.dk>
Precedence: bulk

On Thu, Feb 17, 2022 at 07:55:16AM -0700, Jens Axboe wrote:
> > from Jan 22.  Although it would need a rebase so it can be applied
> > without the preceding patches.
> 
> Can someone respin that for 5.17 then?

I looked at it and it I don't think we can do that without a lot of
the prep patches.

That being said I think this version of the patch also is buggy, we
want the policies shut down in del_gendisk with the queue frozen for
normal operation.

I guess until we can move the initialization and teardown entirely
to the gendisk as in Ming's more complex series we need to keep the
call in del_gendisk and also do it in blk_cleanup_queue.  For the
normal shutdown on disk that were life del_gendisk does the all the
work on the frozen queue, while for queues that never had a disk
blk_cleanup_queue will clean up the unused rq_qos.