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;
Subject: Re: [PATCH] block: Limit number of items taken from the I/O scheduler
 in one go
To:     Salman Qazi <sqazi@google.com>, Jens Axboe <axboe@kernel.dk>,
        linux-block@vger.kernel.org, linux-kernel@vger.kernel.org
Cc:     Jesse Barnes <jsbarnes@google.com>,
        Gwendal Grignou <gwendal@google.com>,
        Christoph Hellwig <hch@lst.de>, Ming Lei <ming.lei@redhat.com>,
        Hannes Reinecke <hare@suse.com>
References: <CAKUOC8U03G27b6E7Z6mBo6RB=D7bKS_MQPwexEZiA7SOt_Lyvw@mail.gmail.com>
 <20200203205950.127629-1-sqazi@google.com>
From:   Bart Van Assche <bvanassche@acm.org>
Autocrypt: addr=bvanassche@acm.org; prefer-encrypt=mutual; keydata=
 mQENBFSOu4oBCADcRWxVUvkkvRmmwTwIjIJvZOu6wNm+dz5AF4z0FHW2KNZL3oheO3P8UZWr
 LQOrCfRcK8e/sIs2Y2D3Lg/SL7qqbMehGEYcJptu6mKkywBfoYbtBkVoJ/jQsi2H0vBiiCOy
 fmxMHIPcYxaJdXxrOG2UO4B60Y/BzE6OrPDT44w4cZA9DH5xialliWU447Bts8TJNa3lZKS1
 AvW1ZklbvJfAJJAwzDih35LxU2fcWbmhPa7EO2DCv/LM1B10GBB/oQB5kvlq4aA2PSIWkqz4
 3SI5kCPSsygD6wKnbRsvNn2mIACva6VHdm62A7xel5dJRfpQjXj2snd1F/YNoNc66UUTABEB
 AAG0JEJhcnQgVmFuIEFzc2NoZSA8YnZhbmFzc2NoZUBhY20ub3JnPokBOQQTAQIAIwUCVI67
 igIbAwcLCQgHAwIBBhUIAgkKCwQWAgMBAh4BAheAAAoJEHFcPTXFzhAJ8QkH/1AdXblKL65M
 Y1Zk1bYKnkAb4a98LxCPm/pJBilvci6boefwlBDZ2NZuuYWYgyrehMB5H+q+Kq4P0IBbTqTa
 jTPAANn62A6jwJ0FnCn6YaM9TZQjM1F7LoDX3v+oAkaoXuq0dQ4hnxQNu792bi6QyVdZUvKc
 macVFVgfK9n04mL7RzjO3f+X4midKt/s+G+IPr4DGlrq+WH27eDbpUR3aYRk8EgbgGKvQFdD
 CEBFJi+5ZKOArmJVBSk21RHDpqyz6Vit3rjep7c1SN8s7NhVi9cjkKmMDM7KYhXkWc10lKx2
 RTkFI30rkDm4U+JpdAd2+tP3tjGf9AyGGinpzE2XY1K5AQ0EVI67igEIAKiSyd0nECrgz+H5
 PcFDGYQpGDMTl8MOPCKw/F3diXPuj2eql4xSbAdbUCJzk2ETif5s3twT2ER8cUTEVOaCEUY3
 eOiaFgQ+nGLx4BXqqGewikPJCe+UBjFnH1m2/IFn4T9jPZkV8xlkKmDUqMK5EV9n3eQLkn5g
 lco+FepTtmbkSCCjd91EfThVbNYpVQ5ZjdBCXN66CKyJDMJ85HVr5rmXG/nqriTh6cv1l1Js
 T7AFvvPjUPknS6d+BETMhTkbGzoyS+sywEsQAgA+BMCxBH4LvUmHYhpS+W6CiZ3ZMxjO8Hgc
 ++w1mLeRUvda3i4/U8wDT3SWuHcB3DWlcppECLkAEQEAAYkBHwQYAQIACQUCVI67igIbDAAK
 CRBxXD01xc4QCZ4dB/0QrnEasxjM0PGeXK5hcZMT9Eo998alUfn5XU0RQDYdwp6/kMEXMdmT
 oH0F0xB3SQ8WVSXA9rrc4EBvZruWQ+5/zjVrhhfUAx12CzL4oQ9Ro2k45daYaonKTANYG22y
 //x8dLe2Fv1By4SKGhmzwH87uXxbTJAUxiWIi1np0z3/RDnoVyfmfbbL1DY7zf2hYXLLzsJR
 mSsED/1nlJ9Oq5fALdNEPgDyPUerqHxcmIub+pF0AzJoYHK5punqpqfGmqPbjxrJLPJfHVKy
 goMj5DlBMoYqEgpbwdUYkH6QdizJJCur4icy8GUNbisFYABeoJ91pnD4IGei3MTdvINSZI5e
Message-ID: <6499bf21-df25-3881-4252-8c51a9785bb7@acm.org>
Date:   Mon, 3 Feb 2020 19:47:02 -0800
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101
 Thunderbird/68.4.1
MIME-Version: 1.0
In-Reply-To: <20200203205950.127629-1-sqazi@google.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 2020-02-03 12:59, Salman Qazi wrote:
> We observed that it is possible for a flush to bypass the I/O
> scheduler and get added to hctx->dispatch in blk_mq_sched_bypass_insert.
> This can happen while a kworker is running blk_mq_do_dispatch_sched call
> in blk_mq_sched_dispatch_requests.
> 
> However, the blk_mq_do_dispatch_sched call doesn't end in bounded time.
> As a result, the flush can sit there indefinitely, as the I/O scheduler
> feeds an arbitrary number of requests to the hardware.
> 
> The solution is to periodically poll hctx->dispatch in
> blk_mq_do_dispatch_sched, to put a bound on the latency of the commands
> sitting there.

(added Christoph, Ming and Hannes to the Cc-list)

Thank you for having posted a patch; that really helps.

I see that my name occurs first in the "To:" list. Since Jens is the
block layer maintainer I think Jens should have been mentioned first.

In version v4.20 of the Linux kernel I found the following in the legacy
block layer code:
* From blk_insert_flush():
	list_add_tail(&rq->queuelist, &q->queue_head);
* From elv_next_request():
	list_for_each_entry(rq, &q->queue_head, queuelist)

I think this means that the legacy block layer sent flush requests to
the scheduler instead of directly to the block driver. How about
modifying the blk-mq code such that it mimics that approach? I'm asking
this because this patch, although the code looks clean, doesn't seem the
best solution to me.

> +		if (count > 0 && count % q->max_sched_batch == 0 &&
> +		    !list_empty_careful(&hctx->dispatch))
> +			break;

A modulo operation in the hot path? Please don't do that.

> +static ssize_t queue_max_sched_batch_store(struct request_queue *q,
> +					   const char *page,
> +					   size_t count)
> +{
> +	int err, val;
> +
> +	if (!q->mq_ops)
> +		return -EINVAL;
> +
> +	err = kstrtoint(page, 10, &val);
> +	if (err < 0)
> +		return err;
> +
> +	if (val <= 0)
> +		return -EINVAL;
> +
> +	q->max_sched_batch = val;
> +
> +	return count;
> +}

Has it been considered to use kstrtouint() instead of checking whether
the value returned by kstrtoint() is positive?

> +	int			max_sched_batch;

unsigned int?

Thanks,

Bart.