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 v3] blk-mq: fix io hung due to missing commit_rqs
To:     Yu Kuai <yukuai1@huaweicloud.com>, axboe@kernel.dk
Cc:     Ming Lei <ming.lei@redhat.com>, osandov@fb.com,
        linux-block@vger.kernel.org, linux-kernel@vger.kernel.org,
        yi.zhang@huawei.com, "yukuai (C)" <yukuai3@huawei.com>
References: <20220726122224.1790882-1-yukuai1@huaweicloud.com>
 <YuCsNuVuVppgn9FL@T590>
 <1ca6db8a-5780-45c6-064d-dd72c9f274c3@huaweicloud.com>
From:   Yu Kuai <yukuai1@huaweicloud.com>
Message-ID: <f2fe1889-e381-e29a-14c7-3e57c4488dca@huaweicloud.com>
Date:   Sat, 13 Aug 2022 14:00:36 +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: <1ca6db8a-5780-45c6-064d-dd72c9f274c3@huaweicloud.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk

在 2022/07/29 14:35, Yu Kuai 写道:
> Hi, Jens
> 
> Can you please consider this patch for v5.20?
> 
friendly ping ...
> Thanks,
> Kuai
> 
> 在 2022/07/27 11:08, Ming Lei 写道:
>> On Tue, Jul 26, 2022 at 08:22:24PM +0800, Yu Kuai wrote:
>>> From: Yu Kuai <yukuai3@huawei.com>
>>>
>>> Currently, in virtio_scsi, if 'bd->last' is not set to true while
>>> dispatching request, such io will stay in driver's queue, and driver
>>> will wait for block layer to dispatch more rqs. However, if block
>>> layer failed to dispatch more rq, it should trigger commit_rqs to
>>> inform driver.
>>>
>>> There is a problem in blk_mq_try_issue_list_directly() that commit_rqs
>>> won't be called:
>>>
>>> // assume that queue_depth is set to 1, list contains two rq
>>> blk_mq_try_issue_list_directly
>>>   blk_mq_request_issue_directly
>>>   // dispatch first rq
>>>   // last is false
>>>    __blk_mq_try_issue_directly
>>>     blk_mq_get_dispatch_budget
>>>     // succeed to get first budget
>>>     __blk_mq_issue_directly
>>>      scsi_queue_rq
>>>       cmd->flags |= SCMD_LAST
>>>        virtscsi_queuecommand
>>>         kick = (sc->flags & SCMD_LAST) != 0
>>>         // kick is false, first rq won't issue to disk
>>>   queued++
>>>
>>>   blk_mq_request_issue_directly
>>>   // dispatch second rq
>>>    __blk_mq_try_issue_directly
>>>     blk_mq_get_dispatch_budget
>>>     // failed to get second budget
>>>   ret == BLK_STS_RESOURCE
>>>    blk_mq_request_bypass_insert
>>>   // errors is still 0
>>>
>>>   if (!list_empty(list) || errors && ...)
>>>    // won't pass, commit_rqs won't be called
>>>
>>> In this situation, first rq relied on second rq to dispatch, while
>>> second rq relied on first rq to complete, thus they will both hung.
>>>
>>> Fix the problem by also treat 'BLK_STS_*RESOURCE' as 'errors' since
>>> it means that request is not queued successfully.
>>>
>>> Same problem exists in blk_mq_dispatch_rq_list(), 'BLK_STS_*RESOURCE'
>>> can't be treated as 'errors' here, fix the problem by calling
>>> commit_rqs if queue_rq return 'BLK_STS_*RESOURCE'.
>>>
>>> Fixes: d666ba98f849 ("blk-mq: add mq_ops->commit_rqs()")
>>> Signed-off-by: Yu Kuai <yukuai3@huawei.com>
>>
>> Reviewed-by: Ming Lei <ming.lei@redhat.com>
>>
>> Thanks,
>> Ming
>>
>> .
>>
> 
> .
>