From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <jianchao.w.wang@oracle.com>
Subject: Re: [PATCH V5 0/9] nvme: pci: fix & improve timeout handling
To: Ming Lei <ming.lei@redhat.com>
Cc: Jens Axboe <axboe@kernel.dk>, linux-block@vger.kernel.org,
        Laurence Oberman <loberman@redhat.com>,
        Sagi Grimberg <sagi@grimberg.me>,
        James Smart <james.smart@broadcom.com>, linux-nvme@lists.infradead.org,
        Keith Busch <keith.busch@intel.com>, Christoph Hellwig <hch@lst.de>
References: <20180511122933.27155-1-ming.lei@redhat.com>
 <776f21e1-dc19-1b77-9ba4-44f0b8366625@oracle.com>
 <20180514093850.GA807@ming.t460p>
 <008cb38d-aa91-6ab7-64d9-417d6c53a1eb@oracle.com>
 <20180514122211.GB807@ming.t460p> <20180515003332.GB21743@ming.t460p>
From: "jianchao.wang" <jianchao.w.wang@oracle.com>
Message-ID: <d97b375f-9b47-6b95-bd5b-8166dae93676@oracle.com>
Date: Tue, 15 May 2018 17:56:14 +0800
MIME-Version: 1.0
In-Reply-To: <20180515003332.GB21743@ming.t460p>
Content-Type: text/plain; charset=utf-8
List-ID: <linux-block@vger.kernel.org>

Hi ming

On 05/15/2018 08:33 AM, Ming Lei wrote:
> We still have to quiesce admin queue before canceling request, so looks
> the following patch is better, so please ignore the above patch and try
> the following one and see if your hang can be addressed:
> 
> diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
> index f509d37b2fb8..c2adc76472a8 100644
> --- a/drivers/nvme/host/pci.c
> +++ b/drivers/nvme/host/pci.c
> @@ -1741,8 +1741,7 @@ static int nvme_alloc_admin_tags(struct nvme_dev *dev)
>  			dev->ctrl.admin_q = NULL;
>  			return -ENODEV;
>  		}
> -	} else
> -		blk_mq_unquiesce_queue(dev->ctrl.admin_q);
> +	}
>  
>  	return 0;
>  }
> @@ -2520,6 +2519,12 @@ static void nvme_dev_disable(struct nvme_dev *dev, bool shutdown, bool
>  	 */
>  	if (shutdown)
>  		nvme_start_queues(&dev->ctrl);
> +
> +	/*
> +	 * Avoid to suck reset because timeout may happen during reset and
> +	 * reset may hang forever if admin queue is kept as quiesced
> +	 */
> +	blk_mq_unquiesce_queue(dev->ctrl.admin_q);
>  	mutex_unlock(&dev->shutdown_lock);
>  }

w/ patch above and patch below, both the warning and io hung issue didn't reproduce till now.


@@ -1450,6 +1648,7 @@ static int nvme_create_queue(struct nvme_queue *nvmeq, int qid)
 {
        struct nvme_dev *dev = nvmeq->dev;
        int result;
+       int cq_vector;
 
        if (dev->cmb && use_cmb_sqes && (dev->cmbsz & NVME_CMBSZ_SQS)) {
                unsigned offset = (qid - 1) * roundup(SQ_SIZE(nvmeq->q_depth),
@@ -1462,15 +1661,16 @@ static int nvme_create_queue(struct nvme_queue *nvmeq, int qid)
         * A queue's vector matches the queue identifier unless the controller
         * has only one vector available.
         */
-       nvmeq->cq_vector = dev->num_vecs == 1 ? 0 : qid;
-       result = adapter_alloc_cq(dev, qid, nvmeq);
+       cq_vector = dev->num_vecs == 1 ? 0 : qid;
+       result = adapter_alloc_cq(dev, qid, nvmeq, cq_vector);
        if (result < 0)
-               goto release_vector;
+               goto out;
 
        result = adapter_alloc_sq(dev, qid, nvmeq);
        if (result < 0)
                goto release_cq;
-
+       
+       nvmeq->cq_vector = cq_vector;
        nvme_init_queue(nvmeq, qid);
        result = queue_request_irq(nvmeq);
        if (result < 0)
@@ -1479,12 +1679,12 @@ static int nvme_create_queue(struct nvme_queue *nvmeq, int qid)
        return result;
 
  release_sq:
+       nvmeq->cq_vector = -1;
        dev->online_queues--;
        adapter_delete_sq(dev, qid);
  release_cq:
        adapter_delete_cq(dev, qid);
- release_vector:
-       nvmeq->cq_vector = -1;
+ out:
        return result;
 }

Thanks
Jianchao