Re: [PATCH -next v4 3/6] nbd: don't clear 'NBD_CMD_INFLIGHT' flag if request is not completed

[Yu Kuai]

在 2022/05/28 10:12, Yu Kuai 写道:
> Otherwise io will hung because request will only be completed if the
> cmd has the flag 'NBD_CMD_INFLIGHT'.
>
> Fixes: 07175cb1baf4 ("nbd: make sure request completion won't concurrent")
> Signed-off-by: Yu Kuai <yukuai3@xxxxxxxxxx>
> ---
>   drivers/block/nbd.c | 13 +++++++++----
>   1 file changed, 9 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
> index 2ee1e376d5c4..a673a97b9b6b 100644
> --- a/drivers/block/nbd.c
> +++ b/drivers/block/nbd.c
> @@ -403,13 +403,14 @@ static enum blk_eh_timer_return nbd_xmit_timeout(struct request *req,
>   	if (!mutex_trylock(&cmd->lock))
>   		return BLK_EH_RESET_TIMER;
>   
> -	if (!__test_and_clear_bit(NBD_CMD_INFLIGHT, &cmd->flags)) {
> +	if (!test_bit(NBD_CMD_INFLIGHT, &cmd->flags)) {
>   		mutex_unlock(&cmd->lock);
>   		return BLK_EH_DONE;
>   	}
>   
>   	if (!refcount_inc_not_zero(&nbd->config_refs)) {
>   		cmd->status = BLK_STS_TIMEOUT;
> +		__clear_bit(NBD_CMD_INFLIGHT, &cmd->flags);
>   		mutex_unlock(&cmd->lock);
>   		goto done;
>   	}
> @@ -478,6 +479,7 @@ static enum blk_eh_timer_return nbd_xmit_timeout(struct request *req,
>   	dev_err_ratelimited(nbd_to_dev(nbd), "Connection timed out\n");
>   	set_bit(NBD_RT_TIMEDOUT, &config->runtime_flags);
>   	cmd->status = BLK_STS_IOERR;
> +	__clear_bit(NBD_CMD_INFLIGHT, &cmd->flags);
>   	mutex_unlock(&cmd->lock);
>   	sock_shutdown(nbd);
>   	nbd_config_put(nbd);
> @@ -745,7 +747,7 @@ static struct nbd_cmd *nbd_handle_reply(struct nbd_device *nbd, int index,
>   	cmd = blk_mq_rq_to_pdu(req);
>   
>   	mutex_lock(&cmd->lock);
> -	if (!__test_and_clear_bit(NBD_CMD_INFLIGHT, &cmd->flags)) {
> +	if (!test_bit(NBD_CMD_INFLIGHT, &cmd->flags)) {
>   		dev_err(disk_to_dev(nbd->disk), "Suspicious reply %d (status %u flags %lu)",
>   			tag, cmd->status, cmd->flags);
>   		ret = -ENOENT;
> @@ -813,6 +815,10 @@ static struct nbd_cmd *nbd_handle_reply(struct nbd_device *nbd, int index,
>   	}
>   out:
>   	trace_nbd_payload_received(req, handle);
> +	if (!ret && likely(!blk_should_fake_timeout(req->q)))
> +		__clear_bit(NBD_CMD_INFLIGHT, &cmd->flags);
> +	else if (!ret)
> +		ret = -ENOENT;

Hi, Josef

Do you prefer to move blk_should_fake_time() into nbd_handle_reply()
like this ? Code is a litter simplier.

Thanks,
Kuai
>   	mutex_unlock(&cmd->lock);
>   	return ret ? ERR_PTR(ret) : cmd;
>   }
> @@ -854,8 +860,7 @@ static void recv_work(struct work_struct *work)
>   		}
>   
>   		rq = blk_mq_rq_from_pdu(cmd);
> -		if (likely(!blk_should_fake_timeout(rq->q)))
> -			blk_mq_complete_request(rq);
> +		blk_mq_complete_request(rq);
>   		percpu_ref_put(&q->q_usage_counter);
>   	}
>