[PATCH] nvme-host tcp: check if the queue is allocated before stopping it

Sagi Grimberg sagi at grimberg.me
Wed Aug 10 06:05:10 PDT 2022


> 
>> diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c
>> index ac74dce3e95b..ab9bb5c98c78 100644
>> --- a/drivers/nvme/host/tcp.c
>> +++ b/drivers/nvme/host/tcp.c
>> @@ -1619,6 +1619,9 @@ static void nvme_tcp_stop_queue(struct nvme_ctrl 
>> *nctrl, int qid)
>>       struct nvme_tcp_ctrl *ctrl = to_tcp_ctrl(nctrl);
>>       struct nvme_tcp_queue *queue = &ctrl->queues[qid];
>> +    if (!test_bit(NVME_TCP_Q_ALLOCATED, &queue->flags))
>> +        return;
>> +
>>       mutex_lock(&queue->queue_lock);
>>       if (test_and_clear_bit(NVME_TCP_Q_LIVE, &queue->flags))
>>           __nvme_tcp_stop_queue(queue);
> 
> This looks correct to me,
> 
> Reviewed-by: Sagi Grimberg <sagi at grimberg.me>

Just one nit,

The patch title should start with nvme-tcp: and not nvme-host tcp:
This can be easily addressed when applying the patch



More information about the Linux-nvme mailing list