[PATCH 1/3] nvme: also check for RESETTING state in nvmf_check_if_ready()

[hch@lst.de (Christoph Hellwig)]

On Thu, Jun 07, 2018@11:41:11AM +0300, Sagi Grimberg wrote:
> Thank you Hannes - just spotted that...
>
> On 06/07/2018 11:38 AM, Hannes Reinecke wrote:
>> When resetting the controller some transports go into 'RESETTING' state
>> before issuing the actual 'reset' command. So add the 'RESETTING' state
>> to nvmf_check_if_ready() to give them a chance to submit it.
>>
>> Signed-off-by: Hannes Reinecke <hare at suse.com>
>> ---
>>   drivers/nvme/host/fabrics.c | 1 +
>>   1 file changed, 1 insertion(+)
>>
>> diff --git a/drivers/nvme/host/fabrics.c b/drivers/nvme/host/fabrics.c
>> index fa32c1216409..909dd337221a 100644
>> --- a/drivers/nvme/host/fabrics.c
>> +++ b/drivers/nvme/host/fabrics.c
>> @@ -548,6 +548,7 @@ blk_status_t nvmf_check_if_ready(struct nvme_ctrl *ctrl, struct request *rq,
>>   	case NVME_CTRL_NEW:
>>   	case NVME_CTRL_CONNECTING:
>>   	case NVME_CTRL_DELETING:
>> +	case NVME_CTRL_RESETTING:
>
> I guess ADMIN_ONLY should also allow commands to pass.

Only admin commands, though.

>
>>   		/*
>>   		 * This is the case of starting a new or deleting an association
>>   		 * but connectivity was lost before it was fully created or torn
>
> Looks like we have almost all the states? Shouldn't we reverse such that
> we check only for DEAD (and WARN_ON for state NEW)?

Sounds like an idea.  I'd like to see it in patch form first before
agreeing..