[RFC] nvme-mpath: delete disk after last connection

[Sagi Grimberg]

>> Something here is not clear to me, we are not really talking about "all
>> paths down" but rather "all paths lost", which should take the gendisk
>> down AFAIK.
>>
>> Keith, shouldn't we modify the gendisk reference manipulation in
>> nvme_mpath_remove_disk instead of moving it to the call sites?
>>
>> I'm still unsure why we need a timer for this. If a path is removed
>> (e.g. disconnected, we shouldn't keep the gendisk up).
> 
> True.  And I guess the right place to deal with timers is on how long
> we are going to keep retrying (or at least the underlying disks alive).

Which is orthogonal to this specific patch right?

As for timers, I wouldn't want to introduce a modparam for that. Perhaps
the correct way to do this is with what Victor suggested where mpath
fastfail is really the longest controller fastfail...