Not saying that this should be a solution though.Our controller only support one io queue currently.
How many queues does your controller support that you happen to use
queue 0 ?
Ouch.
Remember, NVMe gets most of the performance improvements by using several queues, and be able to bind the queues to cpu sets.
Exposing just one queue will be invalidating any assumptions we do,
and trying to improve interrupt steering won't work anyway.
I sincerely doubt we should try to 'optimize' for this rather peculiar setup.