[PATCH] nvme: reject completions for requests that are not in flight

[Jens Axboe]

> I agree that a device in this state can't be used successfully, but the
> kernel should, IMHO. 
> Say, in case the device hits this due to firmware which needs an update,
> the kernel crash might prevent the user from applying it, in the first
> place.

OK, now we're really reaching... If the device is so buggy that it can't
even do basic operations without violating protocol and crashing the
kernel, then there's nothing the kernel can (or should) do to try and
prevent that. We don't cater to severely broken firmware or devices.

Please stop all of this nonsense.

-- 
Jens Axboe