On 12/4/18 8:05 AM, Christoph Hellwig wrote:
> On Mon, Dec 03, 2018 at 05:00:59PM -0800, Sagi Grimberg wrote:
>>
>>> @@ -2428,7 +2426,8 @@ static void nvme_dev_disable(struct nvme_dev *dev,
>>> bool shutdown)
>>> nvme_stop_queues(&dev->ctrl);
>>> if (!dead && dev->ctrl.queue_count > 0) {
>>> - nvme_disable_io_queues(dev);
>>> + if (nvme_disable_io_queues(dev, nvme_admin_delete_sq))
>>> +
>>
>> Would be nice if the opcode change would be kept inside but still
>> split like:
>
> I actually like not having another wrapper to stop through..
>
> Keith, Jens, any preference?
Fine either way, prefer not having a wrapper.
--
Jens Axboe
