nvme_stop_queues quiesce queues for all name spaces, now quiesce one by
one, if there is lots of name spaces, sync wait long time(more than 10s).
Multipath can not fail over to retry quickly, cause io pause long time.
This is not expected.
To reduce quiesce time, we introduce async mechanism for sync SRCUs
and quiesce queue.
Frankly speaking, I prefer to replace SRCU with percpu_refcount:
- percpu_refcount has much less memory footprint than SRCU, so we can simply
move percpu_refcount into request_queue, instead of adding more bytes
into each hctx by this patch
- percpu_ref_get()/percpu_ref_put() isn't slower than srcu_read_lock()/srcu_read_unlock().
- with percpu_refcount, we can remove 'srcu_idx' from hctx_lock/hctx_unlock()
Ming, I don't have an issue with your preference, but I don't want to
tie it to what Chao is attempting to fix.
So we should either move forward with an srcu approach, or you please
provide evidence to your claims if you are replacing the fast-path
synchronization mechanism, as this warrants a very good justification.
