On Wed, Dec 09, 2020 at 09:55:30AM +0000, John Garry wrote: > On 09/12/2020 01:01, Ming Lei wrote: > > blk_mq_queue_tag_busy_iter() can be run on another request queue just > > between one driver tag is allocated and updating the request map, so one > > extra request reference still can be grabbed. > > > > So looks only holding one queue's usage_counter doesn't help this issue, > > since > > bt_for_each() always iterates on driver tags wide. > > > > > But I didn't see such a guard for blk_mq_tagset_busy_iter(). > > IMO there isn't real difference between the two iteration. > > ok, I see. Let me try to recreate that one, which will prob again require > artificial delays added. > > Apart from this, my concern is that we come with for a solution, but it's a > complicated solution and may not be accepted as this issue is not seen as a > problem in practice.
If that is the case, I'd suggest to consider the solution in the following link: https://lore.kernel.org/linux-block/20200820180335.3109216-1-ming....@redhat.com/ At least, the idea is simple, which can be extended to support allocate driver tags request pool dynamically. Thanks, Ming
