On 6/8/23 9:12 AM, Bart Van Assche wrote: > On 6/7/23 22:44, Christoph Hellwig wrote: >>>> Thanks. This wasn't obvious to me from the current code. I'll add a >>>> comment in the next version. >>> >>> The crucial question is now, is it sufficient to call >>> blk_mq_quiesce_queue_nowait() under the mutex, or does the call to >>> blk_mq_wait_quiesce_done() have to be under the mutex, too? >>> The latter would actually kill off our attempt to fix the delay >>> in fc_remote_port_delete() that was caused by repeated >>> synchronize_rcu() calls. >>> >>> But if I understand you correctly, moving the wait out of the mutex >>> should be ok. I'll update the series accordingly. >> >> I can't think of a reason we'd want to lock over the wait, but Bart >> knows this code way better than I do. > > Unless deep changes would be made in the block layer blk_mq_quiesce_queue_nowait() and/or blk_mq_wait_quiesce_done() functions, moving blk_mq_wait_quiesce_done() outside the critical section seems fine to me. If it helps, I tested with iscsi and the mutex changes discussed above and it worked ok for me. Also thanks for fixing this Martin.
