173 /* Idling period duration, in ns. */
322  * b) make BFQ uselessly perform device idling; device idling results
1292  * or device idling to their queues, unless these queues must be
1304  * which also implies idling the device for it, is almost always
1331  * idling depending on which choice boosts the throughput more. The
1504  * - BFQQE_NO_MORE_REQUESTS bfqq did not enjoy any device idling
1508  * - BFQQE_TOO_IDLE bfqq did enjoy device idling, but did not issue
1509  *   a new request before the expiration of the idling-time.
1515  * idling, I/O throttling somewhere in the path from the process to
2158  	 * since the first detection. If wakeups are rare, pointless idling  in bfq_check_waker()
2744 	 * time idling, and 2) serve the union of their requests in  in bfq_find_close_cooperator()
3397 	 * is seeky. A long idling is preserved for a weight-raised  in bfq_arm_slice_timer()
3399 	 * because a long idling is needed for guaranteeing to a queue  in bfq_arm_slice_timer()
3689  * There is a case where idling does not have to be performed for
3729  * idling (I/O-dispatch plugging) is certainly needed to guarantee
3733  * The problem is that idling may significantly reduce throughput with
3736  * queueing. So, unless bfqq falls in cases where idling also boosts
3738  * (ii) accurately, so as to avoid idling when not strictly needed for
3759  * idling may not be performed for the descendant processes of the
3776  * idling is not performed, regardless of whether condition (ii)
3778  * hold, then idling is allowed, and the device tends to be prevented
3801  * the mid term, even without idling. In fact, even if not
3802  * idling allows the internal queues of the device to contain
3824  * The motivation for using preemption instead of idling (for
3825  * queues with the same weight) is that, by not idling,
3864  * device-idling countermeasures may however fail in the following
3894 	/* No point in idling for bfqq if it won't get requests any longer */  in idling_needed_for_service_guarantees()
4052 			 * idling timeout expire. Hence it is not  in __bfq_bfqq_recalc_budget()
4086 			 * Since bfqq does not enjoy device idling,  in __bfq_bfqq_recalc_budget()
4186 			  * give same worst-case guarantees as idling  in bfq_bfqq_is_slow()
4328  * @compensate: if true, compensate for the time spent idling.
4510 	/* No point in idling for bfqq if it won't get requests any longer */  in idling_boosts_thr_without_issues()
4518 	 * The next variable takes into account the cases where idling  in idling_boosts_thr_without_issues()
4522 	 * idling is virtually always beneficial for the throughput if:  in idling_boosts_thr_without_issues()
4530 	 * Secondly, and in contrast to the above item (b), idling an  in idling_boosts_thr_without_issues()
4546 	 * special case, described below, idling may cause problems to  in idling_boosts_thr_without_issues()
4565 	 * then idling will be guaranteed by another variable, see  in idling_boosts_thr_without_issues()
4582  * For a queue that becomes empty, device idling is allowed only if
4584  * device idling plays a critical role for both throughput boosting
4588  * In a nutshell, this function returns true only if idling is
4590  * idling is however necessary to preserve service guarantees (low
4607 	/* No point in idling for bfqq if it won't get requests any longer */  in bfq_better_to_idle()
4615 	 * Idling is performed only if slice_idle > 0. In addition, we  in bfq_better_to_idle()
4634 	 * return value of the function, which is true only if idling  in bfq_better_to_idle()
4649  * why performing device idling is the best choice to boost the throughput
4828 	 * to enjoy device idling. The reason why, in this case, we  in bfq_select_queue()
4872 			 * not disable disk idling even when a new request  in bfq_select_queue()
4887 				 * So we disable idling.  in bfq_select_queue()
4897 	 * No requests pending. However, if the in-service queue is idling  in bfq_select_queue()
5627 		 * idle_class, because no device idling is performed  in bfq_init_bfqq()
5968 	 * no device idling is performed for bfqq in this case.  in bfq_update_has_short_ttime()
6107 		 * - we are idling to boost throughput, and  in bfq_rq_enqueued()
6125 		 * A large enough request arrived, or idling is being  in bfq_rq_enqueued()
6128 		 * cases disk idling is to be stopped, so clear  in bfq_rq_enqueued()
6471 			 * bfqq will start enjoying device idling  in bfq_completed_request()
6475 			 * not have the chance to enjoy device idling  in bfq_completed_request()
7053 		 * during disk idling.  in bfq_idle_slice_timer_body()
7068  * is idling inside its time slice.
7078 	 * different from the queue that was idling if a new request  in bfq_idle_slice_timer()