21  * The timer migration mechanism is built on a hierarchy of groups. The
49  * hierarchy up to the point where it is assigned the migrator role or if
76  * propagated up the hierarchy so hand over from other leaves can happen at
77  * all hierarchy levels w/o doing a search.
80  * duties up to the top level of the hierarchy (LVL2 in the example). It
90  * just won't participate in the hierarchy management anymore. Destroying
137  *    walk the hierarchy.
152  *    has to walk the hierarchy. Both CPUs (CPU0 and CPU1) now walk the
153  *    hierarchy to perform the needed update from their point of view. The
168  *    step 2) through the hierarchy to GRP1:0 before CPU0 (step 1) did. The
334  * After expiring timers of a remote CPU, a walk through the hierarchy and
385  *    been processed. So it is required to walk the hierarchy from CPU0's point
412  * The hierarchy walk in step 3 can be skipped if the migrator notices that a
414  * of the group as migrator and any needed updates within the hierarchy.
479  * struct tmigr_walk - data required for walking the hierarchy
484  *			hierarchy is completely idle.  When CPU itself was the
490  *			new top level of the hierarchy between the 'top level
506  *			the hierarchy walk is !idle in the timer migration
507  *			hierarchy. When the CPU is idle and the whole hierarchy is
688  * tmigr_cpu_activate() - set this CPU active in timer migration hierarchy
712  * the) hierarchy, but only when hierarchy is completely idle.
763 		 * Walking the hierarchy is required in any case when a  in tmigr_update_events()
771 		 * the hierarchy by the return:  in tmigr_update_events()
781 		 * However if there is no parent, ie: the hierarchy has only a  in tmigr_update_events()
815 		 * hierarchy even if there is a parent group.  in tmigr_update_events()
819 		 * is not active, walking the hierarchy is required to not miss  in tmigr_update_events()
878  * returned, if an active CPU will handle all the timer migration hierarchy
926 	 * updated the event takes care when hierarchy is completely  in tmigr_handle_remote_cpu()
952 	 * the hierarchy and goes back to idle.  in tmigr_handle_remote_cpu()
958 	 * several (unnecessary) locks during walking the hierarchy for updating  in tmigr_handle_remote_cpu()
966 	 * When the CPU went offline in the meantime, no hierarchy walk has to  in tmigr_handle_remote_cpu()
968 	 * already done during marking the CPU offline in the hierarchy.  in tmigr_handle_remote_cpu()
971 	 * also of the timers in the hierarchy.  in tmigr_handle_remote_cpu()
1119 	 * hierarchy walk is not active, proceed the walk to reach the top level  in tmigr_requires_handle_remote_up()
1173 	 * If the CPU is active, walk the hierarchy to check whether a remote  in tmigr_requires_handle_remote()
1205  * tmigr_cpu_new_timer() - enqueue next global timer into hierarchy (idle tmc)
1209  * hierarchy. tick_nohz_get_sleep_length() calls tick_nohz_next_event()
1211  * holds the first timer, when the timer migration hierarchy is
1347  * from the hierarchy if this CPU is the top level migrator or the hierarchy is
1431  * last active CPU in the hierarchy is offlining. With this, it is ensured that
1549 		 * hierarchy. Rely on the topology sibling mask would be a  in tmigr_get_group()
1567 	/* Setup successful. Add it to the hierarchy */  in tmigr_get_group()
1667 		 * available, not all calculated hierarchy levels are required.  in tmigr_setup_groups()
1805 	 * Calculate the required hierarchy levels. Unfortunately there is no  in tmigr_init()
1816 	/* Calc the hierarchy levels required to hold the CPUs of a node */  in tmigr_init()
1828 	 * level(s) of the hierarchy contains groups which handle all CPU groups  in tmigr_init()
1842 	pr_info("Timer migration: %d hierarchy levels; %d children per group;"  in tmigr_init()