include/linux/damon.h

49  * struct damon_region - Represents a monitoring target region.
69  * for it before getting the access monitoring results.
89  * struct damon_target - Represents a monitoring target.
91  * @nr_regions:		Number of monitoring target regions of this target.
92  * @regions_list:	Head of the monitoring target regions of this target.
95  * Each monitoring context could have multiple targets.  For example, a context
98  * virtual address spaces monitoring operations.
108  * enum damos_action - Represents an action of a Data Access Monitoring-based
225  * &weight_nr_accesses, and &weight_age, because monitoring operations are
279  * means DAMON does monitoring and applying the action of the scheme to
343  * @DAMOS_FILTER_TYPE_TARGET:	Data Access Monitoring target.
445  * struct damos - Represents a Data Access Monitoring-based Operation Scheme.
466  * monitoring context are inactive, DAMON stops monitoring either, and just
524  * enum damon_ops_id - Identifier for each monitoring operations implementation
526  * @DAMON_OPS_VADDR:	Monitoring operations for virtual address spaces
527  * @DAMON_OPS_FVADDR:	Monitoring operations for only fixed ranges of virtual
529  * @DAMON_OPS_PADDR:	Monitoring operations for the physical address space
530  * @NR_DAMON_OPS:	Number of monitoring operations implementations
540  * struct damon_operations - Monitoring operations for given use cases.
554  * space and usecase via the &damon_ctx.ops.  Then, the monitoring thread
556  * the monitoring, @update after each &damon_attrs.ops_update_interval, and
563  * this could be used to construct proper monitoring target regions and link
566  * this could be used to update monitoring target regions for current status.
567  * @prepare_access_checks should manipulate the monitoring regions to be
581  * monitoring.
601  * struct damon_callback - Monitoring events notification callbacks.
605  * @before_terminate:	Called before terminating the monitoring.
607  * The monitoring thread (&damon_ctx.kdamond) calls @before_terminate just
608  * before finishing the monitoring.
610  * The monitoring thread calls @after_wmarks_check after each DAMON-based
612  * attributes of the monitoring context while it's deactivated due to the
615  * The monitoring thread calls @after_aggregation for each of the aggregation
616  * intervals.  Therefore, users can safely access the monitoring results
620  * If any callback returns non-zero, monitoring stops.
650  * struct damon_intervals_goal - Monitoring intervals auto-tuning goal.
675  * struct damon_attrs - Monitoring attributes for accuracy/overhead control.
679  * @ops_update_interval:	The time between monitoring operations updates.
681  * @min_nr_regions:		The minimum number of adaptive monitoring
683  * @max_nr_regions:		The maximum number of adaptive monitoring
691  * ``mmap()`` calls from the application, in case of virtual memory monitoring)
717  * struct damon_ctx - Represents a context for each monitoring.  This is the
719  * of the monitoring.
721  * @attrs:		Monitoring attributes for accuracy/overhead control.
722  * @kdamond:		Kernel thread who does the monitoring.
725  * For each monitoring context, one kernel thread for the monitoring is
728  * Once started, the monitoring thread runs until explicitly required to be
729  * terminated or every monitoring target is invalid.  The validity of the
733  * know whether the monitoring is ongoing or terminated by reading @kdamond.
734  * Reads and writes to @kdamond from outside of the monitoring thread must
737  * Note that the monitoring thread protects only @kdamond via @kdamond_lock.
740  * @ops:	Set of monitoring operations for given use cases.
741  * @callback:	Set of callbacks for monitoring events notifications.
743  * @adaptive_targets:	Head of monitoring targets (&damon_target) list.