include/linux/damon.h

38  * struct damon_region - Represents a monitoring target region.
62  * struct damon_target - Represents a monitoring target.
64  * @nr_regions:		Number of monitoring target regions of this target.
65  * @regions_list:	Head of the monitoring target regions of this target.
68  * Each monitoring context could have multiple targets.  For example, a context
71  * example, in the virtual address monitoring context, it could be a pidfd or
82  * enum damos_action - Represents an action of a Data Access Monitoring-based
126  * &weight_nr_accesses, and &weight_age, because monitoring primitives are
176  * means DAMON does monitoring and applying the action of the scheme to
212  * struct damos - Represents a Data Access Monitoring-based Operation Scheme.
233  * monitoring context are inactive, DAMON stops monitoring either, and just
237  * stops monitoring and just repeatedly checks the watermarks.
260  * struct damon_primitive - Monitoring primitives for given use cases.
266  * @reset_aggregated:		Reset aggregated accesses monitoring results.
274  * space and usecase via the &damon_ctx.primitive.  Then, the monitoring thread
276  * the monitoring, @update after each &damon_ctx.primitive_update_interval, and
282  * this could be used to construct proper monitoring target regions and link
285  * this could be used to update monitoring target regions for current status.
286  * @prepare_access_checks should manipulate the monitoring regions to be
292  * @reset_aggregated should reset the access monitoring results that aggregated
301  * monitoring.
321  * struct damon_callback - Monitoring events notification callbacks.
323  * @before_start:	Called before starting the monitoring.
326  * @before_terminate:	Called before terminating the monitoring.
329  * The monitoring thread (&damon_ctx.kdamond) calls @before_start and
330  * @before_terminate just before starting and finishing the monitoring,
334  * The monitoring thread calls @after_sampling and @after_aggregation for each
336  * Therefore, users can safely access the monitoring results without additional
340  * If any callback returns non-zero, monitoring stops.
352  * struct damon_ctx - Represents a context for each monitoring.  This is the
354  * of the monitoring.
358  * @primitive_update_interval:	The time between monitoring primitive updates.
364  * in case of virtual memory monitoring) and applies the changes for each
369  * @kdamond:		Kernel thread who does the monitoring.
373  * For each monitoring context, one kernel thread for the monitoring is
376  * Once started, the monitoring thread runs until explicitly required to be
377  * terminated or every monitoring target is invalid.  The validity of the
381  * Therefore, users can know whether the monitoring is ongoing or terminated by
383  * outside of the monitoring thread must be protected by @kdamond_lock.
385  * Note that the monitoring thread protects only @kdamond and @kdamond_stop via
388  * @primitive:	Set of monitoring primitives for given use cases.
389  * @callback:	Set of callbacks for monitoring events notifications.
391  * @min_nr_regions:	The minimum number of adaptive monitoring regions.
392  * @max_nr_regions:	The maximum number of adaptive monitoring regions.
393  * @adaptive_targets:	Head of monitoring targets (&damon_target) list.