mm/damon/usage.rst

162 ``contexts`` directory contains files for controlling the monitoring contexts
172 ``0`` to ``N-1``.  Each directory represents each monitoring context (refer to
186 DAMON supports multiple types of :ref:`monitoring operations
189 monitoring operations set on the currently running kernel by reading
195 You can set and get what type of monitoring operations DAMON will use for the
204 Files for specifying attributes of the monitoring including required quality
205 and efficiency of the monitoring are in ``monitoring_attrs`` directory.
215 of DAMON's monitoring regions (``min`` and ``max``, respectively), which
216 controls the monitoring overhead, exist.  You can set and get the values by
219 For more details about the intervals and monitoring regions range, please refer
229 to ``N-1``.  Each directory represents each monitoring target.
248 In case of ``fvaddr`` or ``paddr`` monitoring operations sets, users are
249 required to set the monitoring target address ranges.  In case of ``vaddr``
251 monitoring region to specific address ranges.  Please refer to the :ref:`design
254 For such cases, users can explicitly set the initial monitoring target regions
259 to ``N-1``.  Each directory represents each initial monitoring target region.
267 can set and get the start and end addresses of the initial monitoring target
420 DAMON monitoring target filtering.  In case of the memory cgroup filtering, you
425 monitoring target filtering, you can specify the index of the target between
426 the list of the DAMON context's monitoring targets list to ``target_idx`` file.
495 and query-like efficient data access monitoring results retrievals.  For the
517 free memory rate of the system every 5 seconds, start the monitoring and paging
554 Users can get the monitoring results via the :ref:`tried_regions
556 snapshot, but it could be inefficient for fully recording all the monitoring
559 provides the whole monitoring results, while ``damon:damos_before_apply``
560 provides the monitoring results for regions that each DAMON-based Operation
565 monitoring results recording.
567 While the monitoring is turned on, you could record the tracepoint events and
579 Each line of the perf script output represents each monitoring region.  The
581 (``target_id=X``) shows the ide of the monitoring target of the region.  The
582 seventh field (``nr_regions=X``) shows the total number of monitoring regions
597 Each line of the output represents each monitoring region that each DAMON-based
636 ``update interval``, and min/max number of monitoring target regions by
637 reading from and writing to the ``attrs`` file.  To know about the monitoring
651 Some types of address spaces supports multiple monitoring target.  For example,
652 the virtual memory address spaces monitoring can have multiple processes as the
653 monitoring targets.  Users can set the targets by writing relevant id values of
655 ``target_ids`` file.  In case of the virtual address spaces monitoring, the
656 values should be pids of the monitoring target processes.  For example, below
657 commands set processes having pids 42 and 4242 as the monitoring targets and
667 space monitoring doesn't support multiple targets, reading the file will show a
675 Note that setting the target ids doesn't start the monitoring.
681 In case of the virtual address space monitoring, DAMON automatically sets and
682 updates the monitoring target regions so that entire memory mappings of target
683 processes can be covered.  However, users can want to limit the monitoring
689 In contrast, DAMON do not automatically sets and updates the monitoring target
690 regions in case of physical memory monitoring.  Therefore, users should set the
691 monitoring target regions by themselves.
693 In such cases, users can explicitly set the initial monitoring target regions
703 ``100-200`` as the initial monitoring target region of pid 42, which is the
716 Note that this sets the initial monitoring target regions only.  In case of
717 virtual memory monitoring, DAMON will automatically updates the boundary of the
828 free memory rate of the system every 5 seconds, start the monitoring and paging
844 start the monitoring.  You can start, stop, and check the current status of the
845 monitoring by writing to and reading from the ``monitor_on_DEPRECATED`` file.
846 Writing ``on`` to the file starts the monitoring of the targets with the
858 the monitoring is turned on.  If you write to the files while DAMON is running,
865 DAMON does requested monitoring with a kernel thread called ``kdamond``.  You
867 monitoring is turned off, reading the file returns ``none``. ::
882 One ``kdamond`` thread is created for each monitoring context.  You can create
883 and remove monitoring contexts for multiple ``kdamond`` required use case using