[PATCH 0/8] Introduce Data Access MONitor (DAMON)

[PATCH 0/8] Introduce Data Access MONitor (DAMON)

[SeongJae Park]

From: SeongJae Park <sjpark@amazon.de>

This patchset introduces a new kernel module for practical monitoring of
data accesses, namely DAMON.

The patches are organized in the following sequence.  The first four
patches implements the core logic of DAMON one by one.  After that, the
second patch allow users to use DAMON by implementing a debugfs based
low level interface.  To provide a minimal reference to the low level
interfaces and for more convenient test of the DAMON itself, the third
patch implements an user space tool for the DAMON.  The fourth patch
adds a document for administrators of DAMON, and finally the fifth patch
provides DAMON's unit tests, which is using the kunit framework.

The patches are based on the v5.4 plus the back-ported kunit, which
retrieved from v5.5-rc1.  You can also clone the complete git tree by:

    $ git clone git://github.com/sjp38/linux -b damon/patches/v1

The web is also available:
https://github.com/sjp38/linux/releases/tag/damon/patches/v1


Patch History
-------------

Changes from RFC v1
(https://lore.kernel.org/linux-mm/20200110131522.29964-1-sjpark@amazon.com/)
 - Split core logic of DAMON for more convenient review
 - Specify an ambiguous plan of access pattern based mm optimizations
 - Support loadable module build
 - Cleanup code

----

DAMON is a kernel module that allows users to monitor the actual memory
access pattern of specific user-space processes.  It aims to be 1)
accurate enough to be useful for performance-centric domains, and 2)
sufficiently light-weight so that it can be applied online.

For the goals, DAMON utilizes its two core mechanisms, called
region-based sampling and adaptive regions adjustment.  The region-based
sampling allows users to make their own trade-off between the quality
and the overhead of the monitoring and set the upperbound of the
monitoring overhead.  Further, the adaptive regions adjustment mechanism
makes DAMON to maximize the quality and minimize the overhead with its
best efforts while preserving the users configured trade-off.

Please note that the term 'memory' in this document means 'main memory'.
It also assumes that it would usually utilizes the middle level speed
memory devices such as DRAMs or NVRAMs.  CPU caches or storage devices
are not our concern, as those are too fast or too slow to be in DAMON's
scope.


Background
==========

For performance-centric analysis and optimizations of memory management
schemes (either that of kernel space or user space), the actual data
access pattern of the workloads is highly useful.  The information need
to be only reasonable rather than strictly correct, because some level
of incorrectness can be handled in many performance-centric domains.  It
also need to be taken within reasonably short time with only
light-weight overhead.

Manually extracting such data is not easy and time consuming if the
target workload is huge and complex, even for the developers of the
programs.  There are a range of tools and techniques developed for
general memory access investigations, and some of those could be
partially used for this purpose.  However, most of those are not
practical or unscalable, mainly because those are designed with no
consideration about the trade-off between the accuracy of the output and
the overhead.

The memory access instrumentation techniques which is applied to many
tools such as Intel PIN is essential for correctness required cases such
as invalid memory access bug detections.  However, those usually incur
high overhead which is unacceptable for many of the performance-centric
domains.  Periodic access checks based on H/W or S/W access counting
features (e.g., the Accessed bits of PTEs or the PG_Idle flags of pages)
can dramatically decrease the overhead by forgiving some of the quality,
compared to the instrumentation based techniques.  The reduced quality
is still reasonable for many of the domains, but the overhead can
arbitrarily increase as the size of the target workload grows.
Miniature-like static region based sampling can set the upperbound of
the overhead, but it will now decrease the quality of the output as the
size of the workload grows.


Related Works
=============

There are a number of researches[1,2,3,4,5,6] optimizing memory
management mechanisms based on the actual memory access patterns that
shows impressive results.  However, most of those has no deep
consideration about the monitoring of the accesses itself.  Some of
those focused on the overhead of the monitoring, but does not consider
the accuracy scalability[6] or has additional dependencies[7].  Indeed,
one recent research[5] about the proactive reclamation has also
proposed[8] to the kernel community but the monitoring overhead was
considered a main problem.

[1] Subramanya R Dulloor, Amitabha Roy, Zheguang Zhao, Narayanan
    Sundaram, Nadathur Satish, Rajesh Sankaran, Jeff Jackson, and
    Karsten Schwan. 2016.  Data tiering in heterogeneous memory systems.
    In Proceedings of the 11th European Conference on Computer Systems
    (EuroSys). ACM, 15.
[2] Youngjin Kwon, Hangchen Yu, Simon Peter, Christopher J Rossbach, and
    Emmett Witchel. 2016. Coordinated and efficient huge page management
    with ingens.  In 12th USENIX Symposium on Operating Systems Design
    and Implementation (OSDI).  705–721.
[3] Harald Servat, Antonio J Peña, Germán Llort, Estanislao Mercadal,
    HansChristian Hoppe, and Jesús Labarta. 2017. Automating the
    application data placement in hybrid memory systems. In 2017 IEEE
    International Conference on Cluster Computing (CLUSTER). IEEE, 126–136.
[4] Vlad Nitu, Boris Teabe, Alain Tchana, Canturk Isci, and Daniel
    Hagimont.  2018. Welcome to zombieland: practical and
    energy-efficient memory disaggregation in a datacenter. In
    Proceedings of the 13th European Conference on Computer Systems
    (EuroSys). ACM, 16.
[5] Andres Lagar-Cavilla, Junwhan Ahn, Suleiman Souhlal, Neha Agarwal,
    Radoslaw Burny, Shakeel Butt, Jichuan Chang, Ashwin Chaugule, Nan
    Deng, Junaid Shahid, Greg Thelen, Kamil Adam Yurtsever, Yu Zhao, and
    Parthasarathy Ranganathan.  2019. Software-Defined Far Memory in
    Warehouse-Scale Computers.  In Proceedings of the 24th International
    Conference on Architectural Support for Programming Languages and
    Operating Systems (ASPLOS).  ACM, New York, NY, USA, 317–330.
    DOI:https://doi.org/10.1145/3297858.3304053
[6] Carl Waldspurger, Trausti Saemundsson, Irfan Ahmad, and Nohhyun
    Park.  2017. Cache Modeling and Optimization using Miniature
    Simulations. In 2017 USENIX Annual Technical Conference (ATC).
    USENIX Association, Santa Clara, CA, 487–498.
    https://www.usenix.org/conference/atc17/technical-sessions/
[7] Haojie Wang, Jidong Zhai, Xiongchao Tang, Bowen Yu, Xiaosong Ma, and
    Wenguang Chen. 2018. Spindle: Informed Memory Access Monitoring. In
    2018 USENIX Annual Technical Conference (ATC). USENIX Association,
    Boston, MA, 561–574.
    https://www.usenix.org/conference/atc18/presentation/wang-haojie
[8] Jonathan Corbet. 2019. Proactively reclaiming idle memory. (2019).
    https://lwn.net/Articles/787611/.


Expected Use-cases
==================

A straightforward usecase of DAMON would be the program behavior
analysis.  With the DAMON output, users can confirm whether the program
is running as intended or not.  This will be useful for debuggings and
tests of design points.

The monitored results can also be useful for counting the dynamic
working set size of workloads.  For the administration of memory
overcommitted systems or selection of the environments (e.g., containers
providing different amount of memory) for your workloads, this will be
useful.

If you are a programmer, you can optimize your program by managing the
memory based on the actual data access pattern.  For example, you can
identify the dynamic hotness of your data using DAMON and call
``mlock()`` to keep your hot data in DRAM, or call ``madvise()`` with
``MADV_PAGEOUT`` to proactively reclaim cold data.  Even though your
program is guaranteed to not encounter memory pressure, you can still
improve the performance by applying the DAMON outputs for call of
``MADV_HUGEPAGE`` and ``MADV_NOHUGEPAGE``.  More creative optimizations
would be possible.  Our evaluations of DAMON includes a straightforward
optimization using the ``mlock()``.  Please refer to the below
Evaluation section for more detail.

As DAMON incurs very low overhead, such optimizations can be applied not
only offline, but also online.  Also, there is no reason to limit such
optimizations to the user space.  Several parts of the kernel's memory
management mechanisms could be also optimized using DAMON. The
reclamation, the THP (de)promotion decisions, and the compaction would
be such a candidates.  Nevertheless, current version of DAMON is not
highly optimized for the online/in-kernel uses.


A Future Plan: Data Access Based Optimizations Support
------------------------------------------------------

As described in the above section, DAMON could be helpful for actual
access based memory management optimizations.  Nevertheless, users who
want to do such optimizations should run DAMON, read the traced data
(either online or offline), analyze it, plan a new memory management
scheme, and apply the new scheme by themselves.  It must be easier than
the past, but could still require some level of efforts.  In its next
development stage, DAMON will reduce some of such efforts by allowing
users to specify some access based memory management rules for their
specific processes.

Because this is just a plan, the specific interface is not fixed yet,
but for example, users will be allowed to write their desired memory
management rules to a special file in a DAMON specific format.  The
rules will be something like 'if a memory region of size in a range is
keeping a range of hotness for more than a duration, apply specific
memory management rule using madvise() or mlock() to the region'.  For
example, we can imagine rules like below:

    $ cat <debugfs>/damon/rules
    # format is: <min/max size> <min/max frequency> <duration> <action>
    # Below rules assume the frequency score is in [0,20].

    # if a region of a size keeps a very high access frequency for more
    # than 100ms, lock the region in the main memory (call mlock()).
    # But, if the region is larger than 500 MiB, skip it.  The exception
    # might be helpful if the system has only, say, 600 MiB of DRAM, a
    # region of size larger than 600 MiB cannot be locked in the DRAM at
    # all.
    na 500M 15 20 100ms mlock

    # if a region keeps a high access frequency for more than 100ms, put
    # the region in the head of the LRU list (call madvise() with
    # MADV_WILLNEED).
    na na 12 15 100ms madv_willneed

    # if a region keeps a low access frequency for more than 100ms, put
    # the region in the tail of the LRU list (call madvise() with
    # MADV_COLD).
    na na 3  5  100ms madv_cold

    # if a region keeps a very low access frequency for more than 100ms,
    # swap out the region immediately (call madvise() with
    # MADV_PAGEOUT).
    na na 0  3  100ms madv_pageout

    # if a region of size bigger than 2MB keeps a very high access
    # frequency for more than 100ms, let the region to use huge pages
    # (call madvise() with MADV_HUGEPAGE).
    2M na 15 20 100ms madv_hugepage

    # If a regions of size bigger than > 2MB keeps no high access
    # frequency for more than 100ms, avoid the region from using huge
    # pages (call madvise() with MADV_NOHUGEPAGE).
    2M na 0 15 100ms madv_nohugepage


Mechanisms of DAMON
===================


Basic Access Check
------------------

DAMON basically reports what pages are how frequently accessed.  The
report is passed to users in binary format via a ``result file`` which
users can set it's path.  Note that the frequency is not an absolute
number of accesses, but a relative frequency among the pages of the
target workloads.

Users can also control the resolution of the reports by setting two time
intervals, ``sampling interval`` and ``aggregation interval``.  In
detail, DAMON checks access to each page per ``sampling interval``,
aggregates the results (counts the number of the accesses to each page),
and reports the aggregated results per ``aggregation interval``.  For
the access check of each page, DAMON uses the Accessed bits of PTEs.

This is thus similar to the previously mentioned periodic access checks
based mechanisms, which overhead is increasing as the size of the target
process grows.


Region Based Sampling
---------------------

To avoid the unbounded increase of the overhead, DAMON groups a number
of adjacent pages that assumed to have same access frequencies into a
region.  As long as the assumption (pages in a region have same access
frequencies) is kept, only one page in the region is required to be
checked.  Thus, for each ``sampling interval``, DAMON randomly picks one
page in each region and clears its Accessed bit.  After one more
``sampling interval``, DAMON reads the Accessed bit of the page and
increases the access frequency of the region if the bit has set
meanwhile.  Therefore, the monitoring overhead is controllable by
setting the number of regions.  DAMON allows users to set the minimal
and maximum number of regions for the trade-off.

Except the assumption, this is almost same with the above-mentioned
miniature-like static region based sampling.  In other words, this
scheme cannot preserve the quality of the output if the assumption is
not guaranteed.


Adaptive Regions Adjustment
---------------------------

At the beginning of the monitoring, DAMON constructs the initial regions
by evenly splitting the memory mapped address space of the process into
the user-specified minimal number of regions.  In this initial state,
the assumption is normally not kept and thus the quality could be low.
To keep the assumption as much as possible, DAMON adaptively merges and
splits each region.  For each ``aggregation interval``, it compares the
access frequencies of adjacent regions and merges those if the frequency
difference is small.  Then, after it reports and clears the aggregated
access frequency of each region, it splits each region into two regions
if the total number of regions is smaller than the half of the
user-specified maximum number of regions.

In this way, DAMON provides its best-effort quality and minimal overhead
while keeping the bounds users set for their trade-off.


Applying Dynamic Memory Mappings
--------------------------------

Only a number of small parts in the super-huge virtual address space of
the processes is mapped to physical memory and accessed.  Thus, tracking
the unmapped address regions is just wasteful.  However, tracking every
memory mapping change might incur an overhead.  For the reason, DAMON
applies the dynamic memory mapping changes to the tracking regions only
for each of an user-specified time interval (``regions update
interval``).


Evaluations
===========

A prototype of DAMON has evaluated on an Intel Xeon E7-8837 machine
using 20 benchmarks that picked from SPEC CPU 2006, NAS, Tensorflow
Benchmark, SPLASH-2X, and PARSEC 3 benchmark suite.  Nonethless, this
section provides only summary of the results.  For more detail, please
refer to the slides used for the introduction of DAMON at the Linux
Plumbers Conference 2019[1] or the MIDDLEWARE'19 industrial track
paper[2].


Quality
-------

We first traced and visualized the data access pattern of each workload.
We were able to confirm that the visualized results are reasonably
accurate by manually comparing those with the source code of the
workloads.

To see the usefulness of the monitoring, we optimized 9 memory intensive
workloads among them for memory pressure situations using the DAMON
outputs.  In detail, we identified frequently accessed memory regions in
each workload based on the DAMON results and protected them with
``mlock()`` system calls.  The optimized versions consistently show
speedup (2.55x in best case, 1.65x in average) under memory pressure
situation.


Overhead
--------

We also measured the overhead of DAMON.  It was not only under the
upperbound we set, but was much lower (0.6 percent of the bound in best
case, 13.288 percent of the bound in average).  This reduction of the
overhead is mainly resulted from the adaptive regions adjustment.  We
also compared the overhead with that of the straightforward periodic
Accessed bit check-based monitoring, which checks the access of every
page frame.  DAMON's overhead was much smaller than the straightforward
mechanism by 94,242.42x in best case, 3,159.61x in average.


References
==========

Prototypes of DAMON have introduced by an LPC kernel summit track
talk[1] and two academic papers[2,3].  Please refer to those for more
detailed information, especially the evaluations.

[1] SeongJae Park, Tracing Data Access Pattern with Bounded Overhead and
    Best-effort Accuracy. In The Linux Kernel Summit, September 2019.
    https://linuxplumbersconf.org/event/4/contributions/548/
[2] SeongJae Park, Yunjae Lee, Heon Y. Yeom, Profiling Dynamic Data
    Access Patterns with Controlled Overhead and Quality. In 20th
    ACM/IFIP International Middleware Conference Industry, December
    2019.  https://dl.acm.org/doi/10.1145/3366626.3368125
[3] SeongJae Park, Yunjae Lee, Yunhee Kim, Heon Y. Yeom, Profiling
    Dynamic Data Access Patterns with Bounded Overhead and Accuracy. In
    IEEE International Workshop on Foundations and Applications of Self-
    Systems (FAS 2019), June 2019.

SeongJae Park (8):
  mm: Introduce Data Access MONitor (DAMON)
  mm/damon: Implement region based sampling
  mm/damon: Adaptively adjust regions
  mm/damon: Apply dynamic memory mapping changes
  mm/damon: Add debugfs interface
  mm/damon: Add minimal user-space tools
  Documentation/admin-guide/mm: Add a document for DAMON
  mm/damon: Add kunit tests

 .../admin-guide/mm/data_access_monitor.rst    |  240 ++++
 Documentation/admin-guide/mm/index.rst        |    1 +
 MAINTAINERS                                   |    9 +
 mm/Kconfig                                    |   23 +
 mm/Makefile                                   |    1 +
 mm/damon-test.h                               |  571 ++++++++
 mm/damon.c                                    | 1279 +++++++++++++++++
 tools/damon/.gitignore                        |    1 +
 tools/damon/_dist.py                          |   35 +
 tools/damon/bin2txt.py                        |   64 +
 tools/damon/damo                              |   37 +
 tools/damon/heats.py                          |  358 +++++
 tools/damon/nr_regions.py                     |   88 ++
 tools/damon/record.py                         |  193 +++
 tools/damon/report.py                         |   45 +
 tools/damon/wss.py                            |   94 ++
 16 files changed, 3039 insertions(+)
 create mode 100644 Documentation/admin-guide/mm/data_access_monitor.rst
 create mode 100644 mm/damon-test.h
 create mode 100644 mm/damon.c
 create mode 100644 tools/damon/.gitignore
 create mode 100644 tools/damon/_dist.py
 create mode 100644 tools/damon/bin2txt.py
 create mode 100755 tools/damon/damo
 create mode 100644 tools/damon/heats.py
 create mode 100644 tools/damon/nr_regions.py
 create mode 100644 tools/damon/record.py
 create mode 100644 tools/damon/report.py
 create mode 100644 tools/damon/wss.py

-- 
2.17.1

[PATCH 1/8] mm: Introduce Data Access MONitor (DAMON)

[SeongJae Park]

From: SeongJae Park <sjpark@amazon.de>

This commit introduces a new kernel module named DAMON.  Note that this
commit is implementing only the stub for the module load/unload, basic
data structures, and simple manipulation functions of the structures to
keep the size of commit small.  The core mechanisms of DAMON will be
implemented one by one by following commits.

Brief Introduction
==================

DAMON is a kernel module that would allow users to monitor the actual
memory access pattern of specific user-space processes.  It aims to be
1) accurate enough to be useful for performance-centric domains, and 2)
sufficiently light-weight so that it can be applied online.

For the goals, DAMON will utilize its two core mechanisms, called
region-based sampling and adaptive regions adjustment.  The region-based
sampling allows users to make their own trade-off between the quality
and the overhead of the monitoring and set the upperbound of the
monitoring overhead.  Further, the adaptive regions adjustment mechanism
makes DAMON to maximize the quality and minimize the overhead with its
best efforts while preserving the users configured trade-off.

Please note that the term 'memory' in this context means 'main memory'.
It also assumes that it would usually utilizes the middle level speed
memory devices such as DRAMs or NVRAMs.  CPU caches or storage devices
are not DAMON's concern, as those are too fast or too slow to be in
DAMON's scope.

Expected Use-cases
==================

A straightforward usecase of DAMON would be the program behavior
analysis.  With the DAMON output, users can confirm whether the program
is running as intended or not.  This will be useful for debuggings and
tests of design points.

The monitored results can also be useful for counting the dynamic
working set size of workloads.  For the administration of memory
overcommitted systems or selection of the environments (e.g., containers
providing different amount of memory) for your workloads, this will be
useful.

If you are a programmer, you can optimize your program by managing the
memory based on the actual data access pattern.  For example, you can
identify the dynamic hotness of your data using DAMON and call
``mlock()`` to keep your hot data in DRAM, or call ``madvise()`` with
``MADV_PAGEOUT`` to proactively reclaim cold data.  Even though your
program is guaranteed to not encounter memory pressure, you can still
improve the performance by applying the DAMON outputs for call of
``MADV_HUGEPAGE`` and ``MADV_NOHUGEPAGE``.  More creative optimizations
would be possible.  Our evaluations of DAMON includes a straightforward
optimization using the ``mlock()``.  Please refer to the below
Evaluation section for more detail.

As DAMON incurs very low overhead, such optimizations can be applied not
only offline, but also online.  Also, there is no reason to limit such
optimizations to the user space.  Several parts of the kernel's memory
management mechanisms could be also optimized using DAMON. The
reclamation, the THP (de)promotion decisions, and the compaction would
be such a candidates.  Nevertheless, current version of DAMON is not
highly optimized for the online/in-kernel uses.

Signed-off-by: SeongJae Park <sjpark@amazon.de>
---
 MAINTAINERS |   6 ++
 mm/Kconfig  |  12 +++
 mm/Makefile |   1 +
 mm/damon.c  | 223 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 242 insertions(+)
 create mode 100644 mm/damon.c

diff --git a/MAINTAINERS b/MAINTAINERS
index 97e3ce930083..c8db500a182f 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4583,6 +4583,12 @@ F:	net/ax25/ax25_out.c
 F:	net/ax25/ax25_timer.c
 F:	net/ax25/sysctl_net_ax25.c
 
+DATA ACCESS MONITOR
+M:	SeongJae Park <sjpark@amazon.de>
+L:	linux-mm@kvack.org
+S:	Maintained
+F:	mm/damon.c
+
 DAVICOM FAST ETHERNET (DMFE) NETWORK DRIVER
 L:	netdev@vger.kernel.org
 S:	Orphan
diff --git a/mm/Kconfig b/mm/Kconfig
index a5dae9a7eb51..35813e0d0e28 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -736,4 +736,16 @@ config ARCH_HAS_PTE_SPECIAL
 config ARCH_HAS_HUGEPD
 	bool
 
+config DAMON
+	tristate "Data Access Monitor"
+	depends on MMU
+	default y
+	help
+	  Provides data access monitoring.
+
+	  DAMON is a kernel module that allows users to monitor the actual
+	  memory access pattern of specific user-space processes.  It aims to
+	  be 1) accurate enough to be useful for performance-centric domains,
+	  and 2) sufficiently light-weight so that it can be applied online.
+
 endmenu
diff --git a/mm/Makefile b/mm/Makefile
index d996846697ef..6c3c7c364015 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -107,3 +107,4 @@ obj-$(CONFIG_PERCPU_STATS) += percpu-stats.o
 obj-$(CONFIG_ZONE_DEVICE) += memremap.o
 obj-$(CONFIG_HMM_MIRROR) += hmm.o
 obj-$(CONFIG_MEMFD_CREATE) += memfd.o
+obj-$(CONFIG_DAMON) += damon.o
diff --git a/mm/damon.c b/mm/damon.c
new file mode 100644
index 000000000000..064ec1f6ded9
--- /dev/null
+++ b/mm/damon.c
@@ -0,0 +1,223 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Data Access Monitor
+ *
+ * Copyright 2019 Amazon.com, Inc. or its affiliates.  All rights reserved.
+ *
+ * Author: SeongJae Park <sjpark@amazon.de>
+ */
+
+#define pr_fmt(fmt) "damon: " fmt
+
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/random.h>
+#include <linux/slab.h>
+
+#define damon_get_task_struct(t) \
+	(get_pid_task(find_vpid(t->pid), PIDTYPE_PID))
+
+#define damon_next_region(r) \
+	(container_of(r->list.next, struct damon_region, list))
+
+#define damon_prev_region(r) \
+	(container_of(r->list.prev, struct damon_region, list))
+
+#define damon_for_each_region(r, t) \
+	list_for_each_entry(r, &t->regions_list, list)
+
+#define damon_for_each_region_safe(r, next, t) \
+	list_for_each_entry_safe(r, next, &t->regions_list, list)
+
+#define damon_for_each_task(t) \
+	list_for_each_entry(t, &damon_tasks_list, list)
+
+#define damon_for_each_task_safe(t, next) \
+	list_for_each_entry_safe(t, next, &damon_tasks_list, list)
+
+/* Represents a monitoring target region on the virtual address space */
+struct damon_region {
+	unsigned long vm_start;
+	unsigned long vm_end;
+	unsigned long sampling_addr;
+	unsigned int nr_accesses;
+	struct list_head list;
+};
+
+/* Represents a monitoring target task */
+struct damon_task {
+	unsigned long pid;
+	struct list_head regions_list;
+	struct list_head list;
+};
+
+/* List of damon_task objects */
+static LIST_HEAD(damon_tasks_list);
+
+static struct rnd_state rndseed;
+/* Get a random number in [l, r) */
+#define damon_rand(l, r) (l + prandom_u32_state(&rndseed) % (r - l))
+
+/*
+ * Construct a damon_region struct
+ *
+ * Returns the pointer to the new struct if success, or NULL otherwise
+ */
+static struct damon_region *damon_new_region(unsigned long vm_start,
+					unsigned long vm_end)
+{
+	struct damon_region *ret;
+
+	ret = kmalloc(sizeof(struct damon_region), GFP_KERNEL);
+	if (!ret)
+		return NULL;
+	ret->vm_start = vm_start;
+	ret->vm_end = vm_end;
+	ret->nr_accesses = 0;
+	ret->sampling_addr = damon_rand(vm_start, vm_end);
+	INIT_LIST_HEAD(&ret->list);
+
+	return ret;
+}
+
+/*
+ * Add a region between two other regions
+ */
+static inline void damon_add_region(struct damon_region *r,
+		struct damon_region *prev, struct damon_region *next)
+{
+	__list_add(&r->list, &prev->list, &next->list);
+}
+
+/*
+ * Append a region to a task's list of regions
+ */
+static void damon_add_region_tail(struct damon_region *r, struct damon_task *t)
+{
+	list_add_tail(&r->list, &t->regions_list);
+}
+
+/*
+ * Delete a region from its list
+ */
+static void damon_del_region(struct damon_region *r)
+{
+	list_del(&r->list);
+}
+
+/*
+ * De-allocate a region
+ */
+static void damon_free_region(struct damon_region *r)
+{
+	kfree(r);
+}
+
+static void damon_destroy_region(struct damon_region *r)
+{
+	damon_del_region(r);
+	damon_free_region(r);
+}
+
+/*
+ * Construct a damon_task struct
+ *
+ * Returns the pointer to the new struct if success, or NULL otherwise
+ */
+static struct damon_task *damon_new_task(unsigned long pid)
+{
+	struct damon_task *t;
+
+	t = kmalloc(sizeof(struct damon_task), GFP_KERNEL);
+	if (!t)
+		return NULL;
+	t->pid = pid;
+	INIT_LIST_HEAD(&t->regions_list);
+
+	return t;
+}
+
+/* Returns n-th damon_region of the given task */
+struct damon_region *damon_nth_region_of(struct damon_task *t, unsigned int n)
+{
+	struct damon_region *r;
+	unsigned int i;
+
+	i = 0;
+	damon_for_each_region(r, t) {
+		if (i++ == n)
+			return r;
+	}
+	return NULL;
+}
+
+static void damon_add_task_tail(struct damon_task *t)
+{
+	list_add_tail(&t->list, &damon_tasks_list);
+}
+
+static void damon_del_task(struct damon_task *t)
+{
+	list_del(&t->list);
+}
+
+static void damon_free_task(struct damon_task *t)
+{
+	struct damon_region *r, *next;
+
+	damon_for_each_region_safe(r, next, t)
+		damon_free_region(r);
+	kfree(t);
+}
+
+static void damon_destroy_task(struct damon_task *t)
+{
+	damon_del_task(t);
+	damon_free_task(t);
+}
+
+/*
+ * Returns number of monitoring target tasks
+ */
+static unsigned int nr_damon_tasks(void)
+{
+	struct damon_task *t;
+	unsigned int ret = 0;
+
+	damon_for_each_task(t)
+		ret++;
+	return ret;
+}
+
+/*
+ * Returns the number of target regions for a given target task
+ */
+static unsigned int nr_damon_regions(struct damon_task *t)
+{
+	struct damon_region *r;
+	unsigned int ret = 0;
+
+	damon_for_each_region(r, t)
+		ret++;
+	return ret;
+}
+
+static int __init damon_init(void)
+{
+	pr_info("init\n");
+
+	prandom_seed_state(&rndseed, 42);
+	return 0;
+}
+
+static void __exit damon_exit(void)
+{
+	pr_info("exit\n");
+}
+
+module_init(damon_init);
+module_exit(damon_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("SeongJae Park <sjpark@amazon.de>");
+MODULE_DESCRIPTION("DAMON: Data Access MONitor");
-- 
2.17.1

[PATCH 2/8] mm/damon: Implement region based sampling

[SeongJae Park]

From: SeongJae Park <sjpark@amazon.de>

This commit implements DAMON's basic access check and region based
sampling mechanisms.

Basic Access Check
------------------

DAMON basically reports what pages are how frequently accessed.  Note
that the frequency is not an absolute number of accesses, but a relative
frequency among the pages of the target workloads.

Users can control the resolution of the reports by setting two time
intervals, ``sampling interval`` and ``aggregation interval``.  In
detail, DAMON checks access to each page per ``sampling interval``,
aggregates the results (counts the number of the accesses to each page),
and reports the aggregated results per ``aggregation interval``.  For
the access check of each page, DAMON uses the Accessed bits of PTEs.

This is thus similar to common periodic access checks based access
tracking mechanisms, which overhead is increasing as the size of the
target process grows.

Region Based Sampling
---------------------

To avoid the unbounded increase of the overhead, DAMON groups a number
of adjacent pages that assumed to have same access frequencies into a
region.  As long as the assumption (pages in a region have same access
frequencies) is kept, only one page in the region is required to be
checked.  Thus, for each ``sampling interval``, DAMON randomly picks one
page in each region and clears its Accessed bit.  After one more
``sampling interval``, DAMON reads the Accessed bit of the page and
increases the access frequency of the region if the bit has set
meanwhile.  Therefore, the monitoring overhead is controllable by
setting the number of regions.

Nonetheless, this scheme cannot preserve the quality of the output if
the assumption is not kept.

Signed-off-by: SeongJae Park <sjpark@amazon.de>
---
 mm/damon.c | 599 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 599 insertions(+)

diff --git a/mm/damon.c b/mm/damon.c
index 064ec1f6ded9..2a0c010291f8 100644
--- a/mm/damon.c
+++ b/mm/damon.c
@@ -9,9 +9,14 @@
 
 #define pr_fmt(fmt) "damon: " fmt
 
+#include <linux/delay.h>
+#include <linux/kthread.h>
 #include <linux/mm.h>
 #include <linux/module.h>
+#include <linux/page_idle.h>
 #include <linux/random.h>
+#include <linux/sched/mm.h>
+#include <linux/sched/task.h>
 #include <linux/slab.h>
 
 #define damon_get_task_struct(t) \
@@ -54,6 +59,36 @@ struct damon_task {
 /* List of damon_task objects */
 static LIST_HEAD(damon_tasks_list);
 
+/*
+ * For each 'sample_interval', DAMON checks whether each region is accessed or
+ * not.  It aggregates and keeps the access information (number of accesses to
+ * each region) for 'aggr_interval' and then flushes it to the result buffer if
+ * an 'aggr_interval' surpassed.
+ *
+ * All time intervals are in micro-seconds.
+ */
+static unsigned long sample_interval = 5 * 1000;
+static unsigned long aggr_interval = 100 * 1000;
+
+static struct timespec64 last_aggregate_time;
+
+static unsigned long min_nr_regions = 10;
+
+/* result buffer */
+#define DAMON_LEN_RBUF	(1024 * 1024 * 4)
+static char damon_rbuf[DAMON_LEN_RBUF];
+static unsigned int damon_rbuf_offset;
+
+/* result file */
+#define LEN_RES_FILE_PATH	256
+static char rfile_path[LEN_RES_FILE_PATH] = "/damon.data";
+
+static struct task_struct *kdamond;
+static bool kdamond_stop;
+
+/* Protects read/write of kdamond and kdamond_stop */
+static DEFINE_SPINLOCK(kdamond_lock);
+
 static struct rnd_state rndseed;
 /* Get a random number in [l, r) */
 #define damon_rand(l, r) (l + prandom_u32_state(&rndseed) % (r - l))
@@ -202,16 +237,580 @@ static unsigned int nr_damon_regions(struct damon_task *t)
 	return ret;
 }
 
+/*
+ * Get the mm_struct of the given task
+ *
+ * Callser should put the mm_struct after use, unless it is NULL.
+ *
+ * Returns the mm_struct of the task on success, NULL on failure
+ */
+static struct mm_struct *damon_get_mm(struct damon_task *t)
+{
+	struct task_struct *task;
+	struct mm_struct *mm;
+
+	task = damon_get_task_struct(t);
+	if (!task)
+		return NULL;
+
+	mm = get_task_mm(task);
+	put_task_struct(task);
+	return mm;
+}
+
+/*
+ * Size-evenly split a region into 'nr_pieces' small regions
+ *
+ * Returns 0 on success, or negative error code otherwise.
+ */
+static int damon_split_region_evenly(struct damon_region *r,
+				unsigned int nr_pieces)
+{
+	unsigned long sz_orig, sz_piece, orig_end;
+	struct damon_region *piece = NULL, *next;
+	unsigned long start;
+
+	if (!r || !nr_pieces)
+		return -EINVAL;
+
+	orig_end = r->vm_end;
+	sz_orig = r->vm_end - r->vm_start;
+	sz_piece = sz_orig / nr_pieces;
+
+	if (!sz_piece)
+		return -EINVAL;
+
+	r->vm_end = r->vm_start + sz_piece;
+	next = damon_next_region(r);
+	for (start = r->vm_end; start + sz_piece <= orig_end;
+			start += sz_piece) {
+		piece = damon_new_region(start, start + sz_piece);
+		damon_add_region(piece, r, next);
+		r = piece;
+	}
+	if (piece)
+		piece->vm_end = orig_end;
+	return 0;
+}
+
+struct region {
+	unsigned long start;
+	unsigned long end;
+};
+
+static unsigned long sz_region(struct region *r)
+{
+	return r->end - r->start;
+}
+
+static void swap_regions(struct region *r1, struct region *r2)
+{
+	struct region tmp;
+
+	tmp = *r1;
+	*r1 = *r2;
+	*r2 = tmp;
+}
+
+/*
+ * Find the three regions in an address space
+ *
+ * vma		the head vma of the target address space
+ * regions	an array of three 'struct region's that results will be saved
+ *
+ * This function receives an address space and finds three regions in it which
+ * separated by the two biggest unmapped regions in the space.
+ *
+ * Returns 0 if success, or negative error code otherwise.
+ */
+static int damon_three_regions_in_vmas(struct vm_area_struct *vma,
+		struct region regions[3])
+{
+	struct region gap = {0,}, first_gap = {0,}, second_gap = {0,};
+	struct vm_area_struct *last_vma = NULL;
+	unsigned long start = 0;
+
+	/* Find two biggest gaps so that first_gap > second_gap > others */
+	for (; vma; vma = vma->vm_next) {
+		if (!last_vma) {
+			start = vma->vm_start;
+			last_vma = vma;
+			continue;
+		}
+		gap.start = last_vma->vm_end;
+		gap.end = vma->vm_start;
+		if (sz_region(&gap) > sz_region(&second_gap)) {
+			swap_regions(&gap, &second_gap);
+			if (sz_region(&second_gap) > sz_region(&first_gap))
+				swap_regions(&second_gap, &first_gap);
+		}
+		last_vma = vma;
+	}
+
+	if (!sz_region(&second_gap) || !sz_region(&first_gap))
+		return -EINVAL;
+
+	/* Sort the two biggest gaps by address */
+	if (first_gap.start > second_gap.start)
+		swap_regions(&first_gap, &second_gap);
+
+	/* Store the result */
+	regions[0].start = start;
+	regions[0].end = first_gap.start;
+	regions[1].start = first_gap.end;
+	regions[1].end = second_gap.start;
+	regions[2].start = second_gap.end;
+	regions[2].end = last_vma->vm_end;
+
+	return 0;
+}
+
+/*
+ * Get the three regions in the given task
+ *
+ * Returns 0 on success, negative error code otherwise.
+ */
+static int damon_three_regions_of(struct damon_task *t,
+				struct region regions[3])
+{
+	struct mm_struct *mm;
+	int ret;
+
+	mm = damon_get_mm(t);
+	if (!mm)
+		return -EINVAL;
+
+	down_read(&mm->mmap_sem);
+	ret = damon_three_regions_in_vmas(mm->mmap, regions);
+	up_read(&mm->mmap_sem);
+
+	mmput(mm);
+	return ret;
+}
+
+/*
+ * Initialize the monitoring target regions for the given task
+ *
+ * t	the given target task
+ *
+ * Because only a number of small portions of the entire address space
+ * is acutally mapped to the memory and accessed, monitoring the unmapped
+ * regions is wasteful.  That said, because we can deal with small noises,
+ * tracking every mapping is not strictly required but could even incur a high
+ * overhead if the mapping frequently changes or the number of mappings is
+ * high.
+ *
+ * As usual memory map of processes is as below, the gap between the heap and
+ * the uppermost mmap()-ed region, and the gap between the lowermost mmap()-ed
+ * region and the stack will be two biggest unmapped regions.  Because these
+ * gaps are outliers between the mapped and unmapped regions in the address
+ * space in terms of the size, excluding these two biggest unmapped regions
+ * will be sufficient to make a trade-off.
+ *
+ *   <heap>
+ *   <BIG UNMAPPED REGION 1>
+ *   <uppermost mmap()-ed region>
+ *   (other mmap()-ed regions and small unmapped regions)
+ *   <lowermost mmap()-ed region>
+ *   <BIG UNMAPPED REGION 2>
+ *   <stack>
+ *
+ * For the reason, this function converts the original address space of the
+ * given task to a simplified address space, that is constructed with three
+ * regions separated by the two biggest unmapped regions and stores those in
+ * the given task.
+ */
+static void damon_init_regions_of(struct damon_task *t)
+{
+	struct damon_region *r;
+	struct region regions[3];
+	int i;
+
+	if (damon_three_regions_of(t, regions)) {
+		pr_err("Failed to get three regions of task %lu\n", t->pid);
+		return;
+	}
+
+	/* Set the initial three regions of the task */
+	for (i = 0; i < 3; i++) {
+		r = damon_new_region(regions[i].start, regions[i].end);
+		damon_add_region_tail(r, t);
+	}
+
+	/* Split the middle region into 'min_nr_regions - 2' regions */
+	r = damon_nth_region_of(t, 1);
+	if (damon_split_region_evenly(r, min_nr_regions - 2))
+		pr_warn("Init middle region failed to be split\n");
+}
+
+/* Initialize '->regions_list' of every task */
+static void kdamond_init_regions(void)
+{
+	struct damon_task *t;
+
+	damon_for_each_task(t)
+		damon_init_regions_of(t);
+}
+
+/*
+ * Check whether the given region has accessed since the last check
+ *
+ * mm	'mm_struct' for the given virtual address space
+ * r	the region to be checked
+ */
+static void kdamond_check_access(struct mm_struct *mm, struct damon_region *r)
+{
+	pte_t *pte = NULL;
+	pmd_t *pmd = NULL;
+	spinlock_t *ptl;
+
+	if (follow_pte_pmd(mm, r->sampling_addr, NULL, &pte, &pmd, &ptl))
+		goto mkold;
+
+	/* Read the page table access bit of the page */
+	if (pte && pte_young(*pte))
+		r->nr_accesses++;
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+	else if (pmd && pmd_young(*pmd))
+		r->nr_accesses++;
+#endif	/* CONFIG_TRANSPARENT_HUGEPAGE */
+
+	spin_unlock(ptl);
+
+mkold:
+	/* mkold next target */
+	r->sampling_addr = damon_rand(r->vm_start, r->vm_end);
+
+	if (follow_pte_pmd(mm, r->sampling_addr, NULL, &pte, &pmd, &ptl))
+		return;
+
+	if (pte) {
+		if (pte_young(*pte))
+			clear_page_idle(pte_page(*pte));
+		*pte = pte_mkold(*pte);
+	}
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+	else if (pmd) {
+		if (pmd_young(*pmd))
+			clear_page_idle(pmd_page(*pmd));
+		*pmd = pmd_mkold(*pmd);
+	}
+#endif
+
+	spin_unlock(ptl);
+}
+
+/*
+ * Check whether a time interval is elapsed
+ *
+ * baseline	the time to check whether the interval has elapsed since
+ * interval	the time interval (microseconds)
+ *
+ * See whether the given time interval has passed since the given baseline
+ * time.  If so, it also updates the baseline to current time for next check.
+ *
+ * Returns true if the time interval has passed, or false otherwise.
+ */
+static bool damon_check_reset_time_interval(struct timespec64 *baseline,
+		unsigned long interval)
+{
+	struct timespec64 now;
+
+	ktime_get_coarse_ts64(&now);
+	if ((timespec64_to_ns(&now) - timespec64_to_ns(baseline)) / 1000 <
+			interval)
+		return false;
+	*baseline = now;
+	return true;
+}
+
+/*
+ * Check whether it is time to flush the aggregated information
+ */
+static bool kdamond_aggregate_interval_passed(void)
+{
+	return damon_check_reset_time_interval(&last_aggregate_time,
+			aggr_interval);
+}
+
+/*
+ * Flush the content in the result buffer to the result file
+ */
+static void damon_flush_rbuffer(void)
+{
+	ssize_t sz;
+	loff_t pos;
+	struct file *rfile;
+
+	while (damon_rbuf_offset) {
+		pos = 0;
+		rfile = filp_open(rfile_path, O_CREAT | O_RDWR | O_APPEND,
+				0644);
+		if (IS_ERR(rfile)) {
+			pr_err("Cannot open the result file %s\n", rfile_path);
+			return;
+		}
+
+		sz = kernel_write(rfile, damon_rbuf, damon_rbuf_offset, &pos);
+		filp_close(rfile, NULL);
+
+		damon_rbuf_offset -= sz;
+	}
+}
+
+/*
+ * Write a data into the result buffer
+ */
+static void damon_write_rbuf(void *data, ssize_t size)
+{
+	if (damon_rbuf_offset + size > DAMON_LEN_RBUF)
+		damon_flush_rbuffer();
+
+	memcpy(&damon_rbuf[damon_rbuf_offset], data, size);
+	damon_rbuf_offset += size;
+}
+
+/*
+ * Flush the aggregated monitoring results to the result buffer
+ *
+ * Stores current tracking results to the result buffer and reset 'nr_accesses'
+ * of each regions.  The format for the result buffer is as below:
+ *
+ *   <time> <number of tasks> <array of task infos>
+ *
+ *   task info: <pid> <number of regions> <array of region infos>
+ *   region info: <start address> <end address> <nr_accesses>
+ */
+static void kdamond_flush_aggregated(void)
+{
+	struct damon_task *t;
+	struct timespec64 now;
+	unsigned int nr;
+
+	ktime_get_coarse_ts64(&now);
+
+	damon_write_rbuf(&now, sizeof(struct timespec64));
+	nr = nr_damon_tasks();
+	damon_write_rbuf(&nr, sizeof(nr));
+
+	damon_for_each_task(t) {
+		struct damon_region *r;
+
+		damon_write_rbuf(&t->pid, sizeof(t->pid));
+		nr = nr_damon_regions(t);
+		damon_write_rbuf(&nr, sizeof(nr));
+		damon_for_each_region(r, t) {
+			damon_write_rbuf(&r->vm_start, sizeof(r->vm_start));
+			damon_write_rbuf(&r->vm_end, sizeof(r->vm_end));
+			damon_write_rbuf(&r->nr_accesses,
+					sizeof(r->nr_accesses));
+			r->nr_accesses = 0;
+		}
+	}
+}
+
+/*
+ * Check whether current monitoring should be stopped
+ *
+ * If users asked to stop, need stop.  Even though no user has asked to stop,
+ * need stop if every target task has dead.
+ *
+ * Returns true if need to stop current monitoring.
+ */
+static bool kdamond_need_stop(void)
+{
+	struct damon_task *t;
+	struct task_struct *task;
+	bool stop;
+
+	spin_lock(&kdamond_lock);
+	stop = kdamond_stop;
+	spin_unlock(&kdamond_lock);
+	if (stop)
+		return true;
+
+	damon_for_each_task(t) {
+		task = damon_get_task_struct(t);
+		if (task) {
+			put_task_struct(task);
+			return false;
+		}
+	}
+
+	return true;
+}
+
+/*
+ * The monitoring daemon that runs as a kernel thread
+ */
+static int kdamond_fn(void *data)
+{
+	struct damon_task *t;
+	struct damon_region *r, *next;
+	struct mm_struct *mm;
+
+	pr_info("kdamond (%d) starts\n", kdamond->pid);
+	kdamond_init_regions();
+	while (!kdamond_need_stop()) {
+		damon_for_each_task(t) {
+			mm = damon_get_mm(t);
+			if (!mm)
+				continue;
+			damon_for_each_region(r, t)
+				kdamond_check_access(mm, r);
+			mmput(mm);
+		}
+
+		if (kdamond_aggregate_interval_passed())
+			kdamond_flush_aggregated();
+
+		usleep_range(sample_interval, sample_interval + 1);
+	}
+	damon_flush_rbuffer();
+	damon_for_each_task(t) {
+		damon_for_each_region_safe(r, next, t)
+			damon_destroy_region(r);
+	}
+	pr_info("kdamond (%d) finishes\n", kdamond->pid);
+	spin_lock(&kdamond_lock);
+	kdamond = NULL;
+	spin_unlock(&kdamond_lock);
+	return 0;
+}
+
+/*
+ * Controller functions
+ */
+
+/*
+ * Start or stop the kdamond
+ *
+ * Returns 0 if success, negative error code otherwise.
+ */
+static int damon_turn_kdamond(bool on)
+{
+	spin_lock(&kdamond_lock);
+	kdamond_stop = !on;
+	if (!kdamond && on) {
+		kdamond = kthread_run(kdamond_fn, NULL, "kdamond");
+		if (!kdamond)
+			goto fail;
+		goto success;
+	}
+	if (kdamond && !on) {
+		spin_unlock(&kdamond_lock);
+		while (true) {
+			spin_lock(&kdamond_lock);
+			if (!kdamond)
+				goto success;
+			spin_unlock(&kdamond_lock);
+
+			usleep_range(sample_interval, sample_interval * 2);
+		}
+	}
+
+	/* tried to turn on while turned on, or turn off while turned off */
+
+fail:
+	spin_unlock(&kdamond_lock);
+	return -EINVAL;
+
+success:
+	spin_unlock(&kdamond_lock);
+	return 0;
+}
+
+static inline bool damon_is_target_pid(unsigned long pid)
+{
+	struct damon_task *t;
+
+	damon_for_each_task(t) {
+		if (t->pid == pid)
+			return true;
+	}
+	return false;
+}
+
+/*
+ * This function should not be called while the kdamond is running.
+ */
+static long damon_set_pids(unsigned long *pids, ssize_t nr_pids)
+{
+	ssize_t i;
+	struct damon_task *t, *next;
+
+	/* Remove unselected tasks */
+	damon_for_each_task_safe(t, next) {
+		for (i = 0; i < nr_pids; i++) {
+			if (pids[i] == t->pid)
+				break;
+		}
+		if (i != nr_pids)
+			continue;
+		damon_destroy_task(t);
+	}
+
+	/* Add new tasks */
+	for (i = 0; i < nr_pids; i++) {
+		if (damon_is_target_pid(pids[i]))
+			continue;
+		t = damon_new_task(pids[i]);
+		if (!t) {
+			pr_err("Failed to alloc damon_task\n");
+			return -ENOMEM;
+		}
+		damon_add_task_tail(t);
+	}
+
+	return 0;
+}
+
+/*
+ * Set attributes for the monitoring
+ *
+ * sample_int		time interval between samplings
+ * aggr_int		time interval between aggregations
+ * min_nr_reg		minimal number of regions
+ * path_to_rfile	path to the monitor result files
+ *
+ * This function should not be called while the kdamond is running.
+ * Every time interval is in micro-seconds.
+ *
+ * Returns 0 on success, negative error code otherwise.
+ */
+static long damon_set_attrs(unsigned long sample_int, unsigned long aggr_int,
+		unsigned long min_nr_reg, char *path_to_rfile)
+{
+	if (strnlen(path_to_rfile, LEN_RES_FILE_PATH) >= LEN_RES_FILE_PATH) {
+		pr_err("too long (>%d) result file path %s\n",
+				LEN_RES_FILE_PATH, path_to_rfile);
+		return -EINVAL;
+	}
+	if (min_nr_reg < 3) {
+		pr_err("min_nr_regions (%lu) should be bigger than 2\n",
+				min_nr_reg);
+		return -EINVAL;
+	}
+
+	sample_interval = sample_int;
+	aggr_interval = aggr_int;
+	min_nr_regions = min_nr_reg;
+	strncpy(rfile_path, path_to_rfile, LEN_RES_FILE_PATH);
+	return 0;
+}
+
 static int __init damon_init(void)
 {
 	pr_info("init\n");
 
 	prandom_seed_state(&rndseed, 42);
+	ktime_get_coarse_ts64(&last_aggregate_time);
 	return 0;
 }
 
 static void __exit damon_exit(void)
 {
+	damon_turn_kdamond(false);
 	pr_info("exit\n");
 }
 
-- 
2.17.1

[PATCH 3/8] mm/damon: Adaptively adjust regions

[SeongJae Park]

From: SeongJae Park <sjpark@amazon.de>

At the beginning of the monitoring, DAMON constructs the initial regions
by evenly splitting the memory mapped address space of the process into
the user-specified minimal number of regions.  In this initial state,
the assumption of the regions (pages in same region have similar access
frequencies) is normally not kept and thus the monitoring quality could
be low.  To keep the assumption as much as possible, DAMON adaptively
merges and splits each region.

For each ``aggregation interval``, it compares the access frequencies of
adjacent regions and merges those if the frequency difference is small.
Then, after it reports and clears the aggregated access frequency of
each region, it splits each region into two regions if the total number
of regions is smaller than the half of the user-specified maximum number
of regions.

In this way, DAMON provides its best-effort quality and minimal overhead
while keeping the bounds users set for their trade-off.

Signed-off-by: SeongJae Park <sjpark@amazon.de>
---
 mm/damon.c | 142 +++++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 138 insertions(+), 4 deletions(-)

diff --git a/mm/damon.c b/mm/damon.c
index 2a0c010291f8..5012d8c25b8b 100644
--- a/mm/damon.c
+++ b/mm/damon.c
@@ -73,6 +73,7 @@ static unsigned long aggr_interval = 100 * 1000;
 static struct timespec64 last_aggregate_time;
 
 static unsigned long min_nr_regions = 10;
+static unsigned long max_nr_regions = 1000;
 
 /* result buffer */
 #define DAMON_LEN_RBUF	(1024 * 1024 * 4)
@@ -609,6 +610,122 @@ static void kdamond_flush_aggregated(void)
 	}
 }
 
+#define sz_damon_region(r) (r->vm_end - r->vm_start)
+
+/*
+ * Merge two adjacent regions into one region
+ */
+static void damon_merge_two_regions(struct damon_region *l,
+				struct damon_region *r)
+{
+	l->nr_accesses = (l->nr_accesses * sz_damon_region(l) +
+			r->nr_accesses * sz_damon_region(r)) /
+			(sz_damon_region(l) + sz_damon_region(r));
+	l->vm_end = r->vm_end;
+	damon_destroy_region(r);
+}
+
+#define diff_of(a, b) (a > b ? a - b : b - a)
+
+/*
+ * Merge adjacent regions having similar access frequencies
+ *
+ * t		task that merge operation will make change
+ * thres	merge regions having '->nr_accesses' diff smaller than this
+ */
+static void damon_merge_regions_of(struct damon_task *t, unsigned int thres)
+{
+	struct damon_region *r, *prev = NULL, *next;
+
+	damon_for_each_region_safe(r, next, t) {
+		if (!prev || prev->vm_end != r->vm_start)
+			goto next;
+		if (diff_of(prev->nr_accesses, r->nr_accesses) > thres)
+			goto next;
+		damon_merge_two_regions(prev, r);
+		continue;
+next:
+		prev = r;
+	}
+}
+
+/*
+ * Merge adjacent regions having similar access frequencies
+ *
+ * threshold	merge regions havind nr_accesses diff larger than this
+ *
+ * This function merges monitoring target regions which are adjacent and their
+ * access frequencies are similar.  This is for minimizing the monitoring
+ * overhead under the dynamically changeable access pattern.  If a merge was
+ * unnecessarily made, later 'kdamond_split_regions()' will revert it.
+ */
+static void kdamond_merge_regions(unsigned int threshold)
+{
+	struct damon_task *t;
+
+	damon_for_each_task(t)
+		damon_merge_regions_of(t, threshold);
+}
+
+/*
+ * Split a region into two small regions
+ *
+ * r		the region to be split
+ * sz_r		size of the first sub-region that will be made
+ */
+static void damon_split_region_at(struct damon_region *r, unsigned long sz_r)
+{
+	struct damon_region *new;
+
+	new = damon_new_region(r->vm_start + sz_r, r->vm_end);
+	r->vm_end = new->vm_start;
+
+	damon_add_region(new, r, damon_next_region(r));
+}
+
+static void damon_split_regions_of(struct damon_task *t)
+{
+	struct damon_region *r, *next;
+	unsigned long sz_left_region;
+
+	damon_for_each_region_safe(r, next, t) {
+		/*
+		 * Randomly select size of left sub-region to be at least
+		 * 10 percent and at most 90% of original region
+		 */
+		sz_left_region = (prandom_u32_state(&rndseed) % 9 + 1) *
+			(r->vm_end - r->vm_start) / 10;
+		/* Do not allow blank region */
+		if (sz_left_region == 0)
+			continue;
+		damon_split_region_at(r, sz_left_region);
+	}
+}
+
+/*
+ * splits every target regions into two randomly-sized regions
+ *
+ * This function splits every target regions into two random-sized regions if
+ * current total number of the regions is smaller than the half of the
+ * user-specified maximum number of regions.  This is for maximizing the
+ * monitoring accuracy under the dynamically changeable access patterns.  If a
+ * split was unnecessarily made, later 'kdamond_merge_regions()' will revert
+ * it.
+ */
+static void kdamond_split_regions(void)
+{
+	struct damon_task *t;
+	unsigned int nr_regions = 0;
+
+	damon_for_each_task(t)
+		nr_regions += nr_damon_regions(t);
+	if (nr_regions > max_nr_regions / 2)
+		return;
+
+	damon_for_each_task(t)
+		damon_split_regions_of(t);
+}
+
 /*
  * Check whether current monitoring should be stopped
  *
@@ -648,21 +765,29 @@ static int kdamond_fn(void *data)
 	struct damon_task *t;
 	struct damon_region *r, *next;
 	struct mm_struct *mm;
+	unsigned long max_nr_accesses;
 
 	pr_info("kdamond (%d) starts\n", kdamond->pid);
 	kdamond_init_regions();
 	while (!kdamond_need_stop()) {
+		max_nr_accesses = 0;
 		damon_for_each_task(t) {
 			mm = damon_get_mm(t);
 			if (!mm)
 				continue;
-			damon_for_each_region(r, t)
+			damon_for_each_region(r, t) {
 				kdamond_check_access(mm, r);
+				if (r->nr_accesses > max_nr_accesses)
+					max_nr_accesses = r->nr_accesses;
+			}
 			mmput(mm);
 		}
 
-		if (kdamond_aggregate_interval_passed())
+		if (kdamond_aggregate_interval_passed()) {
+			kdamond_merge_regions(max_nr_accesses / 10);
 			kdamond_flush_aggregated();
+			kdamond_split_regions();
+		}
 
 		usleep_range(sample_interval, sample_interval + 1);
 	}
@@ -771,6 +896,7 @@ static long damon_set_pids(unsigned long *pids, ssize_t nr_pids)
  * sample_int		time interval between samplings
  * aggr_int		time interval between aggregations
  * min_nr_reg		minimal number of regions
+ * max_nr_reg		maximum number of regions
  * path_to_rfile	path to the monitor result files
  *
  * This function should not be called while the kdamond is running.
@@ -778,8 +904,10 @@ static long damon_set_pids(unsigned long *pids, ssize_t nr_pids)
  *
  * Returns 0 on success, negative error code otherwise.
  */
-static long damon_set_attrs(unsigned long sample_int, unsigned long aggr_int,
-		unsigned long min_nr_reg, char *path_to_rfile)
+static long damon_set_attrs(unsigned long sample_int,
+		unsigned long aggr_int,
+		unsigned long min_nr_reg, unsigned long max_nr_reg,
+		char *path_to_rfile)
 {
 	if (strnlen(path_to_rfile, LEN_RES_FILE_PATH) >= LEN_RES_FILE_PATH) {
 		pr_err("too long (>%d) result file path %s\n",
@@ -791,10 +919,16 @@ static long damon_set_attrs(unsigned long sample_int, unsigned long aggr_int,
 				min_nr_reg);
 		return -EINVAL;
 	}
+	if (min_nr_reg >= max_nr_regions) {
+		pr_err("invalid nr_regions.  min (%lu) >= max (%lu)\n",
+				min_nr_reg, max_nr_reg);
+		return -EINVAL;
+	}
 
 	sample_interval = sample_int;
 	aggr_interval = aggr_int;
 	min_nr_regions = min_nr_reg;
+	max_nr_regions = max_nr_reg;
 	strncpy(rfile_path, path_to_rfile, LEN_RES_FILE_PATH);
 	return 0;
 }
-- 
2.17.1

[PATCH 4/8] mm/damon: Apply dynamic memory mapping changes

[SeongJae Park]

From: SeongJae Park <sjpark@amazon.de>

Only a number of parts in the virtual address space of the processes is
mapped to physical memory and accessed.  Thus, tracking the unmapped
address regions is just wasteful.  However, tracking every memory
mapping change might incur an overhead.  For the reason, DAMON applies
the dynamic memory mapping changes to the tracking regions only for each
of a user-specified time interval (``regions update interval``).

Signed-off-by: SeongJae Park <sjpark@amazon.de>
---
 mm/damon.c | 95 ++++++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 93 insertions(+), 2 deletions(-)

diff --git a/mm/damon.c b/mm/damon.c
index 5012d8c25b8b..9252a2ba825d 100644
--- a/mm/damon.c
+++ b/mm/damon.c
@@ -63,14 +63,18 @@ static LIST_HEAD(damon_tasks_list);
  * For each 'sample_interval', DAMON checks whether each region is accessed or
  * not.  It aggregates and keeps the access information (number of accesses to
  * each region) for 'aggr_interval' and then flushes it to the result buffer if
- * an 'aggr_interval' surpassed.
+ * an 'aggr_interval' surpassed.  And for each 'regions_update_interval', damon
+ * checks whether the memory mapping of the target tasks has changed (e.g., by
+ * mmap() calls from the applications) and applies the changes.
  *
  * All time intervals are in micro-seconds.
  */
 static unsigned long sample_interval = 5 * 1000;
 static unsigned long aggr_interval = 100 * 1000;
+static unsigned long regions_update_interval = 1000 * 1000;
 
 static struct timespec64 last_aggregate_time;
+static struct timespec64 last_regions_update_time;
 
 static unsigned long min_nr_regions = 10;
 static unsigned long max_nr_regions = 1000;
@@ -726,6 +730,87 @@ static void kdamond_split_regions(void)
 		damon_split_regions_of(t);
 }
 
+/*
+ * Check whether it is time to check and apply the dynamic mmap changes
+ *
+ * Returns true if it is.
+ */
+static bool kdamond_need_update_regions(void)
+{
+	return damon_check_reset_time_interval(&last_regions_update_time,
+			regions_update_interval);
+}
+
+static bool damon_intersect(struct damon_region *r, struct region *re)
+{
+	return !(r->vm_end <= re->start || re->end <= r->vm_start);
+}
+
+/*
+ * Update damon regions for the three big regions of the given task
+ *
+ * t		the given task
+ * bregions	the three big regions of the task
+ */
+static void damon_apply_three_regions(struct damon_task *t,
+				struct region bregions[3])
+{
+	struct damon_region *r, *next;
+	unsigned int i = 0;
+
+	/* Remove regions which isn't in the three big regions now */
+	damon_for_each_region_safe(r, next, t) {
+		for (i = 0; i < 3; i++) {
+			if (damon_intersect(r, &bregions[i]))
+				break;
+		}
+		if (i == 3)
+			damon_destroy_region(r);
+	}
+
+	/* Adjust intersecting regions to fit with the threee big regions */
+	for (i = 0; i < 3; i++) {
+		struct damon_region *first = NULL, *last;
+		struct damon_region *newr;
+		struct region *br;
+
+		br = &bregions[i];
+		/* Get the first and last regions which intersects with br */
+		damon_for_each_region(r, t) {
+			if (damon_intersect(r, br)) {
+				if (!first)
+					first = r;
+				last = r;
+			}
+			if (r->vm_start >= br->end)
+				break;
+		}
+		if (!first) {
+			/* no damon_region intersects with this big region */
+			newr = damon_new_region(br->start, br->end);
+			damon_add_region(newr, damon_prev_region(r), r);
+		} else {
+			first->vm_start = br->start;
+			last->vm_end = br->end;
+		}
+	}
+}
+
+/*
+ * Update regions for current memory mappings
+ */
+static void kdamond_update_regions(void)
+{
+	struct region three_regions[3];
+	struct damon_task *t;
+
+	damon_for_each_task(t) {
+		if (damon_three_regions_of(t, three_regions))
+			continue;
+		damon_apply_three_regions(t, three_regions);
+	}
+}
+
 /*
  * Check whether current monitoring should be stopped
  *
@@ -789,6 +874,9 @@ static int kdamond_fn(void *data)
 			kdamond_split_regions();
 		}
 
+		if (kdamond_need_update_regions())
+			kdamond_update_regions();
+
 		usleep_range(sample_interval, sample_interval + 1);
 	}
 	damon_flush_rbuffer();
@@ -895,6 +983,7 @@ static long damon_set_pids(unsigned long *pids, ssize_t nr_pids)
  *
  * sample_int		time interval between samplings
  * aggr_int		time interval between aggregations
+ * regions_update_int	time interval between vma update checks
  * min_nr_reg		minimal number of regions
  * max_nr_reg		maximum number of regions
  * path_to_rfile	path to the monitor result files
@@ -905,7 +994,7 @@ static long damon_set_pids(unsigned long *pids, ssize_t nr_pids)
  * Returns 0 on success, negative error code otherwise.
  */
 static long damon_set_attrs(unsigned long sample_int,
-		unsigned long aggr_int,
+		unsigned long aggr_int, unsigned long regions_update_int,
 		unsigned long min_nr_reg, unsigned long max_nr_reg,
 		char *path_to_rfile)
 {
@@ -927,6 +1016,7 @@ static long damon_set_attrs(unsigned long sample_int,
 
 	sample_interval = sample_int;
 	aggr_interval = aggr_int;
+	regions_update_interval = regions_update_int;
 	min_nr_regions = min_nr_reg;
 	max_nr_regions = max_nr_reg;
 	strncpy(rfile_path, path_to_rfile, LEN_RES_FILE_PATH);
@@ -939,6 +1029,7 @@ static int __init damon_init(void)
 
 	prandom_seed_state(&rndseed, 42);
 	ktime_get_coarse_ts64(&last_aggregate_time);
+	last_regions_update_time = last_aggregate_time;
 	return 0;
 }
 
-- 
2.17.1

[PATCH 5/8] mm/damon: Add debugfs interface

[SeongJae Park]

From: SeongJae Park <sjpark@amazon.de>

This commit adds a debugfs interface for DAMON.

DAMON exports three files, ``attrs``, ``pids``, and ``monitor_on`` under
its debugfs directory, ``<debugfs>/damon/``.

Attributes
----------

Users can read and write the ``sampling interval``, ``aggregation
interval``, ``regions update interval``, min/max number of regions, and
the path to ``result file`` by reading from and writing to the ``attrs``
file.  For example, below commands set those values to 5 ms, 100 ms,
1,000 ms, 10, 1000, and ``/damon.data`` and check it again::

    # cd <debugfs>/damon
    # echo 5000 100000 1000000 10 1000 /damon.data > attrs
    # cat attrs
    5000 100000 1000000 10 1000 /damon.data

Target PIDs
-----------

Users can read and write the pids of current monitoring target processes
by reading from and writing to the `pids` file.  For example, below
commands set processes having pids 42 and 4242 as the processes to be
monitored and check it again::

    # cd <debugfs>/damon
    # echo 42 4242 > pids
    # cat pids
    42 4242

Note that setting the pids doesn't starts the monitoring.

Turning On/Off
--------------

You can check current status, start and stop the monitoring by reading
from and writing to the ``monitor_on`` file.  Writing ``on`` to the file
starts DAMON to monitor the target processes with the attributes.
Writing ``off`` to the file stops DAMON.  DAMON also stops if every
target processes is be terminated.  Below example commands turn on, off,
and check status of DAMON::

    # cd <debugfs>/damon
    # echo on > monitor_on
    # echo off > monitor_on
    # cat monitor_on
    off

Please note that you cannot write to the ``attrs`` and ``pids`` files
while the monitoring is turned on.  If you write to the files while
DAMON is running, ``-EINVAL`` will be returned.

Signed-off-by: SeongJae Park <sjpark@amazon.de>
---
 mm/damon.c | 232 ++++++++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 231 insertions(+), 1 deletion(-)

diff --git a/mm/damon.c b/mm/damon.c
index 9252a2ba825d..e2492344b58c 100644
--- a/mm/damon.c
+++ b/mm/damon.c
@@ -9,6 +9,7 @@
 
 #define pr_fmt(fmt) "damon: " fmt
 
+#include <linux/debugfs.h>
 #include <linux/delay.h>
 #include <linux/kthread.h>
 #include <linux/mm.h>
@@ -1023,6 +1024,233 @@ static long damon_set_attrs(unsigned long sample_int,
 	return 0;
 }
 
+/*
+ * debugfs functions
+ */
+
+static ssize_t debugfs_monitor_on_read(struct file *file,
+		char __user *buf, size_t count, loff_t *ppos)
+{
+	char monitor_on_buf[5];
+	bool monitor_on;
+
+	spin_lock(&kdamond_lock);
+	monitor_on = kdamond != NULL;
+	spin_unlock(&kdamond_lock);
+
+	snprintf(monitor_on_buf, 5, monitor_on ? "on\n" : "off\n");
+
+	return simple_read_from_buffer(buf, count, ppos, monitor_on_buf,
+			monitor_on ? 3 : 4);
+}
+
+static ssize_t debugfs_monitor_on_write(struct file *file,
+		const char __user *buf, size_t count, loff_t *ppos)
+{
+	ssize_t ret;
+	bool on = false;
+	char cmdbuf[5];
+
+	ret = simple_write_to_buffer(cmdbuf, 5, ppos, buf, count);
+	if (ret < 0)
+		return ret;
+
+	if (sscanf(cmdbuf, "%s", cmdbuf) != 1)
+		return -EINVAL;
+	if (!strncmp(cmdbuf, "on", 5))
+		on = true;
+	else if (!strncmp(cmdbuf, "off", 5))
+		on = false;
+	else
+		return -EINVAL;
+
+	if (damon_turn_kdamond(on))
+		return -EINVAL;
+
+	return ret;
+}
+
+static ssize_t damon_sprint_pids(char *buf, ssize_t len)
+{
+	char *cursor = buf;
+	struct damon_task *t;
+
+	damon_for_each_task(t) {
+		snprintf(cursor, len, "%lu ", t->pid);
+		cursor += strnlen(cursor, len);
+	}
+	if (cursor != buf)
+		cursor--;
+	snprintf(cursor, len, "\n");
+	return strnlen(buf, len);
+}
+
+static ssize_t debugfs_pids_read(struct file *file,
+		char __user *buf, size_t count, loff_t *ppos)
+{
+	ssize_t len;
+	char pids_buf[512];
+
+	len = damon_sprint_pids(pids_buf, 512);
+
+	return simple_read_from_buffer(buf, count, ppos, pids_buf, len);
+}
+
+/*
+ * Converts a string into an array of unsigned long integers
+ *
+ * Returns an array of unsigned long integers that converted, or NULL if the
+ * input is wrong.
+ */
+static unsigned long *str_to_pids(const char *str, ssize_t len,
+				ssize_t *nr_pids)
+{
+	unsigned long *pids;
+	unsigned long pid;
+	int pos = 0, parsed, ret;
+
+	*nr_pids = 0;
+	pids = kmalloc_array(256, sizeof(unsigned long), GFP_KERNEL);
+	while (*nr_pids < 256 && pos < len) {
+		ret = sscanf(&str[pos], "%lu%n", &pid, &parsed);
+		pos += parsed;
+		if (ret != 1)
+			break;
+		pids[*nr_pids] = pid;
+		*nr_pids += 1;
+	}
+	if (*nr_pids == 0) {
+		kfree(pids);
+		pids = NULL;
+	}
+
+	return pids;
+}
+
+static ssize_t debugfs_pids_write(struct file *file,
+		const char __user *buf, size_t count, loff_t *ppos)
+{
+	ssize_t ret;
+	unsigned long *targets;
+	ssize_t nr_targets;
+	char pids_buf[512];
+
+	ret = simple_write_to_buffer(pids_buf, 512, ppos, buf, count);
+	if (ret < 0)
+		return ret;
+
+	targets = str_to_pids(pids_buf, ret, &nr_targets);
+
+	spin_lock(&kdamond_lock);
+	if (kdamond)
+		goto monitor_running;
+
+	damon_set_pids(targets, nr_targets);
+	spin_unlock(&kdamond_lock);
+	kfree(targets);
+
+	return ret;
+
+monitor_running:
+	spin_unlock(&kdamond_lock);
+	pr_err("%s: kdamond is running. Turn it off first.\n", __func__);
+	return -EINVAL;
+}
+
+static ssize_t debugfs_attrs_read(struct file *file,
+		char __user *buf, size_t count, loff_t *ppos)
+{
+	char attrs_buf[512];
+
+	snprintf(attrs_buf, 512, "%lu %lu %lu %lu %lu %s\n",
+			sample_interval, aggr_interval,
+			regions_update_interval, min_nr_regions,
+			max_nr_regions, rfile_path);
+
+	return simple_read_from_buffer(buf, count, ppos, attrs_buf,
+			strnlen(attrs_buf, 512));
+}
+
+static ssize_t debugfs_attrs_write(struct file *file,
+		const char __user *buf, size_t count, loff_t *ppos)
+{
+	unsigned long s, a, r, minr, maxr;
+	char attrs_buf[512];
+	char res_file_path[LEN_RES_FILE_PATH];
+	ssize_t ret;
+
+	if (count > 512) {
+		pr_err("attributes stream is too large: %s\n", buf);
+		return -ENOMEM;
+	}
+
+	ret = simple_write_to_buffer(attrs_buf, 512, ppos, buf, count);
+	if (ret < 0)
+		return ret;
+
+	if (sscanf(attrs_buf, "%lu %lu %lu %lu %lu %s",
+				&s, &a, &r, &minr, &maxr, res_file_path) != 6)
+		return -EINVAL;
+
+	spin_lock(&kdamond_lock);
+	if (kdamond)
+		goto monitor_running;
+
+	damon_set_attrs(s, a, r, minr, maxr, res_file_path);
+	spin_unlock(&kdamond_lock);
+
+	return ret;
+
+monitor_running:
+	spin_unlock(&kdamond_lock);
+	pr_err("%s: kdamond is running. Turn it off first.\n", __func__);
+	return -EINVAL;
+}
+
+static const struct file_operations monitor_on_fops = {
+	.owner = THIS_MODULE,
+	.read = debugfs_monitor_on_read,
+	.write = debugfs_monitor_on_write,
+};
+
+static const struct file_operations pids_fops = {
+	.owner = THIS_MODULE,
+	.read = debugfs_pids_read,
+	.write = debugfs_pids_write,
+};
+
+static const struct file_operations attrs_fops = {
+	.owner = THIS_MODULE,
+	.read = debugfs_attrs_read,
+	.write = debugfs_attrs_write,
+};
+
+static struct dentry *debugfs_root;
+
+static int __init debugfs_init(void)
+{
+	const char * const file_names[] = {"attrs", "pids", "monitor_on"};
+	const struct file_operations *fops[] = {&attrs_fops, &pids_fops,
+		&monitor_on_fops};
+	int i;
+
+	debugfs_root = debugfs_create_dir("damon", NULL);
+	if (!debugfs_root) {
+		pr_err("failed to create the debugfs dir\n");
+		return -ENOMEM;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(file_names); i++) {
+		if (!debugfs_create_file(file_names[i], 0600, debugfs_root,
+					NULL, fops[i])) {
+			pr_err("failed to create %s file\n", file_names[i]);
+			return -ENOMEM;
+		}
+	}
+
+	return 0;
+}
+
 static int __init damon_init(void)
 {
 	pr_info("init\n");
@@ -1030,12 +1258,14 @@ static int __init damon_init(void)
 	prandom_seed_state(&rndseed, 42);
 	ktime_get_coarse_ts64(&last_aggregate_time);
 	last_regions_update_time = last_aggregate_time;
-	return 0;
+
+	return debugfs_init();
 }
 
 static void __exit damon_exit(void)
 {
 	damon_turn_kdamond(false);
+	debugfs_remove_recursive(debugfs_root);
 	pr_info("exit\n");
 }
 
-- 
2.17.1

[PATCH 6/8] mm/damon: Add minimal user-space tools

[SeongJae Park]

From: SeongJae Park <sjpark@amazon.de>

This commit adds a shallow wrapper python script, ``/tools/damon/damo``
that provides more convenient interface.  Note that it is only aimed to
be used for minimal reference of the DAMON's raw interfaces and for
debugging of the DAMON itself.  Based on the debugfs interface, you can
create another cool and more convenient user space tools.

Quick Tutorial
--------------

1. Ensure your kernel is built with CONFIG_DAMON turned on, and debugfs
   is mounted at ``/sys/kernel/debug/``.
2. ``<your kernel source tree>/tools/damon/damo -h``

Signed-off-by: SeongJae Park <sjpark@amazon.de>
---
 MAINTAINERS               |   1 +
 tools/damon/.gitignore    |   1 +
 tools/damon/_dist.py      |  35 ++++
 tools/damon/bin2txt.py    |  64 +++++++
 tools/damon/damo          |  37 ++++
 tools/damon/heats.py      | 358 ++++++++++++++++++++++++++++++++++++++
 tools/damon/nr_regions.py |  88 ++++++++++
 tools/damon/record.py     | 193 ++++++++++++++++++++
 tools/damon/report.py     |  45 +++++
 tools/damon/wss.py        |  94 ++++++++++
 10 files changed, 916 insertions(+)
 create mode 100644 tools/damon/.gitignore
 create mode 100644 tools/damon/_dist.py
 create mode 100644 tools/damon/bin2txt.py
 create mode 100755 tools/damon/damo
 create mode 100644 tools/damon/heats.py
 create mode 100644 tools/damon/nr_regions.py
 create mode 100644 tools/damon/record.py
 create mode 100644 tools/damon/report.py
 create mode 100644 tools/damon/wss.py

diff --git a/MAINTAINERS b/MAINTAINERS
index c8db500a182f..fb41236f3aed 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4588,6 +4588,7 @@ M:	SeongJae Park <sjpark@amazon.de>
 L:	linux-mm@kvack.org
 S:	Maintained
 F:	mm/damon.c
+F:	tools/damon/*
 
 DAVICOM FAST ETHERNET (DMFE) NETWORK DRIVER
 L:	netdev@vger.kernel.org
diff --git a/tools/damon/.gitignore b/tools/damon/.gitignore
new file mode 100644
index 000000000000..96403d36ff93
--- /dev/null
+++ b/tools/damon/.gitignore
@@ -0,0 +1 @@
+__pycache__/*
diff --git a/tools/damon/_dist.py b/tools/damon/_dist.py
new file mode 100644
index 000000000000..f26409cf9232
--- /dev/null
+++ b/tools/damon/_dist.py
@@ -0,0 +1,35 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0
+
+import os
+import struct
+import subprocess
+
+def access_patterns(f):
+    nr_regions = struct.unpack('I', f.read(4))[0]
+
+    patterns = []
+    for r in range(nr_regions):
+        saddr = struct.unpack('L', f.read(8))[0]
+        eaddr = struct.unpack('L', f.read(8))[0]
+        nr_accesses = struct.unpack('I', f.read(4))[0]
+        patterns.append([eaddr - saddr, nr_accesses])
+    return patterns
+
+def plot_dist(data_file, output_file, xlabel):
+    terminal = output_file.split('.')[-1]
+    if not terminal in ['pdf', 'jpeg', 'png', 'svg']:
+        os.remove(data_file)
+        print("Unsupported plot output type.")
+        exit(-1)
+
+    gnuplot_cmd = """
+    set term %s;
+    set output '%s';
+    set key off;
+    set ylabel 'working set size (bytes)';
+    set xlabel '%s';
+    plot '%s' with linespoints;""" % (terminal, output_file, xlabel, data_file)
+    subprocess.call(['gnuplot', '-e', gnuplot_cmd])
+    os.remove(data_file)
+
diff --git a/tools/damon/bin2txt.py b/tools/damon/bin2txt.py
new file mode 100644
index 000000000000..d5ffac60e02c
--- /dev/null
+++ b/tools/damon/bin2txt.py
@@ -0,0 +1,64 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0
+
+import argparse
+import os
+import struct
+import sys
+
+def parse_time(bindat):
+    "bindat should be 16 bytes"
+    sec = struct.unpack('l', bindat[0:8])[0]
+    nsec = struct.unpack('l', bindat[8:16])[0]
+    return sec * 1000000000 + nsec;
+
+def pr_region(f):
+    saddr = struct.unpack('L', f.read(8))[0]
+    eaddr = struct.unpack('L', f.read(8))[0]
+    nr_accesses = struct.unpack('I', f.read(4))[0]
+    print("%012x-%012x(%10d):\t%d" %
+            (saddr, eaddr, eaddr - saddr, nr_accesses))
+
+def pr_task_info(f):
+    pid = struct.unpack('L', f.read(8))[0]
+    print("pid: ", pid)
+    nr_regions = struct.unpack('I', f.read(4))[0]
+    print("nr_regions: ", nr_regions)
+    for r in range(nr_regions):
+        pr_region(f)
+
+def set_argparser(parser):
+    parser.add_argument('--input', '-i', type=str, metavar='<file>',
+            default='damon.data', help='input file name')
+
+def main(args=None):
+    if not args:
+        parser = argparse.ArgumentParser()
+        set_argparser(parser)
+        args = parser.parse_args()
+
+    file_path = args.input
+
+    if not os.path.isfile(file_path):
+        print('input file (%s) is not exist' % file_path)
+        exit(1)
+
+    with open(file_path, 'rb') as f:
+        start_time = None
+        while True:
+            timebin = f.read(16)
+            if len(timebin) != 16:
+                break
+            time = parse_time(timebin)
+            if not start_time:
+                start_time = time
+                print("start_time: ", start_time)
+            print("rel time: %16d" % (time - start_time))
+            nr_tasks = struct.unpack('I', f.read(4))[0]
+            print("nr_tasks: ", nr_tasks)
+            for t in range(nr_tasks):
+                pr_task_info(f)
+                print("")
+
+if __name__ == '__main__':
+    main()
diff --git a/tools/damon/damo b/tools/damon/damo
new file mode 100755
index 000000000000..58e1099ae5fc
--- /dev/null
+++ b/tools/damon/damo
@@ -0,0 +1,37 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0
+
+import argparse
+
+import record
+import report
+
+class SubCmdHelpFormatter(argparse.RawDescriptionHelpFormatter):
+    def _format_action(self, action):
+        parts = super(argparse.RawDescriptionHelpFormatter,
+                self)._format_action(action)
+        # skip sub parsers help
+        if action.nargs == argparse.PARSER:
+            parts = '\n'.join(parts.split('\n')[1:])
+        return parts
+
+parser = argparse.ArgumentParser(formatter_class=SubCmdHelpFormatter)
+
+subparser = parser.add_subparsers(title='command', dest='command',
+        metavar='<command>')
+subparser.required = True
+
+parser_record = subparser.add_parser('record',
+        help='record data accesses of the given target processes')
+record.set_argparser(parser_record)
+
+parser_report = subparser.add_parser('report',
+        help='report the recorded data accesses in the specified form')
+report.set_argparser(parser_report)
+
+args = parser.parse_args()
+
+if args.command == 'record':
+    record.main(args)
+elif args.command == 'report':
+    report.main(args)
diff --git a/tools/damon/heats.py b/tools/damon/heats.py
new file mode 100644
index 000000000000..48e966c5ca02
--- /dev/null
+++ b/tools/damon/heats.py
@@ -0,0 +1,358 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0
+
+"""
+Transform binary trace data into human readable text that can be used for
+heatmap drawing, or directly plot the data in a heatmap format.
+
+Format of the text is:
+
+    <time> <space> <heat>
+    ...
+
+"""
+
+import argparse
+import os
+import struct
+import subprocess
+import sys
+import tempfile
+
+class HeatSample:
+    space_idx = None
+    sz_time_space = None
+    heat = None
+
+    def __init__(self, space_idx, sz_time_space, heat):
+        if sz_time_space < 0:
+            raise RuntimeError()
+        self.space_idx = space_idx
+        self.sz_time_space = sz_time_space
+        self.heat = heat
+
+    def total_heat(self):
+        return self.heat * self.sz_time_space
+
+    def merge(self, sample):
+        "sample must have a space idx that same to self"
+        heat_sum = self.total_heat() + sample.total_heat()
+        self.heat = heat_sum / (self.sz_time_space + sample.sz_time_space)
+        self.sz_time_space += sample.sz_time_space
+
+def pr_samples(samples, time_idx, time_unit, region_unit):
+    display_time = time_idx * time_unit
+    for idx, sample in enumerate(samples):
+        display_addr = idx * region_unit
+        if not sample:
+            print("%s\t%s\t%s" % (display_time, display_addr, 0.0))
+            continue
+        print("%s\t%s\t%s" % (display_time, display_addr, sample.total_heat() /
+            time_unit / region_unit))
+
+def to_idx(value, min_, unit):
+    return (value - min_) // unit
+
+def read_task_heats(f, pid, aunit, amin, amax):
+    pid_ = struct.unpack('L', f.read(8))[0]
+    nr_regions = struct.unpack('I', f.read(4))[0]
+    if pid_ != pid:
+        f.read(20 * nr_regions)
+        return None
+    samples = []
+    for i in range(nr_regions):
+        saddr = struct.unpack('L', f.read(8))[0]
+        eaddr = struct.unpack('L', f.read(8))[0]
+        eaddr = min(eaddr, amax - 1)
+        heat = struct.unpack('I', f.read(4))[0]
+
+        if eaddr <= amin:
+            continue
+        if saddr >= amax:
+            continue
+        saddr = max(amin, saddr)
+        eaddr = min(amax, eaddr)
+
+        sidx = to_idx(saddr, amin, aunit)
+        eidx = to_idx(eaddr - 1, amin, aunit)
+        for idx in range(sidx, eidx + 1):
+            sa = max(amin + idx * aunit, saddr)
+            ea = min(amin + (idx + 1) * aunit, eaddr)
+            sample = HeatSample(idx, (ea - sa), heat)
+            samples.append(sample)
+    return samples
+
+def parse_time(bindat):
+    sec = struct.unpack('l', bindat[0:8])[0]
+    nsec = struct.unpack('l', bindat[8:16])[0]
+    return sec * 1000000000 + nsec
+
+def apply_samples(target_samples, samples, start_time, end_time, aunit, amin):
+    for s in samples:
+        sample = HeatSample(s.space_idx,
+                s.sz_time_space * (end_time - start_time), s.heat)
+        idx = sample.space_idx
+        if not target_samples[idx]:
+            target_samples[idx] = sample
+        else:
+            target_samples[idx].merge(sample)
+
+def __pr_heats(f, pid, tunit, tmin, tmax, aunit, amin, amax):
+    heat_samples = [None] * ((amax - amin) // aunit)
+
+    start_time = 0
+    end_time = 0
+    last_flushed = -1
+    while True:
+        start_time = end_time
+        timebin = f.read(16)
+        if (len(timebin)) != 16:
+            break
+        end_time = parse_time(timebin)
+        nr_tasks = struct.unpack('I', f.read(4))[0]
+        samples_set = {}
+        for t in range(nr_tasks):
+            samples = read_task_heats(f, pid, aunit, amin, amax)
+            if samples:
+                samples_set[pid] = samples
+        if not pid in samples_set:
+            continue
+        if start_time >= tmax:
+            continue
+        if end_time <= tmin:
+            continue
+        start_time = max(start_time, tmin)
+        end_time = min(end_time, tmax)
+
+        sidx = to_idx(start_time, tmin, tunit)
+        eidx = to_idx(end_time - 1, tmin, tunit)
+        for idx in range(sidx, eidx + 1):
+            if idx != last_flushed:
+                pr_samples(heat_samples, idx, tunit, aunit)
+                heat_samples = [None] * ((amax - amin) // aunit)
+                last_flushed = idx
+            st = max(start_time, tmin + idx * tunit)
+            et = min(end_time, tmin + (idx + 1) * tunit)
+            apply_samples(heat_samples, samples_set[pid], st, et, aunit, amin)
+
+def pr_heats(args):
+    binfile = args.input
+    pid = args.pid
+    tres = args.tres
+    tmin = args.tmin
+    ares = args.ares
+    amin = args.amin
+
+    tunit = (args.tmax - tmin) // tres
+    aunit = (args.amax - amin) // ares
+
+    # Compensate the values so that those fit with the resolution
+    tmax = tmin + tunit * tres
+    amax = amin + aunit * ares
+
+    with open(binfile, 'rb') as f:
+        __pr_heats(f, pid, tunit, tmin, tmax, aunit, amin, amax)
+
+class GuideInfo:
+    pid = None
+    start_time = None
+    end_time = None
+    lowest_addr = None
+    highest_addr = None
+    gaps = None
+
+    def __init__(self, pid, start_time):
+        self.pid = pid
+        self.start_time = start_time
+        self.gaps = []
+
+    def regions(self):
+        regions = []
+        region = [self.lowest_addr]
+        for gap in self.gaps:
+            for idx, point in enumerate(gap):
+                if idx == 0:
+                    region.append(point)
+                    regions.append(region)
+                else:
+                    region = [point]
+        region.append(self.highest_addr)
+        regions.append(region)
+        return regions
+
+    def total_space(self):
+        ret = 0
+        for r in self.regions():
+            ret += r[1] - r[0]
+        return ret
+
+    def __str__(self):
+        lines = ['pid:%d' % self.pid]
+        lines.append('time: %d-%d (%d)' % (self.start_time, self.end_time,
+                    self.end_time - self.start_time))
+        for idx, region in enumerate(self.regions()):
+            lines.append('region\t%2d: %020d-%020d (%d)' %
+                    (idx, region[0], region[1], region[1] - region[0]))
+        return '\n'.join(lines)
+
+def is_overlap(region1, region2):
+    if region1[1] < region2[0]:
+        return False
+    if region2[1] < region1[0]:
+        return False
+    return True
+
+def overlap_region_of(region1, region2):
+    return [max(region1[0], region2[0]), min(region1[1], region2[1])]
+
+def overlapping_regions(regions1, regions2):
+    overlap_regions = []
+    for r1 in regions1:
+        for r2 in regions2:
+            if is_overlap(r1, r2):
+                r1 = overlap_region_of(r1, r2)
+        if r1:
+            overlap_regions.append(r1)
+    return overlap_regions
+
+def get_guide_info(binfile):
+    "Read file, return the set of guide information objects of the data"
+    guides = {}
+    with open(binfile, 'rb') as f:
+        while True:
+            timebin = f.read(16)
+            if len(timebin) != 16:
+                break
+            monitor_time = parse_time(timebin)
+            nr_tasks = struct.unpack('I', f.read(4))[0]
+            for t in range(nr_tasks):
+                pid = struct.unpack('L', f.read(8))[0]
+                nr_regions = struct.unpack('I', f.read(4))[0]
+                if not pid in guides:
+                    guides[pid] = GuideInfo(pid, monitor_time)
+                guide = guides[pid]
+                guide.end_time = monitor_time
+
+                last_addr = None
+                gaps = []
+                for r in range(nr_regions):
+                    saddr = struct.unpack('L', f.read(8))[0]
+                    eaddr = struct.unpack('L', f.read(8))[0]
+                    f.read(4)
+
+                    if not guide.lowest_addr or saddr < guide.lowest_addr:
+                        guide.lowest_addr = saddr
+                    if not guide.highest_addr or eaddr > guide.highest_addr:
+                        guide.highest_addr = eaddr
+
+                    if not last_addr:
+                        last_addr = eaddr
+                        continue
+                    if last_addr != saddr:
+                        gaps.append([last_addr, saddr])
+                    last_addr = eaddr
+
+                if not guide.gaps:
+                    guide.gaps = gaps
+                else:
+                    guide.gaps = overlapping_regions(guide.gaps, gaps)
+    return sorted(list(guides.values()), key=lambda x: x.total_space(),
+                    reverse=True)
+
+def pr_guide(binfile):
+    for guide in get_guide_info(binfile):
+        print(guide)
+
+def region_sort_key(region):
+    return region[1] - region[0]
+
+def set_missed_args(args):
+    if args.pid and args.tmin and args.tmax and args.amin and args.amax:
+        return
+    guides = get_guide_info(args.input)
+    guide = guides[0]
+    if not args.pid:
+        args.pid = guide.pid
+    for g in guides:
+        if g.pid == args.pid:
+            guide = g
+            break
+
+    if not args.tmin:
+        args.tmin = guide.start_time
+    if not args.tmax:
+        args.tmax = guide.end_time
+
+    if not args.amin or not args.amax:
+        region = sorted(guide.regions(), key=lambda x: x[1] - x[0],
+                reverse=True)[0]
+        args.amin = region[0]
+        args.amax = region[1]
+
+def plot_heatmap(data_file, output_file):
+    terminal = output_file.split('.')[-1]
+    if not terminal in ['pdf', 'jpeg', 'png', 'svg']:
+        os.remove(data_file)
+        print("Unsupported plot output type.")
+        exit(-1)
+
+    gnuplot_cmd = """
+    set term %s;
+    set output '%s';
+    set key off;
+    set xrange [0:];
+    set yrange [0:];
+    set xlabel 'Time (ns)';
+    set ylabel 'Virtual Address (bytes)';
+    plot '%s' using 1:2:3 with image;""" % (terminal, output_file, data_file)
+    subprocess.call(['gnuplot', '-e', gnuplot_cmd])
+    os.remove(data_file)
+
+def set_argparser(parser):
+    parser.add_argument('--input', '-i', type=str, metavar='<file>',
+            default='damon.data', help='input file name')
+    parser.add_argument('--pid', metavar='<pid>', type=int,
+            help='pid of target task')
+    parser.add_argument('--tres', metavar='<resolution>', type=int,
+            default=500, help='time resolution of the output')
+    parser.add_argument('--tmin', metavar='<time>', type=lambda x: int(x,0),
+            help='minimal time of the output')
+    parser.add_argument('--tmax', metavar='<time>', type=lambda x: int(x,0),
+            help='maximum time of the output')
+    parser.add_argument('--ares', metavar='<resolution>', type=int, default=500,
+            help='space address resolution of the output')
+    parser.add_argument('--amin', metavar='<address>', type=lambda x: int(x,0),
+            help='minimal space address of the output')
+    parser.add_argument('--amax', metavar='<address>', type=lambda x: int(x,0),
+            help='maximum space address of the output')
+    parser.add_argument('--guide', action='store_true',
+            help='print a guidance for the min/max/resolution settings')
+    parser.add_argument('--heatmap', metavar='<file>', type=str,
+            help='heatmap image file to create')
+
+def main(args=None):
+    if not args:
+        parser = argparse.ArgumentParser()
+        set_argparser(parser)
+        args = parser.parse_args()
+
+    if args.guide:
+        pr_guide(args.input)
+    else:
+        set_missed_args(args)
+        orig_stdout = sys.stdout
+        if args.heatmap:
+            tmp_path = tempfile.mkstemp()[1]
+            tmp_file = open(tmp_path, 'w')
+            sys.stdout = tmp_file
+
+        pr_heats(args)
+
+        if args.heatmap:
+            sys.stdout = orig_stdout
+            tmp_file.flush()
+            tmp_file.close()
+            plot_heatmap(tmp_path, args.heatmap)
+
+if __name__ == '__main__':
+    main()
diff --git a/tools/damon/nr_regions.py b/tools/damon/nr_regions.py
new file mode 100644
index 000000000000..4eb7e824f7b6
--- /dev/null
+++ b/tools/damon/nr_regions.py
@@ -0,0 +1,88 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0
+
+"Print out distribution of the number of regions in the given record"
+
+import argparse
+import struct
+import sys
+import tempfile
+
+import _dist
+
+def set_argparser(parser):
+    parser.add_argument('--input', '-i', type=str, metavar='<file>',
+            default='damon.data', help='input file name')
+    parser.add_argument('--range', '-r', type=int, nargs=3,
+            metavar=('<start>', '<stop>', '<step>'),
+            help='range of percentiles to print')
+    parser.add_argument('--sortby', '-s', choices=['time', 'size'],
+            help='the metric to be used for sorting the number of regions')
+    parser.add_argument('--plot', '-p', type=str, metavar='<file>',
+            help='plot the distribution to an image file')
+
+def main(args=None):
+    if not args:
+        parser = argparse.ArgumentParser()
+        set_argparser(parser)
+        args = parser.parse_args()
+
+    percentiles = [0, 25, 50, 75, 100]
+
+    file_path = args.input
+    if args.range:
+        percentiles = range(args.range[0], args.range[1], args.range[2])
+    nr_regions_sort = True
+    if args.sortby == 'time':
+        nr_regions_sort = False
+
+    pid_pattern_map = {}
+    with open(file_path, 'rb') as f:
+        start_time = None
+        while True:
+            timebin = f.read(16)
+            if len(timebin) != 16:
+                break
+            nr_tasks = struct.unpack('I', f.read(4))[0]
+            for t in range(nr_tasks):
+                pid = struct.unpack('L', f.read(8))[0]
+                if not pid in pid_pattern_map:
+                    pid_pattern_map[pid] = []
+                pid_pattern_map[pid].append(_dist.access_patterns(f))
+
+    orig_stdout = sys.stdout
+    if args.plot:
+        tmp_path = tempfile.mkstemp()[1]
+        tmp_file = open(tmp_path, 'w')
+        sys.stdout = tmp_file
+
+    print('# <percentile> <# regions>')
+    for pid in pid_pattern_map.keys():
+        # Skip firs 20 regions as those would not adaptively adjusted
+        snapshots = pid_pattern_map[pid][20:]
+        nr_regions_dist = []
+        for snapshot in snapshots:
+            nr_regions_dist.append(len(snapshot))
+        if nr_regions_sort:
+            nr_regions_dist.sort(reverse=False)
+
+        print('# pid\t%s' % pid)
+        print('# avr:\t%d' % (sum(nr_regions_dist) / len(nr_regions_dist)))
+        for percentile in percentiles:
+            thres_idx = int(percentile / 100.0 * len(nr_regions_dist))
+            if thres_idx == len(nr_regions_dist):
+                thres_idx -= 1
+            threshold = nr_regions_dist[thres_idx]
+            print('%d\t%d' % (percentile, nr_regions_dist[thres_idx]))
+
+    if args.plot:
+        sys.stdout = orig_stdout
+        tmp_file.flush()
+        tmp_file.close()
+        xlabel = 'runtime (percent)'
+        if nr_regions_sort:
+            xlabel = 'percentile'
+        _dist.plot_dist(tmp_path, args.plot, xlabel)
+
+if __name__ == '__main__':
+    main()
diff --git a/tools/damon/record.py b/tools/damon/record.py
new file mode 100644
index 000000000000..4a3bf876ad8e
--- /dev/null
+++ b/tools/damon/record.py
@@ -0,0 +1,193 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0
+
+"""
+Record data access patterns of the target process.
+"""
+
+import argparse
+import copy
+import os
+import signal
+import subprocess
+import time
+
+debugfs_attrs = None
+debugfs_pids = None
+debugfs_monitor_on = None
+
+def set_target_pid(pid):
+    return subprocess.call('echo %s > %s' % (pid, debugfs_pids), shell=True,
+            executable='/bin/bash')
+
+def turn_damon(on_off):
+    return subprocess.call("echo %s > %s" % (on_off, debugfs_monitor_on),
+            shell=True, executable="/bin/bash")
+
+def is_damon_running():
+    with open(debugfs_monitor_on, 'r') as f:
+        return f.read().strip() == 'on'
+
+def do_record(target, is_target_cmd, attrs, old_attrs):
+    if os.path.isfile(attrs.rfile_path):
+        os.rename(attrs.rfile_path, attrs.rfile_path + '.old')
+
+    if attrs.apply():
+        print('attributes (%s) failed to be applied' % attrs)
+        cleanup_exit(old_attrs, -1)
+    print('# damon attrs: %s' % attrs)
+    if is_target_cmd:
+        p = subprocess.Popen(target, shell=True, executable='/bin/bash')
+        target = p.pid
+    if set_target_pid(target):
+        print('pid setting (%s) failed' % target)
+        cleanup_exit(old_attrs, -2)
+    if turn_damon('on'):
+        print('could not turn on damon' % target)
+        cleanup_exit(old_attrs, -3)
+    if is_target_cmd:
+        p.wait()
+    while True:
+        # damon will turn it off by itself if the target tasks are terminated.
+        if not is_damon_running():
+            break
+        time.sleep(1)
+
+    cleanup_exit(old_attrs, 0)
+
+class Attrs:
+    sample_interval = None
+    aggr_interval = None
+    regions_update_interval = None
+    min_nr_regions = None
+    max_nr_regions = None
+    rfile_path = None
+
+    def __init__(self, s, a, r, n, x, f):
+        self.sample_interval = s
+        self.aggr_interval = a
+        self.regions_update_interval = r
+        self.min_nr_regions = n
+        self.max_nr_regions = x
+        self.rfile_path = f
+
+    def __str__(self):
+        return "%s %s %s %s %s %s" % (self.sample_interval, self.aggr_interval,
+                self.regions_update_interval, self.min_nr_regions,
+                self.max_nr_regions, self.rfile_path)
+
+    def apply(self):
+        return subprocess.call('echo %s > %s' % (self, debugfs_attrs),
+                shell=True, executable='/bin/bash')
+
+def current_attrs():
+    with open(debugfs_attrs, 'r') as f:
+        attrs = f.read().split()
+    atnrs = [int(x) for x in attrs[0:5]]
+    attrs = atnrs + [attrs[5]]
+    return Attrs(*attrs)
+
+def cmd_args_to_attrs(args):
+    "Generate attributes based on current attributes and command arguments"
+    a = current_attrs()
+    if args.sample:
+        a.sample_interval = args.sample
+    if args.aggr:
+        a.aggr_interval = args.aggr
+    if args.updr:
+        a.regions_update_interval = args.updr
+    if args.minr:
+        a.min_nr_regions = args.minr
+    if args.maxr:
+        a.max_nr_regions = args.maxr
+    if args.out:
+        if not os.path.isabs(args.out):
+            args.out = os.path.join(os.getcwd(), args.out)
+        a.rfile_path = args.out
+    return a
+
+def cleanup_exit(orig_attrs, exit_code):
+    if is_damon_running():
+        if turn_damon('off'):
+            print('failed to turn damon off!')
+    if orig_attrs:
+        if orig_attrs.apply():
+            print('original attributes (%s) restoration failed!' % orig_attrs)
+    exit(exit_code)
+
+def sighandler(signum, frame):
+    print('\nsignal %s received' % signum)
+    cleanup_exit(orig_attrs, signum)
+
+def chk_update_debugfs(debugfs):
+    global debugfs_attrs
+    global debugfs_pids
+    global debugfs_monitor_on
+
+    debugfs_damon = os.path.join(debugfs, 'damon')
+    debugfs_attrs = os.path.join(debugfs_damon, 'attrs')
+    debugfs_pids = os.path.join(debugfs_damon, 'pids')
+    debugfs_monitor_on = os.path.join(debugfs_damon, 'monitor_on')
+
+    if not os.path.isdir(debugfs_damon):
+        print("damon debugfs dir (%s) not found", debugfs_damon)
+        exit(1)
+
+    for f in [debugfs_attrs, debugfs_pids, debugfs_monitor_on]:
+        if not os.path.isfile(f):
+            print("damon debugfs file (%s) not found" % f)
+            exit(1)
+
+def chk_permission():
+    if os.geteuid() != 0:
+        print("Run as root")
+        exit(1)
+
+def set_argparser(parser):
+    parser.add_argument('target', type=str, metavar='<target>',
+            help='the target command or the pid to record')
+    parser.add_argument('-s', '--sample', metavar='<interval>', type=int,
+            help='sampling interval')
+    parser.add_argument('-a', '--aggr', metavar='<interval>', type=int,
+            help='aggregate interval')
+    parser.add_argument('-u', '--updr', metavar='<interval>', type=int,
+            help='regions update interval')
+    parser.add_argument('-n', '--minr', metavar='<# regions>', type=int,
+            help='minimal number of regions')
+    parser.add_argument('-m', '--maxr', metavar='<# regions>', type=int,
+            help='maximum number of regions')
+    parser.add_argument('-o', '--out', metavar='<file path>', type=str,
+            default='damon.data', help='output file path')
+    parser.add_argument('-d', '--debugfs', metavar='<debugfs>', type=str,
+            default='/sys/kernel/debug', help='debugfs mounted path')
+
+def main(args=None):
+    if not args:
+        parser = argparse.ArgumentParser()
+        set_argparser(parser)
+        args = parser.parse_args()
+
+    chk_permission()
+    chk_update_debugfs(args.debugfs)
+
+    signal.signal(signal.SIGINT, sighandler)
+    signal.signal(signal.SIGTERM, sighandler)
+    orig_attrs = current_attrs()
+
+    new_attrs = cmd_args_to_attrs(args)
+    target = args.target
+
+    target_fields = target.split()
+    if not subprocess.call('which %s > /dev/null' % target_fields[0],
+            shell=True, executable='/bin/bash'):
+        do_record(target, True, new_attrs, orig_attrs)
+    else:
+        try:
+            pid = int(target)
+        except:
+            print('target \'%s\' is neither a command, nor a pid' % target)
+            exit(1)
+        do_record(target, False, new_attrs, orig_attrs)
+
+if __name__ == '__main__':
+    main()
diff --git a/tools/damon/report.py b/tools/damon/report.py
new file mode 100644
index 000000000000..c661c7b2f1af
--- /dev/null
+++ b/tools/damon/report.py
@@ -0,0 +1,45 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0
+
+import argparse
+
+import bin2txt
+import heats
+import nr_regions
+import wss
+
+def set_argparser(parser):
+    subparsers = parser.add_subparsers(title='report type', dest='report_type',
+            metavar='<report type>', help='the type of the report to generate')
+    subparsers.required = True
+
+    parser_raw = subparsers.add_parser('raw', help='human readable raw data')
+    bin2txt.set_argparser(parser_raw)
+
+    parser_heats = subparsers.add_parser('heats', help='heats of regions')
+    heats.set_argparser(parser_heats)
+
+    parser_wss = subparsers.add_parser('wss', help='working set size')
+    wss.set_argparser(parser_wss)
+
+    parser_nr_regions = subparsers.add_parser('nr_regions',
+            help='number of regions')
+    nr_regions.set_argparser(parser_nr_regions)
+
+def main(args=None):
+    if not args:
+        parser = argparse.ArgumentParser()
+        set_argparser(parser)
+        args = parser.parse_args()
+
+    if args.report_type == 'raw':
+        bin2txt.main(args)
+    elif args.report_type == 'heats':
+        heats.main(args)
+    elif args.report_type == 'wss':
+        wss.main(args)
+    elif args.report_type == 'nr_regions':
+        nr_regions.main(args)
+
+if __name__ == '__main__':
+    main()
diff --git a/tools/damon/wss.py b/tools/damon/wss.py
new file mode 100644
index 000000000000..fd5a0320070d
--- /dev/null
+++ b/tools/damon/wss.py
@@ -0,0 +1,94 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0
+
+"Print out the distribution of the working set sizes of the given trace"
+
+import argparse
+import struct
+import sys
+import tempfile
+
+import _dist
+
+def set_argparser(parser):
+    parser.add_argument('--input', '-i', type=str, metavar='<file>',
+            default='damon.data', help='input file name')
+    parser.add_argument('--range', '-r', type=int, nargs=3,
+            metavar=('<start>', '<stop>', '<step>'),
+            help='range of wss percentiles to print')
+    parser.add_argument('--sortby', '-s', choices=['time', 'size'],
+            help='the metric to be used for the sort of the working set sizes')
+    parser.add_argument('--plot', '-p', type=str, metavar='<file>',
+            help='plot the distribution to an image file')
+
+def main(args=None):
+    if not args:
+        parser = argparse.ArgumentParser()
+        set_argparser(parser)
+        args = parser.parse_args()
+
+    percentiles = [0, 25, 50, 75, 100]
+
+    file_path = args.input
+    if args.range:
+        percentiles = range(args.range[0], args.range[1], args.range[2])
+    wss_sort = True
+    if args.sortby == 'time':
+        wss_sort = False
+
+    pid_pattern_map = {}
+    with open(file_path, 'rb') as f:
+        start_time = None
+        while True:
+            timebin = f.read(16)
+            if len(timebin) != 16:
+                break
+            nr_tasks = struct.unpack('I', f.read(4))[0]
+            for t in range(nr_tasks):
+                pid = struct.unpack('L', f.read(8))[0]
+                if not pid in pid_pattern_map:
+                    pid_pattern_map[pid] = []
+                pid_pattern_map[pid].append(_dist.access_patterns(f))
+
+    orig_stdout = sys.stdout
+    if args.plot:
+        tmp_path = tempfile.mkstemp()[1]
+        tmp_file = open(tmp_path, 'w')
+        sys.stdout = tmp_file
+
+    print('# <percentile> <wss>')
+    for pid in pid_pattern_map.keys():
+        # Skip first 20 snapshots as regions may not adjusted yet.
+        snapshots = pid_pattern_map[pid][20:]
+        wss_dist = []
+        for snapshot in snapshots:
+            wss = 0
+            for p in snapshot:
+                # Ignore regions not accessed
+                if p[1] <= 0:
+                    continue
+                wss += p[0]
+            wss_dist.append(wss)
+        if wss_sort:
+            wss_dist.sort(reverse=False)
+
+        print('# pid\t%s' % pid)
+        print('# avr:\t%d' % (sum(wss_dist) / len(wss_dist)))
+        for percentile in percentiles:
+            thres_idx = int(percentile / 100.0 * len(wss_dist))
+            if thres_idx == len(wss_dist):
+                thres_idx -= 1
+            threshold = wss_dist[thres_idx]
+            print('%d\t%d' % (percentile, wss_dist[thres_idx]))
+
+    if args.plot:
+        sys.stdout = orig_stdout
+        tmp_file.flush()
+        tmp_file.close()
+        xlabel = 'runtime (percent)'
+        if wss_sort:
+            xlabel = 'percentile'
+        _dist.plot_dist(tmp_path, args.plot, xlabel)
+
+if __name__ == '__main__':
+    main()
-- 
2.17.1

[PATCH 7/8] Documentation/admin-guide/mm: Add a document for DAMON

[SeongJae Park]

From: SeongJae Park <sjpark@amazon.de>

This commit adds a simple document for DAMON under
`Documentation/admin-guide/mm`.

Signed-off-by: SeongJae Park <sjpark@amazon.de>
---
 .../admin-guide/mm/data_access_monitor.rst    | 240 ++++++++++++++++++
 Documentation/admin-guide/mm/index.rst        |   1 +
 MAINTAINERS                                   |   1 +
 3 files changed, 242 insertions(+)
 create mode 100644 Documentation/admin-guide/mm/data_access_monitor.rst

diff --git a/Documentation/admin-guide/mm/data_access_monitor.rst b/Documentation/admin-guide/mm/data_access_monitor.rst
new file mode 100644
index 000000000000..7a4d7ce88c20
--- /dev/null
+++ b/Documentation/admin-guide/mm/data_access_monitor.rst
@@ -0,0 +1,240 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+==========================
+DAMON: Data Access MONitor
+==========================
+
+
+Too Long; Don't Read
+====================
+
+DAMON is a kernel module that allows users to monitor the actual memory access
+pattern of specific user-space processes.  It aims to be 1) accurate enough to
+be useful for performance-centric domains, and 2) sufficiently light-weight so
+that it can be applied online.
+
+For the goals, DAMON utilizes its two core mechanisms, called region-based
+sampling and adaptive regions adjustment.  The region-based sampling allows
+users to make their own trade-off between the quality and the overhead of the
+monitoring and set the upperbound of the monitoring overhead.  Further, the
+adaptive regions adjustment mechanism makes DAMON to maximize the quality and
+minimize the overhead with its best efforts while preserving the users
+configured trade-off.
+
+Please note that the term 'memory' in this document means 'main memory'.  It
+also assumes that it would usually utilizes the middle level speed memory
+devices such as DRAMs or NVRAMs.  CPU caches or storage devices are not our
+concern, as those are too fast or too slow to be in DAMON's scope.
+
+
+Background
+==========
+
+For performance-centric analysis and optimizations of memory management schemes
+(either that of kernel space or user space), the actual data access pattern of
+the workloads is highly useful.  The information need to be only reasonable
+rather than strictly correct, because some level of incorrectness can be
+handled in many performance-centric domains.  It also need to be taken within
+reasonably short time with only light-weight overhead.
+
+Manually extracting such data is not easy and time consuming if the target
+workload is huge and complex, even for the developers of the programs.  There
+are a range of tools and techniques developed for general memory access
+investigations, and some of those could be partially used for this purpose.
+However, most of those are not practical or unscalable, mainly because those
+are designed with no consideration about the trade-off between the accuracy of
+the output and the overhead.
+
+The memory access instrumentation techniques which is applied to many tools
+such as Intel PIN is essential for correctness required cases such as invalid
+memory access bug detections.  However, those usually incur high overhead which
+is unacceptable for many of the performance-centric domains.  Periodic access
+checks based on H/W or S/W access counting features (e.g., the Accessed bits of
+PTEs or the PG_Idle flags of pages) can dramatically decrease the overhead by
+forgiving some of the quality, compared to the instrumentation based
+techniques.  The reduced quality is still reasonable for many of the domains,
+but the overhead can arbitrarily increase as the size of the target workload
+grows.  Miniature-like static region based sampling can set the upperbound of
+the overhead, but it will now decrease the quality of the output as the size of
+the workload grows.
+
+
+Expected Use-cases
+==================
+
+A straightforward usecase of DAMON would be the program behavior analysis.
+With the DAMON output, users can confirm whether the program is running as
+intended or not.  This will be useful for debuggings and tests of design
+points.
+
+The monitored results can also be useful for counting the dynamic working set
+size of workloads.  For the administration of memory overcommitted systems or
+selection of the environments (e.g., containers providing different amount of
+memory) for your workloads, this will be useful.
+
+If you are a programmer, you can optimize your program by managing the memory
+based on the actual data access pattern.  For example, you can identify the
+dynamic hotness of your data using DAMON and call ``mlock()`` to keep your hot
+data in DRAM, or call ``madvise()`` with ``MADV_PAGEOUT`` to proactively
+reclaim cold data.  Even though your program is guaranteed to not encounter
+memory pressure, you can still improve the performance by applying the DAMON
+outputs for call of ``MADV_HUGEPAGE`` and ``MADV_NOHUGEPAGE``.  More creative
+optimizations would be possible.  Our evaluations of DAMON includes a
+straightforward optimization using the ``mlock()``.  Please refer to the below
+Evaluation section for more detail.
+
+As DAMON incurs very low overhead, such optimizations can be applied not only
+offline, but also online.  Also, there is no reason to limit such optimizations
+to the user space.  Several parts of the kernel's memory management mechanisms
+could be also optimized using DAMON. The reclamation, the THP (de)promotion
+decisions, and the compaction would be such a candidates.
+
+
+Mechanisms of DAMON
+===================
+
+
+Basic Access Check
+------------------
+
+DAMON basically reports what pages are how frequently accessed.  The report is
+passed to users in binary format via a ``result file`` which users can set it's
+path.  Note that the frequency is not an absolute number of accesses, but a
+relative frequency among the pages of the target workloads.
+
+Users can also control the resolution of the reports by setting two time
+intervals, ``sampling interval`` and ``aggregation interval``.  In detail,
+DAMON checks access to each page per ``sampling interval``, aggregates the
+results (counts the number of the accesses to each page), and reports the
+aggregated results per ``aggregation interval``.  For the access check of each
+page, DAMON uses the Accessed bits of PTEs.
+
+This is thus similar to the previously mentioned periodic access checks based
+mechanisms, which overhead is increasing as the size of the target process
+grows.
+
+
+Region Based Sampling
+---------------------
+
+To avoid the unbounded increase of the overhead, DAMON groups a number of
+adjacent pages that assumed to have same access frequencies into a region.  As
+long as the assumption (pages in a region have same access frequencies) is
+kept, only one page in the region is required to be checked.  Thus, for each
+``sampling interval``, DAMON randomly picks one page in each region and clears
+its Accessed bit.  After one more ``sampling interval``, DAMON reads the
+Accessed bit of the page and increases the access frequency of the region if
+the bit has set meanwhile.  Therefore, the monitoring overhead is controllable
+by setting the number of regions.  DAMON allows users to set the minimal and
+maximum number of regions for the trade-off.
+
+Except the assumption, this is almost same with the above-mentioned
+miniature-like static region based sampling.  In other words, this scheme
+cannot preserve the quality of the output if the assumption is not guaranteed.
+
+
+Adaptive Regions Adjustment
+---------------------------
+
+At the beginning of the monitoring, DAMON constructs the initial regions by
+evenly splitting the memory mapped address space of the process into the
+user-specified minimal number of regions.  In this initial state, the
+assumption is normally not kept and thus the quality could be low.  To keep the
+assumption as much as possible, DAMON adaptively merges and splits each region.
+For each ``aggregation interval``, it compares the access frequencies of
+adjacent regions and merges those if the frequency difference is small.  Then,
+after it reports and clears the aggregated access frequency of each region, it
+splits each region into two regions if the total number of regions is smaller
+than the half of the user-specified maximum number of regions.
+
+In this way, DAMON provides its best-effort quality and minimal overhead while
+keeping the bounds users set for their trade-off.
+
+
+Applying Dynamic Memory Mappings
+--------------------------------
+
+Only a number of small parts in the super-huge virtual address space of the
+processes is mapped to physical memory and accessed.  Thus, tracking the
+unmapped address regions is just wasteful.  However, tracking every memory
+mapping change might incur an overhead.  For the reason, DAMON applies the
+dynamic memory mapping changes to the tracking regions only for each of an
+user-specified time interval (``regions update interval``).
+
+
+User Interface
+==============
+
+DAMON exports three files, ``attrs``, ``pids``, and ``monitor_on`` under its
+debugfs directory, ``<debugfs>/damon/``.
+
+
+Attributes
+----------
+
+Users can read and write the ``sampling interval``, ``aggregation interval``,
+``regions update interval``, min/max number of regions, and the path to
+``result file`` by reading from and writing to the ``attrs`` file.  For
+example, below commands set those values to 5 ms, 100 ms, 1,000 ms, 10, 1000,
+and ``/damon.data`` and check it again::
+
+    # cd <debugfs>/damon
+    # echo 5000 100000 1000000 10 1000 /damon.data > attrs
+    # cat attrs
+    5000 100000 1000000 10 1000 /damon.data
+
+
+Target PIDs
+-----------
+
+Users can read and write the pids of current monitoring target processes by
+reading from and writing to the `pids` file.  For example, below commands set
+processes having pids 42 and 4242 as the processes to be monitored and check
+it again::
+
+    # cd <debugfs>/damon
+    # echo 42 4242 > pids
+    # cat pids
+    42 4242
+
+Note that setting the pids doesn't starts the monitoring.
+
+
+Turning On/Off
+--------------
+
+You can check current status, start and stop the monitoring by reading from and
+writing to the ``monitor_on`` file.  Writing ``on`` to the file starts DAMON to
+monitor the target processes with the attributes.  Writing ``off`` to the file
+stops DAMON.  DAMON also stops if every target processes is be terminated.
+Below example commands turn on, off, and check status of DAMON::
+
+    # cd <debugfs>/damon
+    # echo on > monitor_on
+    # echo off > monitor_on
+    # cat monitor_on
+    off
+
+Please note that you cannot write to the ``attrs`` and ``pids`` files while the
+monitoring is turned on.  If you write to the files while DAMON is running,
+``-EINVAL`` will be returned.
+
+
+User Space Wrapper
+------------------
+
+DAMON has a shallow wrapper python script, ``/tools/damon/damo`` that provides
+more convenient interface.  Note that it is only aimed to be used for minimal
+reference of the DAMON's raw interfaces and for debugging of the DAMON itself.
+Based on the debugfs interface, you can create another cool and more convenient
+user space tools.
+
+
+Quick Tutorial
+--------------
+
+To test DAMON on your system,
+
+1. Ensure your kernel is built with CONFIG_DAMON turned on, and debugfs is
+   mounted at ``/sys/kernel/debug/``.
+2. ``<your kernel source tree>/tools/damon/damo -h``
diff --git a/Documentation/admin-guide/mm/index.rst b/Documentation/admin-guide/mm/index.rst
index 11db46448354..d3d0ba373eb6 100644
--- a/Documentation/admin-guide/mm/index.rst
+++ b/Documentation/admin-guide/mm/index.rst
@@ -27,6 +27,7 @@ the Linux memory management.
 
    concepts
    cma_debugfs
+   data_access_monitor
    hugetlbpage
    idle_page_tracking
    ksm
diff --git a/MAINTAINERS b/MAINTAINERS
index fb41236f3aed..71116be4701b 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4589,6 +4589,7 @@ L:	linux-mm@kvack.org
 S:	Maintained
 F:	mm/damon.c
 F:	tools/damon/*
+F:	Documentation/admin-guide/mm/data_access_monitor.rst
 
 DAVICOM FAST ETHERNET (DMFE) NETWORK DRIVER
 L:	netdev@vger.kernel.org
-- 
2.17.1

[PATCH 8/8] mm/damon: Add kunit tests

[SeongJae Park]

From: SeongJae Park <sjpark@amazon.de>

This commit adds kunit based unit tests for DAMON.

Signed-off-by: SeongJae Park <sjpark@amazon.de>
---
 MAINTAINERS     |   1 +
 mm/Kconfig      |  11 +
 mm/damon-test.h | 571 ++++++++++++++++++++++++++++++++++++++++++++++++
 mm/damon.c      |   2 +
 4 files changed, 585 insertions(+)
 create mode 100644 mm/damon-test.h

diff --git a/MAINTAINERS b/MAINTAINERS
index 71116be4701b..7bd9dcfe50cb 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4588,6 +4588,7 @@ M:	SeongJae Park <sjpark@amazon.de>
 L:	linux-mm@kvack.org
 S:	Maintained
 F:	mm/damon.c
+F:	mm/damon-test.h
 F:	tools/damon/*
 F:	Documentation/admin-guide/mm/data_access_monitor.rst
 
diff --git a/mm/Kconfig b/mm/Kconfig
index 35813e0d0e28..c999c1adfb0f 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -748,4 +748,15 @@ config DAMON
 	  be 1) accurate enough to be useful for performance-centric domains,
 	  and 2) sufficiently light-weight so that it can be applied online.
 
+config DAMON_TEST
+	bool "Test for damon"
+	depends on DAMON && KUNIT
+	help
+	  This builds the DAMON Kunit test suite.
+
+	  For more information on KUnit and unit tests in general, please refer
+	  to the KUnit documentation.
+
+	  If unsure, say N.
+
 endmenu
diff --git a/mm/damon-test.h b/mm/damon-test.h
new file mode 100644
index 000000000000..0d94910b8fe5
--- /dev/null
+++ b/mm/damon-test.h
@@ -0,0 +1,571 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Data Access Monitor Unit Tests
+ *
+ * Copyright 2019 Amazon.com, Inc. or its affiliates.  All rights reserved.
+ *
+ * Author: SeongJae Park <sjpark@amazon.de>
+ */
+
+#ifdef CONFIG_DAMON_TEST
+
+#ifndef _DAMON_TEST_H
+#define _DAMON_TEST_H
+
+#include <kunit/test.h>
+
+static void damon_test_str_to_pids(struct kunit *test)
+{
+	char *question;
+	unsigned long *answers;
+	unsigned long expected[] = {12, 35, 46};
+	ssize_t nr_integers = 0, i;
+
+	question = "123";
+	answers = str_to_pids(question, strnlen(question, 128), &nr_integers);
+	KUNIT_EXPECT_EQ(test, 1l, nr_integers);
+	KUNIT_EXPECT_EQ(test, 123ul, answers[0]);
+	kfree(answers);
+
+	question = "123abc";
+	answers = str_to_pids(question, strnlen(question, 128), &nr_integers);
+	KUNIT_EXPECT_EQ(test, 1l, nr_integers);
+	KUNIT_EXPECT_EQ(test, 123ul, answers[0]);
+	kfree(answers);
+
+	question = "a123";
+	answers = str_to_pids(question, strnlen(question, 128), &nr_integers);
+	KUNIT_EXPECT_EQ(test, 0l, nr_integers);
+	KUNIT_EXPECT_PTR_EQ(test, answers, (unsigned long *)NULL);
+
+	question = "12 35";
+	answers = str_to_pids(question, strnlen(question, 128), &nr_integers);
+	KUNIT_EXPECT_EQ(test, 2l, nr_integers);
+	for (i = 0; i < nr_integers; i++)
+		KUNIT_EXPECT_EQ(test, expected[i], answers[i]);
+	kfree(answers);
+
+	question = "12 35 46";
+	answers = str_to_pids(question, strnlen(question, 128), &nr_integers);
+	KUNIT_EXPECT_EQ(test, 3l, nr_integers);
+	for (i = 0; i < nr_integers; i++)
+		KUNIT_EXPECT_EQ(test, expected[i], answers[i]);
+	kfree(answers);
+
+	question = "12 35 abc 46";
+	answers = str_to_pids(question, strnlen(question, 128), &nr_integers);
+	KUNIT_EXPECT_EQ(test, 2l, nr_integers);
+	for (i = 0; i < 2; i++)
+		KUNIT_EXPECT_EQ(test, expected[i], answers[i]);
+	kfree(answers);
+
+	question = "";
+	answers = str_to_pids(question, strnlen(question, 128), &nr_integers);
+	KUNIT_EXPECT_EQ(test, 0l, nr_integers);
+	KUNIT_EXPECT_PTR_EQ(test, (unsigned long *)NULL, answers);
+	kfree(answers);
+
+	question = "\n";
+	answers = str_to_pids(question, strnlen(question, 128), &nr_integers);
+	KUNIT_EXPECT_EQ(test, 0l, nr_integers);
+	KUNIT_EXPECT_PTR_EQ(test, (unsigned long *)NULL, answers);
+	kfree(answers);
+}
+
+static void damon_test_regions(struct kunit *test)
+{
+	struct damon_region *r;
+	struct damon_task *t;
+
+	r = damon_new_region(1, 2);
+	KUNIT_EXPECT_EQ(test, 1ul, r->vm_start);
+	KUNIT_EXPECT_EQ(test, 2ul, r->vm_end);
+	KUNIT_EXPECT_EQ(test, 0u, r->nr_accesses);
+	KUNIT_EXPECT_TRUE(test, r->sampling_addr >= r->vm_start);
+	KUNIT_EXPECT_TRUE(test, r->sampling_addr < r->vm_end);
+
+	t = damon_new_task(42);
+	KUNIT_EXPECT_EQ(test, 0u, nr_damon_regions(t));
+
+	damon_add_region_tail(r, t);
+	KUNIT_EXPECT_EQ(test, 1u, nr_damon_regions(t));
+
+	damon_del_region(r);
+	KUNIT_EXPECT_EQ(test, 0u, nr_damon_regions(t));
+
+	damon_free_task(t);
+}
+
+static void damon_test_tasks(struct kunit *test)
+{
+	struct damon_task *t;
+
+	t = damon_new_task(42);
+	KUNIT_EXPECT_EQ(test, 42ul, t->pid);
+	KUNIT_EXPECT_EQ(test, 0u, nr_damon_tasks());
+
+	damon_add_task_tail(t);
+	KUNIT_EXPECT_EQ(test, 1u, nr_damon_tasks());
+
+	damon_destroy_task(t);
+	KUNIT_EXPECT_EQ(test, 0u, nr_damon_tasks());
+}
+
+static void damon_test_set_pids(struct kunit *test)
+{
+	unsigned long pids[] = {1, 2, 3};
+	char buf[64];
+
+	damon_set_pids(pids, 3);
+	damon_sprint_pids(buf, 64);
+	pr_info("buf: %s (%zu)\n", buf, strlen(buf));
+	KUNIT_EXPECT_EQ(test, 0, strncmp(buf, "1 2 3\n", 64));
+
+	damon_set_pids(NULL, 0);
+	damon_sprint_pids(buf, 64);
+	KUNIT_EXPECT_EQ(test, 0, strncmp(buf, "\n", 64));
+
+	damon_set_pids((unsigned long []){1, 2}, 2);
+	damon_sprint_pids(buf, 64);
+	KUNIT_EXPECT_EQ(test, 0, strncmp(buf, "1 2\n", 64));
+
+	damon_set_pids((unsigned long []){2}, 1);
+	damon_sprint_pids(buf, 64);
+	KUNIT_EXPECT_EQ(test, 0, strncmp(buf, "2\n", 64));
+
+	damon_set_pids(NULL, 0);
+	damon_sprint_pids(buf, 64);
+	KUNIT_EXPECT_EQ(test, 0, strncmp(buf, "\n", 64));
+}
+
+static void damon_test_three_regions_in_vmas(struct kunit *test)
+{
+	struct region regions[3] = {0,};
+
+	struct vm_area_struct vmas[] = {
+		(struct vm_area_struct) {.vm_start = 10, .vm_end = 20},
+		(struct vm_area_struct) {.vm_start = 20, .vm_end = 25},
+		(struct vm_area_struct) {.vm_start = 200, .vm_end = 210},
+		(struct vm_area_struct) {.vm_start = 210, .vm_end = 220},
+		(struct vm_area_struct) {.vm_start = 300, .vm_end = 305},
+		(struct vm_area_struct) {.vm_start = 307, .vm_end = 330},
+	};
+	vmas[0].vm_next = &vmas[1];
+	vmas[1].vm_next = &vmas[2];
+	vmas[2].vm_next = &vmas[3];
+	vmas[3].vm_next = &vmas[4];
+	vmas[4].vm_next = &vmas[5];
+	vmas[5].vm_next = NULL;
+
+	damon_three_regions_in_vmas(&vmas[0], regions);
+
+	KUNIT_EXPECT_EQ(test, 10ul, regions[0].start);
+	KUNIT_EXPECT_EQ(test, 25ul, regions[0].end);
+	KUNIT_EXPECT_EQ(test, 200ul, regions[1].start);
+	KUNIT_EXPECT_EQ(test, 220ul, regions[1].end);
+	KUNIT_EXPECT_EQ(test, 300ul, regions[2].start);
+	KUNIT_EXPECT_EQ(test, 330ul, regions[2].end);
+}
+
+/* Clean up global state of damon */
+static void damon_cleanup_global_state(void)
+{
+	struct damon_task *t, *next;
+
+	damon_for_each_task_safe(t, next)
+		damon_destroy_task(t);
+
+	damon_rbuf_offset = 0;
+}
+
+static void damon_test_aggregate(struct kunit *test)
+{
+	unsigned long pids[] = {1, 2, 3};
+	unsigned long saddr[][3] = {{10, 20, 30}, {5, 42, 49}, {13, 33, 55} };
+	unsigned long eaddr[][3] = {{15, 27, 40}, {31, 45, 55}, {23, 44, 66} };
+	unsigned long accesses[][3] = {{42, 95, 84}, {10, 20, 30}, {0, 1, 2} };
+	struct damon_task *t;
+	struct damon_region *r;
+	int it, ir;
+	ssize_t sz, sr, sp;
+
+	damon_set_pids(pids, 3);
+
+	it = 0;
+	damon_for_each_task(t) {
+		for (ir = 0; ir < 3; ir++) {
+			r = damon_new_region(saddr[it][ir], eaddr[it][ir]);
+			r->nr_accesses = accesses[it][ir];
+			damon_add_region_tail(r, t);
+		}
+		it++;
+	}
+	kdamond_flush_aggregated();
+	it = 0;
+	damon_for_each_task(t) {
+		ir = 0;
+		damon_for_each_region(r, t) {
+			KUNIT_EXPECT_EQ(test, 0u, r->nr_accesses);
+			ir++;
+		}
+		KUNIT_EXPECT_EQ(test, 3, ir);
+		it++;
+	}
+	KUNIT_EXPECT_EQ(test, 3, it);
+
+	sr = sizeof(r->vm_start) + sizeof(r->vm_end) + sizeof(r->nr_accesses);
+	sp = sizeof(t->pid) + sizeof(unsigned int) + 3 * sr;
+	sz = sizeof(struct timespec64) + sizeof(unsigned int) + 3 * sp;
+	KUNIT_EXPECT_EQ(test, (unsigned int)sz, damon_rbuf_offset);
+
+	damon_cleanup_global_state();
+}
+
+static void damon_test_write_rbuf(struct kunit *test)
+{
+	char *data;
+
+	data = "hello";
+	damon_write_rbuf(data, strnlen(data, 256));
+	KUNIT_EXPECT_EQ(test, damon_rbuf_offset, 5u);
+
+	damon_write_rbuf(data, 0);
+	KUNIT_EXPECT_EQ(test, damon_rbuf_offset, 5u);
+
+	KUNIT_EXPECT_EQ(test, strncmp(damon_rbuf, data, 5), 0);
+}
+
+static void damon_test_update_two_gaps(struct kunit *test)
+{
+	struct damon_task *t;
+	struct damon_region *r, *prev = NULL;
+	unsigned long regions[] = {10, 20, 20, 30,
+		50, 55, 55, 57, 57, 59,
+		70, 80, 80, 90, 90, 100};	/* 10-30, 50-59, 70-100 */
+	struct region new_regions[3] = {
+		(struct region){.start = 5, .end = 27},
+		(struct region){.start = 45, .end = 55},
+		(struct region){.start = 73, .end = 104} };
+	int i;
+	bool first_gap = true;
+
+	t = damon_new_task(42);
+	for (i = 0; i < ARRAY_SIZE(regions) / 2; i++) {
+		r = damon_new_region(regions[i * 2], regions[i * 2 + 1]);
+		damon_add_region_tail(r, t);
+	}
+	damon_add_task_tail(t);
+
+	damon_apply_three_regions(t, new_regions);
+
+	damon_for_each_region(r, t) {
+		if (prev == NULL) {
+			KUNIT_EXPECT_EQ(test, r->vm_start, 5ul);
+			goto next;
+		}
+
+		if (prev->vm_end != r->vm_start && first_gap) {
+			KUNIT_EXPECT_EQ(test, prev->vm_end, 27ul);
+			KUNIT_EXPECT_EQ(test, r->vm_start, 45ul);
+			first_gap = false;
+			goto next;
+		}
+
+		if (prev->vm_end != r->vm_start && !first_gap) {
+			KUNIT_EXPECT_EQ(test, prev->vm_end, 55ul);
+			KUNIT_EXPECT_EQ(test, r->vm_start, 73ul);
+			goto next;
+		}
+
+next:
+		prev = r;
+	}
+
+	damon_cleanup_global_state();
+}
+
+static void damon_test_update_two_gaps2(struct kunit *test)
+{
+	struct damon_task *t;
+	struct damon_region *r;
+	/* 10-20-30, 50-55-57-59, 70-80-90-100 */
+	unsigned long regions[] = {10, 20, 20, 30,
+		50, 55, 55, 57, 57, 59,
+		70, 80, 80, 90, 90, 100};
+	struct region new_regions[3] = {
+		(struct region){.start = 5, .end = 27},
+		(struct region){.start = 56, .end = 57},
+		(struct region){.start = 65, .end = 104} };
+	/* expect 5-27, 56-57, 65-80-90-104 */
+	unsigned long answers[] = {5, 20, 20, 27,
+		56, 57,
+		65, 80, 80, 90, 90, 104};
+	int i;
+
+	t = damon_new_task(42);
+	for (i = 0; i < ARRAY_SIZE(regions) / 2; i++) {
+		r = damon_new_region(regions[i * 2], regions[i * 2 + 1]);
+		damon_add_region_tail(r, t);
+	}
+	damon_add_task_tail(t);
+
+	damon_apply_three_regions(t, new_regions);
+
+	for (i = 0; i < ARRAY_SIZE(answers) / 2; i++) {
+		r = damon_nth_region_of(t, i);
+		KUNIT_EXPECT_EQ(test, r->vm_start, answers[i * 2]);
+		KUNIT_EXPECT_EQ(test, r->vm_end, answers[i++ * 2 + 1]);
+	}
+
+	damon_cleanup_global_state();
+}
+
+static void damon_test_update_two_gaps3(struct kunit *test)
+{
+	struct damon_task *t;
+	struct damon_region *r;
+	/* 10-20-30, 50-55-57-59, 70-80-90-100 */
+	unsigned long regions[] = {10, 20, 20, 30,
+		50, 55, 55, 57, 57, 59,
+		70, 80, 80, 90, 90, 100};
+	struct region new_regions[3] = {
+		(struct region){.start = 5, .end = 27},
+		(struct region){.start = 61, .end = 63},
+		(struct region){.start = 65, .end = 104} };
+	/* expect 5-27, 56-57, 65-80-90-104 */
+	unsigned long answers[] = {5, 20, 20, 27,
+		61, 63,
+		65, 80, 80, 90, 90, 104};
+	int i;
+
+	t = damon_new_task(42);
+	for (i = 0; i < ARRAY_SIZE(regions) / 2; i++) {
+		r = damon_new_region(regions[i * 2], regions[i * 2 + 1]);
+		damon_add_region_tail(r, t);
+	}
+	damon_add_task_tail(t);
+
+	damon_apply_three_regions(t, new_regions);
+
+	for (i = 0; i < ARRAY_SIZE(answers) / 2; i++) {
+		r = damon_nth_region_of(t, i);
+		KUNIT_EXPECT_EQ(test, r->vm_start, answers[i * 2]);
+		KUNIT_EXPECT_EQ(test, r->vm_end, answers[i++ * 2 + 1]);
+	}
+
+	damon_cleanup_global_state();
+}
+
+static void damon_test_update_two_gaps4(struct kunit *test)
+{
+	struct damon_task *t;
+	struct damon_region *r;
+	/* 10-20-30, 50-55-57-59, 70-80-90-100 */
+	unsigned long regions[] = {10, 20, 20, 30,
+		50, 55, 55, 57, 57, 59,
+		70, 80, 80, 90, 90, 100};
+	struct region new_regions[3] = {
+		(struct region){.start = 5, .end = 7},
+		(struct region){.start = 30, .end = 32},
+		(struct region){.start = 65, .end = 68} };
+	/* expect 5-27, 56-57, 65-80-90-104 */
+	unsigned long answers[] = {5, 7, 30, 32, 65, 68};
+	int i;
+
+	t = damon_new_task(42);
+	for (i = 0; i < ARRAY_SIZE(regions) / 2; i++) {
+		r = damon_new_region(regions[i * 2], regions[i * 2 + 1]);
+		damon_add_region_tail(r, t);
+	}
+	damon_add_task_tail(t);
+
+	damon_apply_three_regions(t, new_regions);
+
+	for (i = 0; i < ARRAY_SIZE(answers) / 2; i++) {
+		r = damon_nth_region_of(t, i);
+		KUNIT_EXPECT_EQ(test, r->vm_start, answers[i * 2]);
+		KUNIT_EXPECT_EQ(test, r->vm_end, answers[i++ * 2 + 1]);
+	}
+
+	damon_cleanup_global_state();
+}
+
+static void damon_test_split_evenly(struct kunit *test)
+{
+	struct damon_task *t;
+	struct damon_region *r;
+	unsigned long i;
+
+	KUNIT_EXPECT_EQ(test, damon_split_region_evenly(NULL, 5), -EINVAL);
+
+	t = damon_new_task(42);
+	r = damon_new_region(0, 100);
+	KUNIT_EXPECT_EQ(test, damon_split_region_evenly(r, 0), -EINVAL);
+
+	damon_add_region_tail(r, t);
+	KUNIT_EXPECT_EQ(test, damon_split_region_evenly(r, 10), 0);
+	KUNIT_EXPECT_EQ(test, nr_damon_regions(t), 10u);
+
+	i = 0;
+	damon_for_each_region(r, t) {
+		KUNIT_EXPECT_EQ(test, r->vm_start, i++ * 10);
+		KUNIT_EXPECT_EQ(test, r->vm_end, i * 10);
+	}
+	damon_free_task(t);
+
+	t = damon_new_task(42);
+	r = damon_new_region(5, 59);
+	damon_add_region_tail(r, t);
+	KUNIT_EXPECT_EQ(test, damon_split_region_evenly(r, 5), 0);
+	KUNIT_EXPECT_EQ(test, nr_damon_regions(t), 5u);
+
+	i = 0;
+	damon_for_each_region(r, t) {
+		if (i == 4)
+			break;
+		KUNIT_EXPECT_EQ(test, r->vm_start, 5 + 10 * i++);
+		KUNIT_EXPECT_EQ(test, r->vm_end, 5 + 10 * i);
+	}
+	KUNIT_EXPECT_EQ(test, r->vm_start, 5 + 10 * i);
+	KUNIT_EXPECT_EQ(test, r->vm_end, 59ul);
+	damon_free_task(t);
+
+	t = damon_new_task(42);
+	r = damon_new_region(5, 6);
+	damon_add_region_tail(r, t);
+	KUNIT_EXPECT_EQ(test, damon_split_region_evenly(r, 2), -EINVAL);
+	KUNIT_EXPECT_EQ(test, nr_damon_regions(t), 1u);
+
+	damon_for_each_region(r, t) {
+		KUNIT_EXPECT_EQ(test, r->vm_start, 5ul);
+		KUNIT_EXPECT_EQ(test, r->vm_end, 6ul);
+	}
+	damon_free_task(t);
+}
+
+static void damon_test_split_at(struct kunit *test)
+{
+	struct damon_task *t;
+	struct damon_region *r;
+
+	t = damon_new_task(42);
+	r = damon_new_region(0, 100);
+	damon_add_region_tail(r, t);
+	damon_split_region_at(r, 25);
+	KUNIT_EXPECT_EQ(test, r->vm_start, 0ul);
+	KUNIT_EXPECT_EQ(test, r->vm_end, 25ul);
+
+	r = damon_next_region(r);
+	KUNIT_EXPECT_EQ(test, r->vm_start, 25ul);
+	KUNIT_EXPECT_EQ(test, r->vm_end, 100ul);
+
+	damon_free_task(t);
+}
+
+static void damon_test_merge_two(struct kunit *test)
+{
+	struct damon_task *t;
+	struct damon_region *r, *r2, *r3;
+	int i;
+
+	t = damon_new_task(42);
+	r = damon_new_region(0, 100);
+	r->nr_accesses = 10;
+	damon_add_region_tail(r, t);
+	r2 = damon_new_region(100, 300);
+	r2->nr_accesses = 20;
+	damon_add_region_tail(r2, t);
+
+	damon_merge_two_regions(r, r2);
+	KUNIT_EXPECT_EQ(test, r->vm_start, 0ul);
+	KUNIT_EXPECT_EQ(test, r->vm_end, 300ul);
+	KUNIT_EXPECT_EQ(test, r->nr_accesses, 16u);
+
+	i = 0;
+	damon_for_each_region(r3, t) {
+		KUNIT_EXPECT_PTR_EQ(test, r, r3);
+		i++;
+	}
+	KUNIT_EXPECT_EQ(test, i, 1);
+
+	damon_free_task(t);
+}
+
+static void damon_test_merge_regions_of(struct kunit *test)
+{
+	struct damon_task *t;
+	struct damon_region *r;
+	unsigned long sa[] = {0, 100, 114, 122, 130, 156, 170, 184};
+	unsigned long ea[] = {100, 112, 122, 130, 156, 170, 184, 230};
+	unsigned int nrs[] = {0, 0, 10, 10, 20, 30, 1, 2};
+
+	unsigned long saddrs[] = {0, 114, 130, 156, 170};
+	unsigned long eaddrs[] = {112, 130, 156, 170, 230};
+	int i;
+
+	t = damon_new_task(42);
+	for (i = 0; i < ARRAY_SIZE(sa); i++) {
+		r = damon_new_region(sa[i], ea[i]);
+		r->nr_accesses = nrs[i];
+		damon_add_region_tail(r, t);
+	}
+
+	damon_merge_regions_of(t, 9);
+	/* 0-112, 114-130, 130-156, 156-170 */
+	KUNIT_EXPECT_EQ(test, nr_damon_regions(t), 5u);
+	for (i = 0; i < 5; i++) {
+		r = damon_nth_region_of(t, i);
+		KUNIT_EXPECT_EQ(test, r->vm_start, saddrs[i]);
+		KUNIT_EXPECT_EQ(test, r->vm_end, eaddrs[i]);
+	}
+	damon_free_task(t);
+}
+
+static void damon_test_split_regions_of(struct kunit *test)
+{
+	struct damon_task *t;
+	struct damon_region *r;
+
+	t = damon_new_task(42);
+	r = damon_new_region(0, 22);
+	damon_add_region_tail(r, t);
+	damon_split_regions_of(t);
+	KUNIT_EXPECT_EQ(test, nr_damon_regions(t), 2u);
+	damon_free_task(t);
+}
+
+static void damon_test_kdamond_need_stop(struct kunit *test)
+{
+	KUNIT_EXPECT_TRUE(test, kdamond_need_stop());
+}
+
+static struct kunit_case damon_test_cases[] = {
+	KUNIT_CASE(damon_test_str_to_pids),
+	KUNIT_CASE(damon_test_tasks),
+	KUNIT_CASE(damon_test_regions),
+	KUNIT_CASE(damon_test_set_pids),
+	KUNIT_CASE(damon_test_three_regions_in_vmas),
+	KUNIT_CASE(damon_test_aggregate),
+	KUNIT_CASE(damon_test_write_rbuf),
+	KUNIT_CASE(damon_test_update_two_gaps),
+	KUNIT_CASE(damon_test_update_two_gaps2),
+	KUNIT_CASE(damon_test_update_two_gaps3),
+	KUNIT_CASE(damon_test_update_two_gaps4),
+	KUNIT_CASE(damon_test_split_evenly),
+	KUNIT_CASE(damon_test_split_at),
+	KUNIT_CASE(damon_test_merge_two),
+	KUNIT_CASE(damon_test_merge_regions_of),
+	KUNIT_CASE(damon_test_split_regions_of),
+	KUNIT_CASE(damon_test_kdamond_need_stop),
+	{},
+};
+
+static struct kunit_suite damon_test_suite = {
+	.name = "damon",
+	.test_cases = damon_test_cases,
+};
+kunit_test_suite(damon_test_suite);
+
+#endif /* _DAMON_TEST_H */
+
+#endif	/* CONFIG_DAMON_TEST */
diff --git a/mm/damon.c b/mm/damon.c
index e2492344b58c..8067ea916f81 100644
--- a/mm/damon.c
+++ b/mm/damon.c
@@ -1275,3 +1275,5 @@ module_exit(damon_exit);
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("SeongJae Park <sjpark@amazon.de>");
 MODULE_DESCRIPTION("DAMON: Data Access MONitor");
+
+#include "damon-test.h"
-- 
2.17.1

Re: [PATCH 0/8] Introduce Data Access MONitor (DAMON)

[Kirill A. Shutemov]

On Mon, Jan 20, 2020 at 05:27:49PM +0100, SeongJae Park wrote:
> This patchset introduces a new kernel module for practical monitoring of
> data accesses, namely DAMON.

Looks like it's not integrated with perf at all right? Why?
Correlating measurements from different domains would help to see the
bigger picture.

-- 
 Kirill A. Shutemov

Re: Re: [PATCH 0/8] Introduce Data Access MONitor (DAMON)

[SeongJae Park]

On Mon, 20 Jan 2020 19:55:51 +0300 "Kirill A. Shutemov" <kirill@shutemov.name> wrote:

> On Mon, Jan 20, 2020 at 05:27:49PM +0100, SeongJae Park wrote:
> > This patchset introduces a new kernel module for practical monitoring of
> > data accesses, namely DAMON.
> 
> Looks like it's not integrated with perf at all right? Why?
> Correlating measurements from different domains would help to see the
> bigger picture.

Right, it's not integrated with perf.  DAMON provides only its debugfs
interface.  Also I agree that correlating measurments will be helpful.  I do
not integrate DAMON with perf with this patchset for following reasons.

First of all, I think I have no deep understanding of perf, yet.  Partly for
the reason, I couldn't figure out the best way to integrate DAMON with perf.
Especially, I couldn't straightforwardly classify DAMON providing information
into one of the categories of the perf events I know.

Therefore, rather than integrating DAMON with perf in my arguable way and
increasing the complexity of the code, I decided to keep the interface as
simple and flexible as it can be for the first stage.  That said, I believe it
would be not too hard to integrate DAMON with perf in a future.


Thanks,
SeongJae Park

> 
> -- 
>  Kirill A. Shutemov
> 
> 

Re: [PATCH 2/8] mm/damon: Implement region based sampling

[SeongJae Park]

On Mon, 20 Jan 2020 17:27:51 +0100 SeongJae Park <sjpark@amazon.com> wrote:

> From: SeongJae Park <sjpark@amazon.de>
> 
> This commit implements DAMON's basic access check and region based
> sampling mechanisms.
> 
> Basic Access Check
> ------------------
> 
> DAMON basically reports what pages are how frequently accessed.  Note
> that the frequency is not an absolute number of accesses, but a relative
> frequency among the pages of the target workloads.
> 
> Users can control the resolution of the reports by setting two time
> intervals, ``sampling interval`` and ``aggregation interval``.  In
> detail, DAMON checks access to each page per ``sampling interval``,
> aggregates the results (counts the number of the accesses to each page),
> and reports the aggregated results per ``aggregation interval``.  For
> the access check of each page, DAMON uses the Accessed bits of PTEs.
> 
> This is thus similar to common periodic access checks based access
> tracking mechanisms, which overhead is increasing as the size of the
> target process grows.
> 
> Region Based Sampling
> ---------------------
> 
> To avoid the unbounded increase of the overhead, DAMON groups a number
> of adjacent pages that assumed to have same access frequencies into a
> region.  As long as the assumption (pages in a region have same access
> frequencies) is kept, only one page in the region is required to be
> checked.  Thus, for each ``sampling interval``, DAMON randomly picks one
> page in each region and clears its Accessed bit.  After one more
> ``sampling interval``, DAMON reads the Accessed bit of the page and
> increases the access frequency of the region if the bit has set
> meanwhile.  Therefore, the monitoring overhead is controllable by
> setting the number of regions.
> 
> Nonetheless, this scheme cannot preserve the quality of the output if
> the assumption is not kept.
> 
> Signed-off-by: SeongJae Park <sjpark@amazon.de>
> ---
>  mm/damon.c | 599 +++++++++++++++++++++++++++++++++++++++++++++++++++++
>  1 file changed, 599 insertions(+)
> 
> diff --git a/mm/damon.c b/mm/damon.c
> index 064ec1f6ded9..2a0c010291f8 100644
> --- a/mm/damon.c
> +++ b/mm/damon.c
> @@ -9,9 +9,14 @@
>  
>  #define pr_fmt(fmt) "damon: " fmt
> [...]
> +
> +/*
> + * Check whether the given region has accessed since the last check
> + *
> + * mm	'mm_struct' for the given virtual address space
> + * r	the region to be checked
> + */
> +static void kdamond_check_access(struct mm_struct *mm, struct damon_region *r)
> +{
> +	pte_t *pte = NULL;
> +	pmd_t *pmd = NULL;
> +	spinlock_t *ptl;
> +
> +	if (follow_pte_pmd(mm, r->sampling_addr, NULL, &pte, &pmd, &ptl))
> +		goto mkold;
> +
> +	/* Read the page table access bit of the page */
> +	if (pte && pte_young(*pte))
> +		r->nr_accesses++;
> +#ifdef CONFIG_TRANSPARENT_HUGEPAGE
> +	else if (pmd && pmd_young(*pmd))
> +		r->nr_accesses++;
> +#endif	/* CONFIG_TRANSPARENT_HUGEPAGE */
> +
> +	spin_unlock(ptl);
> +
> +mkold:
> +	/* mkold next target */
> +	r->sampling_addr = damon_rand(r->vm_start, r->vm_end);
> +
> +	if (follow_pte_pmd(mm, r->sampling_addr, NULL, &pte, &pmd, &ptl))
> +		return;
> +
> +	if (pte) {
> +		if (pte_young(*pte))
> +			clear_page_idle(pte_page(*pte));

Yunjae has personally pointed me out that this could interfere with reclamation
logic because page_referenced_one() checks the pte Accessed bits.  As the
function also checks PG_Young, we agreed to adjust PG_Young in addition to the
PG_Idle here, as below:

    diff --git a/mm/damon.c b/mm/damon.c
    index 8067ea916f81..55b89a2c0140 100644
    --- a/mm/damon.c
    +++ b/mm/damon.c
    @@ -491,14 +491,18 @@ static void kdamond_check_access(struct mm_struct *mm, struct damon_region *r)
                    return;
    
            if (pte) {
    -               if (pte_young(*pte))
    +               if (pte_young(*pte)) {
                            clear_page_idle(pte_page(*pte));
    +                       set_page_young(pte_page(*pte));
    +               }
                    *pte = pte_mkold(*pte);
            }
     #ifdef CONFIG_TRANSPARENT_HUGEPAGE
            else if (pmd) {
    -               if (pmd_young(*pmd))
    +               if (pmd_young(*pmd)) {
                            clear_page_idle(pmd_page(*pmd));
    +                       set_page_young(pte_page(*pte));
    +               }
                    *pmd = pmd_mkold(*pmd);
            }
     #endif

This change will be merged into this patch by next spin.  Also, adding CC for
page_idle.c related people.


Thanks,
SeongJae Park


> +		*pte = pte_mkold(*pte);
> +	}
> +#ifdef CONFIG_TRANSPARENT_HUGEPAGE
> +	else if (pmd) {
> +		if (pmd_young(*pmd))
> +			clear_page_idle(pmd_page(*pmd));
> +		*pmd = pmd_mkold(*pmd);
> +	}
> +#endif
> +
> +	spin_unlock(ptl);
> +}
> +
> +/*
> + * Check whether a time interval is elapsed
> + *
> + * baseline	the time to check whether the interval has elapsed since
> + * interval	the time interval (microseconds)
> + *
> + * See whether the given time interval has passed since the given baseline
> + * time.  If so, it also updates the baseline to current time for next check.
> + *
> + * Returns true if the time interval has passed, or false otherwise.
> + */
> +static bool damon_check_reset_time_interval(struct timespec64 *baseline,
> +		unsigned long interval)
> +{
> +	struct timespec64 now;
> +
> +	ktime_get_coarse_ts64(&now);
> +	if ((timespec64_to_ns(&now) - timespec64_to_ns(baseline)) / 1000 <
> +			interval)
> +		return false;
> +	*baseline = now;
> +	return true;
> +}
> +
> +/*
> + * Check whether it is time to flush the aggregated information
> + */
> +static bool kdamond_aggregate_interval_passed(void)
> +{
> +	return damon_check_reset_time_interval(&last_aggregate_time,
> +			aggr_interval);
> +}
> +
> +/*
> + * Flush the content in the result buffer to the result file
> + */
> +static void damon_flush_rbuffer(void)
> +{
> +	ssize_t sz;
> +	loff_t pos;
> +	struct file *rfile;
> +
> +	while (damon_rbuf_offset) {
> +		pos = 0;
> +		rfile = filp_open(rfile_path, O_CREAT | O_RDWR | O_APPEND,
> +				0644);
> +		if (IS_ERR(rfile)) {
> +			pr_err("Cannot open the result file %s\n", rfile_path);
> +			return;
> +		}
> +
> +		sz = kernel_write(rfile, damon_rbuf, damon_rbuf_offset, &pos);
> +		filp_close(rfile, NULL);
> +
> +		damon_rbuf_offset -= sz;
> +	}
> +}
> +
> +/*
> + * Write a data into the result buffer
> + */
> +static void damon_write_rbuf(void *data, ssize_t size)
> +{
> +	if (damon_rbuf_offset + size > DAMON_LEN_RBUF)
> +		damon_flush_rbuffer();
> +
> +	memcpy(&damon_rbuf[damon_rbuf_offset], data, size);
> +	damon_rbuf_offset += size;
> +}
> +
> +/*
> + * Flush the aggregated monitoring results to the result buffer
> + *
> + * Stores current tracking results to the result buffer and reset 'nr_accesses'
> + * of each regions.  The format for the result buffer is as below:
> + *
> + *   <time> <number of tasks> <array of task infos>
> + *
> + *   task info: <pid> <number of regions> <array of region infos>
> + *   region info: <start address> <end address> <nr_accesses>
> + */
> +static void kdamond_flush_aggregated(void)
> +{
> +	struct damon_task *t;
> +	struct timespec64 now;
> +	unsigned int nr;
> +
> +	ktime_get_coarse_ts64(&now);
> +
> +	damon_write_rbuf(&now, sizeof(struct timespec64));
> +	nr = nr_damon_tasks();
> +	damon_write_rbuf(&nr, sizeof(nr));
> +
> +	damon_for_each_task(t) {
> +		struct damon_region *r;
> +
> +		damon_write_rbuf(&t->pid, sizeof(t->pid));
> +		nr = nr_damon_regions(t);
> +		damon_write_rbuf(&nr, sizeof(nr));
> +		damon_for_each_region(r, t) {
> +			damon_write_rbuf(&r->vm_start, sizeof(r->vm_start));
> +			damon_write_rbuf(&r->vm_end, sizeof(r->vm_end));
> +			damon_write_rbuf(&r->nr_accesses,
> +					sizeof(r->nr_accesses));
> +			r->nr_accesses = 0;
> +		}
> +	}
> +}
> +
> +/*
> + * Check whether current monitoring should be stopped
> + *
> + * If users asked to stop, need stop.  Even though no user has asked to stop,
> + * need stop if every target task has dead.
> + *
> + * Returns true if need to stop current monitoring.
> + */
> +static bool kdamond_need_stop(void)
> +{
> +	struct damon_task *t;
> +	struct task_struct *task;
> +	bool stop;
> +
> +	spin_lock(&kdamond_lock);
> +	stop = kdamond_stop;
> +	spin_unlock(&kdamond_lock);
> +	if (stop)
> +		return true;
> +
> +	damon_for_each_task(t) {
> +		task = damon_get_task_struct(t);
> +		if (task) {
> +			put_task_struct(task);
> +			return false;
> +		}
> +	}
> +
> +	return true;
> +}
> +
> +/*
> + * The monitoring daemon that runs as a kernel thread
> + */
> +static int kdamond_fn(void *data)
> +{
> +	struct damon_task *t;
> +	struct damon_region *r, *next;
> +	struct mm_struct *mm;
> +
> +	pr_info("kdamond (%d) starts\n", kdamond->pid);
> +	kdamond_init_regions();
> +	while (!kdamond_need_stop()) {
> +		damon_for_each_task(t) {
> +			mm = damon_get_mm(t);
> +			if (!mm)
> +				continue;
> +			damon_for_each_region(r, t)
> +				kdamond_check_access(mm, r);
> +			mmput(mm);
> +		}
> +
> +		if (kdamond_aggregate_interval_passed())
> +			kdamond_flush_aggregated();
> +
> +		usleep_range(sample_interval, sample_interval + 1);
> +	}
> +	damon_flush_rbuffer();
> +	damon_for_each_task(t) {
> +		damon_for_each_region_safe(r, next, t)
> +			damon_destroy_region(r);
> +	}
> +	pr_info("kdamond (%d) finishes\n", kdamond->pid);
> +	spin_lock(&kdamond_lock);
> +	kdamond = NULL;
> +	spin_unlock(&kdamond_lock);
> +	return 0;
> +}
> +
> +/*
> + * Controller functions
> + */
> +
> +/*
> + * Start or stop the kdamond
> + *
> + * Returns 0 if success, negative error code otherwise.
> + */
> +static int damon_turn_kdamond(bool on)
> +{
> +	spin_lock(&kdamond_lock);
> +	kdamond_stop = !on;
> +	if (!kdamond && on) {
> +		kdamond = kthread_run(kdamond_fn, NULL, "kdamond");
> +		if (!kdamond)
> +			goto fail;
> +		goto success;
> +	}
> +	if (kdamond && !on) {
> +		spin_unlock(&kdamond_lock);
> +		while (true) {
> +			spin_lock(&kdamond_lock);
> +			if (!kdamond)
> +				goto success;
> +			spin_unlock(&kdamond_lock);
> +
> +			usleep_range(sample_interval, sample_interval * 2);
> +		}
> +	}
> +
> +	/* tried to turn on while turned on, or turn off while turned off */
> +
> +fail:
> +	spin_unlock(&kdamond_lock);
> +	return -EINVAL;
> +
> +success:
> +	spin_unlock(&kdamond_lock);
> +	return 0;
> +}
> +
> +static inline bool damon_is_target_pid(unsigned long pid)
> +{
> +	struct damon_task *t;
> +
> +	damon_for_each_task(t) {
> +		if (t->pid == pid)
> +			return true;
> +	}
> +	return false;
> +}
> +
> +/*
> + * This function should not be called while the kdamond is running.
> + */
> +static long damon_set_pids(unsigned long *pids, ssize_t nr_pids)
> +{
> +	ssize_t i;
> +	struct damon_task *t, *next;
> +
> +	/* Remove unselected tasks */
> +	damon_for_each_task_safe(t, next) {
> +		for (i = 0; i < nr_pids; i++) {
> +			if (pids[i] == t->pid)
> +				break;
> +		}
> +		if (i != nr_pids)
> +			continue;
> +		damon_destroy_task(t);
> +	}
> +
> +	/* Add new tasks */
> +	for (i = 0; i < nr_pids; i++) {
> +		if (damon_is_target_pid(pids[i]))
> +			continue;
> +		t = damon_new_task(pids[i]);
> +		if (!t) {
> +			pr_err("Failed to alloc damon_task\n");
> +			return -ENOMEM;
> +		}
> +		damon_add_task_tail(t);
> +	}
> +
> +	return 0;
> +}
> +
> +/*
> + * Set attributes for the monitoring
> + *
> + * sample_int		time interval between samplings
> + * aggr_int		time interval between aggregations
> + * min_nr_reg		minimal number of regions
> + * path_to_rfile	path to the monitor result files
> + *
> + * This function should not be called while the kdamond is running.
> + * Every time interval is in micro-seconds.
> + *
> + * Returns 0 on success, negative error code otherwise.
> + */
> +static long damon_set_attrs(unsigned long sample_int, unsigned long aggr_int,
> +		unsigned long min_nr_reg, char *path_to_rfile)
> +{
> +	if (strnlen(path_to_rfile, LEN_RES_FILE_PATH) >= LEN_RES_FILE_PATH) {
> +		pr_err("too long (>%d) result file path %s\n",
> +				LEN_RES_FILE_PATH, path_to_rfile);
> +		return -EINVAL;
> +	}
> +	if (min_nr_reg < 3) {
> +		pr_err("min_nr_regions (%lu) should be bigger than 2\n",
> +				min_nr_reg);
> +		return -EINVAL;
> +	}
> +
> +	sample_interval = sample_int;
> +	aggr_interval = aggr_int;
> +	min_nr_regions = min_nr_reg;
> +	strncpy(rfile_path, path_to_rfile, LEN_RES_FILE_PATH);
> +	return 0;
> +}
> +
>  static int __init damon_init(void)
>  {
>  	pr_info("init\n");
>  
>  	prandom_seed_state(&rndseed, 42);
> +	ktime_get_coarse_ts64(&last_aggregate_time);
>  	return 0;
>  }
>  
>  static void __exit damon_exit(void)
>  {
> +	damon_turn_kdamond(false);
>  	pr_info("exit\n");
>  }
>  
> -- 
> 2.17.1
> 

Re: Re: Re: [PATCH 0/8] Introduce Data Access MONitor (DAMON)

[sjpark]

On Mon, 20 Jan 2020 19:14:40 +0100 SeongJae Park <sj38.park@gmail.com> wrote:

> On Mon, 20 Jan 2020 19:55:51 +0300 "Kirill A. Shutemov" <kirill@shutemov.name> wrote:
> 
> > On Mon, Jan 20, 2020 at 05:27:49PM +0100, SeongJae Park wrote:
> > > This patchset introduces a new kernel module for practical monitoring of
> > > data accesses, namely DAMON.
> > 
> > Looks like it's not integrated with perf at all right? Why?
> > Correlating measurements from different domains would help to see the
> > bigger picture.
> 
> Right, it's not integrated with perf.  DAMON provides only its debugfs
> interface.  Also I agree that correlating measurments will be helpful.  I do
> not integrate DAMON with perf with this patchset for following reasons.
> 
> First of all, I think I have no deep understanding of perf, yet.  Partly for
> the reason, I couldn't figure out the best way to integrate DAMON with perf.
> Especially, I couldn't straightforwardly classify DAMON providing information
> into one of the categories of the perf events I know.
> 
> Therefore, rather than integrating DAMON with perf in my arguable way and
> increasing the complexity of the code, I decided to keep the interface as
> simple and flexible as it can be for the first stage.  That said, I believe it
> would be not too hard to integrate DAMON with perf in a future.

I think I was unnecessarily worry about the best way.  I added a tracepoint to
the monitoring result writing logic of DAMON.  Other tracers including perf
would easily integrate DAMON using it, though the tracepoint is not highly
optimized yet.  Will post next version soon.


Thanks,
SeongJae Park

> 
> 
> Thanks,
> SeongJae Park
> 
> > 
> > -- 
> >  Kirill A. Shutemov
> > 
> >