[RFC PATCH 00/16] mm/damon: introduce data attributes only monitoring

[RFC PATCH 00/16] mm/damon: introduce data attributes only monitoring

[SJ Park]

TL;DR: Introduce a way to get DAMON's best effort accuracy monitoring of
user-demanding non-access data attributes.

Background
==========

DAMON was initially designed for only access monitoring.  It turned out
users want to get the information together with more data attributes.
For example, some users want to know how much of a hot memory region
belongs to huge pages or specific cgroups.  Page level properties based
monitoring was introduced with commit 626ffabe67c2 ("mm/damon: clarify
trying vs applying on damos_stat kernel-doc comment") to fill the gap.
Because it works only at snapshot level and snapshot capturing in the
mode can induce high overhead, commit 45c49d9fd608 ("mm/damon/core:
introduce struct damon_probe") introduced data attributes monitoring.

Data attributes monitoring treats the attributes as only additional and
subordinate information.  Data access monitoring is always turned on,
and regions are adjusted for best accuracy of the access information. In
some cases, users may be primarily interested in the attributes more
than the access.  They might even not care about the access information
at all.  Because DAMON treats data accesses as the only primary
information, such users cannot get high quality attributes information.

Design and Implementation
=========================

Introduce another way for treating data attributes as the primary
information.  Add 'weight' property to each data attribute probe.  When
any of the weights are set, the mode is enabled.  Data access monitoring
is completely turned off in the mode.  For region  adjustment, the
weighted sum of probe hit counters is used instead of the nr_accesses.

Using the weights, users can specify to what attributes they are
interested in to what degree.  DAMON will adjust the regions and provide
the best-effort quality monitoring that is optimized for the user
demands.

Extend damon_operations for efficient use of probe hits.  Update regions
merge and kdamond main logic to support the new mode.  Add a new struct
field and a sysfs file for API callers and ABI users, respectively.

Test
====

On ~7 GiB memory idle system, run a simple AI-assisted program.  The
program allocates and faults 2 GiB anonymous pages.  Then, it does
nothing but wait until the user terminates it.  Hence, the system ~2 GiB
of anonymous pages with no active accesses.

Monitor the distribution of the anonymous pages using DAMON attributes
monitoring mode, using DAMON user-space tool, damo [1].

    $ sudo ./damo start --probe_filter allow anon
    $ sudo ./damo report access --dont_merge_regions
    heatmap: 00000000000000000000000000000000000000000000000399999995111111146666666666666666
    # min/max temperatures: -2,470,000,000, -1,620,000,000, column size: 99.800 MiB
    intervals: sample 5 ms aggr 100 ms (max access hz 200)
    #   <start>      <size>       <freq> <age>         <probe hits>
    0   4.000 KiB    79.840 MiB   0 hz   24.700 s      2
    1   79.844 MiB   718.562 MiB  0 hz   24.700 s      8
    2   798.406 MiB  793.148 MiB  0 hz   24.700 s      7
    3   1.554 GiB    797.828 MiB  0 hz   24.700 s      7
    4   2.333 GiB    794.668 MiB  0 hz   24.600 s      8
    5   3.109 GiB    791.117 MiB  0 hz   24.500 s      0
    6   3.882 GiB    785.312 MiB  0 hz   24 s          2
    7   4.649 GiB    787.867 MiB  0 hz   16.200 s      6
    8   5.418 GiB    784.477 MiB  0 hz   23.300 s      6
    9   6.184 GiB    783.820 MiB  0 hz   18.200 s      9
    10  6.950 GiB    797.730 MiB  0 hz   18.900 s      7
    11  7.729 GiB    69.625 MiB   0 hz   18.900 s      0
    memory bw estimate: 0 B per second
    total size: 7.797 GiB
    record DAMON intervals: sample 5 ms, aggr 100 ms

Note that the line after the line starting with "intervals:" is not
provided by the current version of 'damo'.  I manually added the legends
line for easier understanding of these results.

Each of the 12 lines after the legend line shows the DAMON-found
regions.  Each line shows 1) index of the region, 2) start address of
the region, 3) size of the region, 4) access frequency of the region, 5)
age (how long the access frequency on the region was kept) of the
region, and finally 6) the probe hit count.

Because data access is the primary information that adjusts region for,
and there is only nearly zero access on the system, regions are naively
adjusted with the same size.  Still <probe hits> show different
distribution of the anonymous pages, but it is obviously very rough
information.

Switch to the attributes only mode and show how it changes the picture:

    $ sudo ./damo tune --probe_filter allow anon --probe_weight 100
    $ sudo ./damo report access --dont_merge_regions
    heatmap: 88888888888888888889888999999889999999000004888888888888889999988888898888888888
    # min/max temperatures: -4,430,000,000, 0, column size: 99.800 MiB
    intervals: sample 5 ms aggr 100 ms (max access hz 200)
    #   <start>      <size>       <freq> <age>         <probe hits>
    0   4.000 KiB    60.445 MiB   0 hz   700 ms        0
    1   60.449 MiB   1.363 MiB    0 hz   600 ms        18
    2   61.812 MiB   144.000 KiB  0 hz   0 ns          1
    3   61.953 MiB   1.922 MiB    0 hz   2.400 s       19
    4   63.875 MiB   12.133 MiB   0 hz   200 ms        0
    [...]
    500 5.132 GiB    8.000 KiB    0 hz   2 m 15.800 s  20
    501 5.132 GiB    8.000 KiB    0 hz   2 m 16.200 s  0
    502 5.132 GiB    16.000 KiB   0 hz   2 m 16.900 s  20
    503 5.132 GiB    24.000 KiB   0 hz   2 m 14.200 s  0
    504 5.132 GiB    8.000 KiB    0 hz   2 m 14.900 s  20
    [...]
    923 7.534 GiB    126.637 MiB  0 hz   0 ns          6
    924 7.658 GiB    252.000 KiB  0 hz   54.800 s      20
    925 7.658 GiB    142.242 MiB  0 hz   300 ms        0
    memory bw estimate: 0 B per second
    total size: 7.797 GiB
    record DAMON intervals: sample 5 ms, aggr 100 ms

As expected, regions are adjusted to provide the best accurate picture
for the anonymous pages distribution (<probe hits>).  The region 0
(60.445 MiB memory from the address 4.000 KiB) has nearly zero anonymous
pages.  The region 1 (1.363 MiB memory from the address 60.449 MiB) is
nearly full with anonymous pages.  Region 500 (8 KiB memory from the
address 5.132 GiB) is certainly two anonymous pages.

Future Work
===========

Attributes only monitoring disables access monitoring.  We will enable
that in future, by extending the supported attributes to include data
accesses.  This patch series, and the future work are parts of the
ongoing project [2] for extending DAMON.  The project aims to extend
DAMON with primitives other than page table accessed bits such as AMD
IBS, Intel PEBS, and Arm SPE, to provide more powerful and detailed
information like per-CPUs/threads/reads/writes monitoring.

Patches Sequence
================

Patch 1 introduces damon_probe->weight for specifying the weights of
each attribute.  Patches 2-6 extends apply_probe() damon_ops callback to
efficiently support the new mode.  Patch 7 extends regions merge
function to work with probe hits in the mode.  Patch 7 also introduces
the function for detecting the mode enablement but always returns false,
for safe and incremental changes.  Patches 8-11 incrementally update
kdamond_fn() to support the mode.  Patch 12 completes the mode detection
function implementation, so that the new mode really works.  Patch 13
introduces a new sysfs file for ABI users.  Finally, patches 14-16
respectively updates design, usage and ABI documents for the new feature
and interfaces.

[1] https://github.com/damonitor/damo
[2] https://lore.kernel.org/20260525225208.1179-1-sj@kernel.org/

SJ Park (16):
  mm/damon/core: introduce damon_probe->weight
  mm/damon/core: ask apply_probes() ops callback to set sampling address
  mm/damon/paddr: set samples in apply_probes() if requested
  mm/damon/core: ask apply_probe() to return max probe hits weighted sum
  mm/damon/core: implement damon_probe_hits_wsum()
  mm/damon/paddr: respect return_max_wsum
  mm/damon/core: extend merge function to work with probe hits
  mm/damon/core: disable access monitoring when probe weights are set
  mm/damon/core: set samples in apply_probes() if probe weights are set
  mm/damon/core: s/max_nr_accesses/max_merge_score/ in kdamond_fn()
  mm/damon/core: get merge threshold from probe hits when weights are
    set
  mm/damon/core: implement damon_has_probe_weight()
  mm/damon/sysfs: implement probe/weight file
  Docs/mm/damon/design: document attrs-only monitoring
  Docs/admin-guide/mm/damon/usage: document weight sysfs file
  Docs/ABI/damon: document probe weight file

 .../ABI/testing/sysfs-kernel-mm-damon         |   6 +
 Documentation/admin-guide/mm/damon/usage.rst  |  11 +-
 Documentation/mm/damon/design.rst             |  25 ++++
 include/linux/damon.h                         |  11 +-
 mm/damon/core.c                               | 107 +++++++++++++++---
 mm/damon/paddr.c                              |  11 +-
 mm/damon/sysfs.c                              |  25 ++++
 mm/damon/tests/core-kunit.h                   |  13 ++-
 8 files changed, 184 insertions(+), 25 deletions(-)


base-commit: 3783e23372c397a29574c34283e1f07aa420db74
-- 
2.47.3

[RFC PATCH 01/16] mm/damon/core: introduce damon_probe->weight

[SJ Park]

Add a new field, weight to damon_probe struct.  The field is used to
specify the degree of the API caller's interest to the data attribute
of the probe.

Signed-off-by: SJ Park <sj@kernel.org>
---
 include/linux/damon.h | 1 +
 mm/damon/core.c       | 3 +++
 2 files changed, 4 insertions(+)

diff --git a/include/linux/damon.h b/include/linux/damon.h
index 19b7e839bde0b..c9c1de7ec0e7b 100644
--- a/include/linux/damon.h
+++ b/include/linux/damon.h
@@ -764,6 +764,7 @@ struct damon_filter {
  * @list:	Siblings list.
  */
 struct damon_probe {
+	unsigned int weight;
 	struct list_head filters;
 	struct list_head list;
 };
diff --git a/mm/damon/core.c b/mm/damon/core.c
index 390e00b3685ef..ca5c3f3e45fc6 100644
--- a/mm/damon/core.c
+++ b/mm/damon/core.c
@@ -166,6 +166,7 @@ struct damon_probe *damon_new_probe(void)
 	p = kmalloc_obj(*p);
 	if (!p)
 		return NULL;
+	p->weight = 0;
 	INIT_LIST_HEAD(&p->filters);
 	INIT_LIST_HEAD(&p->list);
 	return p;
@@ -1641,6 +1642,7 @@ static int damon_commit_probes(struct damon_ctx *dst, struct damon_ctx *src)
 	damon_for_each_probe_safe(dst_probe, next, dst) {
 		src_probe = damon_nth_probe(i++, src);
 		if (src_probe) {
+			dst_probe->weight = src_probe->weight;
 			err = damon_commit_filters(dst_probe, src_probe);
 			if (err)
 				return err;
@@ -1657,6 +1659,7 @@ static int damon_commit_probes(struct damon_ctx *dst, struct damon_ctx *src)
 		if (!new_probe)
 			return -ENOMEM;
 		damon_add_probe(dst, new_probe);
+		new_probe->weight = src_probe->weight;
 		err = damon_commit_filters(new_probe, src_probe);
 		if (err)
 			return err;
-- 
2.47.3

[RFC PATCH 02/16] mm/damon/core: ask apply_probes() ops callback to set sampling address

[SJ Park]

prepare_access_checks() DAMON ops callback sets the monitoring sampling
address per region.  In future, DAMON will be able to call only
apply_probes().  In this case, applyy_probes() may need to do the
sampling address setup, to minimize unnecessary regions iteration.
Update the protocol for the request.

Signed-off-by: SJ Park <sj@kernel.org>
---
 include/linux/damon.h | 5 +++--
 mm/damon/core.c       | 2 +-
 mm/damon/paddr.c      | 2 +-
 3 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/include/linux/damon.h b/include/linux/damon.h
index c9c1de7ec0e7b..4dec5d314c782 100644
--- a/include/linux/damon.h
+++ b/include/linux/damon.h
@@ -647,7 +647,8 @@ enum damon_ops_id {
  * It should also return max number of observed accesses that made as a result
  * of its update.  The value will be used for regions adjustment threshold.
  * @apply_probes should apply the data attribute probes to each region and
- * accordingly update the probe hits counter of the region.
+ * accordingly update the probe hits counter of the region.  It should also
+ * &set sampling_addr or each region if &set_samples is true.
  * @get_scheme_score should return the priority score of a region for a scheme
  * as an integer in [0, &DAMOS_MAX_SCORE].
  * @apply_scheme is called from @kdamond when a region for user provided
@@ -665,7 +666,7 @@ struct damon_operations {
 	void (*update)(struct damon_ctx *context);
 	void (*prepare_access_checks)(struct damon_ctx *context);
 	unsigned int (*check_accesses)(struct damon_ctx *context);
-	void (*apply_probes)(struct damon_ctx *context);
+	void (*apply_probes)(struct damon_ctx *context, bool set_samples);
 	int (*get_scheme_score)(struct damon_ctx *context,
 			struct damon_region *r, struct damos *scheme);
 	unsigned long (*apply_scheme)(struct damon_ctx *context,
diff --git a/mm/damon/core.c b/mm/damon/core.c
index ca5c3f3e45fc6..7c0184cb78a50 100644
--- a/mm/damon/core.c
+++ b/mm/damon/core.c
@@ -3675,7 +3675,7 @@ static int kdamond_fn(void *data)
 		if (ctx->ops.check_accesses)
 			max_nr_accesses = ctx->ops.check_accesses(ctx);
 		if (ctx->ops.apply_probes)
-			ctx->ops.apply_probes(ctx);
+			ctx->ops.apply_probes(ctx, false);
 
 		if (time_after_eq(ctx->passed_sample_intervals,
 					next_aggregation_sis)) {
diff --git a/mm/damon/paddr.c b/mm/damon/paddr.c
index 6172464981730..b13bf7c6eade3 100644
--- a/mm/damon/paddr.c
+++ b/mm/damon/paddr.c
@@ -166,7 +166,7 @@ static bool damon_pa_filter_pass(phys_addr_t pa, struct folio *folio,
 	return pass;
 }
 
-static void damon_pa_apply_probes(struct damon_ctx *ctx)
+static void damon_pa_apply_probes(struct damon_ctx *ctx, bool set_samples)
 {
 	struct damon_target *t;
 	struct damon_region *r;
-- 
2.47.3

[RFC PATCH 03/16] mm/damon/paddr: set samples in apply_probes() if requested

[SJ Park]

apply_probe() callback implementation in DAMON_PADDR is ignoring
set_samples parameter.  Respect it.

Signed-off-by: SJ Park <sj@kernel.org>
---
 mm/damon/paddr.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/mm/damon/paddr.c b/mm/damon/paddr.c
index b13bf7c6eade3..ff4fcca949660 100644
--- a/mm/damon/paddr.c
+++ b/mm/damon/paddr.c
@@ -178,6 +178,10 @@ static void damon_pa_apply_probes(struct damon_ctx *ctx, bool set_samples)
 			phys_addr_t pa;
 			struct folio *folio;
 
+			if (set_samples)
+				r->sampling_addr = damon_rand(ctx, r->ar.start,
+						r->ar.end);
+
 			pa = damon_pa_phys_addr(r->sampling_addr,
 					ctx->addr_unit);
 			folio = damon_get_folio(PHYS_PFN(pa));
-- 
2.47.3

[RFC PATCH 04/16] mm/damon/core: ask apply_probe() to return max probe hits weighted sum

[SJ Park]

check_accesses() DAMON ops callback returns the maximum nr_accesses of
regions.  DAMON core uses it to calculate a reasonable region merge
threshold.  The core will need to adjust regions for not nr_accesses but
probe hits weighted sum in future.  For that, the core needs to know the
maximum weighted sum of the regions.  Update the protocol for the task.

Signed-off-by: SJ Park <sj@kernel.org>
---
 include/linux/damon.h | 7 +++++--
 mm/damon/core.c       | 2 +-
 mm/damon/paddr.c      | 4 +++-
 3 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/include/linux/damon.h b/include/linux/damon.h
index 4dec5d314c782..62caa67a3c40c 100644
--- a/include/linux/damon.h
+++ b/include/linux/damon.h
@@ -648,7 +648,9 @@ enum damon_ops_id {
  * of its update.  The value will be used for regions adjustment threshold.
  * @apply_probes should apply the data attribute probes to each region and
  * accordingly update the probe hits counter of the region.  It should also
- * &set sampling_addr or each region if &set_samples is true.
+ * &set sampling_addr or each region if &set_samples is true.  It should also
+ * return maximum probe hits weighted sum of regions if &return_max_wsum is
+ * true.
  * @get_scheme_score should return the priority score of a region for a scheme
  * as an integer in [0, &DAMOS_MAX_SCORE].
  * @apply_scheme is called from @kdamond when a region for user provided
@@ -666,7 +668,8 @@ struct damon_operations {
 	void (*update)(struct damon_ctx *context);
 	void (*prepare_access_checks)(struct damon_ctx *context);
 	unsigned int (*check_accesses)(struct damon_ctx *context);
-	void (*apply_probes)(struct damon_ctx *context, bool set_samples);
+	unsigned int (*apply_probes)(struct damon_ctx *context,
+			bool set_samples, bool return_max_wsum);
 	int (*get_scheme_score)(struct damon_ctx *context,
 			struct damon_region *r, struct damos *scheme);
 	unsigned long (*apply_scheme)(struct damon_ctx *context,
diff --git a/mm/damon/core.c b/mm/damon/core.c
index 7c0184cb78a50..e42f512d87a11 100644
--- a/mm/damon/core.c
+++ b/mm/damon/core.c
@@ -3675,7 +3675,7 @@ static int kdamond_fn(void *data)
 		if (ctx->ops.check_accesses)
 			max_nr_accesses = ctx->ops.check_accesses(ctx);
 		if (ctx->ops.apply_probes)
-			ctx->ops.apply_probes(ctx, false);
+			ctx->ops.apply_probes(ctx, false, false);
 
 		if (time_after_eq(ctx->passed_sample_intervals,
 					next_aggregation_sis)) {
diff --git a/mm/damon/paddr.c b/mm/damon/paddr.c
index ff4fcca949660..3eba3564a31fc 100644
--- a/mm/damon/paddr.c
+++ b/mm/damon/paddr.c
@@ -166,7 +166,8 @@ static bool damon_pa_filter_pass(phys_addr_t pa, struct folio *folio,
 	return pass;
 }
 
-static void damon_pa_apply_probes(struct damon_ctx *ctx, bool set_samples)
+static unsigned int damon_pa_apply_probes(struct damon_ctx *ctx,
+		bool set_samples, bool return_max_wsum)
 {
 	struct damon_target *t;
 	struct damon_region *r;
@@ -194,6 +195,7 @@ static void damon_pa_apply_probes(struct damon_ctx *ctx, bool set_samples)
 				folio_put(folio);
 		}
 	}
+	return 0;
 }
 
 /*
-- 
2.47.3

[RFC PATCH 05/16] mm/damon/core: implement damon_probe_hits_wsum()

[SJ Park]

When damon_probe->weight is set, the weighted sum of probe hits will be
useful.  It will be useful for not only the users but also DAMON
internal logics like regions merging.  Implement a function for
calculating it.

Signed-off-by: SJ Park <sj@kernel.org>
---
 include/linux/damon.h |  2 ++
 mm/damon/core.c       | 24 ++++++++++++++++++++++++
 2 files changed, 26 insertions(+)

diff --git a/include/linux/damon.h b/include/linux/damon.h
index 62caa67a3c40c..c43986a3aa16c 100644
--- a/include/linux/damon.h
+++ b/include/linux/damon.h
@@ -1011,6 +1011,8 @@ unsigned int damon_nr_accesses_mvsum(struct damon_region *r,
 		struct damon_ctx *ctx);
 unsigned char damon_probe_hits_mvsum(int probe_idx, struct damon_region *r,
 		struct damon_ctx *ctx);
+unsigned int damon_probe_hits_wsum(struct damon_region *r, bool last,
+		struct damon_ctx *ctx);
 
 int damon_set_regions(struct damon_target *t, struct damon_addr_range *ranges,
 		unsigned int nr_ranges, unsigned long min_region_sz);
diff --git a/mm/damon/core.c b/mm/damon/core.c
index e42f512d87a11..d7d1e27d767eb 100644
--- a/mm/damon/core.c
+++ b/mm/damon/core.c
@@ -390,6 +390,30 @@ static bool damon_is_last_region(struct damon_region *r,
 	return list_is_last(&r->list, &t->regions_list);
 }
 
+/**
+ * damon_probe_hits_wsum() - Returns probe hits weighted sum of a region.
+ * @r:		region to get the weigted sum of.
+ * @last:	if the request is for last-window agggregated probe hits.
+ * @ctx:	context of &r.
+ *
+ * Return: the weighted sum of probe hits of the region.
+ */
+unsigned int damon_probe_hits_wsum(struct damon_region *r, bool last,
+		struct damon_ctx *ctx)
+{
+	struct damon_probe *probe;
+	unsigned int sum = 0;
+	int i = 0;
+
+	damon_for_each_probe(probe, ctx) {
+		if (last)
+			sum += r->last_probe_hits[i++] * probe->weight;
+		else
+			sum += r->probe_hits[i++] * probe->weight;
+	}
+	return sum;
+}
+
 /*
  * Check whether a region is intersecting an address range
  *
-- 
2.47.3

[RFC PATCH 06/16] mm/damon/paddr: respect return_max_wsum

[SJ Park]

apply_probes() ops implementation in DAMON_PADDR is ignoring
return_max_wsum.  Respect it.

Signed-off-by: SJ Park <sj@kernel.org>
---
 mm/damon/paddr.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/mm/damon/paddr.c b/mm/damon/paddr.c
index 3eba3564a31fc..b85f88a7a38f4 100644
--- a/mm/damon/paddr.c
+++ b/mm/damon/paddr.c
@@ -172,6 +172,7 @@ static unsigned int damon_pa_apply_probes(struct damon_ctx *ctx,
 	struct damon_target *t;
 	struct damon_region *r;
 	struct damon_probe *p;
+	unsigned int max_wsum = 0;
 
 	damon_for_each_target(t, ctx) {
 		damon_for_each_region(r, t) {
@@ -182,7 +183,6 @@ static unsigned int damon_pa_apply_probes(struct damon_ctx *ctx,
 			if (set_samples)
 				r->sampling_addr = damon_rand(ctx, r->ar.start,
 						r->ar.end);
-
 			pa = damon_pa_phys_addr(r->sampling_addr,
 					ctx->addr_unit);
 			folio = damon_get_folio(PHYS_PFN(pa));
@@ -193,9 +193,12 @@ static unsigned int damon_pa_apply_probes(struct damon_ctx *ctx,
 			}
 			if (folio)
 				folio_put(folio);
+			if (return_max_wsum)
+				max_wsum = max(damon_probe_hits_wsum(r, false,
+							ctx), max_wsum);
 		}
 	}
-	return 0;
+	return max_wsum;
 }
 
 /*
-- 
2.47.3

[RFC PATCH 07/16] mm/damon/core: extend merge function to work with probe hits

[SJ Park]

When probe weights are set, users may want DAMON monitoring results to
be optimized for the weights.  For that, regions adjustment should work
for the weighted sum of probe hits.  Extend damon_merge_regions_of() to
detect if the weights are set, and work with probe hits in the case.

The weights setup detection function is incomplete.  It always returns
false.  It is intentional, so that more changes to completely support
weights can be made in an incremental but safe way.  Until the function
is completed, all changes depend on it is no-op, so DAMON works in the
current mode.

Signed-off-by: SJ Park <sj@kernel.org>
---
 mm/damon/core.c             | 48 +++++++++++++++++++++++++++++--------
 mm/damon/tests/core-kunit.h | 13 ++++++++--
 2 files changed, 49 insertions(+), 12 deletions(-)

diff --git a/mm/damon/core.c b/mm/damon/core.c
index d7d1e27d767eb..9bb884ab135ec 100644
--- a/mm/damon/core.c
+++ b/mm/damon/core.c
@@ -209,6 +209,11 @@ static struct damon_probe *damon_nth_probe(int n, struct damon_ctx *ctx)
 	return NULL;
 }
 
+static bool damon_has_probe_weights(struct damon_ctx *c)
+{
+	return false;
+}
+
 /*
  * damon_mvsum() - Returns pseudo moving sum value for a time window.
  * @current_nr:		The value of the current aggregation window.
@@ -3231,6 +3236,16 @@ static void damon_merge_two_regions(struct damon_target *t,
 	damon_destroy_region(r, t);
 }
 
+static unsigned int damon_merge_score(struct damon_region *r, bool last,
+		struct damon_ctx *ctx, bool use_probe_hits)
+{
+	if (use_probe_hits)
+		return damon_probe_hits_wsum(r, last, ctx);
+	if (last)
+		return r->last_nr_accesses;
+	return r->nr_accesses;
+}
+
 /*
  * Merge adjacent regions having similar access frequencies
  *
@@ -3239,24 +3254,37 @@ static void damon_merge_two_regions(struct damon_target *t,
  * sz_limit	size upper limit of each region
  */
 static void damon_merge_regions_of(struct damon_target *t, unsigned int thres,
-				   unsigned long sz_limit)
+		unsigned long sz_limit, struct damon_ctx *ctx)
 {
 	struct damon_region *r, *prev = NULL, *next;
+	bool use_probe_hits = damon_has_probe_weights(ctx);
 
 	damon_for_each_region_safe(r, next, t) {
-		if (abs(r->nr_accesses - r->last_nr_accesses) > thres)
+		unsigned int score, last_score;
+
+		score = damon_merge_score(r, false, ctx, use_probe_hits);
+		last_score = damon_merge_score(r, true, ctx, use_probe_hits);
+
+		if (abs(score - last_score) > thres)
 			r->age = 0;
-		else if ((r->nr_accesses == 0) != (r->last_nr_accesses == 0))
+		else if ((score == 0) != (last_score == 0))
 			r->age = 0;
 		else
 			r->age++;
 
-		if (prev && prev->ar.end == r->ar.start &&
-		    abs(prev->nr_accesses - r->nr_accesses) <= thres &&
-		    damon_sz_region(prev) + damon_sz_region(r) <= sz_limit)
-			damon_merge_two_regions(t, prev, r);
-		else
-			prev = r;
+		if (!prev)
+			goto set_prev_continue;
+		if (prev->ar.end != r->ar.start)
+			goto set_prev_continue;
+		if (abs(damon_merge_score(prev, false, ctx, use_probe_hits) -
+					score) > thres)
+			goto set_prev_continue;
+		if (damon_sz_region(prev) + damon_sz_region(r) > sz_limit)
+			goto set_prev_continue;
+		damon_merge_two_regions(t, prev, r);
+		continue;
+set_prev_continue:
+		prev = r;
 	}
 }
 
@@ -3289,7 +3317,7 @@ static void kdamond_merge_regions(struct damon_ctx *c, unsigned int threshold,
 	do {
 		nr_regions = 0;
 		damon_for_each_target(t, c) {
-			damon_merge_regions_of(t, threshold, sz_limit);
+			damon_merge_regions_of(t, threshold, sz_limit, c);
 			nr_regions += damon_nr_regions(t);
 		}
 		threshold = max(1, threshold * 2);
diff --git a/mm/damon/tests/core-kunit.h b/mm/damon/tests/core-kunit.h
index 0124f83b39b83..fe948000b72cf 100644
--- a/mm/damon/tests/core-kunit.h
+++ b/mm/damon/tests/core-kunit.h
@@ -226,6 +226,7 @@ static struct damon_region *__nth_region_of(struct damon_target *t, int idx)
 
 static void damon_test_merge_regions_of(struct kunit *test)
 {
+	struct damon_ctx *ctx;
 	struct damon_target *t;
 	struct damon_region *r;
 	unsigned long sa[] = {0, 100, 114, 122, 130, 156, 170, 184, 230};
@@ -236,20 +237,27 @@ static void damon_test_merge_regions_of(struct kunit *test)
 	unsigned long eaddrs[] = {112, 130, 156, 170, 230, 10170};
 	int i;
 
+	ctx = damon_new_ctx();
+	if (!ctx)
+		kunit_skip(test, "ctx alloc fail");
+
 	t = damon_new_target();
-	if (!t)
+	if (!t) {
+		damon_destroy_ctx(ctx);
 		kunit_skip(test, "target alloc fail");
+	}
 	for (i = 0; i < ARRAY_SIZE(sa); i++) {
 		r = damon_new_region(sa[i], ea[i]);
 		if (!r) {
 			damon_free_target(t);
+			damon_destroy_ctx(ctx);
 			kunit_skip(test, "region alloc fail");
 		}
 		r->nr_accesses = nrs[i];
 		damon_add_region(r, t);
 	}
 
-	damon_merge_regions_of(t, 9, 9999);
+	damon_merge_regions_of(t, 9, 9999, ctx);
 	/* 0-112, 114-130, 130-156, 156-170, 170-230, 230-10170 */
 	KUNIT_EXPECT_EQ(test, damon_nr_regions(t), 6u);
 	for (i = 0; i < 6; i++) {
@@ -258,6 +266,7 @@ static void damon_test_merge_regions_of(struct kunit *test)
 		KUNIT_EXPECT_EQ(test, r->ar.end, eaddrs[i]);
 	}
 	damon_free_target(t);
+	damon_destroy_ctx(ctx);
 }
 
 static void damon_test_split_regions_of(struct kunit *test)
-- 
2.47.3

[RFC PATCH 08/16] mm/damon/core: disable access monitoring when probe weights are set

[SJ Park]

When users set probe weights, they are interested in the data
attributes.  Not necessarily they are interested in data access
patterns.  Disable the data access monitoring in the case.

Signed-off-by: SJ Park <sj@kernel.org>
---
 mm/damon/core.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/mm/damon/core.c b/mm/damon/core.c
index 9bb884ab135ec..ad0f28469e633 100644
--- a/mm/damon/core.c
+++ b/mm/damon/core.c
@@ -3714,17 +3714,18 @@ static int kdamond_fn(void *data)
 		unsigned long next_aggregation_sis = ctx->next_aggregation_sis;
 		unsigned long next_ops_update_sis = ctx->next_ops_update_sis;
 		unsigned long sample_interval = ctx->attrs.sample_interval;
+		bool access_check_disabled = damon_has_probe_weights(ctx);
 
 		if (kdamond_wait_activation(ctx))
 			break;
 
-		if (ctx->ops.prepare_access_checks)
+		if (!access_check_disabled && ctx->ops.prepare_access_checks)
 			ctx->ops.prepare_access_checks(ctx);
 
 		kdamond_usleep(sample_interval);
 		ctx->passed_sample_intervals++;
 
-		if (ctx->ops.check_accesses)
+		if (!access_check_disabled && ctx->ops.check_accesses)
 			max_nr_accesses = ctx->ops.check_accesses(ctx);
 		if (ctx->ops.apply_probes)
 			ctx->ops.apply_probes(ctx, false, false);
-- 
2.47.3

[RFC PATCH 09/16] mm/damon/core: set samples in apply_probes() if probe weights are set

[SJ Park]

When probe weights are set, the prepare_access_checks() ops callback is
not invoked, so sampling addresses of regions are unset.  Ask
apply_probes() to set those together in the case.

Signed-off-by: SJ Park <sj@kernel.org>
---
 mm/damon/core.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/mm/damon/core.c b/mm/damon/core.c
index ad0f28469e633..71008ee6ca243 100644
--- a/mm/damon/core.c
+++ b/mm/damon/core.c
@@ -3728,7 +3728,8 @@ static int kdamond_fn(void *data)
 		if (!access_check_disabled && ctx->ops.check_accesses)
 			max_nr_accesses = ctx->ops.check_accesses(ctx);
 		if (ctx->ops.apply_probes)
-			ctx->ops.apply_probes(ctx, false, false);
+			ctx->ops.apply_probes(ctx, access_check_disabled,
+					false);
 
 		if (time_after_eq(ctx->passed_sample_intervals,
 					next_aggregation_sis)) {
-- 
2.47.3

[RFC PATCH 10/16] mm/damon/core: s/max_nr_accesses/max_merge_score/ in kdamond_fn()

[SJ Park]

When probe weights are set, DAMON merge logic works with the probe hits
weighted sum.  The core logic is setting the merge threshold based on
the access frequency, though.  Before using probe hits based threshold,
rename the variables to look more general.

Signed-off-by: SJ Park <sj@kernel.org>
---
 mm/damon/core.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/mm/damon/core.c b/mm/damon/core.c
index 71008ee6ca243..a7ca69ea8e565 100644
--- a/mm/damon/core.c
+++ b/mm/damon/core.c
@@ -3681,7 +3681,6 @@ static void kdamond_init_ctx(struct damon_ctx *ctx)
 static int kdamond_fn(void *data)
 {
 	struct damon_ctx *ctx = data;
-	unsigned int max_nr_accesses = 0;
 	unsigned long sz_limit = 0;
 
 	pr_debug("kdamond (%d) starts\n", current->pid);
@@ -3715,6 +3714,7 @@ static int kdamond_fn(void *data)
 		unsigned long next_ops_update_sis = ctx->next_ops_update_sis;
 		unsigned long sample_interval = ctx->attrs.sample_interval;
 		bool access_check_disabled = damon_has_probe_weights(ctx);
+		unsigned int max_merge_score = 0;
 
 		if (kdamond_wait_activation(ctx))
 			break;
@@ -3726,7 +3726,7 @@ static int kdamond_fn(void *data)
 		ctx->passed_sample_intervals++;
 
 		if (!access_check_disabled && ctx->ops.check_accesses)
-			max_nr_accesses = ctx->ops.check_accesses(ctx);
+			max_merge_score = ctx->ops.check_accesses(ctx);
 		if (ctx->ops.apply_probes)
 			ctx->ops.apply_probes(ctx, access_check_disabled,
 					false);
@@ -3734,7 +3734,7 @@ static int kdamond_fn(void *data)
 		if (time_after_eq(ctx->passed_sample_intervals,
 					next_aggregation_sis)) {
 			kdamond_merge_regions(ctx,
-					max_nr_accesses / 10,
+					max_merge_score / 10,
 					sz_limit);
 			/* online updates might be made */
 			sz_limit = damon_apply_min_nr_regions(ctx);
-- 
2.47.3

[RFC PATCH 11/16] mm/damon/core: get merge threshold from probe hits when weights are set

[SJ Park]

When probe weights are set, DAMON merges regions based on their probe
hits weighted sum.  But the merge threshold is calculated based on the
access frequency.  Update it to retrieve the maximum probe hits weighted
sum in the snapshot from apply_probes() ops callback, and generate the
threshold based on it.

Signed-off-by: SJ Park <sj@kernel.org>
---
 mm/damon/core.c | 18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

diff --git a/mm/damon/core.c b/mm/damon/core.c
index a7ca69ea8e565..83db073c74e2b 100644
--- a/mm/damon/core.c
+++ b/mm/damon/core.c
@@ -3714,7 +3714,8 @@ static int kdamond_fn(void *data)
 		unsigned long next_ops_update_sis = ctx->next_ops_update_sis;
 		unsigned long sample_interval = ctx->attrs.sample_interval;
 		bool access_check_disabled = damon_has_probe_weights(ctx);
-		unsigned int max_merge_score = 0;
+		unsigned int max_merge_score = 0, max_wsum;
+		bool get_max_wsum;
 
 		if (kdamond_wait_activation(ctx))
 			break;
@@ -3727,9 +3728,18 @@ static int kdamond_fn(void *data)
 
 		if (!access_check_disabled && ctx->ops.check_accesses)
 			max_merge_score = ctx->ops.check_accesses(ctx);
-		if (ctx->ops.apply_probes)
-			ctx->ops.apply_probes(ctx, access_check_disabled,
-					false);
+		if (ctx->ops.apply_probes) {
+			if (time_after_eq(ctx->passed_sample_intervals,
+						next_aggregation_sis) &&
+					access_check_disabled)
+				get_max_wsum = true;
+			else
+				get_max_wsum = false;
+			max_wsum = ctx->ops.apply_probes(ctx,
+					access_check_disabled, get_max_wsum);
+			if (get_max_wsum)
+				max_merge_score = max_wsum;
+		}
 
 		if (time_after_eq(ctx->passed_sample_intervals,
 					next_aggregation_sis)) {
-- 
2.47.3

[RFC PATCH 12/16] mm/damon/core: implement damon_has_probe_weight()

[SJ Park]

DAMON monitors only data attributes when probe weights are set.
However, the function for detecting it is intentionally returning false
always.  The intention was to make changes for complete support of the
mode in a safe and incremental way.  Now all the building blocks are
made.  Complete the mode detection function so that the mode can really
start working.

Signed-off-by: SJ Park <sj@kernel.org>
---
 mm/damon/core.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/mm/damon/core.c b/mm/damon/core.c
index 83db073c74e2b..b44393ccaf6b1 100644
--- a/mm/damon/core.c
+++ b/mm/damon/core.c
@@ -211,6 +211,12 @@ static struct damon_probe *damon_nth_probe(int n, struct damon_ctx *ctx)
 
 static bool damon_has_probe_weights(struct damon_ctx *c)
 {
+	struct damon_probe *p;
+
+	damon_for_each_probe(p, c) {
+		if (p->weight)
+			return true;
+	}
 	return false;
 }
 
-- 
2.47.3

[RFC PATCH 13/16] mm/damon/sysfs: implement probe/weight file

[SJ Park]

Implement a new sysfs file, 'weight', under probe directory.  Users will
be able to set the probe weight and enable the attributes only
monitoring using it.

Signed-off-by: SJ Park <sj@kernel.org>
---
 mm/damon/sysfs.c | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)

diff --git a/mm/damon/sysfs.c b/mm/damon/sysfs.c
index 6d7e36c9e509d..b5fe036f78015 100644
--- a/mm/damon/sysfs.c
+++ b/mm/damon/sysfs.c
@@ -1065,6 +1065,7 @@ static const struct kobj_type damon_sysfs_filters_ktype = {
 
 struct damon_sysfs_probe {
 	struct kobject kobj;
+	unsigned int weight;
 	struct damon_sysfs_filters *filters;
 };
 
@@ -1100,12 +1101,35 @@ static void damon_sysfs_probe_rm_dirs(struct damon_sysfs_probe *probe)
 	}
 }
 
+static ssize_t weight_show(struct kobject *kobj, struct kobj_attribute *attr,
+		char *buf)
+{
+	struct damon_sysfs_probe *probe = container_of(kobj,
+			struct damon_sysfs_probe, kobj);
+
+	return sysfs_emit(buf, "%u\n", probe->weight);
+}
+
+static ssize_t weight_store(struct kobject *kobj,
+		struct kobj_attribute *attr, const char *buf, size_t count)
+{
+	struct damon_sysfs_probe *probe = container_of(kobj,
+			struct damon_sysfs_probe, kobj);
+	int err = kstrtouint(buf, 0, &probe->weight);
+
+	return err ? err : count;
+}
+
 static void damon_sysfs_probe_release(struct kobject *kobj)
 {
 	kfree(container_of(kobj, struct damon_sysfs_probe, kobj));
 }
 
+static struct kobj_attribute damon_sysfs_probe_weight_attr =
+		__ATTR_RW_MODE(weight, 0600);
+
 static struct attribute *damon_sysfs_probe_attrs[] = {
+	&damon_sysfs_probe_weight_attr.attr,
 	NULL,
 };
 ATTRIBUTE_GROUPS(damon_sysfs_probe);
@@ -1965,6 +1989,7 @@ static int damon_sysfs_set_probes(struct damon_ctx *ctx,
 			return -ENOMEM;
 		damon_add_probe(ctx, p);
 		sys_probe = sys_probes->probes_arr[i];
+		p->weight = sys_probe->weight;
 		err = damon_sysfs_set_probe(p, sys_probe);
 		if (err)
 			return err;
-- 
2.47.3

[RFC PATCH 14/16] Docs/mm/damon/design: document attrs-only monitoring

[SJ Park]

Update DAMON design document for the attributes only monitoring mode.

Signed-off-by: SJ Park <sj@kernel.org>
---
 Documentation/mm/damon/design.rst | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/Documentation/mm/damon/design.rst b/Documentation/mm/damon/design.rst
index 457d6e8bc7878..67398e2133036 100644
--- a/Documentation/mm/damon/design.rst
+++ b/Documentation/mm/damon/design.rst
@@ -316,6 +316,26 @@ Another way to do this for higher accuracy is using :ref:`DAMOS filter
 information in page level.  But, because it is operated in page level, the
 overhead is proportional to the size of the memory.
 
+Data Attributes-only Monitoring
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Data access is the primary monitoring information for DAMON.  Hence it
+:ref:`adjusts <damon_design_data_attrs_monitoring>` regions using the access
+:ref:`counter <damon_design_region_based_sampling>` (``nr_accesses``).  In some
+use cases, however, users may want some of :ref:`attributes
+<damon_design_data_attrs_monitoring>` to be the primary information.
+
+Data attributes-only monitoring mode supports this use case.  For the mode,
+each attribute probe has their priority weight value.  Users can describe by
+what combination of the attributes the primary information is deeecided, by
+setting the priority weight value.  If the total sum of the weights is not
+zero, the mode is enabled.  The regions adjustment mechaanism uses the weighted
+sum of the :ref:`probe hit counts <damon_design_data_attrs_monitoring>` instead
+of ``nr_accesses`` in the case.  When the mode is enabled, access monitoring is
+automatically turned off.  The access counter (``nr_accesses``) will always be
+zero and not updated.  Hence the mode is called Data Attributes "only"
+monitoring.
+
 Dynamic Target Space Updates Handling
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-- 
2.47.3

[RFC PATCH 15/16] Docs/admin-guide/mm/damon/usage: document weight sysfs file

[SJ Park]

Update DAMON usage document for newly added probe weight file.

Signed-off-by: SJ Park <sj@kernel.org>
---
 Documentation/admin-guide/mm/damon/usage.rst | 11 ++++++++---
 Documentation/mm/damon/design.rst            |  5 +++++
 2 files changed, 13 insertions(+), 3 deletions(-)

diff --git a/Documentation/admin-guide/mm/damon/usage.rst b/Documentation/admin-guide/mm/damon/usage.rst
index b2649ea011f93..4515088eba968 100644
--- a/Documentation/admin-guide/mm/damon/usage.rst
+++ b/Documentation/admin-guide/mm/damon/usage.rst
@@ -73,9 +73,10 @@ comma (",").
     │ │ │ │ │ │ │ intervals_goal/access_bp,aggrs,min_sample_us,max_sample_us
     │ │ │ │ │ │ nr_regions/min,max
     │ │ │ │ │ │ :ref:`probes <damon_usage_sysfs_probes>`/nr_probes
-    │ │ │ │ │ │ │ 0/filters/nr_filters
-    │ │ │ │ │ │ │ │ 0/type,matching,allow,path
-    │ │ │ │ │ │ │ │ ...
+    │ │ │ │ │ │ │ 0/weight
+    │ │ │ │ │ │ │ │ filters/nr_filters
+    │ │ │ │ │ │ │ │ │ 0/type,matching,allow,path
+    │ │ │ │ │ │ │ │ │ ...
     │ │ │ │ │ │ │ ...
     │ │ │ │ │ :ref:`targets <sysfs_targets>`/nr_targets
     │ │ │ │ │ │ :ref:`0 <sysfs_target>`/pid_target,obsolete_target
@@ -286,6 +287,10 @@ In each probe directory, one directory, ``filters`` exists.  The directory
 contains files for installing filters for the probe, that is used to determine
 the data attribute for the probe.
 
+Each probe directory also contains ``weight`` file.  Reading from and writing
+to the file gets and sets the :ref:`atttributes-only monitoring
+<damon_design_attrs_only_monitoring>` weight for the attribute of the probe.
+
 In the beginning, ``filters`` directory has only one file, ``nr_filters``.
 Writing a number (``N``) to the file creates the number of child directories
 named ``0`` to ``N-1``.  Each directory represents each filter and works in a
diff --git a/Documentation/mm/damon/design.rst b/Documentation/mm/damon/design.rst
index 67398e2133036..129acb1da4e19 100644
--- a/Documentation/mm/damon/design.rst
+++ b/Documentation/mm/damon/design.rst
@@ -316,6 +316,8 @@ Another way to do this for higher accuracy is using :ref:`DAMOS filter
 information in page level.  But, because it is operated in page level, the
 overhead is proportional to the size of the memory.
 
+.. _damon_design_attrs_only_monitoring:
+
 Data Attributes-only Monitoring
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -336,6 +338,9 @@ automatically turned off.  The access counter (``nr_accesses``) will always be
 zero and not updated.  Hence the mode is called Data Attributes "only"
 monitoring.
 
+Refer to the :ref:`admin guide <damon_usage_sysfs_probes>` to know how users
+can use the mode.
+
 Dynamic Target Space Updates Handling
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-- 
2.47.3

[RFC PATCH 16/16] Docs/ABI/damon: document probe weight file

[SJ Park]

Update DAMON ABI document for the newly added probe weight sysfs file.

Signed-off-by: SJ Park <sj@kernel.org>
---
 Documentation/ABI/testing/sysfs-kernel-mm-damon | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/Documentation/ABI/testing/sysfs-kernel-mm-damon b/Documentation/ABI/testing/sysfs-kernel-mm-damon
index dd6b5bd76e119..907a504fb64c5 100644
--- a/Documentation/ABI/testing/sysfs-kernel-mm-damon
+++ b/Documentation/ABI/testing/sysfs-kernel-mm-damon
@@ -164,6 +164,12 @@ Description:	Writing a number 'N' to this file creates the number of
 		directories for each DAMON probe named '0' to 'N-1' under the
 		probes/ directory.
 
+What:		/sys/kernel/mm/damon/admin/kdamonds/<K>/contexts/<C>/monitoring_attrs/probes/<P>/weight
+Date:		Jul 2026
+Contact:	SJ Park <sj@kernel.org>
+Description:	Writing to and reading from this file sets and gets the
+		per-probe attribute weight.
+
 What:		/sys/kernel/mm/damon/admin/kdamonds/<K>/contexts/<C>/monitoring_attrs/probes/<P>/filters/nr_filters
 Date:		May 2026
 Contact:	SJ Park <sj@kernel.org>
-- 
2.47.3