[RFC PATCH 0/8] mm/damon: introduce hugepage_size probe filter
From: SJ Park
Date: Sun Sep 13 2026 - 13:12:58 EST
Knowing whether a given memory is backed by a hugepage of specific size
is useful for efficient utilization of hugepages. For easy monitoring
of the information, introduce a new data attribute probe filter type,
hugepage_size. It works similar to the DAMOS filter of the same name.
It works for memory that is backed by a hugepage of a given size range.
Patch 1 introduces the new probe filter type to DAMON API and extends
related data structures. Patch 2 updates probe filter commit logic to
handle the size range. Patch 3 Updates the filtering logic to support
the new type. Patch 4 adds new DAMON sysfs files for the size range.
Patch 5 updates DAMON sysfs interface to fully support the new filter
type. Patches 6-8 updates design, usage and ABI documents for the new
feature.
Test
====
Set THP always and start a test workload using masim [1].
$ echo always | sudo tee /sys/kernel/mm/transparent_hugepage/enabled
$ ./masim.py run --config_file ./configs/stairs-500mb.cfg --repeat 10
Using the DAMON user-space tool, damo [2], start DAMON with two probes.
The primary probe monitors access. The secondary probe monitors 2 MiB
sized hugepages. The region adjustment weight is given to only the
primary probe. That is, this setup is for showing access temperature
and additionally how many 2 MiB hugepages are being used for the region
of the access temperature.
$ sudo ./damo start $(pidof masim) \
--probe_prep set_pgidle --probe_filter allow pgidle_unset \
--probe_filter allow hugepage_size 2M 2M --probe_weight 1 0 \
--nr_probe_preps 1 0 --nr_probe_filters 1 1 --nr_probes 2
Show the monitoring results.
$ sudo ./damo report attrs
heatmap: 000004[...]666666668988888888888844444430000000000000000000000000000000000000000000014[...]8
# min/max temperatures: -470,000,000, 10,000,500, column size: 64.119 MiB
probes
- prep: set_pgidle, filter: allow pgidle_unset (weight: 1)
- allow hugepage_size [2.000 MiB, 2.000 MiB] (weight: 0)
intervals: sample 5 ms aggr 100 ms (max probe hits 20)
# size address age probe_hits
0 354.039 MiB 85.833 TiB 4.700 s 0 0
1 1.843 GiB 127.724 TiB 4.700 s 0 6
2 936.781 MiB 127.726 TiB 4.600 s 0 16
[...]
25 44.000 KiB 127.723 TiB 0 ns 15 20
26 11.492 MiB 127.723 TiB 0 ns 16 20
27 2.160 MiB 127.723 TiB 0 ns 19 20
[...]
47 376.000 KiB 127.723 TiB 300 ms 1 20
48 48.000 KiB 127.723 TiB 400 ms 2 20
49 812.000 KiB 127.723 TiB 500 ms 9 20
memory bw estimate: 0 B per second
total size: 5.009 GiB
record DAMON intervals: sample 5 ms, aggr 100 ms
The first probe_hits count and the age show the access temperature of
each region. The second probe_hits count shows the ratio of 2 MiB
hugepages in the region.
For example, region 2, which has 936 MiB size, is quite cold (not
accessed for 4.6 seconds) but uses 2 MiB THP for ~80% (16/20) of the
region. Maybe not a very efficient use of THP.
On the other hand, region 49, which is 812 KiB size, is somewhat hot
(accessed about 9 times per 100 ms for the last 500 ms) and fully backed
by 2 MiB hugepage. Maybe this is a good use of THP.
[1] https://github.com/sjp38/masim
[2] https://github.com/damonitor/damo
SJ Park (8):
mm/damon/api: introduce DAMON_FILTER_TYPE_HUGEPAGE_SIZE
mm/damon/core: commit hugepage_size type damon filter
mm/damon/ops-common: support hugepage_size damon filter matching
mm/damon/sysfs: add min,max files under probe filter directory
mm/damon/sysfs: support hugepage_size probe filter
Docs/mm/damon/design: update for hugepage_size probe filter
Docs/admin-guide/mm/damon/usage: update for hugepage_size
Docs/ABI/damon: update for hugepage_size probe filter
.../ABI/testing/sysfs-kernel-mm-damon | 14 +++++
Documentation/admin-guide/mm/damon/usage.rst | 4 +-
Documentation/mm/damon/design.rst | 2 +
include/linux/damon.h | 8 +++
mm/damon/core.c | 20 +++++--
mm/damon/ops-common.c | 10 ++++
mm/damon/sysfs.c | 55 +++++++++++++++++++
7 files changed, 107 insertions(+), 6 deletions(-)
base-commit: 91c120b49e83b867e29684fd229c3e1f07190071
--
2.47.3