[RFC PATCH 0/4] mm/damon/sysfs: support periodic and automated stats update

[RFC PATCH 0/4] mm/damon/sysfs: support periodic and automated stats update

[SeongJae Park]

DAMON sysfs interface provides files for reading DAMON internal status
including DAMOS stats.  The content of the files are not automatically
updated, though.  Users should manually request updates of the contents
by writing a special command to 'state' file of each kdamond directory.
This interface is good for minimizing overhead, but causes the below
problems.

First, the usage is cumbersome.  This is arguably not a big problem,
since the user-space tool (damo) can do this instead of the user.

Second, it can be too slow.  The update request is not directly handled
by the sysfs interface but kdamond thread.  And kdamond threads wake up
only once per the sampling interval.  Hence if sampling interval is not
short, each update request could take too long time.  The recommended
sampling interval setup is asking DAMON to automatically tune it, within
a range between 5 milliseconds and 10 seconds.  On production systems it
is not very rare to have a few seconds sampling interval as a result of
the auto-tuning, so this can disturb observing DAMON internal status.

Finally, parallel update requests can conflict with each other.  When
parallel update requests are received, DAMON sysfs interface simply
returns -EBUSY to one of the requests.  DAMON user-space tool is hence
implementing its own backoff mechanism, but this can make the operation
even slower.

Introduce a new sysfs file, namely refresh_ms, for asking DAMON sysfs
interface to repeat the essential contents update with a user-specified
time delay.  If non-zero value is written to the file, DAMON sysfs
interface does the updates for essential DAMON internal status including
auto-tuned monitoring intervals, DAMOS stats, and auto-tuned DAMOS
quotas using the user-written value as the time delay.  If zero is
written to the file, the automatic refresh is disabled.

SeongJae Park (4):
  mm/damon/sysfs: implement refresh_ms file under kdamond directory
  mm/damon/sysfs: implement refresh_ms file internal work
  Docs/admin-guide/mm/damon/usage: document refresh_ms file
  Docs/ABI/damon: update for refresh_ms

 .../ABI/testing/sysfs-kernel-mm-damon         |  7 +++
 Documentation/admin-guide/mm/damon/usage.rst  | 13 ++++-
 mm/damon/sysfs.c                              | 58 +++++++++++++++++++
 3 files changed, 75 insertions(+), 3 deletions(-)


base-commit: 982b86e9191292ffcd0f30018981cb16f9fac5c1
-- 
2.39.5

[RFC PATCH 3/4] Docs/admin-guide/mm/damon/usage: document refresh_ms file

[SeongJae Park]

Document the new DAMON sysfs file, refresh_ms, on the usage document.

Signed-off-by: SeongJae Park <sj@kernel.org>
---
 Documentation/admin-guide/mm/damon/usage.rst | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/Documentation/admin-guide/mm/damon/usage.rst b/Documentation/admin-guide/mm/damon/usage.rst
index fc5c962353ed..ff3a2dda1f02 100644
--- a/Documentation/admin-guide/mm/damon/usage.rst
+++ b/Documentation/admin-guide/mm/damon/usage.rst
@@ -59,7 +59,7 @@ comma (",").
 
     :ref:`/sys/kernel/mm/damon <sysfs_root>`/admin
     │ :ref:`kdamonds <sysfs_kdamonds>`/nr_kdamonds
-    │ │ :ref:`0 <sysfs_kdamond>`/state,pid
+    │ │ :ref:`0 <sysfs_kdamond>`/state,pid,refresh_ms
     │ │ │ :ref:`contexts <sysfs_contexts>`/nr_contexts
     │ │ │ │ :ref:`0 <sysfs_context>`/avail_operations,operations
     │ │ │ │ │ :ref:`monitoring_attrs <sysfs_monitoring_attrs>`/
@@ -123,8 +123,8 @@ kdamond.
 kdamonds/<N>/
 -------------
 
-In each kdamond directory, two files (``state`` and ``pid``) and one directory
-(``contexts``) exist.
+In each kdamond directory, three files (``state``, ``pid`` and ``refresh_ms``)
+and one directory (``contexts``) exist.
 
 Reading ``state`` returns ``on`` if the kdamond is currently running, or
 ``off`` if it is not running.
@@ -161,6 +161,13 @@ Users can write below commands for the kdamond to the ``state`` file.
 
 If the state is ``on``, reading ``pid`` shows the pid of the kdamond thread.
 
+Users can ask the kernel to periodically update files showing auto-tuned
+parameters and DAMOS stats instead of manually writing
+``update_tuned_intervals`` like keywords to ``state`` file.  For this, users
+should write the desired update time interval in milliseconds to ``refresh_ms``
+file.  If the interval is zero, the periodic update is disabled.  Reading the
+file shows currently set time interval.
+
 ``contexts`` directory contains files for controlling the monitoring contexts
 that this kdamond will execute.
 
-- 
2.39.5

Re: [RFC PATCH 0/4] mm/damon/sysfs: support periodic and automated stats update

[Honggyu Kim]

Hi SeongJae,

On 7/13/2025 5:46 AM, SeongJae Park wrote:
> DAMON sysfs interface provides files for reading DAMON internal status
> including DAMOS stats.  The content of the files are not automatically
> updated, though.  Users should manually request updates of the contents
> by writing a special command to 'state' file of each kdamond directory.
> This interface is good for minimizing overhead, but causes the below
> problems.
> 
> First, the usage is cumbersome.  This is arguably not a big problem,
> since the user-space tool (damo) can do this instead of the user.
> 
> Second, it can be too slow.  The update request is not directly handled
> by the sysfs interface but kdamond thread.  And kdamond threads wake up
> only once per the sampling interval.  Hence if sampling interval is not
> short, each update request could take too long time.  The recommended
> sampling interval setup is asking DAMON to automatically tune it, within
> a range between 5 milliseconds and 10 seconds.  On production systems it
> is not very rare to have a few seconds sampling interval as a result of
> the auto-tuning, so this can disturb observing DAMON internal status.
> 
> Finally, parallel update requests can conflict with each other.  When
> parallel update requests are received, DAMON sysfs interface simply
> returns -EBUSY to one of the requests.  DAMON user-space tool is hence
> implementing its own backoff mechanism, but this can make the operation
> even slower.
> 
> Introduce a new sysfs file, namely refresh_ms, for asking DAMON sysfs
> interface to repeat the essential contents update with a user-specified
> time delay.

Thanks for working on this, but I have a few questions.
1. Could you please list up what are the "essential contents"?
2. Does it mean that it is different from writing "commit" to "state"?
3. If not, then is there equivalent action to writing something to "state"?

If possible, then this kind of information is better to be documented because
users might get confused if something isn't udpated when "refresh_ms" is set.

Thanks,
Honggyu

> If non-zero value is written to the file, DAMON sysfs
> interface does the updates for essential DAMON internal status including
> auto-tuned monitoring intervals, DAMOS stats, and auto-tuned DAMOS
> quotas using the user-written value as the time delay.  If zero is
> written to the file, the automatic refresh is disabled.
> 
> SeongJae Park (4):
>    mm/damon/sysfs: implement refresh_ms file under kdamond directory
>    mm/damon/sysfs: implement refresh_ms file internal work
>    Docs/admin-guide/mm/damon/usage: document refresh_ms file
>    Docs/ABI/damon: update for refresh_ms
> 
>   .../ABI/testing/sysfs-kernel-mm-damon         |  7 +++
>   Documentation/admin-guide/mm/damon/usage.rst  | 13 ++++-
>   mm/damon/sysfs.c                              | 58 +++++++++++++++++++
>   3 files changed, 75 insertions(+), 3 deletions(-)
> 
> 
> base-commit: 982b86e9191292ffcd0f30018981cb16f9fac5c1
> --
> 2.39.5
> 

Re: [RFC PATCH 0/4] mm/damon/sysfs: support periodic and automated stats update

[SeongJae Park]

On Wed, 16 Jul 2025 07:20:57 +0900 Honggyu Kim <honggyu.kim@sk.com> wrote:

> Hi SeongJae,
> 
> On 7/13/2025 5:46 AM, SeongJae Park wrote:
> > DAMON sysfs interface provides files for reading DAMON internal status
> > including DAMOS stats.  The content of the files are not automatically
> > updated, though.  Users should manually request updates of the contents
> > by writing a special command to 'state' file of each kdamond directory.
> > This interface is good for minimizing overhead, but causes the below
> > problems.
> > 
> > First, the usage is cumbersome.  This is arguably not a big problem,
> > since the user-space tool (damo) can do this instead of the user.
> > 
> > Second, it can be too slow.  The update request is not directly handled
> > by the sysfs interface but kdamond thread.  And kdamond threads wake up
> > only once per the sampling interval.  Hence if sampling interval is not
> > short, each update request could take too long time.  The recommended
> > sampling interval setup is asking DAMON to automatically tune it, within
> > a range between 5 milliseconds and 10 seconds.  On production systems it
> > is not very rare to have a few seconds sampling interval as a result of
> > the auto-tuning, so this can disturb observing DAMON internal status.
> > 
> > Finally, parallel update requests can conflict with each other.  When
> > parallel update requests are received, DAMON sysfs interface simply
> > returns -EBUSY to one of the requests.  DAMON user-space tool is hence
> > implementing its own backoff mechanism, but this can make the operation
> > even slower.
> > 
> > Introduce a new sysfs file, namely refresh_ms, for asking DAMON sysfs
> > interface to repeat the essential contents update with a user-specified
> > time delay.
> 
> Thanks for working on this, but I have a few questions.
> 1. Could you please list up what are the "essential contents"?

Thank you for asking this.  The contents are auto-tuned monitoring intervals,
DAMOS stats, and auto-tuned effective size quota.

I will add these on the next version cover letter.

> 2. Does it mean that it is different from writing "commit" to "state"?
> 3. If not, then is there equivalent action to writing something to "state"?

"refresh_ms" works same to other DAMON parameter files.  You can set it before
starting DAMON, or "commit" new values (including 0 for turning this refresh
off) in runtime.

I'm not that confident if I understood your point very well, especially what
"it"s mean.  Let me know if I'm misunderstanding something.

> 
> If possible, then this kind of information is better to be documented because
> users might get confused if something isn't udpated when "refresh_ms" is set.

You're right!  I'll add above things on the next version of the cover letter.


Thanks,
SJ

[...]

Re: [RFC PATCH 0/4] mm/damon/sysfs: support periodic and automated stats update

[Honggyu Kim]

Hi SeongJae,

On 7/16/2025 8:43 AM, SeongJae Park wrote:
> On Wed, 16 Jul 2025 07:20:57 +0900 Honggyu Kim <honggyu.kim@sk.com> wrote:
> 
>> Hi SeongJae,
>>
>> On 7/13/2025 5:46 AM, SeongJae Park wrote:
>>> DAMON sysfs interface provides files for reading DAMON internal status
>>> including DAMOS stats.  The content of the files are not automatically
>>> updated, though.  Users should manually request updates of the contents
>>> by writing a special command to 'state' file of each kdamond directory.
>>> This interface is good for minimizing overhead, but causes the below
>>> problems.
>>>
>>> First, the usage is cumbersome.  This is arguably not a big problem,
>>> since the user-space tool (damo) can do this instead of the user.
>>>
>>> Second, it can be too slow.  The update request is not directly handled
>>> by the sysfs interface but kdamond thread.  And kdamond threads wake up
>>> only once per the sampling interval.  Hence if sampling interval is not
>>> short, each update request could take too long time.  The recommended
>>> sampling interval setup is asking DAMON to automatically tune it, within
>>> a range between 5 milliseconds and 10 seconds.  On production systems it
>>> is not very rare to have a few seconds sampling interval as a result of
>>> the auto-tuning, so this can disturb observing DAMON internal status.
>>>
>>> Finally, parallel update requests can conflict with each other.  When
>>> parallel update requests are received, DAMON sysfs interface simply
>>> returns -EBUSY to one of the requests.  DAMON user-space tool is hence
>>> implementing its own backoff mechanism, but this can make the operation
>>> even slower.
>>>
>>> Introduce a new sysfs file, namely refresh_ms, for asking DAMON sysfs
>>> interface to repeat the essential contents update with a user-specified
>>> time delay.
>>
>> Thanks for working on this, but I have a few questions.
>> 1. Could you please list up what are the "essential contents"?
> 
> Thank you for asking this.  The contents are auto-tuned monitoring intervals,
> DAMOS stats, and auto-tuned effective size quota.

Thanks, but I meant the specific list of damon knobs refreshed.  If there are
too many knobs, then don't have to list them all.

> I will add these on the next version cover letter.

Thanks.

>> 2. Does it mean that it is different from writing "commit" to "state"?
>> 3. If not, then is there equivalent action to writing something to "state"?
> 
> "refresh_ms" works same to other DAMON parameter files.  You can set it before
> starting DAMON, or "commit" new values (including 0 for turning this refresh
> off) in runtime.
> 
> I'm not that confident if I understood your point very well, especially what
> "it"s mean.  Let me know if I'm misunderstanding something.

I mean writing "commit" to "state" refresh all knobs, but it seems "refresh_ms"
internally refresh paritial knobs so I was wondering what are refreshed and what 
aren't.

Regarding the "equivalent action", I was also wondering if there is a command
that works same as "refresh_ms" internally does among the command below.

   update_tuned_intervals
   commit_schemes_quota_goals
   update_schemes_stats
   update_schemes_tried_regions
   update_schemes_tried_bytes
   clear_schemes_tried_regions
   update_schemes_effective_quotas

https://docs.kernel.org/admin-guide/mm/damon/usage.html#kdamonds-n

In other words, if there is the same command listed above, then users might be
able to run a script that regularaly write the command to the current "state"
even without this "refresh_ms".  I know having "refresh_ms" is much better
though.

Thanks,
Honggyu

> 
>>
>> If possible, then this kind of information is better to be documented because
>> users might get confused if something isn't udpated when "refresh_ms" is set.
> 
> You're right!  I'll add above things on the next version of the cover letter.
> 
> 
> Thanks,
> SJ
> 
> [...]

Re: [RFC PATCH 0/4] mm/damon/sysfs: support periodic and automated stats update

[SeongJae Park]

On Wed, 16 Jul 2025 10:58:06 +0900 Honggyu Kim <honggyu.kim@sk.com> wrote:

> Hi SeongJae,
> 
> On 7/16/2025 8:43 AM, SeongJae Park wrote:
> > On Wed, 16 Jul 2025 07:20:57 +0900 Honggyu Kim <honggyu.kim@sk.com> wrote:
> > 
> >> Hi SeongJae,
> >>
> >> On 7/13/2025 5:46 AM, SeongJae Park wrote:
> >>> DAMON sysfs interface provides files for reading DAMON internal status
> >>> including DAMOS stats.  The content of the files are not automatically
> >>> updated, though.  Users should manually request updates of the contents
> >>> by writing a special command to 'state' file of each kdamond directory.
> >>> This interface is good for minimizing overhead, but causes the below
> >>> problems.
> >>>
> >>> First, the usage is cumbersome.  This is arguably not a big problem,
> >>> since the user-space tool (damo) can do this instead of the user.
> >>>
> >>> Second, it can be too slow.  The update request is not directly handled
> >>> by the sysfs interface but kdamond thread.  And kdamond threads wake up
> >>> only once per the sampling interval.  Hence if sampling interval is not
> >>> short, each update request could take too long time.  The recommended
> >>> sampling interval setup is asking DAMON to automatically tune it, within
> >>> a range between 5 milliseconds and 10 seconds.  On production systems it
> >>> is not very rare to have a few seconds sampling interval as a result of
> >>> the auto-tuning, so this can disturb observing DAMON internal status.
> >>>
> >>> Finally, parallel update requests can conflict with each other.  When
> >>> parallel update requests are received, DAMON sysfs interface simply
> >>> returns -EBUSY to one of the requests.  DAMON user-space tool is hence
> >>> implementing its own backoff mechanism, but this can make the operation
> >>> even slower.
> >>>
> >>> Introduce a new sysfs file, namely refresh_ms, for asking DAMON sysfs
> >>> interface to repeat the essential contents update with a user-specified
> >>> time delay.
> >>
> >> Thanks for working on this, but I have a few questions.
> >> 1. Could you please list up what are the "essential contents"?
> > 
> > Thank you for asking this.  The contents are auto-tuned monitoring intervals,
> > DAMOS stats, and auto-tuned effective size quota.
> 
> Thanks, but I meant the specific list of damon knobs refreshed.  If there are
> too many knobs, then don't have to list them all.

I guess you are saying about sysfs files?  Yes, I think listing those all is
unnecessary.

> 
> > I will add these on the next version cover letter.
> 
> Thanks.
> 
> >> 2. Does it mean that it is different from writing "commit" to "state"?
> >> 3. If not, then is there equivalent action to writing something to "state"?
> > 
> > "refresh_ms" works same to other DAMON parameter files.  You can set it before
> > starting DAMON, or "commit" new values (including 0 for turning this refresh
> > off) in runtime.
> > 
> > I'm not that confident if I understood your point very well, especially what
> > "it"s mean.  Let me know if I'm misunderstanding something.
> 
> I mean writing "commit" to "state" refresh all knobs,

Assuming "knobs" mean sysfs files, this is incorrect.  Writing "commit" to
"state" passes the parameters written on appropriate files to running kdamond.
It doesn't update any sysfs files until a "update_*" keyword is written to the
"state" file.

> but it seems "refresh_ms"
> internally refresh paritial knobs so I was wondering what are refreshed and what 
> aren't.
> 
> Regarding the "equivalent action", I was also wondering if there is a command
> that works same as "refresh_ms" internally does among the command below.
> 
>    update_tuned_intervals
>    commit_schemes_quota_goals
>    update_schemes_stats
>    update_schemes_tried_regions
>    update_schemes_tried_bytes
>    clear_schemes_tried_regions
>    update_schemes_effective_quotas
> 
> https://docs.kernel.org/admin-guide/mm/damon/usage.html#kdamonds-n
> 
> In other words, if there is the same command listed above, then users might be
> able to run a script that regularaly write the command to the current "state"
> even without this "refresh_ms".  I know having "refresh_ms" is much better
> though.

Thank you for this nice question.  Yes.  Enabling "refresh_ms" is roughly same
to periodically writing
"update_{tuned_intervals,schemes_stats,schemes_effective_quotas" to "state"
file.


Thanks,
SJ

[...]

Re: [RFC PATCH 0/4] mm/damon/sysfs: support periodic and automated stats update

[Honggyu Kim]

On 7/16/2025 11:51 AM, SeongJae Park wrote:
> On Wed, 16 Jul 2025 10:58:06 +0900 Honggyu Kim <honggyu.kim@sk.com> wrote:
> 
>> Hi SeongJae,
>>
>> On 7/16/2025 8:43 AM, SeongJae Park wrote:
>>> On Wed, 16 Jul 2025 07:20:57 +0900 Honggyu Kim <honggyu.kim@sk.com> wrote:
>>>
>>>> Hi SeongJae,
>>>>
>>>> On 7/13/2025 5:46 AM, SeongJae Park wrote:
>>>>> DAMON sysfs interface provides files for reading DAMON internal status
>>>>> including DAMOS stats.  The content of the files are not automatically
>>>>> updated, though.  Users should manually request updates of the contents
>>>>> by writing a special command to 'state' file of each kdamond directory.
>>>>> This interface is good for minimizing overhead, but causes the below
>>>>> problems.

[...snip...]

>>>>> Introduce a new sysfs file, namely refresh_ms, for asking DAMON sysfs
>>>>> interface to repeat the essential contents update with a user-specified
>>>>> time delay.
>>>>
>>>> Thanks for working on this, but I have a few questions.
>>>> 1. Could you please list up what are the "essential contents"?
>>>
>>> Thank you for asking this.  The contents are auto-tuned monitoring intervals,
>>> DAMOS stats, and auto-tuned effective size quota.
>>
>> Thanks, but I meant the specific list of damon knobs refreshed.  If there are
>> too many knobs, then don't have to list them all.
> 
> I guess you are saying about sysfs files?  Yes, I think listing those all is
> unnecessary.

OK. No problem then.

>>> I will add these on the next version cover letter.
>>
>> Thanks.
>>
>>>> 2. Does it mean that it is different from writing "commit" to "state"?
>>>> 3. If not, then is there equivalent action to writing something to "state"?
>>>
>>> "refresh_ms" works same to other DAMON parameter files.  You can set it before
>>> starting DAMON, or "commit" new values (including 0 for turning this refresh
>>> off) in runtime.
>>>
>>> I'm not that confident if I understood your point very well, especially what
>>> "it"s mean.  Let me know if I'm misunderstanding something.
>>
>> I mean writing "commit" to "state" refresh all knobs,
> 
> Assuming "knobs" mean sysfs files, this is incorrect.  Writing "commit" to
> "state" passes the parameters written on appropriate files to running kdamond.
> It doesn't update any sysfs files until a "update_*" keyword is written to the
> "state" file.

I thought writing "commit" to "state" applies config changes and updates all the
readable sysfs files together, but it doesn't.  Thanks for the correction.

> 
>> but it seems "refresh_ms"
>> internally refresh paritial knobs so I was wondering what are refreshed and what
>> aren't.
>>
>> Regarding the "equivalent action", I was also wondering if there is a command
>> that works same as "refresh_ms" internally does among the command below.
>>
>>     update_tuned_intervals
>>     commit_schemes_quota_goals
>>     update_schemes_stats
>>     update_schemes_tried_regions
>>     update_schemes_tried_bytes
>>     clear_schemes_tried_regions
>>     update_schemes_effective_quotas
>>
>> https://docs.kernel.org/admin-guide/mm/damon/usage.html#kdamonds-n
>>
>> In other words, if there is the same command listed above, then users might be
>> able to run a script that regularaly write the command to the current "state"
>> even without this "refresh_ms".  I know having "refresh_ms" is much better
>> though.
> 
> Thank you for this nice question.  Yes.  Enabling "refresh_ms" is roughly same
> to periodically writing
> "update_{tuned_intervals,schemes_stats,schemes_effective_quotas" to "state"
> file.

OK. I get that "refresh_ms" does periodic write the combination of them to
"state" without "update_{schemes_tried_regions,schemes_tried_bytes}.

In other words, it is same as periodic execution of the follows.

   echo update_tuned_intervals > state
   echo update_schemes_stats > state
   echo update_schemes_effective_quotas > state

That makes my question clear.  Thanks for the explanation!

Thanks,
Honggyu

> 
> 
> Thanks,
> SJ
> 
> [...]