[RFC PATCH v2 1/1] mm/damon: support MADV_COLLAPSE via DAMOS_COLLAPSE scheme action

[RFC PATCH v2 1/1] mm/damon: support MADV_COLLAPSE via DAMOS_COLLAPSE scheme action

[gutierrez.asier]

From: Asier Gutierrez <gutierrez.asier@huawei-partners.com>

This patch set introces a new action: DAMOS_COLLAPSE.

For DAMOS_HUGEPAGE and DAMOS_NOHUGEPAGE to work, khugepaged should be
working, since it relies on hugepage_madvise to add a new slot. This
slot should be picked up by khugepaged and eventually collapse (or
not, if we are using DAMOS_NOHUGEPAGE) the pages. If THP is not
enabled, khugepaged will not be working, and therefore no collapse
will happen.

DAMOS_COLLAPSE eventually calls madvise_collapse, which will collapse
the address range synchronously.

This new action may be required to support autotuning with hugepage as
a goal.

[1] https://lore.kernel.org/lkml/20260314165156.86647-1-sj@kernel.org/

Signed-off-by: Asier Gutierrez <gutierrez.asier@huawei-partners.com>
Reviewed-by: SeongJae Park <sj@kernel.org>
---
 Documentation/mm/damon/design.rst | 4 ++++
 include/linux/damon.h             | 1 +
 mm/damon/sysfs-schemes.c          | 4 ++++
 mm/damon/vaddr.c                  | 3 +++
 4 files changed, 12 insertions(+)

diff --git a/Documentation/mm/damon/design.rst b/Documentation/mm/damon/design.rst
index 29fff20b3c2a..3b1461f42987 100644
--- a/Documentation/mm/damon/design.rst
+++ b/Documentation/mm/damon/design.rst
@@ -463,6 +463,10 @@ that supports each action are as below.
    Supported by ``vaddr`` and ``fvaddr`` operations set.
  - ``nohugepage``: Call ``madvise()`` for the region with ``MADV_NOHUGEPAGE``.
    Supported by ``vaddr`` and ``fvaddr`` operations set.
+ - ``collapse``: Call ``madvise()`` for the region with ``MADV_COLLAPSE``.
+   Supported by ``vaddr`` and ``fvaddr`` operations set. When
+   TRANSPARENT_HUGEPAGE is disabled, the application of the action will just
+   fail.
  - ``lru_prio``: Prioritize the region on its LRU lists.
    Supported by ``paddr`` operations set.
  - ``lru_deprio``: Deprioritize the region on its LRU lists.
diff --git a/include/linux/damon.h b/include/linux/damon.h
index 3a441fbca170..6720dc70c487 100644
--- a/include/linux/damon.h
+++ b/include/linux/damon.h
@@ -140,6 +140,7 @@ enum damos_action {
 	DAMOS_PAGEOUT,
 	DAMOS_HUGEPAGE,
 	DAMOS_NOHUGEPAGE,
+	DAMOS_COLLAPSE,
 	DAMOS_LRU_PRIO,
 	DAMOS_LRU_DEPRIO,
 	DAMOS_MIGRATE_HOT,
diff --git a/mm/damon/sysfs-schemes.c b/mm/damon/sysfs-schemes.c
index 5186966dafb3..aa08a8f885fb 100644
--- a/mm/damon/sysfs-schemes.c
+++ b/mm/damon/sysfs-schemes.c
@@ -2041,6 +2041,10 @@ static struct damos_sysfs_action_name damos_sysfs_action_names[] = {
 		.action = DAMOS_NOHUGEPAGE,
 		.name = "nohugepage",
 	},
+	{
+		.action = DAMOS_COLLAPSE,
+		.name = "collapse",
+	},
 	{
 		.action = DAMOS_LRU_PRIO,
 		.name = "lru_prio",
diff --git a/mm/damon/vaddr.c b/mm/damon/vaddr.c
index b069dbc7e3d2..dd5f2d7027ac 100644
--- a/mm/damon/vaddr.c
+++ b/mm/damon/vaddr.c
@@ -903,6 +903,9 @@ static unsigned long damon_va_apply_scheme(struct damon_ctx *ctx,
 	case DAMOS_NOHUGEPAGE:
 		madv_action = MADV_NOHUGEPAGE;
 		break;
+	case DAMOS_COLLAPSE:
+		madv_action = MADV_COLLAPSE;
+		break;
 	case DAMOS_MIGRATE_HOT:
 	case DAMOS_MIGRATE_COLD:
 		return damos_va_migrate(t, r, scheme, sz_filter_passed);
-- 
2.43.0

Re: [RFC PATCH v2 1/1] mm/damon: support MADV_COLLAPSE via DAMOS_COLLAPSE scheme action

[SeongJae Park]

Hello Asier,

On Mon, 16 Mar 2026 18:38:05 +0000 <gutierrez.asier@huawei-partners.com> wrote:

> From: Asier Gutierrez <gutierrez.asier@huawei-partners.com>
> 
> This patch set introces a new action: DAMOS_COLLAPSE.
> 
> For DAMOS_HUGEPAGE and DAMOS_NOHUGEPAGE to work, khugepaged should be
> working, since it relies on hugepage_madvise to add a new slot. This
> slot should be picked up by khugepaged and eventually collapse (or
> not, if we are using DAMOS_NOHUGEPAGE) the pages. If THP is not
> enabled, khugepaged will not be working, and therefore no collapse
> will happen.
> 
> DAMOS_COLLAPSE eventually calls madvise_collapse, which will collapse
> the address range synchronously.
> 
> This new action may be required to support autotuning with hugepage as
> a goal.

Above all makes sense.  Thank you for posting this patch.

Do you have some test results that you can also share together?  It would be
nice if it can demonstrate the benefit of DAMOS_COLLAPSE over DAMOS_HUGEPAGE.

> 
> [1] https://lore.kernel.org/lkml/20260314165156.86647-1-sj@kernel.org/

Seems the above link is just added by a mistake?  If not, please clarify.

> 
> Signed-off-by: Asier Gutierrez <gutierrez.asier@huawei-partners.com>
> Reviewed-by: SeongJae Park <sj@kernel.org>
> ---
>  Documentation/mm/damon/design.rst | 4 ++++
>  include/linux/damon.h             | 1 +
>  mm/damon/sysfs-schemes.c          | 4 ++++
>  mm/damon/vaddr.c                  | 3 +++
>  4 files changed, 12 insertions(+)
[...]
> diff --git a/include/linux/damon.h b/include/linux/damon.h
> index 3a441fbca170..6720dc70c487 100644
> --- a/include/linux/damon.h
> +++ b/include/linux/damon.h
> @@ -140,6 +140,7 @@ enum damos_action {
>  	DAMOS_PAGEOUT,
>  	DAMOS_HUGEPAGE,
>  	DAMOS_NOHUGEPAGE,
> +	DAMOS_COLLAPSE,
>  	DAMOS_LRU_PRIO,
>  	DAMOS_LRU_DEPRIO,
>  	DAMOS_MIGRATE_HOT,

sashiko.dev adds [1] below comments.  Let me also add my comments in line.

: This isn't a bug, but should a kernel-doc entry for @DAMOS_COLLAPSE be added
: to the comment block above this enum?

Makes sense.  'make htmldocs' may complain otherwise.  Asier, could you please
add the kernel-doc comment for DAMOS_COLLAPSE in the next spin?

: 
: Also, does inserting DAMOS_COLLAPSE here shift the integer values of the
: subsequent enum entries like DAMOS_STAT?
: 
: The DAMON sysfs selftest script (tools/testing/selftests/damon/sysfs.py) uses
: a hardcoded dictionary action_val to map string names to their integer enum
: values.
: 
: If the enum values shift, the test's assertion:
: 
: assert_true(dump['action'] == action_val[scheme.action])
: 
: might fail when checking the struct memory via drgn. Could the python test
: dictionary be updated to reflect the new values, or could the new action be
: added at the end of the enum list?

There is no test that uses DAMOS actions that defined after DAMOS_NOHUGEPAGE,
so no real test will break.  But this is a good point.  It would be better to
update the hard-coded value together.  Asier, could you also update the
'action_val' dict of assert_scheme_committed() function in
tools/testing/selftets/damon/sysfs.py for the updated enum value in the next
version?

[1] https://sashiko.dev/#/patchset/20260316183805.2090297-1-gutierrez.asier@huawei-partners.com


Thanks,
SJ

[...]

Re: [RFC PATCH v2 1/1] mm/damon: support MADV_COLLAPSE via DAMOS_COLLAPSE scheme action

[Gutierrez Asier]

Hi SJ,

First of all, I just noticed that this was sent as v2 RFC, while
it should be v1. Bear in mind that the next series will also be
v2.

On 3/17/2026 3:32 AM, SeongJae Park wrote:
> Hello Asier,
> 
> On Mon, 16 Mar 2026 18:38:05 +0000 <gutierrez.asier@huawei-partners.com> wrote:
> 
>> From: Asier Gutierrez <gutierrez.asier@huawei-partners.com>
>>
>> This patch set introces a new action: DAMOS_COLLAPSE.
>>
>> For DAMOS_HUGEPAGE and DAMOS_NOHUGEPAGE to work, khugepaged should be
>> working, since it relies on hugepage_madvise to add a new slot. This
>> slot should be picked up by khugepaged and eventually collapse (or
>> not, if we are using DAMOS_NOHUGEPAGE) the pages. If THP is not
>> enabled, khugepaged will not be working, and therefore no collapse
>> will happen.
>>
>> DAMOS_COLLAPSE eventually calls madvise_collapse, which will collapse
>> the address range synchronously.
>>
>> This new action may be required to support autotuning with hugepage as
>> a goal.
> 
> Above all makes sense.  Thank you for posting this patch.
> 
> Do you have some test results that you can also share together?  It would be
> nice if it can demonstrate the benefit of DAMOS_COLLAPSE over DAMOS_HUGEPAGE.
I will run some tests and benchmarks.
> 
>>
>> [1] https://lore.kernel.org/lkml/20260314165156.86647-1-sj@kernel.org/
> 
> Seems the above link is just added by a mistake?  If not, please clarify.
Yes, it looks like I copied the wrong link.
>>
>> Signed-off-by: Asier Gutierrez <gutierrez.asier@huawei-partners.com>
>> Reviewed-by: SeongJae Park <sj@kernel.org>
>> ---
>>  Documentation/mm/damon/design.rst | 4 ++++
>>  include/linux/damon.h             | 1 +
>>  mm/damon/sysfs-schemes.c          | 4 ++++
>>  mm/damon/vaddr.c                  | 3 +++
>>  4 files changed, 12 insertions(+)
> [...]
>> diff --git a/include/linux/damon.h b/include/linux/damon.h
>> index 3a441fbca170..6720dc70c487 100644
>> --- a/include/linux/damon.h
>> +++ b/include/linux/damon.h
>> @@ -140,6 +140,7 @@ enum damos_action {
>>  	DAMOS_PAGEOUT,
>>  	DAMOS_HUGEPAGE,
>>  	DAMOS_NOHUGEPAGE,
>> +	DAMOS_COLLAPSE,
>>  	DAMOS_LRU_PRIO,
>>  	DAMOS_LRU_DEPRIO,
>>  	DAMOS_MIGRATE_HOT,
> 
> sashiko.dev adds [1] below comments.  Let me also add my comments in line.
> 
> : This isn't a bug, but should a kernel-doc entry for @DAMOS_COLLAPSE be added
> : to the comment block above this enum?
> 
> Makes sense.  'make htmldocs' may complain otherwise.  Asier, could you please
> add the kernel-doc comment for DAMOS_COLLAPSE in the next spin?
OK, I will split this patch into 2: one with the code and the other one with
the documentation.
> : 
> : Also, does inserting DAMOS_COLLAPSE here shift the integer values of the
> : subsequent enum entries like DAMOS_STAT?
> : 
> : The DAMON sysfs selftest script (tools/testing/selftests/damon/sysfs.py) uses
> : a hardcoded dictionary action_val to map string names to their integer enum
> : values.
> : 
> : If the enum values shift, the test's assertion:
> : 
> : assert_true(dump['action'] == action_val[scheme.action])
> : 
> : might fail when checking the struct memory via drgn. Could the python test
> : dictionary be updated to reflect the new values, or could the new action be
> : added at the end of the enum list?
> 
> There is no test that uses DAMOS actions that defined after DAMOS_NOHUGEPAGE,
> so no real test will break.  But this is a good point.  It would be better to
> update the hard-coded value together.  Asier, could you also update the
> 'action_val' dict of assert_scheme_committed() function in
> tools/testing/selftets/damon/sysfs.py for the updated enum value in the next
> version?
OK, I will do it.
> 
> [1] https://sashiko.dev/#/patchset/20260316183805.2090297-1-gutierrez.asier@huawei-partners.com
> 
> 
> Thanks,
> SJ
> 
> [...]
> 
Thanks for the feedback.

-- 
Asier Gutierrez
Huawei

Re: [RFC PATCH v2 1/1] mm/damon: support MADV_COLLAPSE via DAMOS_COLLAPSE scheme action

[SeongJae Park]

On Tue, 17 Mar 2026 09:52:09 +0300 Gutierrez Asier <gutierrez.asier@huawei-partners.com> wrote:

> Hi SJ,
> 
> First of all, I just noticed that this was sent as v2 RFC, while
> it should be v1. Bear in mind that the next series will also be
> v2.

No worry :)

> 
> On 3/17/2026 3:32 AM, SeongJae Park wrote:
> > Hello Asier,
> > 
> > On Mon, 16 Mar 2026 18:38:05 +0000 <gutierrez.asier@huawei-partners.com> wrote:
> > 
> >> From: Asier Gutierrez <gutierrez.asier@huawei-partners.com>
> >>
> >> This patch set introces a new action: DAMOS_COLLAPSE.
> >>
> >> For DAMOS_HUGEPAGE and DAMOS_NOHUGEPAGE to work, khugepaged should be
> >> working, since it relies on hugepage_madvise to add a new slot. This
> >> slot should be picked up by khugepaged and eventually collapse (or
> >> not, if we are using DAMOS_NOHUGEPAGE) the pages. If THP is not
> >> enabled, khugepaged will not be working, and therefore no collapse
> >> will happen.
> >>
> >> DAMOS_COLLAPSE eventually calls madvise_collapse, which will collapse
> >> the address range synchronously.
> >>
> >> This new action may be required to support autotuning with hugepage as
> >> a goal.
> > 
> > Above all makes sense.  Thank you for posting this patch.
> > 
> > Do you have some test results that you can also share together?  It would be
> > nice if it can demonstrate the benefit of DAMOS_COLLAPSE over DAMOS_HUGEPAGE.
> I will run some tests and benchmarks.

Looking forward to.

> > 
> >>
> >> [1] https://lore.kernel.org/lkml/20260314165156.86647-1-sj@kernel.org/
> > 
> > Seems the above link is just added by a mistake?  If not, please clarify.
> Yes, it looks like I copied the wrong link.
> >>
> >> Signed-off-by: Asier Gutierrez <gutierrez.asier@huawei-partners.com>
> >> Reviewed-by: SeongJae Park <sj@kernel.org>
> >> ---
> >>  Documentation/mm/damon/design.rst | 4 ++++
> >>  include/linux/damon.h             | 1 +
> >>  mm/damon/sysfs-schemes.c          | 4 ++++
> >>  mm/damon/vaddr.c                  | 3 +++
> >>  4 files changed, 12 insertions(+)
> > [...]
> >> diff --git a/include/linux/damon.h b/include/linux/damon.h
> >> index 3a441fbca170..6720dc70c487 100644
> >> --- a/include/linux/damon.h
> >> +++ b/include/linux/damon.h
> >> @@ -140,6 +140,7 @@ enum damos_action {
> >>  	DAMOS_PAGEOUT,
> >>  	DAMOS_HUGEPAGE,
> >>  	DAMOS_NOHUGEPAGE,
> >> +	DAMOS_COLLAPSE,
> >>  	DAMOS_LRU_PRIO,
> >>  	DAMOS_LRU_DEPRIO,
> >>  	DAMOS_MIGRATE_HOT,
> > 
> > sashiko.dev adds [1] below comments.  Let me also add my comments in line.
> > 
> > : This isn't a bug, but should a kernel-doc entry for @DAMOS_COLLAPSE be added
> > : to the comment block above this enum?
> > 
> > Makes sense.  'make htmldocs' may complain otherwise.  Asier, could you please
> > add the kernel-doc comment for DAMOS_COLLAPSE in the next spin?
> OK, I will split this patch into 2: one with the code and the other one with
> the documentation.

Does the 'documentation' mean the change for 'design.rst'?  Or, the kernel-doc
comment?  If that's the former case (separating 'design.rst' side change as
another patch), that's completely good for me.

If that's the latter case (separating kernel-doc comment addition as another
patch), that will make the commit trigger document build error, and the error
will be fixed only after the followup commit is applied.  Please ensure keeping
'damon.h' side changes as a single patch, and therefore no warning or breaks in
the middle of the patch series is introduced (and later fixed).

> > : 
> > : Also, does inserting DAMOS_COLLAPSE here shift the integer values of the
> > : subsequent enum entries like DAMOS_STAT?
> > : 
> > : The DAMON sysfs selftest script (tools/testing/selftests/damon/sysfs.py) uses
> > : a hardcoded dictionary action_val to map string names to their integer enum
> > : values.
> > : 
> > : If the enum values shift, the test's assertion:
> > : 
> > : assert_true(dump['action'] == action_val[scheme.action])
> > : 
> > : might fail when checking the struct memory via drgn. Could the python test
> > : dictionary be updated to reflect the new values, or could the new action be
> > : added at the end of the enum list?
> > 
> > There is no test that uses DAMOS actions that defined after DAMOS_NOHUGEPAGE,
> > so no real test will break.  But this is a good point.  It would be better to
> > update the hard-coded value together.  Asier, could you also update the
> > 'action_val' dict of assert_scheme_committed() function in
> > tools/testing/selftets/damon/sysfs.py for the updated enum value in the next
> > version?
> OK, I will do it.

Thank you :)

> > 
> > [1] https://sashiko.dev/#/patchset/20260316183805.2090297-1-gutierrez.asier@huawei-partners.com
> > 
> > 
> > Thanks,
> > SJ
> > 
> > [...]
> > 
> Thanks for the feedback.

My pleasure.


Thanks,
SJ

[...]

Re: [RFC PATCH v2 1/1] mm/damon: support MADV_COLLAPSE via DAMOS_COLLAPSE scheme action

[Gutierrez Asier]

On 3/18/2026 3:52 AM, SeongJae Park wrote:
> On Tue, 17 Mar 2026 09:52:09 +0300 Gutierrez Asier <gutierrez.asier@huawei-partners.com> wrote:
> 
>> Hi SJ,
>>
>> First of all, I just noticed that this was sent as v2 RFC, while
>> it should be v1. Bear in mind that the next series will also be
>> v2.
> 
> No worry :)
> 
>>
>> On 3/17/2026 3:32 AM, SeongJae Park wrote:
>>> Hello Asier,
>>>
>>> On Mon, 16 Mar 2026 18:38:05 +0000 <gutierrez.asier@huawei-partners.com> wrote:
>>>
>>>> From: Asier Gutierrez <gutierrez.asier@huawei-partners.com>
>>>>
>>>> This patch set introces a new action: DAMOS_COLLAPSE.
>>>>
>>>> For DAMOS_HUGEPAGE and DAMOS_NOHUGEPAGE to work, khugepaged should be
>>>> working, since it relies on hugepage_madvise to add a new slot. This
>>>> slot should be picked up by khugepaged and eventually collapse (or
>>>> not, if we are using DAMOS_NOHUGEPAGE) the pages. If THP is not
>>>> enabled, khugepaged will not be working, and therefore no collapse
>>>> will happen.
>>>>
>>>> DAMOS_COLLAPSE eventually calls madvise_collapse, which will collapse
>>>> the address range synchronously.
>>>>
>>>> This new action may be required to support autotuning with hugepage as
>>>> a goal.
>>>
>>> Above all makes sense.  Thank you for posting this patch.
>>>
>>> Do you have some test results that you can also share together?  It would be
>>> nice if it can demonstrate the benefit of DAMOS_COLLAPSE over DAMOS_HUGEPAGE.
>> I will run some tests and benchmarks.
> 
> Looking forward to.
> 
>>>
>>>>
>>>> [1] https://lore.kernel.org/lkml/20260314165156.86647-1-sj@kernel.org/
>>>
>>> Seems the above link is just added by a mistake?  If not, please clarify.
>> Yes, it looks like I copied the wrong link.
>>>>
>>>> Signed-off-by: Asier Gutierrez <gutierrez.asier@huawei-partners.com>
>>>> Reviewed-by: SeongJae Park <sj@kernel.org>
>>>> ---
>>>>  Documentation/mm/damon/design.rst | 4 ++++
>>>>  include/linux/damon.h             | 1 +
>>>>  mm/damon/sysfs-schemes.c          | 4 ++++
>>>>  mm/damon/vaddr.c                  | 3 +++
>>>>  4 files changed, 12 insertions(+)
>>> [...]
>>>> diff --git a/include/linux/damon.h b/include/linux/damon.h
>>>> index 3a441fbca170..6720dc70c487 100644
>>>> --- a/include/linux/damon.h
>>>> +++ b/include/linux/damon.h
>>>> @@ -140,6 +140,7 @@ enum damos_action {
>>>>  	DAMOS_PAGEOUT,
>>>>  	DAMOS_HUGEPAGE,
>>>>  	DAMOS_NOHUGEPAGE,
>>>> +	DAMOS_COLLAPSE,
>>>>  	DAMOS_LRU_PRIO,
>>>>  	DAMOS_LRU_DEPRIO,
>>>>  	DAMOS_MIGRATE_HOT,
>>>
>>> sashiko.dev adds [1] below comments.  Let me also add my comments in line.
>>>
>>> : This isn't a bug, but should a kernel-doc entry for @DAMOS_COLLAPSE be added
>>> : to the comment block above this enum?
>>>
>>> Makes sense.  'make htmldocs' may complain otherwise.  Asier, could you please
>>> add the kernel-doc comment for DAMOS_COLLAPSE in the next spin?
>> OK, I will split this patch into 2: one with the code and the other one with
>> the documentation.
> 
> Does the 'documentation' mean the change for 'design.rst'?  Or, the kernel-doc
> comment?  If that's the former case (separating 'design.rst' side change as
> another patch), that's completely good for me.
I meant the doc/mm/damon/design.rst
> 
> If that's the latter case (separating kernel-doc comment addition as another
> patch), that will make the commit trigger document build error, and the error
> will be fixed only after the followup commit is applied.  Please ensure keeping
> 'damon.h' side changes as a single patch, and therefore no warning or breaks in
> the middle of the patch series is introduced (and later fixed).
> 
>>> : 
>>> : Also, does inserting DAMOS_COLLAPSE here shift the integer values of the
>>> : subsequent enum entries like DAMOS_STAT?
>>> : 
>>> : The DAMON sysfs selftest script (tools/testing/selftests/damon/sysfs.py) uses
>>> : a hardcoded dictionary action_val to map string names to their integer enum
>>> : values.
>>> : 
>>> : If the enum values shift, the test's assertion:
>>> : 
>>> : assert_true(dump['action'] == action_val[scheme.action])
>>> : 
>>> : might fail when checking the struct memory via drgn. Could the python test
>>> : dictionary be updated to reflect the new values, or could the new action be
>>> : added at the end of the enum list?
>>>
>>> There is no test that uses DAMOS actions that defined after DAMOS_NOHUGEPAGE,
>>> so no real test will break.  But this is a good point.  It would be better to
>>> update the hard-coded value together.  Asier, could you also update the
>>> 'action_val' dict of assert_scheme_committed() function in
>>> tools/testing/selftets/damon/sysfs.py for the updated enum value in the next
>>> version?
>> OK, I will do it.
> 
> Thank you :)
> 
>>>
>>> [1] https://sashiko.dev/#/patchset/20260316183805.2090297-1-gutierrez.asier@huawei-partners.com
>>>
>>>
>>> Thanks,
>>> SJ
>>>
>>> [...]
>>>
>> Thanks for the feedback.
> 
> My pleasure.
> 
> 
> Thanks,
> SJ
> 
> [...]
> 

-- 
Asier Gutierrez
Huawei

[RFC PATCH v1 1/1] This patch set introces a new action: DAMOS_COLLAPSE.

[gutierrez.asier]

From: Asier Gutierrez <gutierrez.asier@huawei-partners.com>

For DAMOS_HUGEPAGE and DAMOS_NOHUGEPAGE to work, khugepaged should be
working, since it relies on hugepage_madvise to add a new slot. This
slot should be picked up by khugepaged and eventually collapse (or
not, if we are using DAMOS_NOHUGEPAGE) the pages. If THP is not
enabled, khugepaged will not be working, and therefore no collapse
will happen.

DAMOS_COLLAPSE eventually calls madvise_collapse, which will collapse
the address range synchronously.

This new action may be required to support autotuning with hugepage as
a goal[1].

[1]: https://lore.kernel.org/damon/20260313000816.79933-1-sj@kernel.org/

---------
Benchmarks:

T n: THP never
T m: THP madvise
D h: DAMON action hugepage
D c: DAMON action collapse

+------------------+----------+----------+----------+
|                  | T n, D h | T m, D h | T n, D c |
+------------------+----------+----------+----------+
| Total memory use | 2.07     | 2.09     | 2.07     |
| Huge pages       | 0        | 1.3      | 1.25     |
+------------------+----------+----------+----------+

Changes
---------
v1-v2:
Added benchmarks
Added damos_filter_type documentation for new action to fix kernel-doc

Signed-off-by: Asier Gutierrez <gutierrez.asier@huawei-partners.com>
---
 Documentation/mm/damon/design.rst      |  4 ++++
 include/linux/damon.h                  |  2 ++
 mm/damon/sysfs-schemes.c               |  4 ++++
 mm/damon/vaddr.c                       |  3 +++
 tools/testing/selftests/damon/sysfs.py | 11 ++++++-----
 5 files changed, 19 insertions(+), 5 deletions(-)

diff --git a/Documentation/mm/damon/design.rst b/Documentation/mm/damon/design.rst
index 838b14d22519..405142641e55 100644
--- a/Documentation/mm/damon/design.rst
+++ b/Documentation/mm/damon/design.rst
@@ -467,6 +467,10 @@ that supports each action are as below.
    Supported by ``vaddr`` and ``fvaddr`` operations set. When
    TRANSPARENT_HUGEPAGE is disabled, the application of the action will just
    fail.
+ - ``collapse``: Call ``madvise()`` for the region with ``MADV_COLLAPSE``.
+   Supported by ``vaddr`` and ``fvaddr`` operations set. When
+   TRANSPARENT_HUGEPAGE is disabled, the application of the action will just
+   fail.
  - ``lru_prio``: Prioritize the region on its LRU lists.
    Supported by ``paddr`` operations set.
  - ``lru_deprio``: Deprioritize the region on its LRU lists.
diff --git a/include/linux/damon.h b/include/linux/damon.h
index d9a3babbafc1..6941113968ec 100644
--- a/include/linux/damon.h
+++ b/include/linux/damon.h
@@ -121,6 +121,7 @@ struct damon_target {
  * @DAMOS_PAGEOUT:	Reclaim the region.
  * @DAMOS_HUGEPAGE:	Call ``madvise()`` for the region with MADV_HUGEPAGE.
  * @DAMOS_NOHUGEPAGE:	Call ``madvise()`` for the region with MADV_NOHUGEPAGE.
+ * @DAMOS_COLLAPSE:	Call ``madvise()`` for the region with MADV_COLLAPSE.
  * @DAMOS_LRU_PRIO:	Prioritize the region on its LRU lists.
  * @DAMOS_LRU_DEPRIO:	Deprioritize the region on its LRU lists.
  * @DAMOS_MIGRATE_HOT:  Migrate the regions prioritizing warmer regions.
@@ -140,6 +141,7 @@ enum damos_action {
 	DAMOS_PAGEOUT,
 	DAMOS_HUGEPAGE,
 	DAMOS_NOHUGEPAGE,
+	DAMOS_COLLAPSE,
 	DAMOS_LRU_PRIO,
 	DAMOS_LRU_DEPRIO,
 	DAMOS_MIGRATE_HOT,
diff --git a/mm/damon/sysfs-schemes.c b/mm/damon/sysfs-schemes.c
index 5186966dafb3..aa08a8f885fb 100644
--- a/mm/damon/sysfs-schemes.c
+++ b/mm/damon/sysfs-schemes.c
@@ -2041,6 +2041,10 @@ static struct damos_sysfs_action_name damos_sysfs_action_names[] = {
 		.action = DAMOS_NOHUGEPAGE,
 		.name = "nohugepage",
 	},
+	{
+		.action = DAMOS_COLLAPSE,
+		.name = "collapse",
+	},
 	{
 		.action = DAMOS_LRU_PRIO,
 		.name = "lru_prio",
diff --git a/mm/damon/vaddr.c b/mm/damon/vaddr.c
index b069dbc7e3d2..dd5f2d7027ac 100644
--- a/mm/damon/vaddr.c
+++ b/mm/damon/vaddr.c
@@ -903,6 +903,9 @@ static unsigned long damon_va_apply_scheme(struct damon_ctx *ctx,
 	case DAMOS_NOHUGEPAGE:
 		madv_action = MADV_NOHUGEPAGE;
 		break;
+	case DAMOS_COLLAPSE:
+		madv_action = MADV_COLLAPSE;
+		break;
 	case DAMOS_MIGRATE_HOT:
 	case DAMOS_MIGRATE_COLD:
 		return damos_va_migrate(t, r, scheme, sz_filter_passed);
diff --git a/tools/testing/selftests/damon/sysfs.py b/tools/testing/selftests/damon/sysfs.py
index 3aa5c91548a5..c6476e63f4fb 100755
--- a/tools/testing/selftests/damon/sysfs.py
+++ b/tools/testing/selftests/damon/sysfs.py
@@ -123,11 +123,12 @@ def assert_scheme_committed(scheme, dump):
             'pageout': 2,
             'hugepage': 3,
             'nohugeapge': 4,
-            'lru_prio': 5,
-            'lru_deprio': 6,
-            'migrate_hot': 7,
-            'migrate_cold': 8,
-            'stat': 9,
+            'collapse': 5
+            'lru_prio': 6,
+            'lru_deprio': 7,
+            'migrate_hot': 8,
+            'migrate_cold': 9,
+            'stat': 10,
             }
     assert_true(dump['action'] == action_val[scheme.action], 'action', dump)
     assert_true(dump['apply_interval_us'] == scheme. apply_interval_us,
-- 
2.43.0

[RFC PATCH v2 1/1] mm/damon: support MADV_COLLAPSE via DAMOS_COLLAPSE scheme action

[gutierrez.asier]

From: Asier Gutierrez <gutierrez.asier@huawei-partners.com>

This patch set introces a new action:  DAMOS_COLLAPSE.

For DAMOS_HUGEPAGE and DAMOS_NOHUGEPAGE to work, khugepaged should be
working, since it relies on hugepage_madvise to add a new slot. This
slot should be picked up by khugepaged and eventually collapse (or
not, if we are using DAMOS_NOHUGEPAGE) the pages. If THP is not
enabled, khugepaged will not be working, and therefore no collapse
will happen.

DAMOS_COLLAPSE eventually calls madvise_collapse, which will collapse
the address range synchronously.

This new action may be required to support autotuning with hugepage as
a goal[1].

[1]: https://lore.kernel.org/damon/20260313000816.79933-1-sj@kernel.org/

---------
Benchmarks:

T n: THP never
T m: THP madvise
D h: DAMON action hugepage
D c: DAMON action collapse

+------------------+----------+----------+----------+
|                  | T n, D h | T m, D h | T n, D c |
+------------------+----------+----------+----------+
| Total memory use | 2.07     | 2.09     | 2.07     |
| Huge pages       | 0        | 1.3      | 1.25     |
+------------------+----------+----------+----------+

Changes
---------
v1 -> v2:
Added benchmarks
Added damos_filter_type documentation for new action to fix kernel-doc

Signed-off-by: Asier Gutierrez <gutierrez.asier@huawei-partners.com>
---
 Documentation/mm/damon/design.rst      |  4 ++++
 include/linux/damon.h                  |  2 ++
 mm/damon/sysfs-schemes.c               |  4 ++++
 mm/damon/vaddr.c                       |  3 +++
 tools/testing/selftests/damon/sysfs.py | 11 ++++++-----
 5 files changed, 19 insertions(+), 5 deletions(-)

diff --git a/Documentation/mm/damon/design.rst b/Documentation/mm/damon/design.rst
index 838b14d22519..405142641e55 100644
--- a/Documentation/mm/damon/design.rst
+++ b/Documentation/mm/damon/design.rst
@@ -467,6 +467,10 @@ that supports each action are as below.
    Supported by ``vaddr`` and ``fvaddr`` operations set. When
    TRANSPARENT_HUGEPAGE is disabled, the application of the action will just
    fail.
+ - ``collapse``: Call ``madvise()`` for the region with ``MADV_COLLAPSE``.
+   Supported by ``vaddr`` and ``fvaddr`` operations set. When
+   TRANSPARENT_HUGEPAGE is disabled, the application of the action will just
+   fail.
  - ``lru_prio``: Prioritize the region on its LRU lists.
    Supported by ``paddr`` operations set.
  - ``lru_deprio``: Deprioritize the region on its LRU lists.
diff --git a/include/linux/damon.h b/include/linux/damon.h
index d9a3babbafc1..6941113968ec 100644
--- a/include/linux/damon.h
+++ b/include/linux/damon.h
@@ -121,6 +121,7 @@ struct damon_target {
  * @DAMOS_PAGEOUT:	Reclaim the region.
  * @DAMOS_HUGEPAGE:	Call ``madvise()`` for the region with MADV_HUGEPAGE.
  * @DAMOS_NOHUGEPAGE:	Call ``madvise()`` for the region with MADV_NOHUGEPAGE.
+ * @DAMOS_COLLAPSE:	Call ``madvise()`` for the region with MADV_COLLAPSE.
  * @DAMOS_LRU_PRIO:	Prioritize the region on its LRU lists.
  * @DAMOS_LRU_DEPRIO:	Deprioritize the region on its LRU lists.
  * @DAMOS_MIGRATE_HOT:  Migrate the regions prioritizing warmer regions.
@@ -140,6 +141,7 @@ enum damos_action {
 	DAMOS_PAGEOUT,
 	DAMOS_HUGEPAGE,
 	DAMOS_NOHUGEPAGE,
+	DAMOS_COLLAPSE,
 	DAMOS_LRU_PRIO,
 	DAMOS_LRU_DEPRIO,
 	DAMOS_MIGRATE_HOT,
diff --git a/mm/damon/sysfs-schemes.c b/mm/damon/sysfs-schemes.c
index 5186966dafb3..aa08a8f885fb 100644
--- a/mm/damon/sysfs-schemes.c
+++ b/mm/damon/sysfs-schemes.c
@@ -2041,6 +2041,10 @@ static struct damos_sysfs_action_name damos_sysfs_action_names[] = {
 		.action = DAMOS_NOHUGEPAGE,
 		.name = "nohugepage",
 	},
+	{
+		.action = DAMOS_COLLAPSE,
+		.name = "collapse",
+	},
 	{
 		.action = DAMOS_LRU_PRIO,
 		.name = "lru_prio",
diff --git a/mm/damon/vaddr.c b/mm/damon/vaddr.c
index b069dbc7e3d2..dd5f2d7027ac 100644
--- a/mm/damon/vaddr.c
+++ b/mm/damon/vaddr.c
@@ -903,6 +903,9 @@ static unsigned long damon_va_apply_scheme(struct damon_ctx *ctx,
 	case DAMOS_NOHUGEPAGE:
 		madv_action = MADV_NOHUGEPAGE;
 		break;
+	case DAMOS_COLLAPSE:
+		madv_action = MADV_COLLAPSE;
+		break;
 	case DAMOS_MIGRATE_HOT:
 	case DAMOS_MIGRATE_COLD:
 		return damos_va_migrate(t, r, scheme, sz_filter_passed);
diff --git a/tools/testing/selftests/damon/sysfs.py b/tools/testing/selftests/damon/sysfs.py
index 3aa5c91548a5..c6476e63f4fb 100755
--- a/tools/testing/selftests/damon/sysfs.py
+++ b/tools/testing/selftests/damon/sysfs.py
@@ -123,11 +123,12 @@ def assert_scheme_committed(scheme, dump):
             'pageout': 2,
             'hugepage': 3,
             'nohugeapge': 4,
-            'lru_prio': 5,
-            'lru_deprio': 6,
-            'migrate_hot': 7,
-            'migrate_cold': 8,
-            'stat': 9,
+            'collapse': 5
+            'lru_prio': 6,
+            'lru_deprio': 7,
+            'migrate_hot': 8,
+            'migrate_cold': 9,
+            'stat': 10,
             }
     assert_true(dump['action'] == action_val[scheme.action], 'action', dump)
     assert_true(dump['apply_interval_us'] == scheme. apply_interval_us,
-- 
2.43.0