[PATCH] mm: stop recalculating min_free_kbytes once set by userspace

[PATCH] mm: stop recalculating min_free_kbytes once set by userspace

[Nimrod Oren]

min_free_kbytes is recalculated on memory hotplug and on THP sysfs
writes. The new value comes from calculate_min_free_kbytes() or, while
THP is enabled, from set_recommended_min_free_kbytes(). A value written
to the vm.min_free_kbytes sysctl is replaced whenever the new value is
higher, so a lower value does not stick. This matters most when THP is
enabled on configurations with large pageblocks, where the new value
can be excessive [1].

Stop recalculating min_free_kbytes once a value has been written to the
sysctl. Systems where the sysctl is never written keep the current
behavior.

Link: https://lore.kernel.org/r/20260716173504.760369-1-noren@nvidia.com/ [1]
Suggested-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
Link: https://lore.kernel.org/r/d202dcc3-d7d7-4e98-a9a5-3137cfbfd041@kernel.org/
Signed-off-by: Nimrod Oren <noren@nvidia.com>
---
 Documentation/admin-guide/sysctl/vm.rst | 7 +++++++
 mm/khugepaged.c                         | 7 +------
 mm/page_alloc.c                         | 6 +-----
 3 files changed, 9 insertions(+), 11 deletions(-)

diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentation/admin-guide/sysctl/vm.rst
index 5b318d17aa4b..189fc57dac5f 100644
--- a/Documentation/admin-guide/sysctl/vm.rst
+++ b/Documentation/admin-guide/sysctl/vm.rst
@@ -539,6 +539,13 @@ watermark[WMARK_MIN] value for each lowmem zone in the system.
 Each lowmem zone gets a number of reserved free pages based
 proportionally on its size.
 
+By default, the kernel sets this value based on the amount of lowmem
+and recalculates it on memory hotplug.  When Transparent Hugepage (THP)
+support is enabled through global or per-size controls, the kernel may
+also raise it to help keep pageblocks free and reduce fragmentation for
+THP allocations.  Once a value is written to this file, the kernel no
+longer adjusts it automatically.
+
 Some minimal amount of memory is needed to satisfy PF_MEMALLOC
 allocations; if you set this to lower than 1024KB, your system will
 become subtly broken, and prone to deadlock under high loads.
diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index dbae78b68dd8..3fd209f9377d 100644
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -3148,13 +3148,8 @@ void set_recommended_min_free_kbytes(void)
 			      (unsigned long) nr_free_buffer_pages() / 20);
 	recommended_min <<= (PAGE_SHIFT-10);
 
-	if (recommended_min > min_free_kbytes) {
-		if (user_min_free_kbytes >= 0)
-			pr_info_ratelimited("raising min_free_kbytes from %d to %lu to help transparent hugepage allocations\n",
-					    min_free_kbytes, recommended_min);
-
+	if (user_min_free_kbytes < 0 && recommended_min > min_free_kbytes)
 		min_free_kbytes = recommended_min;
-	}
 
 update_wmarks:
 	setup_per_zone_wmarks();
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 6ed36ba4d055..57a33da2562d 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -6737,12 +6737,8 @@ void calculate_min_free_kbytes(void)
 	lowmem_kbytes = nr_free_buffer_pages() * (PAGE_SIZE >> 10);
 	new_min_free_kbytes = int_sqrt(lowmem_kbytes * 16);
 
-	if (new_min_free_kbytes > user_min_free_kbytes)
+	if (user_min_free_kbytes < 0)
 		min_free_kbytes = clamp(new_min_free_kbytes, 128, 262144);
-	else
-		pr_warn_ratelimited("min_free_kbytes is not updated to %d because user defined value %d is preferred\n",
-				    new_min_free_kbytes, user_min_free_kbytes);
-
 }
 
 int __meminit init_per_zone_wmark_min(void)
-- 
2.45.0

Re: [PATCH] mm: stop recalculating min_free_kbytes once set by userspace

[Johannes Weiner]

On Wed, Oct 14, 2026 at 05:08:59PM +0300, Nimrod Oren wrote:
> min_free_kbytes is recalculated on memory hotplug and on THP sysfs
> writes. The new value comes from calculate_min_free_kbytes() or, while
> THP is enabled, from set_recommended_min_free_kbytes(). A value written
> to the vm.min_free_kbytes sysctl is replaced whenever the new value is
> higher, so a lower value does not stick. This matters most when THP is
> enabled on configurations with large pageblocks, where the new value
> can be excessive [1].
> 
> Stop recalculating min_free_kbytes once a value has been written to the
> sysctl. Systems where the sysctl is never written keep the current
> behavior.
> 
> Link: https://lore.kernel.org/r/20260716173504.760369-1-noren@nvidia.com/ [1]
> Suggested-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
> Link: https://lore.kernel.org/r/d202dcc3-d7d7-4e98-a9a5-3137cfbfd041@kernel.org/
> Signed-off-by: Nimrod Oren <noren@nvidia.com>

Reviewed-by: Johannes Weiner <hannes@cmpxchg.org>

Re: [PATCH] mm: stop recalculating min_free_kbytes once set by userspace

[Zi Yan]

On 14 Oct 2026, at 10:08, Nimrod Oren wrote:

> min_free_kbytes is recalculated on memory hotplug and on THP sysfs
> writes. The new value comes from calculate_min_free_kbytes() or, while
> THP is enabled, from set_recommended_min_free_kbytes(). A value written
> to the vm.min_free_kbytes sysctl is replaced whenever the new value is
> higher, so a lower value does not stick. This matters most when THP is
> enabled on configurations with large pageblocks, where the new value
> can be excessive [1].
>
> Stop recalculating min_free_kbytes once a value has been written to the
> sysctl. Systems where the sysctl is never written keep the current
> behavior.
>
> Link: https://lore.kernel.org/r/20260716173504.760369-1-noren@nvidia.com/ [1]
> Suggested-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
> Link: https://lore.kernel.org/r/d202dcc3-d7d7-4e98-a9a5-3137cfbfd041@kernel.org/
> Signed-off-by: Nimrod Oren <noren@nvidia.com>
> ---
>  Documentation/admin-guide/sysctl/vm.rst | 7 +++++++
>  mm/khugepaged.c                         | 7 +------
>  mm/page_alloc.c                         | 6 +-----
>  3 files changed, 9 insertions(+), 11 deletions(-)
>

Acked-by: Zi Yan <ziy@nvidia.com>

Best Regards,
Yan, Zi

Re: [PATCH] mm: stop recalculating min_free_kbytes once set by userspace

[Andrew Morton]

On Wed, 14 Oct 2026 17:08:59 +0300 Nimrod Oren <noren@nvidia.com> wrote:

> min_free_kbytes is recalculated on memory hotplug and on THP sysfs
> writes. The new value comes from calculate_min_free_kbytes() or, while
> THP is enabled, from set_recommended_min_free_kbytes(). A value written
> to the vm.min_free_kbytes sysctl is replaced whenever the new value is
> higher, so a lower value does not stick. This matters most when THP is
> enabled on configurations with large pageblocks, where the new value
> can be excessive [1].
> 
> Stop recalculating min_free_kbytes once a value has been written to the
> sysctl. Systems where the sysctl is never written keep the current
> behavior.

This problem lies somewhere in the range of "minor annoyance" to
"catastrophically horrible", but it isn't clear where in that range it
falls.

So can we please can we have a paragraph which explains the real-world
userspace-visible runtime effects of this problem?

Re: [PATCH] mm: stop recalculating min_free_kbytes once set by userspace

[Nimrod Oren]

On 15/10/2026 0:48, Andrew Morton wrote:
> On Wed, 14 Oct 2026 17:08:59 +0300 Nimrod Oren <noren@nvidia.com> wrote:
> 
>> min_free_kbytes is recalculated on memory hotplug and on THP sysfs
>> writes. The new value comes from calculate_min_free_kbytes() or, while
>> THP is enabled, from set_recommended_min_free_kbytes(). A value written
>> to the vm.min_free_kbytes sysctl is replaced whenever the new value is
>> higher, so a lower value does not stick. This matters most when THP is
>> enabled on configurations with large pageblocks, where the new value
>> can be excessive [1].
>>
>> Stop recalculating min_free_kbytes once a value has been written to the
>> sysctl. Systems where the sysctl is never written keep the current
>> behavior.
> 
> This problem lies somewhere in the range of "minor annoyance" to
> "catastrophically horrible", but it isn't clear where in that range it
> falls.
> 
> So can we please can we have a paragraph which explains the real-world
> userspace-visible runtime effects of this problem?

Sure. Can we add the following after the first paragraph?

  For example, with default arm64 pageblock sizes, the THP
  recommendation can reach 5.5 GiB per eligible zone with 64 KiB
  pages vs 22 MiB with 4 KiB pages. This keeps a significant amount
  of memory free, reducing what applications can use and triggering
  reclaim much earlier. Users can lower vm.min_free_kbytes to avoid
  this, but the lower value only sticks until the next recalculation.

Thanks

Re: [PATCH] mm: stop recalculating min_free_kbytes once set by userspace

[Vlastimil Babka (SUSE)]

On 10/14/26 16:08, Nimrod Oren wrote:
> min_free_kbytes is recalculated on memory hotplug and on THP sysfs
> writes. The new value comes from calculate_min_free_kbytes() or, while
> THP is enabled, from set_recommended_min_free_kbytes(). A value written
> to the vm.min_free_kbytes sysctl is replaced whenever the new value is
> higher, so a lower value does not stick. This matters most when THP is
> enabled on configurations with large pageblocks, where the new value
> can be excessive [1].
> 
> Stop recalculating min_free_kbytes once a value has been written to the
> sysctl. Systems where the sysctl is never written keep the current
> behavior.
> 
> Link: https://lore.kernel.org/r/20260716173504.760369-1-noren@nvidia.com/ [1]
> Suggested-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
> Link: https://lore.kernel.org/r/d202dcc3-d7d7-4e98-a9a5-3137cfbfd041@kernel.org/
> Signed-off-by: Nimrod Oren <noren@nvidia.com>

Reviewed-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>

Thanks!

Re: [PATCH] mm: stop recalculating min_free_kbytes once set by userspace

[Lorenzo Stoakes (ARM)]

On Wed, Oct 14, 2026 at 05:08:59PM +0300, Nimrod Oren wrote:
> min_free_kbytes is recalculated on memory hotplug and on THP sysfs
> writes. The new value comes from calculate_min_free_kbytes() or, while
> THP is enabled, from set_recommended_min_free_kbytes(). A value written
> to the vm.min_free_kbytes sysctl is replaced whenever the new value is
> higher, so a lower value does not stick. This matters most when THP is
> enabled on configurations with large pageblocks, where the new value
> can be excessive [1].
>
> Stop recalculating min_free_kbytes once a value has been written to the
> sysctl. Systems where the sysctl is never written keep the current
> behavior.
>
> Link: https://lore.kernel.org/r/20260716173504.760369-1-noren@nvidia.com/ [1]
> Suggested-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
> Link: https://lore.kernel.org/r/d202dcc3-d7d7-4e98-a9a5-3137cfbfd041@kernel.org/
> Signed-off-by: Nimrod Oren <noren@nvidia.com>

With an added, focused, paragraph about user-visible effects as requested
by Andrew (you can probably just reply to Andrew with something and he can
fold it in for you), LGTM and:

Reviewed-by: Lorenzo Stoakes (ARM) <ljs@kernel.org>

Thanks for sticking with this! :)

> diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentation/admin-guide/sysctl/vm.rst
> index 5b318d17aa4b..189fc57dac5f 100644
> --- a/Documentation/admin-guide/sysctl/vm.rst
> +++ b/Documentation/admin-guide/sysctl/vm.rst

> +By default, the kernel sets this value based on the amount of lowmem
> +and recalculates it on memory hotplug.  When Transparent Hugepage (THP)
> +support is enabled through global or per-size controls, the kernel may
> +also raise it to help keep pageblocks free and reduce fragmentation for
> +THP allocations.  Once a value is written to this file, the kernel no
> +longer adjusts it automatically.
> +

Ah nice to update this also!

--
Cheers, Lorenzo