[PATCH] mm/khugepaged: cap min_free_kbytes recommendation at 1 GiB

[PATCH] mm/khugepaged: cap min_free_kbytes recommendation at 1 GiB

[Nimrod Oren]

When THP is enabled, set_recommended_min_free_kbytes() may raise
min_free_kbytes using a heuristic that scales with pageblock_nr_pages.
With MIGRATE_PCPTYPES equal to 3, the formula accounts for 11
pageblocks for every eligible populated zone before capping the result
at 5% of low memory.

This is reasonable when a pageblock is 2 MiB, as on common 4 KiB page
configurations, but scales poorly with larger base page sizes. With
the default arm64 pageblock sizes, the contribution per eligible zone
before the 5% cap is:

  4 KiB pages:   2 MiB pageblock,   22 MiB per zone
 16 KiB pages:  32 MiB pageblock,  352 MiB per zone
 64 KiB pages: 512 MiB pageblock,  5.5 GiB per zone

Consequently, min_free_kbytes can reach excessive and unwanted levels.

Add an absolute 1 GiB cap to the recommendation, in addition to the
existing percentage cap. This bounds the automatic recommendation to a
sane value on systems with large pageblocks while preserving existing
behavior for typical systems with 2 MiB pageblocks.

Link: https://lore.kernel.org/r/20260716173504.760369-1-noren@nvidia.com/
Suggested-by: Lorenzo Stoakes (ARM) <ljs@kernel.org>
Signed-off-by: Nimrod Oren <noren@nvidia.com>
---
v1:
* Cap the final recommendation at 1 GiB as suggested by Lorenzo.

RFC v1:
https://lore.kernel.org/r/20260716173504.760369-1-noren@nvidia.com/
---
 mm/khugepaged.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index 27e8f3077e80..d9caf10e05b5 100644
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -23,6 +23,7 @@
 #include <linux/ksm.h>
 #include <linux/pgalloc.h>
 #include <linux/backing-dev.h>
+#include <linux/sizes.h>
 
 #include <asm/tlb.h>
 #include "internal.h"
@@ -3096,9 +3097,13 @@ void set_recommended_min_free_kbytes(void)
 	recommended_min += pageblock_nr_pages * nr_zones *
 			   MIGRATE_PCPTYPES * MIGRATE_PCPTYPES;
 
-	/* don't ever allow to reserve more than 5% of the lowmem */
+	/*
+	 * Don't ever allow to reserve more than 5% of lowmem or 1 GiB,
+	 * whichever is smaller.
+	 */
 	recommended_min = min(recommended_min,
 			      (unsigned long) nr_free_buffer_pages() / 20);
+	recommended_min = min(recommended_min, SZ_1G / PAGE_SIZE);
 	recommended_min <<= (PAGE_SHIFT-10);
 
 	if (recommended_min > min_free_kbytes) {
-- 
2.45.0

Re: [PATCH] mm/khugepaged: cap min_free_kbytes recommendation at 1 GiB

[Andrew Morton]

On Tue, 28 Jul 2026 23:20:14 +0300 Nimrod Oren <noren@nvidia.com> wrote:

> When THP is enabled, set_recommended_min_free_kbytes() may raise
> min_free_kbytes using a heuristic that scales with pageblock_nr_pages.
> With MIGRATE_PCPTYPES equal to 3, the formula accounts for 11
> pageblocks for every eligible populated zone before capping the result
> at 5% of low memory.
> 
> This is reasonable when a pageblock is 2 MiB, as on common 4 KiB page
> configurations, but scales poorly with larger base page sizes. With
> the default arm64 pageblock sizes, the contribution per eligible zone
> before the 5% cap is:
> 
>   4 KiB pages:   2 MiB pageblock,   22 MiB per zone
>  16 KiB pages:  32 MiB pageblock,  352 MiB per zone
>  64 KiB pages: 512 MiB pageblock,  5.5 GiB per zone
> 
> Consequently, min_free_kbytes can reach excessive and unwanted levels.
> 
> Add an absolute 1 GiB cap to the recommendation, in addition to the
> existing percentage cap. This bounds the automatic recommendation to a
> sane value on systems with large pageblocks while preserving existing
> behavior for typical systems with 2 MiB pageblocks.

Another hard-coded number isn't pretty :(

>  mm/khugepaged.c | 7 ++++++-

Now, why is a page-allocator function sitting in khugepaged.c?

Perhaps to avoid a #ifdef CONFIG_TRANSPARENT_HUGEPAGE.

Or perhaps so that page-allocator maintainers don't get cc'ed on
page-allocator patches ;)

> 
> diff --git a/mm/khugepaged.c b/mm/khugepaged.c
> index 27e8f3077e80..d9caf10e05b5 100644
> --- a/mm/khugepaged.c
> +++ b/mm/khugepaged.c
> @@ -23,6 +23,7 @@
>  #include <linux/ksm.h>
>  #include <linux/pgalloc.h>
>  #include <linux/backing-dev.h>
> +#include <linux/sizes.h>

Why this?

>  #include <asm/tlb.h>
>  #include "internal.h"
> @@ -3096,9 +3097,13 @@ void set_recommended_min_free_kbytes(void)
>  	recommended_min += pageblock_nr_pages * nr_zones *
>  			   MIGRATE_PCPTYPES * MIGRATE_PCPTYPES;
>  
> -	/* don't ever allow to reserve more than 5% of the lowmem */
> +	/*
> +	 * Don't ever allow to reserve more than 5% of lowmem or 1 GiB,
> +	 * whichever is smaller.
> +	 */
>  	recommended_min = min(recommended_min,
>  			      (unsigned long) nr_free_buffer_pages() / 20);
> +	recommended_min = min(recommended_min, SZ_1G / PAGE_SIZE);
>  	recommended_min <<= (PAGE_SHIFT-10);
>  
>  	if (recommended_min > min_free_kbytes) {

Is a min_free_kbytes documentation update needed? 
Documentation/admin-guide/sysctl/vm.rst.