Re: [PATCH v2 2/2] mm/ksm: prevent zero-page remapping during ksmd-triggered THP splits
From: Usama Arif
Date: Wed Nov 18 2026 - 16:27:49 EST
On Sun, 25 Oct 2026 11:39:47 -0600 "Nico Pache (Red Hat)" <nico.pache@xxxxxxxxx> wrote:
> Since commit b1f202060afe ("mm: remap unused subpages to shared zeropage
> when splitting isolated thp"), splitting an anonymous THP remaps all
> zero-filled subpages to the shared zeropage via TTU_USE_SHARED_ZEROPAGE.
> This flag is set unconditionally for every anonymous folio split,
> including splits triggered by KSM.
>
> When KSM is enabled with THP=always, this causes two regressions:
>
> 1. use_zero_pages=1: KSM calls try_to_merge_one_page() which triggers
> split_huge_page(). The split remaps all 512 zero-filled subpages to
> the shared zeropage at once, freeing the entire 2MB THP when KSM only
> intended to process a single 4KB page. This bypasses KSM's
> pages_to_scan rate limiting, causing ~1GB to be freed almost
> instantly.
Hey Nico,
I think I have had this conversation either on the previous revision
or the mailing list somewhere else before, so sorry for restarting this
again.
Isnt above a good thing? KSM is a memory saving feature, and this
is resulting in 1G being freed instantly?
>
> 2. use_zero_pages=0: The same split side-effect occurs through the
> stable/unstable tree merge paths. Each pages_to_scan iteration
> triggers an expensive split_huge_page() that silently frees 2MB,
> while the scanner wastes cycles on tree searches for zero-filled
> pages that were already freed as a side-effect.
As an alternate to having this patch, should we just have patch 1
and users should turn off split_underused_thp when using KSM.
If you have a large number of THPs on your machine, using KSM is a very
bad idea IMO due to all the splits that happen.
IMO, there are 2 optimum host conditions:
- A large number of THPs, and KSM disabled, hence shrinker can stay on.
- Very low number of THPs, KSM can be enabled or disabled, but as there
are a low number of THPs, the shrinker can be turned off.
>
> Introduce current_is_ksmd(), which uses kthread_func() to check whether
> the current task is the KSM scanning thread, and use it in
> try_to_map_unused_to_zeropage() to skip zero-page remapping when the
> split is triggered by ksmd.
>
> Fixes: b1f202060afe ("mm: remap unused subpages to shared zeropage when splitting isolated thp")
> Signed-off-by: Nico Pache (Red Hat) <nico.pache@xxxxxxxxx>
> ---
> include/linux/ksm.h | 6 ++++++
> mm/ksm.c | 5 +++++
> mm/migrate.c | 3 +++
> 3 files changed, 14 insertions(+)
>
> diff --git a/include/linux/ksm.h b/include/linux/ksm.h
> index d39d0d5483a2..41c06a6abf60 100644
> --- a/include/linux/ksm.h
> +++ b/include/linux/ksm.h
> @@ -17,6 +17,7 @@
> #ifdef CONFIG_KSM
> int ksm_madvise(struct vm_area_struct *vma, unsigned long start,
> unsigned long end, int advice, vm_flags_t *vm_flags);
> +bool current_is_ksmd(void);
> vma_flags_t ksm_vma_flags(struct mm_struct *mm, const struct file *file,
> vma_flags_t vma_flags);
> int ksm_enable_merge_any(struct mm_struct *mm);
> @@ -144,6 +145,11 @@ static inline int ksm_madvise(struct vm_area_struct *vma, unsigned long start,
> return 0;
> }
>
> +static inline bool current_is_ksmd(void)
> +{
> + return false;
> +}
> +
> static inline struct folio *ksm_might_need_to_copy(struct folio *folio,
> struct vm_area_struct *vma, unsigned long addr)
> {
> diff --git a/mm/ksm.c b/mm/ksm.c
> index 910a1307ac9f..e3ab037b1ebe 100644
> --- a/mm/ksm.c
> +++ b/mm/ksm.c
> @@ -2866,6 +2866,11 @@ static int ksm_scan_thread(void *nothing)
> return 0;
> }
>
> +bool current_is_ksmd(void)
> +{
> + return kthread_func(current) == ksm_scan_thread;
> +}
> +
> static bool __ksm_should_add_vma(const struct file *file, vma_flags_t vma_flags)
> {
> if (vma_flags_test(&vma_flags, VMA_MERGEABLE_BIT))
> diff --git a/mm/migrate.c b/mm/migrate.c
> index 7bdcdb57652f..c33ae3fa1043 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -305,6 +305,9 @@ static bool try_to_map_unused_to_zeropage(struct page_vma_mapped_walk *pvmw,
> if (PageCompound(page) || PageHWPoison(page))
> return false;
>
> + if (current_is_ksmd())
> + return false;
> +
> VM_BUG_ON_PAGE(!PageAnon(page), page);
> VM_BUG_ON_PAGE(!PageLocked(page), page);
> VM_BUG_ON_PAGE(pte_present(old_pte), page);
>
> --
> 2.56.0
>
>