Re: [PATCH] mm: synchronize with PTE table reclaim for all folio sizes

From: Baolin Wang

Date: Mon Nov 16 2026 - 08:35:10 EST




On 11/16/26 8:03 PM, David Hildenbrand (Arm) wrote:
On 11/16/26 04:44, Baolin Wang wrote:


On 11/14/26 1:39 AM, David Hildenbrand (Arm) wrote:
On 11/5/26 03:56, Baolin Wang wrote:
Similar to commit 243cf1f67b36 ("mm: memory: fix truncation removing mapped
folios"), the '!pmd_present()' check in page_vma_mapped_walk() can also race
with concurrent page table reclamation via zap_empty_pte_table().

page_vma_mapped_walk() is used by many functions including try_to_unmap(),
try_to_migrate(), folio_mkclean(), pfn_mkclean_range(), remove_migration_ptes(),
page_mapped_in_vma(), and mapping_wrprotect_range(). Except for try_to_unmap()
and try_to_migrate(), the other callers do not perform post-walk accounting
checks, so the race has no observable impact on them.

For migration, the race can cause unnecessary retries. For reclaim via
try_to_unmap(), the race can cause failure to reclaim a folio that is in
fact already unmapped (as shown below), which may lead to unexpected OOM
or increased refaults.

CPU 0 (reclaim) CPU 1 (MADV_DONTNEED)
-------------------------------- --------------------------------
folio_lock(F)
try_to_unmap(F, TTU_SYNC)
zap_pte_range(full PMD range)
spin_lock(ptl)
clear PTEs
zap_empty_pte_table()
spin_trylock(pml)
pmd_clear()
spin_unlock(pml)

page_vma_mapped_walk()
observe !pmd_present(pmde)
nr_pages < HPAGE_PMD_NR
skip synchronization
return false

try_to_unmap() returns
folio_mapped(F) == true
nr_unmap_fail += folio_nr_pages(F)
tlb_flush_mmu_tlbonly()
tlb_flush_rmaps()
remove F's rmaps
spin_unlock(ptl)

Commit 243cf1f67b36 ("mm: memory: fix truncation removing mapped folios")
has already moved the PMD clear operation (via zap_empty_pte_table()) to
after tlb_flush_rmaps(). Similarly, page_vma_mapped_walk() also needs to
use sync_with_folio_pmd_zap() to synchronize and avoid memory ordering
issues on weakly ordered CPUs.

Fixes: 6375e95f381e ("mm: pgtable: reclaim empty PTE page in madvise(MADV_DONTNEED)")
Cc: stable@xxxxxxxxxxxxxxx
Suggested-by: David Hildenbrand (Arm) <david@xxxxxxxxxx>
Signed-off-by: Baolin Wang <baolin.wang@xxxxxxxxxxxxxxxxx>
---
mm/page_vma_mapped.c | 5 +----
1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/mm/page_vma_mapped.c b/mm/page_vma_mapped.c
index e6b6170c2f4f..8797467b6f0c 100644
--- a/mm/page_vma_mapped.c
+++ b/mm/page_vma_mapped.c
@@ -293,10 +293,7 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
spin_unlock(pvmw->ptl);
pvmw->ptl = NULL;
} else if (!pmd_present(pmde)) {
- if ((pvmw->flags & PVMW_SYNC) &&
- thp_vma_suitable_order(vma, pvmw->address,
- PMD_ORDER) &&
- (pvmw->nr_pages >= HPAGE_PMD_NR))
+ if (pvmw->flags & PVMW_SYNC)
sync_with_folio_pmd_zap(mm, pvmw->pmd);
goto next_pmd;

Baolin, how to proceed with that? Are you intending a resend? Including some of
the performance impact and ways forward to optimize?

I hope we can merge this fix first if no other objections. As I replied
to Barry, whether to set TTU_SYNC for small folios should be done in a
separate patch (this is a pre-existing issue) with proper measurement
(though it may be hard to measure the actual impact). What do you think?

Can you update the patch description to include some of the performance aspects
and planned next steps?

Sure.

However, regarding performance aspects, I'm not sure what kind of testing you expect. Since this fixes a uncommon concurrent race, do you want performance data under that race condition, or impact on common paths like try_to_unmap/try_to_migrate()?