Re: [PATCH v4 01/12] mm/khugepaged: deposit a newly allocated page table on collapse

On Wed, Sep 23, 2026 at 10:24:36AM +0800, Lance Yang wrote:

quoted

mm/khugepaged.c | 32 ++++++++++++++++++++++++++++++--
1 file changed, 30 insertions(+), 2 deletions(-)

diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index f49a6710933b..dab421f8233e 100644
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c

@@ -1278,6 +1278,23 @@ static enum scan_result alloc_charge_folio(struct folio **foliop, struct mm_stru

	return SCAN_SUCCEED;
}

+static pgtable_t alloc_deposit_pte(struct mm_struct *mm)
+{
+	/*
+	 * khugepaged is run from a kernel thread, so need to manually set the
+	 * correct memcg so the allocation gets charged correctly.
+	 */
+	struct mem_cgroup *memcg = get_mem_cgroup_from_mm(mm);
+	struct mem_cgroup *old_memcg = set_active_memcg(memcg);
+	pgtable_t pgtable;
+
+	pgtable = pte_alloc_one(mm);
+
+	set_active_memcg(old_memcg);
+	mem_cgroup_put(memcg);
+	return pgtable;
+}
+
/*
 * collapse_huge_page() expects the mmap_lock to be unlocked before entering and
 * will always return with the lock unlocked, to avoid holding the mmap_lock

@@ -1293,7 +1310,7 @@ static enum scan_result collapse_huge_page(struct mm_struct *mm, unsigned long s

	LIST_HEAD(compound_pagelist);
	pmd_t *pmd, _pmd;
	pte_t *pte = NULL;
-	pgtable_t pgtable;
+	pgtable_t pgtable = NULL;
	struct folio *folio;
	spinlock_t *pmd_ptl, *pte_ptl;
	enum scan_result result = SCAN_FAIL;

@@ -1310,6 +1327,12 @@ static enum scan_result collapse_huge_page(struct mm_struct *mm, unsigned long s

		goto out_nolock;
	}

+	if (is_pmd_order(order)) {
+		pgtable = alloc_deposit_pte(mm);
+		if (!pgtable)
+			goto out_nolock;

One small thing: result is still SCAN_SUCCEED after alloc_charge_folio(),

	result = alloc_charge_folio(&folio, mm, cc, order);
	if (result != SCAN_SUCCEED)
		goto out_nolock;

	if (folio_memcg_alloc_deferred(folio)) {
		result = SCAN_ALLOC_HUGE_PAGE_FAIL;
		goto out_nolock;
	}

	if (is_pmd_order(order)) {
		pgtable = alloc_deposit_pte(mm);
		if (!pgtable)
			goto out_nolock;
	}

so if alloc_deposit_pte() fails, collapse_huge_page() returns success
without installing a PMD.

I see sashiko pointed that out too :)

Cheers, Lance

Sigh ack yeah oops :) I think SCAN_ALLOC_HUGE_PAGE_FAIL is the closest though
it's not allocating a huge page so much as the page table, but gets the correct
MADV_COLLAPSE -ENOMEM back.

(Another case of local AI review not catchign something btw! Sashiko does lean
towards over-reporting/hallucinating but catches more)

Will ask Andrew to update in reply to this patch as a quick thing rather than
respin so quick, and fix up on any respin.

--
Cheers, Lorenzo