[RFC PATCH v2 8/9] mm: Handle COW PTE with reclaim algorithm

[Chih-En Lin <shiyn.lin@gmail.com>]

To avoid the PFRA reclaiming the page resided in the COWed PTE table,
break COW when it using rmap to unmap all the processes.

Signed-off-by: Chih-En Lin <shiyn.lin@gmail.com>
---
 include/linux/rmap.h | 2 ++
 mm/page_vma_mapped.c | 5 +++++
 mm/rmap.c            | 2 +-
 mm/swapfile.c        | 1 +
 mm/vmscan.c          | 1 +
 5 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index b89b4b86951f8..5c7e3bedc068b 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -312,6 +312,8 @@ int make_device_exclusive_range(struct mm_struct *mm, unsigned long start,
 #define PVMW_SYNC		(1 << 0)
 /* Look for migration entries rather than present PTEs */
 #define PVMW_MIGRATION		(1 << 1)
+/* Break COW PTE during the walking */
+#define PVMW_COW_PTE		(1 << 2)
 
 struct page_vma_mapped_walk {
 	unsigned long pfn;
diff --git a/mm/page_vma_mapped.c b/mm/page_vma_mapped.c
index 8e9e574d535aa..5008957bbe4a7 100644
--- a/mm/page_vma_mapped.c
+++ b/mm/page_vma_mapped.c
@@ -251,6 +251,11 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 			step_forward(pvmw, PMD_SIZE);
 			continue;
 		}
+
+		/* TODO: Does breaking COW PTE here is correct? */
+		if (pvmw->flags & PVMW_COW_PTE)
+			handle_cow_pte(vma, pvmw->pmd, pvmw->address, false);
+
 		if (!map_pte(pvmw))
 			goto next_pte;
 this_pte:
diff --git a/mm/rmap.c b/mm/rmap.c
index 93d5a6f793d20..8f737cb44e48a 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -1477,7 +1477,7 @@ static bool try_to_unmap_one(struct folio *folio, struct vm_area_struct *vma,
 		     unsigned long address, void *arg)
 {
 	struct mm_struct *mm = vma->vm_mm;
-	DEFINE_FOLIO_VMA_WALK(pvmw, folio, vma, address, 0);
+	DEFINE_FOLIO_VMA_WALK(pvmw, folio, vma, address, PVMW_COW_PTE);
 	pte_t pteval;
 	struct page *subpage;
 	bool anon_exclusive, ret = true;
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 1fdccd2f1422e..ef4d3d81a824b 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -1916,6 +1916,7 @@ static inline int unuse_pmd_range(struct vm_area_struct *vma, pud_t *pud,
 	do {
 		cond_resched();
 		next = pmd_addr_end(addr, end);
+		handle_cow_pte(vma, pmd, addr, false);
 		if (pmd_none_or_trans_huge_or_clear_bad(pmd))
 			continue;
 		ret = unuse_pte_range(vma, pmd, addr, next, type);
diff --git a/mm/vmscan.c b/mm/vmscan.c
index b2b1431352dcd..030fad3d310d9 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1822,6 +1822,7 @@ static unsigned int shrink_page_list(struct list_head *page_list,
 		/*
 		 * The folio is mapped into the page tables of one or more
 		 * processes. Try to unmap it here.
+		 * It will write to the page tables, break COW PTE here.
 		 */
 		if (folio_mapped(folio)) {
 			enum ttu_flags flags = TTU_BATCH_FLUSH;
-- 
2.37.3