[PATCH RESEND v5 4/4] mm/vmscan: avoid split lazyfree THP during shrink_folio_list()

[Lance Yang <ioworker0@gmail.com>]

When the user no longer requires the pages, they would use
madvise(MADV_FREE) to mark the pages as lazy free. Subsequently, they
typically would not re-write to that memory again.

During memory reclaim, if we detect that the large folio and its PMD are
both still marked as clean and there are no unexpected references
(such as GUP), so we can just discard the memory lazily, improving the
efficiency of memory reclamation in this case.

On an Intel i5 CPU, reclaiming 1GiB of lazyfree THPs using
mem_cgroup_force_empty() results in the following runtimes in seconds
(shorter is better):

--------------------------------------------
|     Old       |      New       |  Change  |
--------------------------------------------
|   0.683426    |    0.049197    |  -92.80% |
--------------------------------------------

Suggested-by: Zi Yan <ziy@nvidia.com>
Suggested-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Lance Yang <ioworker0@gmail.com>
---
 include/linux/huge_mm.h |  9 +++++
 mm/huge_memory.c        | 75 +++++++++++++++++++++++++++++++++++++++++
 mm/rmap.c               | 31 ++++++++++-------
 3 files changed, 103 insertions(+), 12 deletions(-)

diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h
index 9fcb0b0b6ed1..cfd7ec2b6d0a 100644
--- a/include/linux/huge_mm.h
+++ b/include/linux/huge_mm.h
@@ -411,6 +411,8 @@ static inline bool thp_migration_supported(void)
 
 void split_huge_pmd_locked(struct vm_area_struct *vma, unsigned long address,
 			   pmd_t *pmd, bool freeze, struct folio *folio);
+bool unmap_huge_pmd_locked(struct vm_area_struct *vma, unsigned long addr,
+			   pmd_t *pmdp, struct folio *folio);
 
 #else /* CONFIG_TRANSPARENT_HUGEPAGE */
 
@@ -478,6 +480,13 @@ static inline void split_huge_pmd_locked(struct vm_area_struct *vma,
 					 unsigned long address, pmd_t *pmd,
 					 bool freeze, struct folio *folio) {}
 
+static inline bool unmap_huge_pmd_locked(struct vm_area_struct *vma,
+					 unsigned long addr, pmd_t *pmdp,
+					 struct folio *folio)
+{
+	return false;
+}
+
 #define split_huge_pud(__vma, __pmd, __address)	\
 	do { } while (0)
 
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 425272c6c50b..3ceeeb2f42d4 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -2687,6 +2687,81 @@ static void unmap_folio(struct folio *folio)
 	try_to_unmap_flush();
 }
 
+static bool __discard_trans_pmd_locked(struct vm_area_struct *vma,
+				       unsigned long addr, pmd_t *pmdp,
+				       struct folio *folio)
+{
+	struct mm_struct *mm = vma->vm_mm;
+	int ref_count, map_count;
+	pmd_t orig_pmd = *pmdp;
+	struct mmu_gather tlb;
+	struct page *page;
+
+	if (pmd_dirty(orig_pmd) || folio_test_dirty(folio))
+		return false;
+	if (unlikely(!pmd_present(orig_pmd) || !pmd_trans_huge(orig_pmd)))
+		return false;
+
+	page = pmd_page(orig_pmd);
+	if (unlikely(page_folio(page) != folio))
+		return false;
+
+	tlb_gather_mmu(&tlb, mm);
+	orig_pmd = pmdp_huge_get_and_clear(mm, addr, pmdp);
+	tlb_remove_pmd_tlb_entry(&tlb, pmdp, addr);
+
+	/*
+	 * Syncing against concurrent GUP-fast:
+	 * - clear PMD; barrier; read refcount
+	 * - inc refcount; barrier; read PMD
+	 */
+	smp_mb();
+
+	ref_count = folio_ref_count(folio);
+	map_count = folio_mapcount(folio);
+
+	/*
+	 * Order reads for folio refcount and dirty flag
+	 * (see comments in __remove_mapping()).
+	 */
+	smp_rmb();
+
+	/*
+	 * If the PMD or folio is redirtied at this point, or if there are
+	 * unexpected references, we will give up to discard this folio
+	 * and remap it.
+	 *
+	 * The only folio refs must be one from isolation plus the rmap(s).
+	 */
+	if (ref_count != map_count + 1 || folio_test_dirty(folio) ||
+	    pmd_dirty(orig_pmd)) {
+		set_pmd_at(mm, addr, pmdp, orig_pmd);
+		return false;
+	}
+
+	folio_remove_rmap_pmd(folio, page, vma);
+	zap_deposited_table(mm, pmdp);
+	add_mm_counter(mm, MM_ANONPAGES, -HPAGE_PMD_NR);
+	if (vma->vm_flags & VM_LOCKED)
+		mlock_drain_local();
+	folio_put(folio);
+
+	return true;
+}
+
+bool unmap_huge_pmd_locked(struct vm_area_struct *vma, unsigned long addr,
+			   pmd_t *pmdp, struct folio *folio)
+{
+	VM_WARN_ON_FOLIO(!folio_test_pmd_mappable(folio), folio);
+	VM_WARN_ON_FOLIO(!folio_test_locked(folio), folio);
+	VM_WARN_ON_ONCE(!IS_ALIGNED(addr, HPAGE_PMD_SIZE));
+
+	if (folio_test_anon(folio) && !folio_test_swapbacked(folio))
+		return __discard_trans_pmd_locked(vma, addr, pmdp, folio);
+
+	return false;
+}
+
 static void remap_page(struct folio *folio, unsigned long nr)
 {
 	int i = 0;
diff --git a/mm/rmap.c b/mm/rmap.c
index 08a93347f283..e09f2141b8dc 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -1677,18 +1677,25 @@ static bool try_to_unmap_one(struct folio *folio, struct vm_area_struct *vma,
 			goto walk_done_err;
 		}
 
-		if (!pvmw.pte && (flags & TTU_SPLIT_HUGE_PMD)) {
-			/*
-			 * We temporarily have to drop the PTL and start once
-			 * again from that now-PTE-mapped page table.
-			 */
-			split_huge_pmd_locked(vma, range.start, pvmw.pmd, false,
-					      folio);
-			pvmw.pmd = NULL;
-			spin_unlock(pvmw.ptl);
-			pvmw.ptl = NULL;
-			flags &= ~TTU_SPLIT_HUGE_PMD;
-			continue;
+		if (!pvmw.pte) {
+			if (unmap_huge_pmd_locked(vma, range.start, pvmw.pmd,
+						  folio))
+				goto walk_done;
+
+			if (flags & TTU_SPLIT_HUGE_PMD) {
+				/*
+				 * We temporarily have to drop the PTL and start
+				 * once again from that now-PTE-mapped page
+				 * table.
+				 */
+				split_huge_pmd_locked(vma, range.start,
+						      pvmw.pmd, false, folio);
+				pvmw.pmd = NULL;
+				spin_unlock(pvmw.ptl);
+				pvmw.ptl = NULL;
+				flags &= ~TTU_SPLIT_HUGE_PMD;
+				continue;
+			}
 		}
 
 		/* Unexpected PMD-mapped THP? */
-- 
2.33.1