mm: handle large folio when large folio in VM_LOCKED VMA range

If large folio is in the range of VM_LOCKED VMA, it should be mlocked to
avoid being picked by page reclaim.  Which may split the large folio and
then mlock each pages again.

Mlock this kind of large folio to prevent them being picked by page
reclaim.

For the large folio which cross the boundary of VM_LOCKED VMA or not fully
mapped to VM_LOCKED VMA, we'd better not to mlock it.  So if the system is
under memory pressure, this kind of large folio will be split and the
pages ouf of VM_LOCKED VMA can be reclaimed.

Ideally, for large folio, we should mlock it when the large folio is fully
mapped to VMA and munlock it if any page are unmampped from VMA.  But it's
not easy to detect whether the large folio is fully mapped to VMA in some
cases (like add/remove rmap).  So we update mlock_vma_folio() and
munlock_vma_folio() to mlock/munlock the folio according to vma->vm_flags.
Let caller to decide whether they should call these two functions.

For add rmap, only mlock normal 4K folio and postpone large folio handling
to page reclaim phase.  It is possible to reuse page table iterator to
detect whether folio is fully mapped or not during page reclaim phase. 
For remove rmap, invoke munlock_vma_folio() to munlock folio unconditionly
because rmap makes folio not fully mapped to VMA.

Link: https://lkml.kernel.org/r/20230918073318.1181104-3-fengwei.yin@intel.com
Signed-off-by: Yin Fengwei <fengwei.yin@intel.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: Yang Shi <shy828301@gmail.com>
Cc: Yosry Ahmed <yosryahmed@google.com>
Cc: Yu Zhao <yuzhao@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

2 files changed, 68 insertions, 21 deletions

diff --git a/mm/internal.h b/mm/internal.h
index 14f358d068ac..9e62c8b952b8 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -644,14 +644,10 @@ folio_within_vma(struct folio *folio, struct vm_area_struct *vma)
  * mlock is usually called at the end of page_add_*_rmap(), munlock at
  * the end of page_remove_rmap(); but new anon folios are managed by
  * folio_add_lru_vma() calling mlock_new_folio().
- *
- * @compound is used to include pmd mappings of THPs, but filter out
- * pte mappings of THPs, which cannot be consistently counted: a pte
- * mapping of the THP head cannot be distinguished by the page alone.
  */
 void mlock_folio(struct folio *folio);
 static inline void mlock_vma_folio(struct folio *folio,
-			struct vm_area_struct *vma, bool compound)
+				struct vm_area_struct *vma)
 {
 	/*
 	 * The VM_SPECIAL check here serves two purposes.
@@ -661,17 +657,24 @@ static inline void mlock_vma_folio(struct folio *folio,
 	 *    file->f_op->mmap() is using vm_insert_page(s), when VM_LOCKED may
 	 *    still be set while VM_SPECIAL bits are added: so ignore it then.
 	 */
-	if (unlikely((vma->vm_flags & (VM_LOCKED|VM_SPECIAL)) == VM_LOCKED) &&
-	    (compound || !folio_test_large(folio)))
+	if (unlikely((vma->vm_flags & (VM_LOCKED|VM_SPECIAL)) == VM_LOCKED))
 		mlock_folio(folio);
 }
 
 void munlock_folio(struct folio *folio);
 static inline void munlock_vma_folio(struct folio *folio,
-			struct vm_area_struct *vma, bool compound)
+					struct vm_area_struct *vma)
 {
-	if (unlikely(vma->vm_flags & VM_LOCKED) &&
-	    (compound || !folio_test_large(folio)))
+	/*
+	 * munlock if the function is called. Ideally, we should only
+	 * do munlock if any page of folio is unmapped from VMA and
+	 * cause folio not fully mapped to VMA.
+	 *
+	 * But it's not easy to confirm that's the situation. So we
+	 * always munlock the folio and page reclaim will correct it
+	 * if it's wrong.
+	 */
+	if (unlikely(vma->vm_flags & VM_LOCKED))
 		munlock_folio(folio);
 }
 
diff --git a/mm/rmap.c b/mm/rmap.c
index d24e2c36372e..c6bb2339e35b 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -798,6 +798,7 @@ struct folio_referenced_arg {
 	unsigned long vm_flags;
 	struct mem_cgroup *memcg;
 };
+
 /*
  * arg: folio_referenced_arg will be passed
  */
@@ -807,17 +808,33 @@ static bool folio_referenced_one(struct folio *folio,
 	struct folio_referenced_arg *pra = arg;
 	DEFINE_FOLIO_VMA_WALK(pvmw, folio, vma, address, 0);
 	int referenced = 0;
+	unsigned long start = address, ptes = 0;
 
 	while (page_vma_mapped_walk(&pvmw)) {
 		address = pvmw.address;
 
-		if ((vma->vm_flags & VM_LOCKED) &&
-		    (!folio_test_large(folio) || !pvmw.pte)) {
-			/* Restore the mlock which got missed */
-			mlock_vma_folio(folio, vma, !pvmw.pte);
-			page_vma_mapped_walk_done(&pvmw);
-			pra->vm_flags |= VM_LOCKED;
-			return false; /* To break the loop */
+		if (vma->vm_flags & VM_LOCKED) {
+			if (!folio_test_large(folio) || !pvmw.pte) {
+				/* Restore the mlock which got missed */
+				mlock_vma_folio(folio, vma);
+				page_vma_mapped_walk_done(&pvmw);
+				pra->vm_flags |= VM_LOCKED;
+				return false; /* To break the loop */
+			}
+			/*
+			 * For large folio fully mapped to VMA, will
+			 * be handled after the pvmw loop.
+			 *
+			 * For large folio cross VMA boundaries, it's
+			 * expected to be picked  by page reclaim. But
+			 * should skip reference of pages which are in
+			 * the range of VM_LOCKED vma. As page reclaim
+			 * should just count the reference of pages out
+			 * the range of VM_LOCKED vma.
+			 */
+			ptes++;
+			pra->mapcount--;
+			continue;
 		}
 
 		if (pvmw.pte) {
@@ -842,6 +859,23 @@ static bool folio_referenced_one(struct folio *folio,
 		pra->mapcount--;
 	}
 
+	if ((vma->vm_flags & VM_LOCKED) &&
+			folio_test_large(folio) &&
+			folio_within_vma(folio, vma)) {
+		unsigned long s_align, e_align;
+
+		s_align = ALIGN_DOWN(start, PMD_SIZE);
+		e_align = ALIGN_DOWN(start + folio_size(folio) - 1, PMD_SIZE);
+
+		/* folio doesn't cross page table boundary and fully mapped */
+		if ((s_align == e_align) && (ptes == folio_nr_pages(folio))) {
+			/* Restore the mlock which got missed */
+			mlock_vma_folio(folio, vma);
+			pra->vm_flags |= VM_LOCKED;
+			return false; /* To break the loop */
+		}
+	}
+
 	if (referenced)
 		folio_clear_idle(folio);
 	if (folio_test_clear_young(folio))
@@ -1254,7 +1288,14 @@ void page_add_anon_rmap(struct page *page, struct vm_area_struct *vma,
 			  (folio_test_large(folio) && folio_entire_mapcount(folio) > 1)) &&
 			 PageAnonExclusive(page), folio);
 
-	mlock_vma_folio(folio, vma, compound);
+	/*
+	 * For large folio, only mlock it if it's fully mapped to VMA. It's
+	 * not easy to check whether the large folio is fully mapped to VMA
+	 * here. Only mlock normal 4K folio and leave page reclaim to handle
+	 * large folio.
+	 */
+	if (!folio_test_large(folio))
+		mlock_vma_folio(folio, vma);
 }
 
 /**
@@ -1354,7 +1395,9 @@ void folio_add_file_rmap_range(struct folio *folio, struct page *page,
 	if (nr)
 		__lruvec_stat_mod_folio(folio, NR_FILE_MAPPED, nr);
 
-	mlock_vma_folio(folio, vma, compound);
+	/* See comments in page_add_anon_rmap() */
+	if (!folio_test_large(folio))
+		mlock_vma_folio(folio, vma);
 }
 
 /**
@@ -1465,7 +1508,7 @@ void page_remove_rmap(struct page *page, struct vm_area_struct *vma,
 	 * it's only reliable while mapped.
 	 */
 
-	munlock_vma_folio(folio, vma, compound);
+	munlock_vma_folio(folio, vma);
 }
 
 /*
@@ -1530,7 +1573,8 @@ static bool try_to_unmap_one(struct folio *folio, struct vm_area_struct *vma,
 		if (!(flags & TTU_IGNORE_MLOCK) &&
 		    (vma->vm_flags & VM_LOCKED)) {
 			/* Restore the mlock which got missed */
-			mlock_vma_folio(folio, vma, false);
+			if (!folio_test_large(folio))
+				mlock_vma_folio(folio, vma);
 			page_vma_mapped_walk_done(&pvmw);
 			ret = false;
 			break;