[Patch v2 0/2] mm/rmap: small cleanup for __folio_remove_rmap()

[Patch v2 0/2] mm/rmap: small cleanup for __folio_remove_rmap()

[Wei Yang]

During code reading, I found some small places where we can improve based on
the knowledge we already have.

Patch 1 remove a mask of on nr, since we are in an if branch conditional which
implies the upper bits are already clear.
Patch 2 use folio_large_nr_pages() to get the nr since we know this is a large
folio.

v2:
  * remove patch 2
  * adjust patch 1 changelog
  * adjust cover letter description

Wei Yang (2):
  mm/rmap: not necessary to mask off FOLIO_PAGES_MAPPED
  mm/rmap: use folio_large_nr_pages() when we are sure it is a large
    folio

 mm/rmap.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

-- 
2.34.1

[Patch v2 1/2] mm/rmap: not necessary to mask off FOLIO_PAGES_MAPPED

[Wei Yang]

At this point, we are in an if branch conditional on
(nr < ENTIRELY_MAPPED), and FOLIO_PAGES_MAPPED is equal to
(ENTIRELY_MAPPED - 1). This means the upper bits are already cleared.

Not necessary to mask off it.

Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Harry Yoo <harry.yoo@oracle.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Acked-by: David Hildenbrand <david@redhat.com>

---
v2:
  * adjust the changelog based on Lorenzo's comment
---
 mm/rmap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/rmap.c b/mm/rmap.c
index 1c5988dbd1e7..a927437a56c2 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -1749,7 +1749,7 @@ static __always_inline void __folio_remove_rmap(struct folio *folio,
 				nr_pages = folio_large_nr_pages(folio);
 				if (level == PGTABLE_LEVEL_PMD)
 					nr_pmdmapped = nr_pages;
-				nr = nr_pages - (nr & FOLIO_PAGES_MAPPED);
+				nr = nr_pages - nr;
 				/* Raced ahead of another remove and an add? */
 				if (unlikely(nr < 0))
 					nr = 0;
-- 
2.34.1

[Patch v2 2/2] mm/rmap: use folio_large_nr_pages() when we are sure it is a large folio

[Wei Yang]

Non-large folio is handled at the beginning, so it is a large folio for
sure.

Use folio_large_nr_pages() here like elsewhere.

Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Harry Yoo <harry.yoo@oracle.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Acked-by: David Hildenbrand <david@redhat.com>
---
 mm/rmap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/rmap.c b/mm/rmap.c
index a927437a56c2..14cf3b24ae50 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -1703,7 +1703,7 @@ static __always_inline void __folio_remove_rmap(struct folio *folio,
 			nr = folio_sub_return_large_mapcount(folio, nr_pages, vma);
 			if (!nr) {
 				/* Now completely unmapped. */
-				nr = folio_nr_pages(folio);
+				nr = folio_large_nr_pages(folio);
 			} else {
 				partially_mapped = nr < folio_large_nr_pages(folio) &&
 						   !folio_entire_mapcount(folio);
-- 
2.34.1