public inbox for ntfs3@lists.linux.dev
 help / color / mirror / Atom feed
* [PATCH] fs/ntfs3: return folios from ntfs_lock_new_page()
@ 2026-02-05  9:49 Sun Jian
  2026-02-07 11:58 ` kernel test robot
                   ` (2 more replies)
  0 siblings, 3 replies; 6+ messages in thread
From: Sun Jian @ 2026-02-05  9:49 UTC (permalink / raw)
  To: Konstantin Komarov; +Cc: ntfs3, linux-kernel, Sun Jian

ntfs_lock_new_page() currently returns a struct page * but it primarily
operates on folios via __filemap_get_folio(). Convert it to return a
struct folio * and use folio_alloc() + __folio_set_locked() for the
temporary page used to avoid data corruption during decompression.

When the cached folio is not uptodate, keep the existing behavior by
calling folio_file_page() and converting the returned page back to a
folio. Defensively handle error pointers from folio_file_page().

ni_readpage_cmpr() is updated to consume the folio return value while
keeping the existing page array and cleanup logic unchanged.

Signed-off-by: Sun Jian <sun.jian.kdev@gmail.com>
---
 fs/ntfs3/frecord.c | 33 +++++++++++++++++++--------------
 1 file changed, 19 insertions(+), 14 deletions(-)

diff --git a/fs/ntfs3/frecord.c b/fs/ntfs3/frecord.c
index 641ddaf8d4a0..a23aaf34f001 100644
--- a/fs/ntfs3/frecord.c
+++ b/fs/ntfs3/frecord.c
@@ -2022,27 +2022,31 @@ int ni_fiemap(struct ntfs_inode *ni, struct fiemap_extent_info *fieinfo,
 	return err;
 }
 
-static struct page *ntfs_lock_new_page(struct address_space *mapping,
-		pgoff_t index, gfp_t gfp)
+static struct folio *ntfs_lock_new_page(struct address_space *mapping,
+					pgoff_t index, gfp_t gfp)
 {
 	struct folio *folio = __filemap_get_folio(mapping, index,
 			FGP_LOCK | FGP_ACCESSED | FGP_CREAT, gfp);
-	struct page *page;
 
 	if (IS_ERR(folio))
-		return ERR_CAST(folio);
+		return folio;
 
-	if (!folio_test_uptodate(folio))
-		return folio_file_page(folio, index);
+	if (!folio_test_uptodate(folio)) {
+		struct page *page = folio_file_page(folio, index);
+
+		if (IS_ERR(page))
+			return ERR_CAST(page);
+		return page_folio(page);
+	}
 
 	/* Use a temporary page to avoid data corruption */
 	folio_unlock(folio);
 	folio_put(folio);
-	page = alloc_page(gfp);
-	if (!page)
+	folio = folio_alloc(gfp, 0);
+	if (!folio)
 		return ERR_PTR(-ENOMEM);
-	__SetPageLocked(page);
-	return page;
+	__folio_set_locked(folio);
+	return folio;
 }
 
 /*
@@ -2064,6 +2068,7 @@ int ni_readpage_cmpr(struct ntfs_inode *ni, struct folio *folio)
 	u32 i, idx, frame_size, pages_per_frame;
 	gfp_t gfp_mask;
 	struct page *pg;
+	struct folio *f;
 
 	if (vbo >= i_size_read(&ni->vfs_inode)) {
 		folio_zero_range(folio, 0, folio_size(folio));
@@ -2099,12 +2104,12 @@ int ni_readpage_cmpr(struct ntfs_inode *ni, struct folio *folio)
 		if (i == idx)
 			continue;
 
-		pg = ntfs_lock_new_page(mapping, index, gfp_mask);
-		if (IS_ERR(pg)) {
-			err = PTR_ERR(pg);
+		f = ntfs_lock_new_page(mapping, index, gfp_mask);
+		if (IS_ERR(f)) {
+			err = PTR_ERR(f);
 			goto out1;
 		}
-		pages[i] = pg;
+		pages[i] = &f->page;
 	}
 
 	err = ni_read_frame(ni, frame_vbo, pages, pages_per_frame, 0);
-- 
2.43.0


^ permalink raw reply related	[flat|nested] 6+ messages in thread
end of thread, other threads:[~2026-02-24 18:49 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-02-05  9:49 [PATCH] fs/ntfs3: return folios from ntfs_lock_new_page() Sun Jian
2026-02-07 11:58 ` kernel test robot
2026-02-07 12:08 ` kernel test robot
2026-02-07 14:59   ` sun jian
2026-02-07 14:45 ` [PATCH v2] " Sun Jian
2026-02-24 18:49   ` Konstantin Komarov

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox