[PATCH 7/8] reiserfs: Convert convert_tail_for_hole() to use folios

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



It's not clear to me that reiserfs will ever support large folios, but
now this function will operate correctly if they are enabled.

Signed-off-by: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx>
---
 fs/reiserfs/inode.c | 45 +++++++++++++++++++++++----------------------
 1 file changed, 23 insertions(+), 22 deletions(-)

diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index b79848111957..008855ddb365 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -571,57 +571,58 @@ static int convert_tail_for_hole(struct inode *inode,
 	unsigned long index;
 	unsigned long tail_end;
 	unsigned long tail_start;
-	struct page *tail_page;
-	struct page *hole_page = bh_result->b_page;
+	struct folio *tail_folio;
+	struct folio *hole_folio = bh_result->b_folio;
 	int retval = 0;
 
 	if ((tail_offset & (bh_result->b_size - 1)) != 1)
 		return -EIO;
 
-	/* always try to read until the end of the block */
-	tail_start = tail_offset & (PAGE_SIZE - 1);
-	tail_end = (tail_start | (bh_result->b_size - 1)) + 1;
-
 	index = tail_offset >> PAGE_SHIFT;
 	/*
-	 * hole_page can be zero in case of direct_io, we are sure
+	 * hole_folio can be zero in case of direct_io, we are sure
 	 * that we cannot get here if we write with O_DIRECT into tail page
 	 */
-	if (!hole_page || index != hole_page->index) {
-		tail_page = grab_cache_page(inode->i_mapping, index);
+	if (!hole_folio || !folio_contains(hole_folio, index)) {
+		tail_folio = __filemap_get_folio(inode->i_mapping, index,
+				FGP_LOCK | FGP_ACCESSED | FGP_CREAT,
+				mapping_gfp_mask(inode->i_mapping));
 		retval = -ENOMEM;
-		if (!tail_page) {
+		if (!tail_folio)
 			goto out;
-		}
 	} else {
-		tail_page = hole_page;
+		tail_folio = hole_folio;
 	}
 
+	/* always try to read until the end of the block */
+	tail_start = offset_in_folio(tail_folio, tail_offset);
+	tail_end = (tail_start | (bh_result->b_size - 1)) + 1;
+
 	/*
 	 * we don't have to make sure the conversion did not happen while
-	 * we were locking the page because anyone that could convert
+	 * we were locking the folio because anyone that could convert
 	 * must first take i_mutex.
 	 *
-	 * We must fix the tail page for writing because it might have buffers
+	 * We must fix the tail folio for writing because it might have buffers
 	 * that are mapped, but have a block number of 0.  This indicates tail
-	 * data that has been read directly into the page, and
+	 * data that has been read directly into the folio, and
 	 * __block_write_begin won't trigger a get_block in this case.
 	 */
-	fix_tail_page_for_writing(tail_page);
-	retval = __reiserfs_write_begin(tail_page, tail_start,
+	fix_tail_page_for_writing(&tail_folio->page);
+	retval = __reiserfs_write_begin(&tail_folio->page, tail_start,
 				      tail_end - tail_start);
 	if (retval)
 		goto unlock;
 
 	/* tail conversion might change the data in the page */
-	flush_dcache_page(tail_page);
+	flush_dcache_folio(tail_folio);
 
-	retval = reiserfs_commit_write(NULL, tail_page, tail_start, tail_end);
+	retval = reiserfs_commit_write(NULL, &tail_folio->page, tail_start, tail_end);
 
 unlock:
-	if (tail_page != hole_page) {
-		unlock_page(tail_page);
-		put_page(tail_page);
+	if (tail_folio != hole_folio) {
+		folio_unlock(tail_folio);
+		folio_put(tail_folio);
 	}
 out:
 	return retval;
-- 
2.35.1




[Index of Archives]     [Linux Ext4 Filesystem]     [Union Filesystem]     [Filesystem Testing]     [Ceph Users]     [Ecryptfs]     [NTFS 3]     [AutoFS]     [Kernel Newbies]     [Share Photos]     [Security]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux Cachefs]     [Reiser Filesystem]     [Linux RAID]     [NTFS 3]     [Samba]     [Device Mapper]     [CEPH Development]

  Powered by Linux