From: Ryusuke Konishi <konishi.ryusuke@xxxxxxxxx>
commit 8cf57c6df818f58fdad16a909506be213623a88e upstream.
In nilfs_rename(), calls to nilfs_put_page() to release pages obtained
with nilfs_find_entry() or nilfs_dotdot() are alternated in the normal
path.
When replacing the kernel memory mapping method from kmap to
kmap_local_{page,folio}, this violates the constraint on the calling order
of kunmap_local().
Swap the order of nilfs_put_page calls where the kmap sections of multiple
pages overlap so that they are nested, allowing direct replacement of
nilfs_put_page() -> unmap_and_put_page().
Without this reordering, that replacement will cause a kernel WARNING in
kunmap_local_indexed() on architectures with high memory mapping.
Link: https://lkml.kernel.org/r/20231127143036.2425-3-konishi.ryusuke@xxxxxxxxx
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@xxxxxxxxx>
Reviewed-by: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Stable-dep-of: ee70999a988b ("nilfs2: handle errors that nilfs_prepare_chunk() may return")
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
fs/nilfs2/namei.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
--- a/fs/nilfs2/namei.c
+++ b/fs/nilfs2/namei.c
@@ -431,13 +431,14 @@ static int nilfs_rename(struct mnt_idmap
inode_set_ctime_current(old_inode);
nilfs_delete_entry(old_de, old_page);
- nilfs_put_page(old_page);
if (dir_de) {
nilfs_set_link(old_inode, dir_de, dir_page, new_dir);
nilfs_put_page(dir_page);
drop_nlink(old_dir);
}
+ nilfs_put_page(old_page);
+
nilfs_mark_inode_dirty(old_dir);
nilfs_mark_inode_dirty(old_inode);
Patches currently in stable-queue which might be from konishi.ryusuke@xxxxxxxxx are
queue-6.6/nilfs2-handle-errors-that-nilfs_prepare_chunk-may-return.patch
queue-6.6/nilfs2-move-page-release-outside-of-nilfs_delete_entry-and-nilfs_set_link.patch
queue-6.6/nilfs2-eliminate-staggered-calls-to-kunmap-in-nilfs_rename.patch
