Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Ryusuke Konishi <konishi.ryusuke@gmail.com>
To:     Andrew Morton <akpm@linux-foundation.org>
Cc:     linux-nilfs <linux-nilfs@vger.kernel.org>,
        LKML <linux-kernel@vger.kernel.org>
Subject: [PATCH] nilfs2: fix incorrect masking of permission flags for symlinks
Date:   Thu, 23 Jun 2022 17:54:01 +0900
Message-Id: <1655974441-5612-1-git-send-email-konishi.ryusuke@gmail.com>
Precedence: bulk

The permission flags of newly created symlinks are wrongly dropped on
nilfs2 with the current umask value even though symlinks should have
777 (rwxrwxrwx) permissions:

 $ umask
 0022
 $ touch file && ln -s file symlink; ls -l file symlink
 -rw-r--r--. 1 root root 0 Jun 23 16:29 file
 lrwxr-xr-x. 1 root root 4 Jun 23 16:29 symlink -> file

This fixes the bug by inserting a missing check that excludes
symlinks.

Reported-by: Tommy Pettersson <ptp@lysator.liu.se>
Reported-by: Ciprian Craciun <ciprian.craciun@gmail.com>
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Tested-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
---
 fs/nilfs2/nilfs.h | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/fs/nilfs2/nilfs.h b/fs/nilfs2/nilfs.h
index 1344f7d475d3..aecda4fc95f5 100644
--- a/fs/nilfs2/nilfs.h
+++ b/fs/nilfs2/nilfs.h
@@ -198,6 +198,9 @@ static inline int nilfs_acl_chmod(struct inode *inode)
 
 static inline int nilfs_init_acl(struct inode *inode, struct inode *dir)
 {
+	if (S_ISLNK(inode->i_mode))
+		return 0;
+
 	inode->i_mode &= ~current_umask();
 	return 0;
 }
-- 
1.8.3.1