linux-kernel - Re: [syzbot] [hfs?] WARNING in hfsplus

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <tencent_C2474B031BD225AABA42BB7D33FC9E861B08@qq.com>
Date: Sun, 24 Nov 2024 22:38:23 +0800
From: Edward Adam Davis <eadavis@...com>
To: syzbot+028180f480a74961919c@...kaller.appspotmail.com
Cc: linux-fsdevel@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	syzkaller-bugs@...glegroups.com
Subject: Re: [syzbot] [hfs?] WARNING in hfsplus_unlink

After the rename syscall fails, the unlink syscall triggers a warning in
drop_nlink() because the i_nlink value is 0.

When unlink succeeds but rename_cat fails in rename, the i_nlink and flags
values of the inode of the new dentry should be restored.

#syz test

diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c
index f5c4b3e31a1c..b489d22409e7 100644
--- a/fs/hfsplus/dir.c
+++ b/fs/hfsplus/dir.c
@@ -534,7 +534,7 @@ static int hfsplus_rename(struct mnt_idmap *idmap,
 			  struct inode *new_dir, struct dentry *new_dentry,
 			  unsigned int flags)
 {
-	int res;
+	int res, unlinked_new = 0;
 
 	if (flags & ~RENAME_NOREPLACE)
 		return -EINVAL;
@@ -543,8 +543,10 @@ static int hfsplus_rename(struct mnt_idmap *idmap,
 	if (d_really_is_positive(new_dentry)) {
 		if (d_is_dir(new_dentry))
 			res = hfsplus_rmdir(new_dir, new_dentry);
-		else
+		else {
 			res = hfsplus_unlink(new_dir, new_dentry);
+			unlinked_new = res == 0 && d_inode(new_dentry)->i_flags & S_DEAD;
+		}
 		if (res)
 			return res;
 	}
@@ -554,6 +556,12 @@ static int hfsplus_rename(struct mnt_idmap *idmap,
 				 new_dir, &new_dentry->d_name);
 	if (!res)
 		new_dentry->d_fsdata = old_dentry->d_fsdata;
+	else if (unlinked_new) {
+		struct inode *inode = d_inode(new_dentry);
+		set_nlink(inode, inode->i_nlink + 1);
+		inode->i_flags &= ~S_DEAD;
+	}
+
 	return res;
 }