On 8/31/22 07:34, Karthik Alapati wrote:
free options in put_ntfs() rather than in ntfs_put_super()
BUG: memory leak
unreferenced object 0xffff8881104f3720 (size 32):
comm "syz-executor634", pid 3605, jiffies 4294945517 (age 12.630s)
hex dump (first 32 bytes):
e0 36 4f 10 81 88 ff ff 80 57 e1 85 ff ff ff ff .6O......W......
00 00 00 00 00 00 00 00 c0 ff c0 ff 00 00 00 00 ................
backtrace:
[<ffffffff81b23ef2>] kmalloc include/linux/slab.h:600 [inline]
[<ffffffff81b23ef2>] kzalloc include/linux/slab.h:733 [inline]
[<ffffffff81b23ef2>] ntfs_init_fs_context+0x22/0x1e0 fs/ntfs3/super.c:1397
[<ffffffff81650045>] alloc_fs_context+0x225/0x3b0 fs/fs_context.c:290
[<ffffffff8162bed3>] do_new_mount fs/namespace.c:3025 [inline]
[<ffffffff8162bed3>] path_mount+0x6f3/0x10d0 fs/namespace.c:3370
[<ffffffff8162cffe>] do_mount fs/namespace.c:3383 [inline]
[<ffffffff8162cffe>] __do_sys_mount fs/namespace.c:3591 [inline]
[<ffffffff8162cffe>] __se_sys_mount fs/namespace.c:3568 [inline]
[<ffffffff8162cffe>] __x64_sys_mount+0x18e/0x1d0 fs/namespace.c:3568
[<ffffffff845e4855>] do_syscall_x64 arch/x86/entry/common.c:50 [inline]
[<ffffffff845e4855>] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
[<ffffffff84800087>] entry_SYSCALL_64_after_hwframe+0x63/0xcd
BUG: memory leak
unreferenced object 0xffff8881104f36e0 (size 32):
comm "syz-executor634", pid 3605, jiffies 4294945517 (age 12.630s)
hex dump (first 32 bytes):
6d 61 63 63 79 72 69 6c 6c 69 63 00 00 00 00 00 maccyrillic.....
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
backtrace:
[<ffffffff814c5b1d>] kmemdup_nul+0x2d/0x70 mm/util.c:152
[<ffffffff8164f66e>] vfs_parse_fs_string+0x6e/0xd0 fs/fs_context.c:178
[<ffffffff8164f7b0>] generic_parse_monolithic+0xe0/0x130 fs/fs_context.c:224
[<ffffffff8162c374>] do_new_mount fs/namespace.c:3036 [inline]
[<ffffffff8162c374>] path_mount+0xb94/0x10d0 fs/namespace.c:3370
[<ffffffff8162cffe>] do_mount fs/namespace.c:3383 [inline]
[<ffffffff8162cffe>] __do_sys_mount fs/namespace.c:3591 [inline]
[<ffffffff8162cffe>] __se_sys_mount fs/namespace.c:3568 [inline]
[<ffffffff8162cffe>] __x64_sys_mount+0x18e/0x1d0 fs/namespace.c:3568
[<ffffffff845e4855>] do_syscall_x64 arch/x86/entry/common.c:50 [inline]
[<ffffffff845e4855>] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
[<ffffffff84800087>] entry_SYSCALL_64_after_hwframe+0x63/0xcd
Link: https://syzkaller.appspot.com/bug?id=332ba47915d0e39e94b42a622f195f0804ecb67f
Reported-by: syzbot+9d67170b20e8f94351c8@xxxxxxxxxxxxxxxxxxxxxxxxx
Signed-off-by: Karthik Alapati <mail@xxxxxxxxxxx>
---
fs/ntfs3/super.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/ntfs3/super.c b/fs/ntfs3/super.c
index 47012c9bf505..ffb76a5bfd4f 100644
--- a/fs/ntfs3/super.c
+++ b/fs/ntfs3/super.c
@@ -439,6 +439,7 @@ static void init_once(void *foo)
*/
static noinline void put_ntfs(struct ntfs_sb_info *sbi)
{
+ put_mount_options(sbi->options);
kfree(sbi->new_rec);
kvfree(ntfs_put_shared(sbi->upcase));
kfree(sbi->def_table);
@@ -482,7 +483,6 @@ static void ntfs_put_super(struct super_block *sb)
/* Mark rw ntfs as clear, if possible. */
ntfs_set_state(sbi, NTFS_DIRTY_CLEAR);
- put_mount_options(sbi->options);
put_ntfs(sbi);
sb->s_fs_info = NULL;
Hello
This patch failed our tests and caused email from kernel test robot,
so I can't accept it.
But the memory leak is a concern, thanks for the report.
