lkml.org 
[lkml]   [2020]   [Nov]   [24]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: general protection fault in reiserfs_security_init
On Monday, September 21, 2020 at 5:32:22 PM UTC+8 syzbot wrote:
>
> Hello,
>
> syzbot found the following issue on:
>
> HEAD commit: 325d0eab Merge branch 'akpm' (patches from Andrew)
> git tree: upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=1671c0e3900000
> kernel config: https://syzkaller.appspot.com/x/.config?x=b12e84189082991c
> dashboard link: https://syzkaller.appspot.com/bug?extid=690cb1e51970435f9775
> compiler: gcc (GCC) 10.1.0-syz 20200507
> syz repro: https://syzkaller.appspot.com/x/repro.syz?x=15705a3d900000
> C reproducer: https://syzkaller.appspot.com/x/repro.c?x=117b3281900000
>
> IMPORTANT: if you fix the issue, please add the following tag to the commit:
> Reported-by: syzbot+690cb1...@syzkaller.appspotmail.com
>
> REISERFS (device loop0): journal params: device loop0, size 15748, journal first block 18, max trans len 256, max batch 225, max commit age 30, max trans age 30
> REISERFS (device loop0): checking transaction log (loop0)
> REISERFS (device loop0): Using tea hash to sort names
> REISERFS (device loop0): using 3.5.x disk format
> general protection fault, probably for non-canonical address 0xdffffc000000000d: 0000 [#1] PREEMPT SMP KASAN
> KASAN: null-ptr-deref in range [0x0000000000000068-0x000000000000006f]
> CPU: 0 PID: 6874 Comm: syz-executor834 Not tainted 5.9.0-rc5-syzkaller #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> RIP: 0010:d_really_is_negative include/linux/dcache.h:472 [inline]
> RIP: 0010:reiserfs_xattr_jcreate_nblocks fs/reiserfs/xattr.h:78 [inline]
> RIP: 0010:reiserfs_security_init+0x285/0x4d0 fs/reiserfs/xattr_security.c:70
> Code: 48 c1 ea 03 80 3c 02 00 0f 85 2b 02 00 00 4d 8b ad a0 05 00 00 48 b8 00 00 00 00 00 fc ff df 49 8d 7d 68 48 89 fa 48 c1 ea 03 <80> 3c 02 00 0f 85 23 02 00 00 49 83 7d 68 00 0f 84 62 01 00 00 48
> RSP: 0018:ffffc90005827980 EFLAGS: 00010202
> RAX: dffffc0000000000 RBX: 0000000000000036 RCX: 000000000000006c
> RDX: 000000000000000d RSI: ffffffff82009dd3 RDI: 0000000000000068
> RBP: ffff88807d8441d0 R08: ffffc90005827a10 R09: ffffc90005827a18
> R10: 0000000000000000 R11: 0000000000000000 R12: 00000000000005fa
> R13: 0000000000000000 R14: ffff888094e60000 R15: 0000000000000000
> FS: 0000000001036880(0000) GS:ffff8880ae400000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 00007f5a6fb90ab4 CR3: 000000009a1ab000 CR4: 00000000001506f0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> Call Trace:
> reiserfs_mkdir+0x2c9/0x980 fs/reiserfs/namei.c:821
> create_privroot fs/reiserfs/xattr.c:882 [inline]
> reiserfs_xattr_init+0x4de/0xb52 fs/reiserfs/xattr.c:1004


int reiserfs_xattr_init(struct super_block *s, int mount_flags)
{
int err = 0;
struct dentry *privroot = REISERFS_SB(s)->priv_root;

err = xattr_mount_check(s);
if (err)
goto error;

if (d_really_is_negative(privroot) && !(mount_flags & SB_RDONLY)) {
inode_lock(d_inode(s->s_root));
err = create_privroot(REISERFS_SB(s)->priv_root);
inode_unlock(d_inode(s->s_root));
}

if (d_really_is_positive(privroot)) {
inode_lock(d_inode(privroot));
if (!REISERFS_SB(s)->xattr_root) {
struct dentry *dentry;

dentry = lookup_one_len(XAROOT_NAME, privroot,
strlen(XAROOT_NAME));
if (!IS_ERR(dentry)) {
pr_alert("assign xattr_root with
dentry = 0x%lx", dentry);
REISERFS_SB(s)->xattr_root = dentry;
}else
err = PTR_ERR(dentry);
}
inode_unlock(d_inode(privroot));
}
......
}
From the implementation of reiserfs_xattr_init, only when
d_really_is_positive(privroot) is true, xattr_root could be assigned
with a dentry obtained from lookup_one_len. In other words,
create_privroot is executed with REISERFS_SB(s)->xattr_root as NULL
pointer. With improper implementation of mkdir operation in reiserfs
filesystem or accessing the xattr_root in reiserfs_mkdir , it can lead
to NULL pointer dereference. If you remove the red code in
reiserfs_xattr_jcreate_nblocks, the crash never occurs, but it may
affect nblocks calculation in the reiserfs filesystem.

static inline size_t reiserfs_xattr_jcreate_nblocks(struct inode *inode)
{
size_t nblocks = JOURNAL_BLOCKS_PER_OBJECT(inode->i_sb);

pr_alert("5: inode = 0x%lx", inode);
if ((REISERFS_I(inode)->i_flags & i_has_xattr_dir) == 0) {
nblocks += JOURNAL_BLOCKS_PER_OBJECT(inode->i_sb);
if (d_really_is_negative(REISERFS_SB(inode->i_sb)->xattr_root))
nblocks += JOURNAL_BLOCKS_PER_OBJECT(inode->i_sb);
}

return nblocks;
}

--
My best regards to you.

No System Is Safe!
Dongliang Mu

\
 
 \ /
  Last update: 2020-11-25 03:35    [W:0.036 / U:0.096 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site