KASAN: use-after-free in __mutex_lock

43 views
Skip to first unread message

Sungwoo Kim

unread,
Sep 3, 2022, 4:54:33 PMSep 3
to mar...@holtmann.org, johan....@gmail.com, luiz....@gmail.com, da...@davemloft.net, edum...@google.com, ku...@kernel.org, pab...@redhat.com, linux-b...@vger.kernel.org, net...@vger.kernel.org, linux-...@vger.kernel.org, syzk...@googlegroups.com
Hello, I am Sungwoo Kim (https://sung-woo.kim) from Purdue University.

We report a bug found by FuzzBT, a modified version of syzkaller.
It looks similar to the recent bug:
https://lore.kernel.org/lkml/20220622082716.4...@linaro.org

We propose to add l2cap_chan_hold_unless_zero() for after calling
__l2cap_get_chan_blah().

Bluetooth: l2cap_core.c:static void l2cap_chan_destroy(struct kref *kref)
Bluetooth: chan 0000000023c4974d
Bluetooth: parent 00000000ae861c08
==================================================================
BUG: KASAN: use-after-free in __mutex_waiter_is_first
kernel/locking/mutex.c:191 [inline]
BUG: KASAN: use-after-free in __mutex_lock_common
kernel/locking/mutex.c:671 [inline]
BUG: KASAN: use-after-free in __mutex_lock+0x278/0x400
kernel/locking/mutex.c:729
Read of size 8 at addr ffff888006a49b08 by task kworker/u3:2/389

CPU: 0 PID: 389 Comm: kworker/u3:2 Not tainted 5.15.0 #14
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
1.13.0-1ubuntu1.1 04/01/2014
Workqueue: hci0 hci_rx_work
Call Trace:
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0x7e/0xb7 lib/dump_stack.c:106
print_address_description+0x88/0x3b0 mm/kasan/report.c:256
__kasan_report mm/kasan/report.c:442 [inline]
kasan_report+0x172/0x1c0 mm/kasan/report.c:459
__asan_load8+0x80/0x90 mm/kasan/generic.c:256
__mutex_waiter_is_first kernel/locking/mutex.c:191 [inline]
__mutex_lock_common kernel/locking/mutex.c:671 [inline]
__mutex_lock+0x278/0x400 kernel/locking/mutex.c:729
__mutex_lock_slowpath+0x13/0x20 kernel/locking/mutex.c:979
mutex_lock+0x91/0xf0 kernel/locking/mutex.c:280
l2cap_chan_lock include/net/bluetooth/l2cap.h:890 [inline]
l2cap_connect_create_rsp net/bluetooth/l2cap_core.c:4431 [inline]
l2cap_bredr_sig_cmd+0xc7c/0x4060 net/bluetooth/l2cap_core.c:5865
l2cap_sig_channel net/bluetooth/l2cap_core.c:6633 [inline]
l2cap_recv_frame+0x71e/0xa00 net/bluetooth/l2cap_core.c:7899
l2cap_recv_acldata+0x3a2/0x6c0 net/bluetooth/l2cap_core.c:8641
hci_acldata_packet net/bluetooth/hci_core.c:5100 [inline]
hci_rx_work+0x39c/0x500 net/bluetooth/hci_core.c:5307
process_one_work+0x28c/0x440 kernel/workqueue.c:2297
worker_thread+0x434/0x5d0 kernel/workqueue.c:2444
kthread+0x214/0x250 kernel/kthread.c:319
ret_from_fork+0x1f/0x30

Allocated by task 389:
kasan_save_stack mm/kasan/common.c:38 [inline]
kasan_set_track mm/kasan/common.c:46 [inline]
set_alloc_info mm/kasan/common.c:434 [inline]
____kasan_kmalloc+0xd6/0x110 mm/kasan/common.c:513
__kasan_kmalloc+0x9/0x10 mm/kasan/common.c:522
kasan_kmalloc include/linux/kasan.h:264 [inline]
kmem_cache_alloc_trace+0x1a9/0x230 mm/slub.c:3240
kmalloc include/linux/slab.h:591 [inline]
kzalloc include/linux/slab.h:721 [inline]
l2cap_chan_create+0x4c/0x2c0 net/bluetooth/l2cap_core.c:481
l2cap_sock_alloc+0xfa/0x150 net/bluetooth/l2cap_sock.c:1867
l2cap_sock_new_connection_cb+0xc9/0x140 net/bluetooth/l2cap_sock.c:1467
l2cap_connect+0x3eb/0x850 net/bluetooth/l2cap_core.c:4274
l2cap_connect_req net/bluetooth/l2cap_core.c:4387 [inline]
l2cap_bredr_sig_cmd+0x13fd/0x4060 net/bluetooth/l2cap_core.c:5860
l2cap_sig_channel net/bluetooth/l2cap_core.c:6633 [inline]
l2cap_recv_frame+0x71e/0xa00 net/bluetooth/l2cap_core.c:7899
l2cap_recv_acldata+0x3a2/0x6c0 net/bluetooth/l2cap_core.c:8641
hci_acldata_packet net/bluetooth/hci_core.c:5100 [inline]
hci_rx_work+0x39c/0x500 net/bluetooth/hci_core.c:5307
process_one_work+0x28c/0x440 kernel/workqueue.c:2297
worker_thread+0x434/0x5d0 kernel/workqueue.c:2444
kthread+0x214/0x250 kernel/kthread.c:319
ret_from_fork+0x1f/0x30

Freed by task 364:
kasan_save_stack mm/kasan/common.c:38 [inline]
kasan_set_track+0x3d/0x70 mm/kasan/common.c:46
kasan_set_free_info+0x23/0x40 mm/kasan/generic.c:360
____kasan_slab_free+0x13a/0x170 mm/kasan/common.c:366
__kasan_slab_free+0x11/0x20 mm/kasan/common.c:374
kasan_slab_free include/linux/kasan.h:230 [inline]
slab_free_hook mm/slub.c:1700 [inline]
slab_free_freelist_hook+0x100/0x1a0 mm/slub.c:1726
slab_free mm/slub.c:3492 [inline]
kfree+0xe0/0x270 mm/slub.c:4552
l2cap_chan_destroy net/bluetooth/l2cap_core.c:524 [inline]
kref_put include/linux/kref.h:65 [inline]
l2cap_chan_put+0x147/0x190 net/bluetooth/l2cap_core.c:540
l2cap_sock_cleanup_listen net/bluetooth/l2cap_sock.c:1449 [inline]
l2cap_sock_teardown_cb+0x33a/0x3e0 net/bluetooth/l2cap_sock.c:1564
l2cap_chan_close+0x297/0x640
l2cap_sock_shutdown+0x6cc/0x8b0 net/bluetooth/l2cap_sock.c:1367
l2cap_sock_release+0x7d/0x130 net/bluetooth/l2cap_sock.c:1411
__sock_release+0x80/0x170 net/socket.c:649
sock_close+0x1e/0x30 net/socket.c:1314
__fput+0x2de/0x5d0 fs/file_table.c:280
____fput+0x1a/0x20 fs/file_table.c:313
task_work_run+0x102/0x150 kernel/task_work.c:164
exit_task_work include/linux/task_work.h:32 [inline]
do_exit+0x511/0x1260 kernel/exit.c:825
do_group_exit+0xd9/0x1a0 kernel/exit.c:922
get_signal+0x397/0x11a0 kernel/signal.c:2855
arch_do_signal_or_restart+0x3b/0x3b0 arch/x86/kernel/signal.c:865
handle_signal_work kernel/entry/common.c:148 [inline]
exit_to_user_mode_loop kernel/entry/common.c:172 [inline]
exit_to_user_mode_prepare+0x103/0x1a0 kernel/entry/common.c:207
__syscall_exit_to_user_mode_work kernel/entry/common.c:289 [inline]
syscall_exit_to_user_mode+0x2a/0x40 kernel/entry/common.c:300
do_syscall_64+0x52/0xc0 arch/x86/entry/common.c:86
entry_SYSCALL_64_after_hwframe+0x44/0xae

The buggy address belongs to the object at ffff888006a49800
which belongs to the cache kmalloc-1k of size 1024
The buggy address is located 776 bytes inside of
1024-byte region [ffff888006a49800, ffff888006a49c00)
The buggy address belongs to the page:
page:000000004d4d912c refcount:1 mapcount:0 mapping:0000000000000000
index:0x0 pfn:0x6a48
head:000000004d4d912c order:2 compound_mapcount:0 compound_pincount:0
flags: 0xfffffc0010200(slab|head|node=0|zone=1|lastcpupid=0x1fffff)
raw: 000fffffc0010200 dead000000000100 dead000000000122 ffff888001041dc0
raw: 0000000000000000 0000000000080008 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected

Memory state around the buggy address:
ffff888006a49a00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
ffff888006a49a80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff888006a49b00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
^
ffff888006a49b80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
ffff888006a49c00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
==================================================================


diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c
index 2c9de67da..0e7978228 100644
--- a/net/bluetooth/l2cap_core.c
+++ b/net/bluetooth/l2cap_core.c
@@ -4309,7 +4309,10 @@ static int l2cap_connect_create_rsp(struct
l2cap_conn *conn,

err = 0;

- l2cap_chan_lock(chan);
+ chan = l2cap_chan_hold_unless_zero(chan);
+ if (chan) {
+ l2cap_chan_lock(chan);
+ }

switch (result) {
case L2CAP_CR_SUCCESS:
@@ -4336,6 +4339,7 @@ static int l2cap_connect_create_rsp(struct
l2cap_conn *conn,
}

l2cap_chan_unlock(chan);
+ l2cap_chan_put(chan);

unlock:
mutex_unlock(&conn->chan_lock);


Best regards,
Sungwoo Kim.

Luiz Augusto von Dentz

unread,
Sep 29, 2022, 4:21:31 PMSep 29
to Sungwoo Kim, mar...@holtmann.org, johan....@gmail.com, da...@davemloft.net, edum...@google.com, ku...@kernel.org, pab...@redhat.com, linux-b...@vger.kernel.org, net...@vger.kernel.org, linux-...@vger.kernel.org, syzk...@googlegroups.com
Hi Kim,
This obviously doesn't work since you will need to bail out if chan is
NULL, so we would need to do something like:

if (!chan) {
err = -EBADSLT;
goto unlock;
}

l2cap_chan_lock(chan);

>
> switch (result) {
> case L2CAP_CR_SUCCESS:
> @@ -4336,6 +4339,7 @@ static int l2cap_connect_create_rsp(struct
> l2cap_conn *conn,
> }
>
> l2cap_chan_unlock(chan);
> + l2cap_chan_put(chan);
>
> unlock:
> mutex_unlock(&conn->chan_lock);
>
>
> Best regards,
> Sungwoo Kim.



--
Luiz Augusto von Dentz

Sungwoo Kim

unread,
Sep 30, 2022, 12:09:46 PMSep 30
to luiz....@gmail.com, da...@davemloft.net, edum...@google.com, i...@sung-woo.kim, johan....@gmail.com, ku...@kernel.org, linux-b...@vger.kernel.org, linux-...@vger.kernel.org, mar...@holtmann.org, net...@vger.kernel.org, pab...@redhat.com, syzk...@googlegroups.com
Hi Dentz,
How about to use l2cap_get_chan_by_scid because it looks resposible to
handle ref_cnt.

Signed-off-by: Sungwoo Kim <i...@sung-woo.kim>
---
net/bluetooth/l2cap_core.c | 24 +++++++-----------------
1 file changed, 7 insertions(+), 17 deletions(-)

diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c
index 2c9de67da..d3a074cbc 100644
--- a/net/bluetooth/l2cap_core.c
+++ b/net/bluetooth/l2cap_core.c
@@ -4291,26 +4291,18 @@ static int l2cap_connect_create_rsp(struct l2cap_conn *conn,
BT_DBG("dcid 0x%4.4x scid 0x%4.4x result 0x%2.2x status 0x%2.2x",
dcid, scid, result, status);

- mutex_lock(&conn->chan_lock);
-
if (scid) {
- chan = __l2cap_get_chan_by_scid(conn, scid);
- if (!chan) {
- err = -EBADSLT;
- goto unlock;
- }
+ chan = l2cap_get_chan_by_scid(conn, scid);
+ if (!chan)
+ return -EBADSLT;
} else {
- chan = __l2cap_get_chan_by_ident(conn, cmd->ident);
- if (!chan) {
- err = -EBADSLT;
- goto unlock;
- }
+ chan = l2cap_get_chan_by_ident(conn, cmd->ident);
+ if (!chan)
+ return -EBADSLT;
}

err = 0;

- l2cap_chan_lock(chan);
-
switch (result) {
case L2CAP_CR_SUCCESS:
l2cap_state_change(chan, BT_CONFIG);
@@ -4336,9 +4328,7 @@ static int l2cap_connect_create_rsp(struct l2cap_conn *conn,
}

l2cap_chan_unlock(chan);
-
-unlock:
- mutex_unlock(&conn->chan_lock);
+ l2cap_chan_put(chan);

return err;
}
--
2.25.1

Luiz Augusto von Dentz

unread,
Sep 30, 2022, 1:46:55 PMSep 30
to Sungwoo Kim, da...@davemloft.net, edum...@google.com, johan....@gmail.com, ku...@kernel.org, linux-b...@vger.kernel.org, linux-...@vger.kernel.org, mar...@holtmann.org, net...@vger.kernel.org, pab...@redhat.com, syzk...@googlegroups.com
Hi Kim,
Ive sent a fix yesterday:

https://patchwork.kernel.org/project/bluetooth/patch/20220929203241.41...@gmail.com/

Both are sorta similar but the one above end up causing less code
changes which might be easier to backport.
Reply all
Reply to author
Forward
0 new messages