KASAN: use-after-free Read in iowarrior_disconnect

[syzbot]

Hello,

syzbot found the following crash on:

HEAD commit: d0847550 usb-fuzzer: main usb gadget fuzzer driver
git tree: https://github.com/google/kasan.git usb-fuzzer
console output: https://syzkaller.appspot.com/x/log.txt?x=139be302600000
kernel config: https://syzkaller.appspot.com/x/.config?x=dbc9c80cc095da19
dashboard link: https://syzkaller.appspot.com/bug?extid=cfe6d93e0abab9a0de05
compiler: gcc (GCC) 9.0.0 20181231 (experimental)
syz repro: https://syzkaller.appspot.com/x/repro.syz?x=12fe6b02600000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=1548189c600000

IMPORTANT: if you fix the bug, please add the following tag to the commit:
Reported-by: syzbot+cfe6d9...@syzkaller.appspotmail.com

==================================================================
BUG: KASAN: use-after-free in __mutex_lock_common
kernel/locking/mutex.c:912 [inline]
BUG: KASAN: use-after-free in __mutex_lock+0xf23/0x1360
kernel/locking/mutex.c:1077
Read of size 8 at addr ffff8881cc866f58 by task kworker/1:5/1755

CPU: 1 PID: 1755 Comm: kworker/1:5 Not tainted 5.3.0-rc4+ #26
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS
Google 01/01/2011
Workqueue: usb_hub_wq hub_event
Call Trace:
__dump_stack lib/dump_stack.c:77 [inline]
dump_stack+0xca/0x13e lib/dump_stack.c:113
print_address_description+0x6a/0x32c mm/kasan/report.c:351
__kasan_report.cold+0x1a/0x33 mm/kasan/report.c:482
kasan_report+0xe/0x12 mm/kasan/common.c:612
__mutex_lock_common kernel/locking/mutex.c:912 [inline]
__mutex_lock+0xf23/0x1360 kernel/locking/mutex.c:1077
iowarrior_disconnect+0xf0/0x2c0 drivers/usb/misc/iowarrior.c:878
usb_unbind_interface+0x1bd/0x8a0 drivers/usb/core/driver.c:423
__device_release_driver drivers/base/dd.c:1134 [inline]
device_release_driver_internal+0x42f/0x500 drivers/base/dd.c:1165
bus_remove_device+0x2dc/0x4a0 drivers/base/bus.c:556
device_del+0x420/0xb10 drivers/base/core.c:2339
usb_disable_device+0x211/0x690 drivers/usb/core/message.c:1237
usb_disconnect+0x284/0x8d0 drivers/usb/core/hub.c:2199
hub_port_connect drivers/usb/core/hub.c:4949 [inline]
hub_port_connect_change drivers/usb/core/hub.c:5213 [inline]
port_event drivers/usb/core/hub.c:5359 [inline]
hub_event+0x1454/0x3640 drivers/usb/core/hub.c:5441
process_one_work+0x92b/0x1530 kernel/workqueue.c:2269
worker_thread+0x96/0xe20 kernel/workqueue.c:2415
kthread+0x318/0x420 kernel/kthread.c:255
ret_from_fork+0x24/0x30 arch/x86/entry/entry_64.S:352

Allocated by task 1755:
save_stack+0x1b/0x80 mm/kasan/common.c:69
set_track mm/kasan/common.c:77 [inline]
__kasan_kmalloc mm/kasan/common.c:487 [inline]
__kasan_kmalloc.constprop.0+0xbf/0xd0 mm/kasan/common.c:460
kmalloc include/linux/slab.h:552 [inline]
kzalloc include/linux/slab.h:748 [inline]
iowarrior_probe+0x7a/0x10b2 drivers/usb/misc/iowarrior.c:753
usb_probe_interface+0x305/0x7a0 drivers/usb/core/driver.c:361
really_probe+0x281/0x6d0 drivers/base/dd.c:548
driver_probe_device+0x101/0x1b0 drivers/base/dd.c:721
__device_attach_driver+0x1c2/0x220 drivers/base/dd.c:828
bus_for_each_drv+0x162/0x1e0 drivers/base/bus.c:454
__device_attach+0x217/0x360 drivers/base/dd.c:894
bus_probe_device+0x1e4/0x290 drivers/base/bus.c:514
device_add+0xae6/0x16f0 drivers/base/core.c:2165
usb_set_configuration+0xdf6/0x1670 drivers/usb/core/message.c:2023
generic_probe+0x9d/0xd5 drivers/usb/core/generic.c:210
usb_probe_device+0x99/0x100 drivers/usb/core/driver.c:266
really_probe+0x281/0x6d0 drivers/base/dd.c:548
driver_probe_device+0x101/0x1b0 drivers/base/dd.c:721
__device_attach_driver+0x1c2/0x220 drivers/base/dd.c:828
bus_for_each_drv+0x162/0x1e0 drivers/base/bus.c:454
__device_attach+0x217/0x360 drivers/base/dd.c:894
bus_probe_device+0x1e4/0x290 drivers/base/bus.c:514
device_add+0xae6/0x16f0 drivers/base/core.c:2165
usb_new_device.cold+0x6a4/0xe79 drivers/usb/core/hub.c:2536
hub_port_connect drivers/usb/core/hub.c:5098 [inline]
hub_port_connect_change drivers/usb/core/hub.c:5213 [inline]
port_event drivers/usb/core/hub.c:5359 [inline]
hub_event+0x1b5c/0x3640 drivers/usb/core/hub.c:5441
process_one_work+0x92b/0x1530 kernel/workqueue.c:2269
worker_thread+0x96/0xe20 kernel/workqueue.c:2415
kthread+0x318/0x420 kernel/kthread.c:255
ret_from_fork+0x24/0x30 arch/x86/entry/entry_64.S:352

Freed by task 1857:
save_stack+0x1b/0x80 mm/kasan/common.c:69
set_track mm/kasan/common.c:77 [inline]
__kasan_slab_free+0x130/0x180 mm/kasan/common.c:449
slab_free_hook mm/slub.c:1423 [inline]
slab_free_freelist_hook mm/slub.c:1474 [inline]
slab_free mm/slub.c:3016 [inline]
kfree+0xe4/0x2f0 mm/slub.c:3957
iowarrior_delete drivers/usb/misc/iowarrior.c:246 [inline]
iowarrior_release+0x187/0x280 drivers/usb/misc/iowarrior.c:670
__fput+0x2d7/0x840 fs/file_table.c:280
task_work_run+0x13f/0x1c0 kernel/task_work.c:113
exit_task_work include/linux/task_work.h:22 [inline]
do_exit+0x8ef/0x2c00 kernel/exit.c:879
do_group_exit+0x125/0x340 kernel/exit.c:983
__do_sys_exit_group kernel/exit.c:994 [inline]
__se_sys_exit_group kernel/exit.c:992 [inline]
__x64_sys_exit_group+0x3a/0x50 kernel/exit.c:992
do_syscall_64+0xb7/0x580 arch/x86/entry/common.c:296
entry_SYSCALL_64_after_hwframe+0x49/0xbe

The buggy address belongs to the object at ffff8881cc866f00
which belongs to the cache kmalloc-512 of size 512
The buggy address is located 88 bytes inside of
512-byte region [ffff8881cc866f00, ffff8881cc867100)
The buggy address belongs to the page:
page:ffffea0007321980 refcount:1 mapcount:0 mapping:ffff8881da002500
index:0x0 compound_mapcount: 0
flags: 0x200000000010200(slab|head)
raw: 0200000000010200 ffffea000733fc80 0000000200000002 ffff8881da002500
raw: 0000000000000000 00000000000c000c 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected

Memory state around the buggy address:
ffff8881cc866e00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
ffff8881cc866e80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
> ffff8881cc866f00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
^
ffff8881cc866f80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
ffff8881cc867000: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
==================================================================


---
This bug is generated by a bot. It may contain errors.
See https://goo.gl/tpsmEJ for more information about syzbot.
syzbot engineers can be reached at syzk...@googlegroups.com.

syzbot will keep track of this bug report. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
syzbot can test patches for this bug, for details see:
https://goo.gl/tpsmEJ#testing-patches

[Oliver Neukum]

Am Montag, den 19.08.2019, 07:48 -0700 schrieb syzbot:
> Hello,
>
> syzbot found the following crash on:
>
> HEAD commit: d0847550 usb-fuzzer: main usb gadget fuzzer driver
> git tree: https://github.com/google/kasan.git usb-fuzzer
> console output: https://syzkaller.appspot.com/x/log.txt?x=139be302600000
> kernel config: https://syzkaller.appspot.com/x/.config?x=dbc9c80cc095da19
> dashboard link: https://syzkaller.appspot.com/bug?extid=cfe6d93e0abab9a0de05
> compiler: gcc (GCC) 9.0.0 20181231 (experimental)
> syz repro: https://syzkaller.appspot.com/x/repro.syz?x=12fe6b02600000
> C reproducer: https://syzkaller.appspot.com/x/repro.c?x=1548189c600000
>
> IMPORTANT: if you fix the bug, please add the following tag to the commit:
> Reported-by: syzbot+cfe6d9...@syzkaller.appspotmail.com
>

#syz test: https://github.com/google/kasan.git d0847550

[syzbot]

Hello,

syzbot has tested the proposed patch but the reproducer still triggered
crash:
possible deadlock in usb_deregister_dev

usb 4-1: USB disconnect, device number 2
======================================================
WARNING: possible circular locking dependency detected
5.3.0-rc4+ #1 Not tainted
------------------------------------------------------
kworker/1:1/21 is trying to acquire lock:
00000000bfac431a (minor_rwsem){++++}, at: usb_deregister_dev
drivers/usb/core/file.c:238 [inline]
00000000bfac431a (minor_rwsem){++++}, at: usb_deregister_dev+0x61/0x270
drivers/usb/core/file.c:230

but task is already holding lock:
000000007638fa06 (iowarrior_open_disc_lock){+.+.}, at:
iowarrior_disconnect+0x45/0x2c0 drivers/usb/misc/iowarrior.c:867

which lock already depends on the new lock.


the existing dependency chain (in reverse order) is:

-> #2 (iowarrior_open_disc_lock){+.+.}:
__mutex_lock_common kernel/locking/mutex.c:930 [inline]
__mutex_lock+0x158/0x1360 kernel/locking/mutex.c:1077
iowarrior_open+0x8a/0x2a0 drivers/usb/misc/iowarrior.c:600
usb_open+0x1df/0x270 drivers/usb/core/file.c:48
chrdev_open+0x219/0x5c0 fs/char_dev.c:414
do_dentry_open+0x494/0x1120 fs/open.c:797
do_last fs/namei.c:3416 [inline]
path_openat+0x1430/0x3f50 fs/namei.c:3533
do_filp_open+0x1a1/0x280 fs/namei.c:3563
do_sys_open+0x3c0/0x580 fs/open.c:1089
do_syscall_64+0xb7/0x580 arch/x86/entry/common.c:296
entry_SYSCALL_64_after_hwframe+0x49/0xbe

-> #1 (iowarrior_mutex){+.+.}:
__mutex_lock_common kernel/locking/mutex.c:930 [inline]
__mutex_lock+0x158/0x1360 kernel/locking/mutex.c:1077
iowarrior_open+0x23/0x2a0 drivers/usb/misc/iowarrior.c:589
usb_open+0x1df/0x270 drivers/usb/core/file.c:48
chrdev_open+0x219/0x5c0 fs/char_dev.c:414
do_dentry_open+0x494/0x1120 fs/open.c:797
do_last fs/namei.c:3416 [inline]
path_openat+0x1430/0x3f50 fs/namei.c:3533
do_filp_open+0x1a1/0x280 fs/namei.c:3563
do_sys_open+0x3c0/0x580 fs/open.c:1089
do_syscall_64+0xb7/0x580 arch/x86/entry/common.c:296
entry_SYSCALL_64_after_hwframe+0x49/0xbe

-> #0 (minor_rwsem){++++}:
check_prev_add kernel/locking/lockdep.c:2405 [inline]
check_prevs_add kernel/locking/lockdep.c:2507 [inline]
validate_chain kernel/locking/lockdep.c:2897 [inline]
__lock_acquire+0x1f7c/0x3b50 kernel/locking/lockdep.c:3880
lock_acquire+0x127/0x320 kernel/locking/lockdep.c:4412
down_write+0x92/0x150 kernel/locking/rwsem.c:1500
usb_deregister_dev drivers/usb/core/file.c:238 [inline]
usb_deregister_dev+0x61/0x270 drivers/usb/core/file.c:230
iowarrior_disconnect+0xa8/0x2c0 drivers/usb/misc/iowarrior.c:873

usb_unbind_interface+0x1bd/0x8a0 drivers/usb/core/driver.c:423
__device_release_driver drivers/base/dd.c:1134 [inline]
device_release_driver_internal+0x42f/0x500 drivers/base/dd.c:1165
bus_remove_device+0x2dc/0x4a0 drivers/base/bus.c:556
device_del+0x420/0xb10 drivers/base/core.c:2339
usb_disable_device+0x211/0x690 drivers/usb/core/message.c:1237
usb_disconnect+0x284/0x8d0 drivers/usb/core/hub.c:2199
hub_port_connect drivers/usb/core/hub.c:4949 [inline]
hub_port_connect_change drivers/usb/core/hub.c:5213 [inline]
port_event drivers/usb/core/hub.c:5359 [inline]
hub_event+0x1454/0x3640 drivers/usb/core/hub.c:5441
process_one_work+0x92b/0x1530 kernel/workqueue.c:2269
worker_thread+0x96/0xe20 kernel/workqueue.c:2415
kthread+0x318/0x420 kernel/kthread.c:255
ret_from_fork+0x24/0x30 arch/x86/entry/entry_64.S:352

other info that might help us debug this:

Chain exists of:
minor_rwsem --> iowarrior_mutex --> iowarrior_open_disc_lock

Possible unsafe locking scenario:

CPU0 CPU1
---- ----
lock(iowarrior_open_disc_lock);
lock(iowarrior_mutex);
lock(iowarrior_open_disc_lock);
lock(minor_rwsem);

*** DEADLOCK ***

6 locks held by kworker/1:1/21:
#0: 00000000ffafc5b3 ((wq_completion)usb_hub_wq){+.+.}, at:
__write_once_size include/linux/compiler.h:226 [inline]
#0: 00000000ffafc5b3 ((wq_completion)usb_hub_wq){+.+.}, at:
arch_atomic64_set arch/x86/include/asm/atomic64_64.h:34 [inline]
#0: 00000000ffafc5b3 ((wq_completion)usb_hub_wq){+.+.}, at: atomic64_set
include/asm-generic/atomic-instrumented.h:855 [inline]
#0: 00000000ffafc5b3 ((wq_completion)usb_hub_wq){+.+.}, at:
atomic_long_set include/asm-generic/atomic-long.h:40 [inline]
#0: 00000000ffafc5b3 ((wq_completion)usb_hub_wq){+.+.}, at: set_work_data
kernel/workqueue.c:620 [inline]
#0: 00000000ffafc5b3 ((wq_completion)usb_hub_wq){+.+.}, at:
set_work_pool_and_clear_pending kernel/workqueue.c:647 [inline]
#0: 00000000ffafc5b3 ((wq_completion)usb_hub_wq){+.+.}, at:
process_one_work+0x827/0x1530 kernel/workqueue.c:2240
#1: 000000005bc0df0d ((work_completion)(&hub->events)){+.+.}, at:
process_one_work+0x85b/0x1530 kernel/workqueue.c:2244
#2: 00000000f73a9504 (&dev->mutex){....}, at: device_lock
include/linux/device.h:1223 [inline]
#2: 00000000f73a9504 (&dev->mutex){....}, at: hub_event+0x17c/0x3640
drivers/usb/core/hub.c:5387
#3: 000000006fe9ca35 (&dev->mutex){....}, at: device_lock
include/linux/device.h:1223 [inline]
#3: 000000006fe9ca35 (&dev->mutex){....}, at: usb_disconnect+0x91/0x8d0
drivers/usb/core/hub.c:2190
#4: 0000000044c331cb (&dev->mutex){....}, at:
device_release_driver_internal+0x23/0x500 drivers/base/dd.c:1162
#5: 000000007638fa06 (iowarrior_open_disc_lock){+.+.}, at:
iowarrior_disconnect+0x45/0x2c0 drivers/usb/misc/iowarrior.c:867

stack backtrace:
CPU: 1 PID: 21 Comm: kworker/1:1 Not tainted 5.3.0-rc4+ #1

Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS
Google 01/01/2011
Workqueue: usb_hub_wq hub_event
Call Trace:
__dump_stack lib/dump_stack.c:77 [inline]
dump_stack+0xca/0x13e lib/dump_stack.c:113

check_noncircular+0x345/0x3e0 kernel/locking/lockdep.c:1741
check_prev_add kernel/locking/lockdep.c:2405 [inline]
check_prevs_add kernel/locking/lockdep.c:2507 [inline]
validate_chain kernel/locking/lockdep.c:2897 [inline]
__lock_acquire+0x1f7c/0x3b50 kernel/locking/lockdep.c:3880
lock_acquire+0x127/0x320 kernel/locking/lockdep.c:4412
down_write+0x92/0x150 kernel/locking/rwsem.c:1500
usb_deregister_dev drivers/usb/core/file.c:238 [inline]
usb_deregister_dev+0x61/0x270 drivers/usb/core/file.c:230
iowarrior_disconnect+0xa8/0x2c0 drivers/usb/misc/iowarrior.c:873

usb_unbind_interface+0x1bd/0x8a0 drivers/usb/core/driver.c:423
__device_release_driver drivers/base/dd.c:1134 [inline]
device_release_driver_internal+0x42f/0x500 drivers/base/dd.c:1165
bus_remove_device+0x2dc/0x4a0 drivers/base/bus.c:556
device_del+0x420/0xb10 drivers/base/core.c:2339
usb_disable_device+0x211/0x690 drivers/usb/core/message.c:1237
usb_disconnect+0x284/0x8d0 drivers/usb/core/hub.c:2199
hub_port_connect drivers/usb/core/hub.c:4949 [inline]
hub_port_connect_change drivers/usb/core/hub.c:5213 [inline]
port_event drivers/usb/core/hub.c:5359 [inline]
hub_event+0x1454/0x3640 drivers/usb/core/hub.c:5441
process_one_work+0x92b/0x1530 kernel/workqueue.c:2269
worker_thread+0x96/0xe20 kernel/workqueue.c:2415
kthread+0x318/0x420 kernel/kthread.c:255
ret_from_fork+0x24/0x30 arch/x86/entry/entry_64.S:352

iowarrior 4-1:0.236: I/O-Warror #0 now disconnected
usb 4-1: new low-speed USB device number 3 using dummy_hcd
usb 4-1: device descriptor read/all, error -71
usb 4-1: new low-speed USB device number 4 using dummy_hcd
usb 4-1: config 0 has an invalid interface number: 236 but max is 2
usb 4-1: config 0 has an invalid descriptor of length 99, skipping
remainder of the config
usb 4-1: config 0 has 1 interface, different from the descriptor's value: 3
usb 4-1: config 0 has no interface number 0
usb 4-1: config 0 interface 236 altsetting 0 endpoint 0x81 is Bulk;
changing to Interrupt
usb 4-1: New USB device found, idVendor=07c0, idProduct=1501,
bcdDevice=74.a0
usb 4-1: New USB device strings: Mfr=0, Product=0, SerialNumber=0
usb 4-1: config 0 descriptor??


Tested on:

commit: d0847550 usb-fuzzer: main usb gadget fuzzer driver
git tree: https://github.com/google/kasan.git

console output: https://syzkaller.appspot.com/x/log.txt?x=154c4522600000
kernel config: https://syzkaller.appspot.com/x/.config?x=dbc9c80cc095da19

compiler: gcc (GCC) 9.0.0 20181231 (experimental)

patch: https://syzkaller.appspot.com/x/patch.diff?x=11898be2600000

[Alan Stern]

There's no need for us to work at cross purposes on this. We can go
with your approach.

However, the code is more complicated than your patch accounts for.
The wait can finish in several different ways:

(1) The control URB succeeds and the interrupt URB gets an
acknowledgment.

(2) The control URB completes with an error.

(3) The wait times out.

(4) A disconnect occurs.

Your patch doesn't handle cases (1) and (3). (And it doesn't get rid
of the dev->waitq field, which is no longer used.)

In fact, (1) is a little ambiguous. When the interrupt URB gets a
command acknowledgment, there's no way (as far as I can tell) to know
which command was acknowledged -- particularly if a prior command URB
had to be cancelled because it timed out.

And as it turns out, the driver neglects to kill the command URB in
case (3). Furthermore, the driver doesn't have mutual exclusion for
writes. So there's nothing to prevent the command URB from being
submitted while it is still active (syzbot's new crash).

I have to wonder if anybody's actually using this driver. It seems to
be pretty broken. Maybe we should just mark it as such and forget
about fixing it.

Alan Stern

[Alan Stern]

Oops, I replied to the wrong email message -- sorry. This was intended
to be about the problem with the yurex driver, not the iowarrior
driver.

Alan Stern

[Oliver Neukum]

I absolutely agree. There is something quite wrong in this driver.
Unfortunately this is likely exploitable by a malicious gadget,
so just ignoring this is a bad option. I will need to go through the
logic. Or do you want to have a shot at it?

The patch was really only for testing. I wanted to know whether
I was hitting this very issue. This driver will need more surgery.

Regards
Oliver

[Oliver Neukum]

PS: Referring to yurex

Regards
Oliver

[Alan Stern]

If you would like to work on it, that's fine with me.

Alan Stern

[Andrey Konovalov]

#syz fix: USB: iowarrior: fix use-after-free on disconnect