Re: usb/serial: use-after-free in usb_serial_disconnect/__lock_acquire

[Johan Hovold]

On Wed, Sep 27, 2017 at 04:36:11PM +0200, Andrey Konovalov wrote:
> Hi!
> 
> I've got the following report while fuzzing the kernel with syzkaller.
> 
> On commit e19b205be43d11bff638cad4487008c48d21c103 (4.14-rc2).
> 
> gadgetfs: bound to dummy_udc driver
> usb 1-1: new full-speed USB device number 2 using dummy_hcd
> gadgetfs: connected
> gadgetfs: disconnected
> gadgetfs: connected
> usb 1-1: config 4 has an invalid interface number: 1 but max is 0
> usb 1-1: config 4 has an invalid interface number: 153 but max is 0
> usb 1-1: config 4 has 2 interfaces, different from the descriptor's value: 1
> usb 1-1: config 4 has no interface number 0
> usb 1-1: config 4 interface 1 altsetting 255 has an invalid endpoint
> with address 0x0, skipping
> usb 1-1: config 4 interface 1 altsetting 255 has an invalid endpoint
> with address 0xFF, skipping
> usb 1-1: config 4 interface 1 altsetting 255 has an invalid endpoint
> with address 0x56, skipping
> usb 1-1: too many endpoints for config 4 interface 153 altsetting 67:
> 174, using maximum allowed: 30
> usb 1-1: config 4 interface 153 altsetting 67 has 0 endpoint
> descriptors, different from the interface d
> escriptor's value: 174
> usb 1-1: config 4 interface 1 has no altsetting 0
> usb 1-1: config 4 interface 153 has no altsetting 0
> usb 1-1: New USB device found, idVendor=1199, idProduct=6832
> usb 1-1: New USB device strings: Mfr=4, Product=20, SerialNumber=3
> usb 1-1: Product: a
> usb 1-1: Manufacturer: a
> usb 1-1: SerialNumber: a
> gadgetfs: configuration #4
> sierra 1-1:4.1: Sierra USB modem converter detected
> usb 1-1: Sierra USB modem converter now attached to ttyUSB0
> sierra 1-1:4.153: Sierra USB modem converter detected
> gadgetfs: disconnected
> usb 1-1: USB disconnect, device number 2
> sierra ttyUSB0: Sierra USB modem converter now disconnected from ttyUSB0
> sierra 1-1:4.1: device disconnected
> ==================================================================
> BUG: KASAN: use-after-free in __lock_acquire+0x4504/0x4550
> Read of size 8 at addr ffff8800674df790 by task kworker/1:2/1846
> 
> CPU: 1 PID: 1846 Comm: kworker/1:2 Not tainted
> 4.14.0-rc2-42660-g24b7bd59eec0 #277
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011
> Workqueue: usb_hub_wq hub_event
> Call Trace:
>  __dump_stack lib/dump_stack.c:16
>  dump_stack+0x292/0x395 lib/dump_stack.c:52
>  print_address_description+0x78/0x280 mm/kasan/report.c:252
>  kasan_report_error mm/kasan/report.c:351
>  kasan_report+0x23d/0x350 mm/kasan/report.c:409
>  __asan_report_load8_noabort+0x19/0x20 mm/kasan/report.c:430
>  __lock_acquire+0x4504/0x4550 kernel/locking/lockdep.c:3376
>  lock_acquire+0x259/0x620 kernel/locking/lockdep.c:4002
>  __mutex_lock_common kernel/locking/mutex.c:756
>  __mutex_lock+0x18e/0x1a50 kernel/locking/mutex.c:893
>  mutex_lock_nested+0x1b/0x20 kernel/locking/mutex.c:908
>  usb_serial_disconnect+0x69/0x2e0 drivers/usb/serial/usb-serial.c:1084
>  usb_unbind_interface+0x21c/0xa90 drivers/usb/core/driver.c:423
>  __device_release_driver drivers/base/dd.c:861
>  device_release_driver_internal+0x4f4/0x5c0 drivers/base/dd.c:893
>  device_release_driver+0x1e/0x30 drivers/base/dd.c:918
>  bus_remove_device+0x2f4/0x4b0 drivers/base/bus.c:565
>  device_del+0x5c4/0xab0 drivers/base/core.c:1985
>  usb_disable_device+0x1e9/0x680 drivers/usb/core/message.c:1170
>  usb_disconnect+0x260/0x7a0 drivers/usb/core/hub.c:2124
>  hub_port_connect drivers/usb/core/hub.c:4754
>  hub_port_connect_change drivers/usb/core/hub.c:5009
>  port_event drivers/usb/core/hub.c:5115
>  hub_event+0x1318/0x3740 drivers/usb/core/hub.c:5195
>  process_one_work+0xc7f/0x1db0 kernel/workqueue.c:2119
>  worker_thread+0x221/0x1850 kernel/workqueue.c:2253
>  kthread+0x3a1/0x470 kernel/kthread.c:231
>  ret_from_fork+0x2a/0x40 arch/x86/entry/entry_64.S:431

Thanks for reporting this. I was able to reproduce the bug from the
information you provided here, and incidentally discovered a
long-standing related issue which I've now also fixed.

Thanks,
Johan