Re: KASAN: null-ptr-deref Write in event_handler

[Shuah Khan <skhan@linuxfoundation.org>]

On 10/5/20 2:44 PM, Shuah Khan wrote:
> On 10/5/20 8:04 AM, Andrey Konovalov wrote:
>> On Mon, Oct 5, 2020 at 3:59 PM syzbot
>> <syzbot+bf1a360e305ee719e364@syzkaller.appspotmail.com> wrote:
>>>
>>> Hello,
>>>
>>> syzbot found the following issue on:
>>>
>>> HEAD commit:    d3d45f82 Merge tag 'pinctrl-v5.9-2' of 
>>> git://git.kernel.or..
>>> git tree:       upstream
>>> console output: https://syzkaller.appspot.com/x/log.txt?x=15781d8f900000
>>> kernel config:  
>>> https://syzkaller.appspot.com/x/.config?x=89ab6a0c48f30b49
>>> dashboard link: 
>>> https://syzkaller.appspot.com/bug?extid=bf1a360e305ee719e364
>>> compiler:       gcc (GCC) 10.1.0-syz 20200507
>>> syz repro:      
>>> https://syzkaller.appspot.com/x/repro.syz?x=16cbaa7d900000
>>> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=1364f367900000
>>>
>>> IMPORTANT: if you fix the issue, please add the following tag to the 
>>> commit:
>>> Reported-by: syzbot+bf1a360e305ee719e364@syzkaller.appspotmail.com
>>>
>>> vhci_hcd: stop threads
>>> vhci_hcd: release socket
>>> vhci_hcd: disconnect device
>>> ==================================================================
>>> BUG: KASAN: null-ptr-deref in instrument_atomic_write 
>>> include/linux/instrumented.h:71 [inline]
>>> BUG: KASAN: null-ptr-deref in atomic_fetch_add_relaxed 
>>> include/asm-generic/atomic-instrumented.h:142 [inline]
>>> BUG: KASAN: null-ptr-deref in refcount_add 
>>> include/linux/refcount.h:201 [inline]
>>> BUG: KASAN: null-ptr-deref in refcount_inc 
>>> include/linux/refcount.h:241 [inline]
>>> BUG: KASAN: null-ptr-deref in get_task_struct 
>>> include/linux/sched/task.h:104 [inline]
>>> BUG: KASAN: null-ptr-deref in kthread_stop+0x90/0x7e0 
>>> kernel/kthread.c:591
>>> Write of size 4 at addr 000000000000001c by task kworker/u4:5/2519
>>>
>>> CPU: 1 PID: 2519 Comm: kworker/u4:5 Not tainted 5.9.0-rc7-syzkaller #0
>>> Hardware name: Google Google Compute Engine/Google Compute Engine, 
>>> BIOS Google 01/01/2011
>>> Workqueue: usbip_event event_handler
>>> Call Trace:
>>>   __dump_stack lib/dump_stack.c:77 [inline]
>>>   dump_stack+0x198/0x1fd lib/dump_stack.c:118
>>>   __kasan_report mm/kasan/report.c:517 [inline]
>>>   kasan_report.cold+0x5/0x37 mm/kasan/report.c:530
>>>   check_memory_region_inline mm/kasan/generic.c:186 [inline]
>>>   check_memory_region+0x13d/0x180 mm/kasan/generic.c:192
>>>   instrument_atomic_write include/linux/instrumented.h:71 [inline]
>>>   atomic_fetch_add_relaxed 
>>> include/asm-generic/atomic-instrumented.h:142 [inline]
>>>   refcount_add include/linux/refcount.h:201 [inline]
>>>   refcount_inc include/linux/refcount.h:241 [inline]
>>>   get_task_struct include/linux/sched/task.h:104 [inline]
>>>   kthread_stop+0x90/0x7e0 kernel/kthread.c:591
>>>   vhci_shutdown_connection+0x170/0x2a0 drivers/usb/usbip/vhci_hcd.c:1015
>>>   event_handler+0x1a5/0x450 drivers/usb/usbip/usbip_event.c:78
>>>   process_one_work+0x94c/0x1670 kernel/workqueue.c:2269
>>>   worker_thread+0x64c/0x1120 kernel/workqueue.c:2415
>>>   kthread+0x3b5/0x4a0 kernel/kthread.c:292
>>>   ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:294
>>> ==================================================================
>>> Kernel panic - not syncing: panic_on_warn set ...
>>> CPU: 1 PID: 2519 Comm: kworker/u4:5 Tainted: G    B             
>>> 5.9.0-rc7-syzkaller #0
>>> Hardware name: Google Google Compute Engine/Google Compute Engine, 
>>> BIOS Google 01/01/2011
>>> Workqueue: usbip_event event_handler
>>> Call Trace:
>>>   __dump_stack lib/dump_stack.c:77 [inline]
>>>   dump_stack+0x198/0x1fd lib/dump_stack.c:118
>>>   panic+0x382/0x7fb kernel/panic.c:231
>>>   end_report+0x4d/0x53 mm/kasan/report.c:104
>>>   __kasan_report mm/kasan/report.c:520 [inline]
>>>   kasan_report.cold+0xd/0x37 mm/kasan/report.c:530
>>>   check_memory_region_inline mm/kasan/generic.c:186 [inline]
>>>   check_memory_region+0x13d/0x180 mm/kasan/generic.c:192
>>>   instrument_atomic_write include/linux/instrumented.h:71 [inline]
>>>   atomic_fetch_add_relaxed 
>>> include/asm-generic/atomic-instrumented.h:142 [inline]
>>>   refcount_add include/linux/refcount.h:201 [inline]
>>>   refcount_inc include/linux/refcount.h:241 [inline]
>>>   get_task_struct include/linux/sched/task.h:104 [inline]
>>>   kthread_stop+0x90/0x7e0 kernel/kthread.c:591
>>>   vhci_shutdown_connection+0x170/0x2a0 drivers/usb/usbip/vhci_hcd.c:1015
>>>   event_handler+0x1a5/0x450 drivers/usb/usbip/usbip_event.c:78
>>>   process_one_work+0x94c/0x1670 kernel/workqueue.c:2269
>>>   worker_thread+0x64c/0x1120 kernel/workqueue.c:2415
>>>   kthread+0x3b5/0x4a0 kernel/kthread.c:292
>>>   ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:294
>>> Kernel Offset: disabled
>>> Rebooting in 86400 seconds..
>>
>> Hi Valentina and Shuah,
>>
>> There appears to be a race condition in the USB/IP vhci_hcd shutdown
>> procedure. It happens quite often during fuzzing with syzkaller, and
>> prevents us from going deeper into the USB/IP code.
>>
>> Could you advise us what would be the best fix for this?
>>
> 
> Hi Andrey,
> 
> Reading the comments for this routine, looks like there is an assumption
> that context begins cleanup and race conditions aren't considered.
> 
> The right fix is holding vhci->lock and vdev->priv_lock to protect
> critical sections in this routine. I will send a patch for this.
> 

Hi Andrey,

I have been unable to reproduce the problem with the reproducer
so far. You mentioned it happens quite often.

- matched config with yours
- load vhci_hcd module and run the reproducer

I do see the messages during shutdown - stop threads etc.

What am I missing?

thanks,
-- Shuah