KASAN: use-after-scope Read in __hrtimer_run_queues

KASAN: use-after-scope Read in __hrtimer_run_queues

[syzbot]

Hello,

syzbot found the following crash on:

HEAD commit:    2bdea157b999 Merge branch 'sctp-fully-support-for-dscp-and..
git tree:       bpf-next
console output: https://syzkaller.appspot.com/x/log.txt?x=14339f42400000
kernel config:  https://syzkaller.appspot.com/x/.config?x=f62553dc846b0692
dashboard link: https://syzkaller.appspot.com/bug?extid=7f73a81393d3b3777fb0
compiler:       gcc (GCC) 8.0.1 20180413 (experimental)

Unfortunately, I don't have any reproducer for this crash yet.

IMPORTANT: if you fix the bug, please add the following tag to the commit:
Reported-by: syzbot+7f73a81393d3b3777fb0@syzkaller.appspotmail.com

==================================================================
BUG: KASAN: use-after-scope in lookup_object lib/debugobjects.c:157 [inline]
BUG: KASAN: use-after-scope in debug_object_deactivate+0x425/0x450  
lib/debugobjects.c:540
Read of size 8 at addr ffff88019d1a5ec0 by task syz-executor3/13445

CPU: 0 PID: 13445 Comm: syz-executor3 Not tainted 4.18.0-rc3+ #45
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS  
Google 01/01/2011
Call Trace:
  <IRQ>
PANIC: double fault, error_code: 0x0
  __dump_stack lib/dump_stack.c:77 [inline]
  dump_stack+0x1c9/0x2b4 lib/dump_stack.c:113
CPU: 1 PID: 13451 Comm: syz-executor7 Not tainted 4.18.0-rc3+ #45
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS  
Google 01/01/2011
RIP: 0010:__lock_acquire+0x2e/0x5020 kernel/locking/lockdep.c:3294
Code:
41 57
  print_address_description+0x6c/0x20b mm/kasan/report.c:256
41 89
cf 41
  kasan_report_error mm/kasan/report.c:354 [inline]
  kasan_report.cold.7+0x242/0x2fe mm/kasan/report.c:412
56 41
  __asan_report_load8_noabort+0x14/0x20 mm/kasan/report.c:433
55
  lookup_object lib/debugobjects.c:157 [inline]
  debug_object_deactivate+0x425/0x450 lib/debugobjects.c:540
49 89
fd
41 54
45
  debug_hrtimer_deactivate kernel/time/hrtimer.c:421 [inline]
  debug_deactivate kernel/time/hrtimer.c:471 [inline]
  __run_hrtimer kernel/time/hrtimer.c:1368 [inline]
  __hrtimer_run_queues+0x2bf/0x10c0 kernel/time/hrtimer.c:1460
89 cc
53
65
4c 8b
34 25
40
ee
01
00
48
  hrtimer_interrupt+0x2f3/0x750 kernel/time/hrtimer.c:1518
83 e4
  local_apic_timer_interrupt arch/x86/kernel/apic/apic.c:1025 [inline]
  smp_apic_timer_interrupt+0x165/0x730 arch/x86/kernel/apic/apic.c:1050
f0
48
81 ec
60
03 00
00 48
  apic_timer_interrupt+0xf/0x20 arch/x86/entry/entry_64.S:863
8b
  </IRQ>
45
RIP: 0010:arch_local_irq_restore arch/x86/include/asm/paravirt.h:783  
[inline]
RIP: 0010:kmem_cache_free+0xb3/0x2d0 mm/slab.c:3757
10 <89>
Code:
94 24
84
80
f1 00
00 00
00
00
00
48
41
ba 00
f7
00
c5
00
00
00
02
00 fc
00
ff
00
df
0f
48
84
89
c2
84 24
00
98
00
00
RSP: 0018:ffff880199163ec0 EFLAGS: 00010086
e8
b0
RAX: 0000000000000000 RBX: 1ffff1003322c855 RCX: 0000000000000002
fb
RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffffffff88f92620
a4
RBP: ffff880199164250 R08: 0000000000000000 R09: 0000000000000000
ff
R10: ffff88019e5171b8 R11: ffff8801daf236b3 R12: 0000000000000000
48
R13: ffffffff88f92620 R14: ffff8801b1e00080 R15: 0000000000000002
83
FS:  00007f6b584e4700(0000) GS:ffff8801daf00000(0000) knlGS:0000000000000000
3d
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
e0
CR2: ffff880199163eb8 CR3: 0000000008e6a000 CR4: 00000000001406e0
03
DR0: 00000000200001c0 DR1: 0000000000000000 DR2: 0000000000000000
38
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600
07
Call Trace:
00


---
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 syzkaller@googlegroups.com.

syzbot will keep track of this bug report. See:
https://goo.gl/tpsmEJ#bug-status-tracking for how to communicate with  
syzbot.

Re: KASAN: use-after-scope Read in __hrtimer_run_queues

[Thomas Gleixner]

On Wed, 4 Jul 2018, syzbot wrote:

> Hello,
> 
> syzbot found the following crash on:
> 
> HEAD commit:    2bdea157b999 Merge branch 'sctp-fully-support-for-dscp-and..
> git tree:       bpf-next
> console output: https://syzkaller.appspot.com/x/log.txt?x=14339f42400000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=f62553dc846b0692
> dashboard link: https://syzkaller.appspot.com/bug?extid=7f73a81393d3b3777fb0
> compiler:       gcc (GCC) 8.0.1 20180413 (experimental)
> 
> Unfortunately, I don't have any reproducer for this crash yet.
> 
> IMPORTANT: if you fix the bug, please add the following tag to the commit:
> Reported-by: syzbot+7f73a81393d3b3777fb0@syzkaller.appspotmail.com
> 
> ==================================================================
> BUG: KASAN: use-after-scope in lookup_object lib/debugobjects.c:157 [inline]
> BUG: KASAN: use-after-scope in debug_object_deactivate+0x425/0x450
> lib/debugobjects.c:540
> Read of size 8 at addr ffff88019d1a5ec0 by task syz-executor3/13445

I don't see how lib/debugobjects.c:157 would be a valid use after scope
issue. The one in line 540 makes no sense either.

Thanks,

	tglx

Re: KASAN: use-after-scope Read in __hrtimer_run_queues

[Dmitry Vyukov]

On Wed, Jul 4, 2018 at 5:20 PM, Thomas Gleixner <tglx@linutronix.de> wrote:
> On Wed, 4 Jul 2018, syzbot wrote:
>
>> Hello,
>>
>> syzbot found the following crash on:
>>
>> HEAD commit:    2bdea157b999 Merge branch 'sctp-fully-support-for-dscp-and..
>> git tree:       bpf-next
>> console output: https://syzkaller.appspot.com/x/log.txt?x=14339f42400000
>> kernel config:  https://syzkaller.appspot.com/x/.config?x=f62553dc846b0692
>> dashboard link: https://syzkaller.appspot.com/bug?extid=7f73a81393d3b3777fb0
>> compiler:       gcc (GCC) 8.0.1 20180413 (experimental)
>>
>> Unfortunately, I don't have any reproducer for this crash yet.
>>
>> IMPORTANT: if you fix the bug, please add the following tag to the commit:
>> Reported-by: syzbot+7f73a81393d3b3777fb0@syzkaller.appspotmail.com
>>
>> ==================================================================
>> BUG: KASAN: use-after-scope in lookup_object lib/debugobjects.c:157 [inline]
>> BUG: KASAN: use-after-scope in debug_object_deactivate+0x425/0x450
>> lib/debugobjects.c:540
>> Read of size 8 at addr ffff88019d1a5ec0 by task syz-executor3/13445
>
> I don't see how lib/debugobjects.c:157 would be a valid use after scope
> issue. The one in line 540 makes no sense either.

Hi Thomas,

They all don't make sense:
https://groups.google.com/d/msg/syzkaller-bugs/A-S5Rfn5bwQ/dFvmNcmyAAAJ

One of them says "stack overflow", so there seems to be a serious
system corruption that now leads to assorted nonsensical crashes. And
this seems to be a very recent change.

Re: KASAN: use-after-scope Read in __hrtimer_run_queues

[Dmitry Vyukov]

On Wed, Jul 4, 2018 at 5:27 PM, Dmitry Vyukov <dvyukov@google.com> wrote:
> On Wed, Jul 4, 2018 at 5:20 PM, Thomas Gleixner <tglx@linutronix.de> wrote:
>> On Wed, 4 Jul 2018, syzbot wrote:
>>
>>> Hello,
>>>
>>> syzbot found the following crash on:
>>>
>>> HEAD commit:    2bdea157b999 Merge branch 'sctp-fully-support-for-dscp-and..
>>> git tree:       bpf-next
>>> console output: https://syzkaller.appspot.com/x/log.txt?x=14339f42400000
>>> kernel config:  https://syzkaller.appspot.com/x/.config?x=f62553dc846b0692
>>> dashboard link: https://syzkaller.appspot.com/bug?extid=7f73a81393d3b3777fb0
>>> compiler:       gcc (GCC) 8.0.1 20180413 (experimental)
>>>
>>> Unfortunately, I don't have any reproducer for this crash yet.
>>>
>>> IMPORTANT: if you fix the bug, please add the following tag to the commit:
>>> Reported-by: syzbot+7f73a81393d3b3777fb0@syzkaller.appspotmail.com
>>>
>>> ==================================================================
>>> BUG: KASAN: use-after-scope in lookup_object lib/debugobjects.c:157 [inline]
>>> BUG: KASAN: use-after-scope in debug_object_deactivate+0x425/0x450
>>> lib/debugobjects.c:540
>>> Read of size 8 at addr ffff88019d1a5ec0 by task syz-executor3/13445
>>
>> I don't see how lib/debugobjects.c:157 would be a valid use after scope
>> issue. The one in line 540 makes no sense either.
>
> Hi Thomas,
>
> They all don't make sense:
> https://groups.google.com/d/msg/syzkaller-bugs/A-S5Rfn5bwQ/dFvmNcmyAAAJ
>
> One of them says "stack overflow", so there seems to be a serious
> system corruption that now leads to assorted nonsensical crashes. And
> this seems to be a very recent change.

#syz dup: KASAN: stack-out-of-bounds Read in timerqueue_add