netbsd boot error: panic: LO0KDE Ucs bute error i_s i: <,528:e,pPn lostehtlds[[ 1.ADDR kc 11: Begif trdceba

[syzbot]

Hello,

syzbot found the following crash on:

HEAD commit: e54bc46e need pfilter.h
git tree: netbsd
console output: https://syzkaller.appspot.com/x/log.txt?x=1419eca6e00000
kernel config: https://syzkaller.appspot.com/x/.config?x=6e4d6bd2b8e377a2
dashboard link: https://syzkaller.appspot.com/bug?extid=0edd82e14b1894916a64
compiler: g++ (Ubuntu 5.4.0-6ubuntu1~16.04.12) 5.4.0 20160609

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+0edd82...@syzkaller.appspotmail.com

[ 1.7795891] panic: LO0KDE Ucs bute error i_s i: <,528:e,pPn
lostehtlds[[ 1.7958811 kc 11: Begif trdceba
1.7795891] ck...
[ 1.8463769] sd0: fabricating a geometry
[ 1.8463769] sd0: 8192 MB, 8192 cyl, 64 head, 32 sec, 512 bytes/sect x
16777216 sectors
[ 1.8686409] sd0: fabricating a geometry
[ 1.8807495] vpanic() at netbsd:vpanic+0x241 sys/kern/subr_prf.c:336
[ 1.9354638] snprintf() at netbsd:snprintf
[ 1.9354638] lockdebug_more() at netbsd:lockdebug_more
[ 1.9354638] lockdebug_barrier() at netbsd:lockdebug_barrier+0xfe
sys/kern/subr_lockdebug.c:684
[ 1.9354638] mi_switch() at netbsd:mi_switch+0xfb
sys/kern/kern_synch.c:532
[ 1.9354638] idle_loop() at netbsd:idle_loop+0x34a x86_curlwp
sys/arch/amd64/compile/obj/GENERIC_SYZKALLER/./machine/cpu.h:79 [inline]
[ 1.9354638] idle_loop() at netbsd:idle_loop+0x34a sys/kern/kern_idle.c:99
[ 1.9354638] cpu_hatch() at netbsd:cpu_hatch+0x270
sys/arch/x86/x86/cpu.c:980
[ 1.9354638] Bad frame pointer: 0xffffa5800f738100
[ 1.9354638] cpu1: End traceback...
[ 1.9354638] fatal breakpoint trap in supervisor mode
[ 1.9354638] trap type 1 code 0 rip 0xffffffff8021ccb5 cs 0x8 rflags
0x246 cr2 0 ilevel 0x8 rsp 0xffffa5816db37b50
[ 1.9354638] curlwp 0xffffa5800f738100 pid 0.20 lowest kstack
0xffffa5816db302c0
Stopped in pid 0.20 (system) at netbsd:breakpoint+0x5: leave
db{1}>


---
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.

[Siddharth Muralee]

I see a lot of similar errors. Is this happening due to improper compilation or something?

--
You received this message because you are subscribed to the Google Groups "syzkaller-netbsd-bugs" group.
To unsubscribe from this group and stop receiving emails from it, send an email to syzkaller-netbsd...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/syzkaller-netbsd-bugs/0000000000007e74cf05993a147d%40google.com.

--

Regards, 

  Siddharth M
  Final Year B.Tech (CSE) Student,
  Amrita School of Engineering, Kollam

  Blog 

---------------------------------------

“Most people get ahead during the time that others waste."

[Dmitry Vyukov]

On Mon, Dec 9, 2019 at 9:16 AM Siddharth Muralee
<siddhart...@gmail.com> wrote:
>
> I see a lot of similar errors. Is this happening due to improper compilation or something?

I would assume that's intermixed output from 2 different threads. Such
intermixed output play bad with automatic crash process (always
requires a human).
There are 2 things we could do:
- detect "corrupted" reports better (syzkaller/syzbot has full
support for this feature and we use it extensively for linux, you may
grep pkg/report for "corrupted")
- fix kernel to not intermix output (it mostly worked for linux, but
we had to do some improvements of various nature; and for Akaros
intermixed output also hit us badly but then kernel was fixed to not
produce mess and it started working very reliably and nicely)

> To view this discussion on the web visit https://groups.google.com/d/msgid/syzkaller-netbsd-bugs/CAB5-aq5MqBxEmdr8wjfp20sXkqj5tv%3DrT88jh2tWRL8OKJmb2A%40mail.gmail.com.

[Kamil Rytarowski]

On 09.12.2019 11:29, Dmitry Vyukov wrote:
> On Mon, Dec 9, 2019 at 9:16 AM Siddharth Muralee
> <siddhart...@gmail.com> wrote:
>>
>> I see a lot of similar errors. Is this happening due to improper compilation or something?
>
> I would assume that's intermixed output from 2 different threads. Such
> intermixed output play bad with automatic crash process (always
> requires a human).
> There are 2 things we could do:
> - detect "corrupted" reports better (syzkaller/syzbot has full
> support for this feature and we use it extensively for linux, you may
> grep pkg/report for "corrupted")
> - fix kernel to not intermix output (it mostly worked for linux, but
> we had to do some improvements of various nature; and for Akaros
> intermixed output also hit us badly but then kernel was fixed to not
> produce mess and it started working very reliably and nicely)
>
>

We introduced changes in the kernel that can hopefully sanitize the
output in multiprocessor setups.

[Andrew Doran]

Hi,

This is an old problem and should be fixed by these recent commits which
help the system to go down in a controlled manner when there is a panic:

http://mail-index.netbsd.org/source-changes/2019/12/09/msg111773.html
http://mail-index.netbsd.org/source-changes/2019/12/09/msg111774.html

Andrew

> >> ?Most people get ahead during the time that others waste."

[Maxime Villard]

dismiss

#syz invalid