Re: 3.11-rc7:BUG: soft lockup

[Ding Tianhong]

On 2013/9/2 14:06, Baoquan He wrote:
> Hi both,
> 
> Thanks for your patches. I tried to test your patches,  first the 2nd 
> one, namely Hillf's patch, it's OK. Then when I wanted to reproduce and 
> test Cong's patch, it failed to happen again. 
> 
> I remember this bug happened randomly at the very beginning, 
> just after kernel compiling  it always happened one day. 
> 
> So maybe when it happened again, I will test your patch separately. 
> 
> Baoquan
> Thanks 
> 
> On 08/31/2013 11:25 AM, Hillf Danton wrote:
>> On Fri, Aug 30, 2013 at 8:18 PM, Cong Wang <xiyou.wangcong@xxxxxxxxx> wrote:
>>> Cc'ing netdev
>>>
>>> On Fri, Aug 30, 2013 at 4:20 PM, Baoquan He <baoquan.he@xxxxxxxxx> wrote:
>>>> Hi,
>>>>
>>>> I tried the 3.11.0-rc7+ on x86_64, and after bootup, the soft lockup bug
>>>> happened.
>>>>
>>>> [   48.895000] BUG: soft lockup - CPU#1 stuck for 22s! [ebtables:444]
>>>> [   48.901191] Modules linked in: bnep(F) bluetooth(F) ebtables(F)
>>>> ip6table_filter(F) ip6_tables(F) rfkill(F) snd_hda_intel(F+)
>>>> snd_hda_codec(F) snd_hwdep(F) snd_seq(F) sn)
>>>> [   48.950034] CPU: 1 PID: 444 Comm: ebtables Tainted: GF     D
>>>> 3.11.0-rc7+ #1
>>>> [   48.957433] Hardware name: Hewlett-Packard HP Z420 Workstation/1589,
>>>> BIOS J61 v01.02 03/09/2012
>>>> [   48.966131] task: ffff88040c2dc650 ti: ffff8804187d2000 task.ti:
>>>> ffff8804187d2000
>>>> [   48.973610] RIP: 0010:[<ffffffff812e57a7>]  [<ffffffff812e57a7>]
>>>> strcmp+0x27/0x40
>>>> [   48.981119] RSP: 0018:ffff8804187d3db8  EFLAGS: 00000246
>>>> [   48.986430] RAX: 0000000000000000 RBX: 00007fffda942730 RCX:
>>>> ffff8804187d3fd8
>>>> [   48.993566] RDX: 0000000000000000 RSI: ffff8804187d3e01 RDI:
>>>> ffffffff81cb8a39
>>>> [   49.000707] RBP: ffff8804187d3db8 R08: 00000000fffffff2 R09:
>>>> 0000000000000000
>>>> [   49.007841] R10: 0000000000000163 R11: 0000000000000000 R12:
>>>> ffffffff8128300c
>>>> [   49.014972] R13: ffff8804187d3d98 R14: ffff8804187d3ef4 R15:
>>>> 0000000000000004
>>>> [   49.022112] FS:  00007faab6589740(0000) GS:ffff88042fc80000(0000)
>>>> knlGS:0000000000000000
>>>> [   49.030194] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>>>> [   49.035942] CR2: 0000003f0d810414 CR3: 000000040d2cc000 CR4:
>>>> 00000000000407e0
>>>> [   49.043077] Stack:
>>>> [   49.045096]  ffff8804187d3de8 ffffffffa0249674 0000000000000080
>>>> ffffffff81cb8180
>>>> [   49.052559]  00007fffda942730 ffff8804187d3ef4 ffff8804187d3ea0
>>>> ffffffffa02497a9
>>>> [   49.060020]  0000000000000000 00007265746c6966 0000003f0d7b92c0
>>>> 00007fffda942850
>>>> [   49.067487] Call Trace:
>>>> [   49.069949]  [<ffffffffa0249674>]
>>>> find_inlist_lock.constprop.16+0x54/0x100 [ebtables]
>>>> [   49.077779]  [<ffffffffa02497a9>] do_ebt_get_ctl+0x89/0x1d0 [ebtables]
>>>> [   49.084306]  [<ffffffff81551ca8>] nf_getsockopt+0x68/0x90
>>>> [   49.089717]  [<ffffffff81560d40>] ip_getsockopt+0x80/0xa0
>>>> [   49.095113]  [<ffffffff815835c5>] raw_getsockopt+0x25/0x50
>>>> [   49.100588]  [<ffffffff8150ddd4>] sock_common_getsockopt+0x14/0x20
>>>> [   49.106766]  [<ffffffff8150d208>] SyS_getsockopt+0x68/0xd0
>>>> [   49.112257]  [<ffffffff8162c682>] system_call_fastpath+0x16/0x1b
>>>> [   49.118260] Code: 00 00 00 00 55 48 89 e5 eb 0e 66 2e 0f 1f 84 00 00
>>>> 00 00 00 84 c0 74 1c 48 83 c7 01 0f b6 47 ff 48 83 c6 01 3a 46 ff 74 eb
>>>> 19 c0 <83> c8 01 5d c3 0f 1
>>>> [   76.925880] BUG: soft lockup - CPU#1 stuck for 22s! [ebtables:444]
>>>> [   76.932069] Modules linked in: bnep(F) bluetooth(F) ebtables(F)
>>>> ip6table_filter(F) ip6_tables(F) rfkill(F) snd_hda_intel(F+)
>>>> snd_hda_codec(F) snd_hwdep(F) snd_seq(F) sn)
>>>> [   76.980847] CPU: 1 PID: 444 Comm: ebtables Tainted: GF     D
>>>> 3.11.0-rc7+ #1
>>>> [   76.988245] Hardware name: Hewlett-Packard HP Z420 Workstation/1589,
>>>> BIOS J61 v01.02 03/09/2012
>>>> [   76.996940] task: ffff88040c2dc650 ti: ffff8804187d2000 task.ti:
>>>> ffff8804187d2000
>>>> [   77.004426] RIP: 0010:[<ffffffff812e5784>]  [<ffffffff812e5784>]
>>>> strcmp+0x4/0x40
>>>> [   77.011849] RSP: 0018:ffff8804187d3db8  EFLAGS: 00000212
>>>> [   77.017163] RAX: 0000000000000001 RBX: 00007fffda942730 RCX:
>>>> ffff8804187d3fd8
>>>> [   77.024304] RDX: 0000000000000000 RSI: ffff8804187d3e00 RDI:
>>>> ffffffff81cb8a38
>>>> [   77.031434] RBP: ffff8804187d3db8 R08: 00000000fffffff2 R09:
>>>> 0000000000000000
>>>> [   77.038566] R10: 0000000000000163 R11: 0000000000000000 R12:
>>>> ffffffff8128300c
>>>> [   77.045699] R13: ffff8804187d3d98 R14: ffff8804187d3ef4 R15:
>>>> 0000000000000004
>>>> [   77.052842] FS:  00007faab6589740(0000) GS:ffff88042fc80000(0000)
>>>> knlGS:0000000000000000
>>>> [   77.060934] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>>>> [   77.066668] CR2: 0000003f0d810414 CR3: 000000040d2cc000 CR4:
>>>> 00000000000407e0
>>>> [   77.073799] Stack:
>>>> [   77.075818]  ffff8804187d3de8 ffffffffa0249674 0000000000000080
>>>> ffffffff81cb8180
>>>> [   77.083287]  00007fffda942730 ffff8804187d3ef4 ffff8804187d3ea0
>>>> ffffffffa02497a9
>>>> [   77.090749]  0000000000000000 00007265746c6966 0000003f0d7b92c0
>>>> 00007fffda942850
>>>> [   77.098215] Call Trace:
>>>> [   77.100668]  [<ffffffffa0249674>]
>>>> find_inlist_lock.constprop.16+0x54/0x100 [ebtables]
>>>> [   77.108500]  [<ffffffffa02497a9>] do_ebt_get_ctl+0x89/0x1d0 [ebtables]
>>>> [   77.115035]  [<ffffffff81551ca8>] nf_getsockopt+0x68/0x90
>>>> [   77.120438]  [<ffffffff81560d40>] ip_getsockopt+0x80/0xa0
>>>> [   77.125845]  [<ffffffff815835c5>] raw_getsockopt+0x25/0x50
>>>> [   77.131328]  [<ffffffff8150ddd4>] sock_common_getsockopt+0x14/0x20
>>>> [   77.137515]  [<ffffffff8150d208>] SyS_getsockopt+0x68/0xd0
>>>> [   77.143011]  [<ffffffff8162c682>] system_call_fastpath+0x16/0x1b
>>>> [   77.149019] Code: 0f 1f 80 00 00 00 00 48 83 c6 01 0f b6 4e ff 48 83
>>>> c2 01 84 c9 88 4a ff 75 ed 5d c3 66 66 2e 0f 1f 84 00 00 00 00 00 55 48
>>>> 89 e5 <eb> 0e 66 2e 0f 1f 8
>>>
>>> Does the following patch help?
>>>
>>>
>>> diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c
>>> index ac78024..4a0ec8f 100644
>>> --- a/net/bridge/netfilter/ebtables.c
>>> +++ b/net/bridge/netfilter/ebtables.c
>>> @@ -1503,6 +1503,10 @@ static int do_ebt_get_ctl(struct sock *sk, int
>>> cmd, void __user *user, int *len)
>>>         if (copy_from_user(&tmp, user, sizeof(tmp)))
>>>                 return -EFAULT;
>>>
>>> +       if (memscan(tmp.name, '\0', EBT_TABLE_MAXNAMELEN) ==
>>> +                   (tmp.name + EBT_TABLE_MAXNAMELEN))
>>> +               return -EINVAL;
>>> +
>>>         t = find_table_lock(net, tmp.name, &ret, &ebt_mutex);
>>>         if (!t)
>>>                 return ret;
>>> --
>>>
>> release lock!!
>>
>> --- a/net/bridge/netfilter/ebtables.c Sat Aug 31 11:12:54 2013
>> +++ b/net/bridge/netfilter/ebtables.c Sat Aug 31 11:15:24 2013
>> @@ -332,8 +332,10 @@ find_inlist_lock_noload(struct list_head
>>   return NULL;
>>
>>   list_for_each_entry(e, head, list) {
>> - if (strcmp(e->name, name) == 0)
>> + if (strcmp(e->name, name) == 0) {
>> + mutex_unlock(mutex);
>>   return e;
>> + }
>>   }
>>   *error = -ENOENT;
>>   mutex_unlock(mutex);
>> --

could not release lock here, otherwise there will unlock twice. I did not meet any problem about the drivers, maybe
it is hard to occur.

> 
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 
> .
> 


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/