Re: INFO: rcu detected stall in ndisc_alloc_skb

[Tetsuo Handa]

On 2019/01/03 2:06, Tetsuo Handa wrote:
> On 2018/12/31 17:24, Dmitry Vyukov wrote:
>>>> Since this involves OOMs and looks like a one-off induced memory corruption:
>>>>
>>>> #syz dup: kernel panic: corrupted stack end in wb_workfn
>>>>
>>>
>>> Why?
>>>
>>> RCU stall in this case is likely to be latency caused by flooding of printk().
>>
>> Just a hypothesis. OOMs lead to arbitrary memory corruptions, so can
>> cause stalls as well. But can be what you said too. I just thought
>> that cleaner dashboard is more useful than a large assorted pile of
>> crashes. If you think it's actionable in some way, feel free to undup.
>>
> 
> We don't know why bpf tree is hitting this problem.
> Let's continue monitoring this problem.
> 
> #syz undup
> 

A report at 2019/01/05 10:08 from "no output from test machine (2)"
( https://syzkaller.appspot.com/text?tag=CrashLog&x=1700726f400000 )
says that there are flood of memory allocation failure messages.
Since continuous memory allocation failure messages itself is not
recognized as a crash, we might be misunderstanding that this problem
is not occurring recently. It will be nice if we can run testcases
which are executed on bpf-next tree.