Re: INFO: rcu detected stall in sys_kill

[Tetsuo Handa]

On 2020/01/08 15:20, Dmitry Vyukov wrote:
> I temporarily re-enabled smack instance and it produced another 50
> stalls all over the kernel, and now keeps spewing a dozen every hour.

Since we can get stall reports rather easily, can we try modifying
kernel command line (e.g. lsm=smack) and/or kernel config (e.g. no kasan) ?

> 
> I've mailed 3 new samples, you can see them here:
> https://syzkaller.appspot.com/bug?extid=de8d933e7d153aa0c1bb
> 
> The config is provided, command line args are here:
> https://github.com/google/syzkaller/blob/master/dashboard/config/upstream-smack.cmdline
> Some non-default sysctls that syzbot sets are here:
> https://github.com/google/syzkaller/blob/master/dashboard/config/upstream.sysctl
> Image can be downloaded from here:
> https://github.com/google/syzkaller/blob/master/docs/syzbot.md#crash-does-not-reproduce
> syzbot uses GCE VMs with 2 CPUs and 7.5GB memory, but this does not
> look to be virtualization-related (?) so probably should reproduce in
> qemu too.

Is it possible to add instance for linux-next.git that uses these configs?
If yes, we could try adding some debug printk() under CONFIG_DEBUG_AID_FOR_SYZBOT=y .