Re: Allocation failure with subsequent kernel crash

From: Daniel Borkmann
Date: Mon Aug 20 2018 - 15:29:51 EST

Hi Matthew,

On 08/20/2018 08:03 AM, tedheadster wrote:
> On Mon, Aug 20, 2018 at 1:22 AM, Alexei Starovoitov
> <alexei.starovoitov@xxxxxxxxx> wrote:
>> I don't remember vzalloc issues that was fixed in this area.
>> 4.14 kernel is quite old. Since then syzbot found few mem
>> related bugs that were fixed.
>> please try to reproduce on the latest tree.
> Alexei,
> I get this panic with two very recent kernels: 4.18.0 and 4.17.14. I
> do not think this has been fixed. I am still trying to bisect it, but
> sometimes it takes 5 hours for the panic to occur.

I've been looking into it a bit today and still am. Given you've seen
this on x86_32 and also on older kernels, I presume JIT was not involved
(/proc/sys/net/core/bpf_jit_enable is 0). Do you run any specific workload
until you trigger this (e.g. fuzzer on BPF), or any specific event that
triggers at that time after ~5hrs? Or only systemd on idle machine? Have
you managed to reproduce this also elsewhere? Bisect seems indeed painful
but would help tremendously; perhaps also dumping the BPF insns that are
loaded at that point in time.

Thanks a lot,

> - Matthew Whitehead