Re: [PATCH] bpf: fix rq lock recursion issue

[Satya Durga Srinivasu Prabhala <quic_satyap@quicinc.com>]

On 6/13/22 2:22 AM, Toke Høiland-Jørgensen wrote:
> Satya Durga Srinivasu Prabhala <quic_satyap@quicinc.com> writes:
>
>> Below recursion is observed in a rare scenario where __schedule()
>> takes rq lock, at around same time task's affinity is being changed,
>> bpf function for tracing sched_switch calls migrate_enabled(),
>> checks for affinity change (cpus_ptr != cpus_mask) lands into
>> __set_cpus_allowed_ptr which tries acquire rq lock and causing the
>> recursion bug.
> So this only affects tracing programs that attach to tasks that can have
> their affinity changed?

That's right.

>   Or do we need to review migrate_enable() vs
> preempt_enable() for networking hooks as well?

I believe networking hooks won't take rq locks, so, we won't see the issue
with networking hooks as far as I can tell.

>
> -Toke
>