Re: [PATCH v2] kernel: bpf: stackmap: fix a possible sleep-in-atomic bug in bpf_mmap_unlock_get_irq_work()

From: John Fastabend
Date: Fri Mar 17 2023 - 14:14:57 EST


Alexei Starovoitov wrote:
> On Fri, Mar 17, 2023 at 03:52:27AM +0000, starmiku1207184332@xxxxxxxxx wrote:
> > context because of its possible sleep operation. However, mmap_read_unlock()
> > is unsafely called in a preempt disabled context when spin_lock() or
> > rcu_read_lock() has been called.
>
> Why is that unsafe?
> See __up_read(). It's doing preempt_disable().

Yep I didn't see the issue either that is why I asked for the stack trace. If
its a bug we would want a reproducer as well seems like it should be trivially
tested in selftests.

>
>
> > - if (irqs_disabled()) {
> > + if (in_atomic() || irqs_disabled()) {
>
> We cannot do this. It will significantly hurt stack traces with build_id.