# Re: \[PATCH bpf\] bpf: fix nested bpf tracepoints with per-cpu data

- From: Steven Rostedt \<rostedt@goodmis.org\>
- Date: 2019-06-07 07:55:37 
- List: netdev
- Page: https://ratatoskr.run/netdev/2019/06/5620614
- Thread: https://ratatoskr.run/bpf/2019/05/5614724/t.md

```text
On Thu, 6 Jun 2019 19:59:18 -0700
Andrii Nakryiko [ref] wrote:

> On Thu, Jun 6, 2019 at 1:17 PM Matt Mullins [ref] wrote:
> >
> > BPF_PROG_TYPE_RAW_TRACEPOINTs can be executed nested on the same CPU, as
> > they do not increment bpf_prog_active while executing.
> >
> > This enables three levels of nesting, to support
> >   - a kprobe or raw tp or perf event,
> >   - another one of the above that irq context happens to call, and
> >   - another one in nmi context  
> 
> Can NMIs be nested?

No, otherwise several things in the kernel will break.

-- Steve
```