Re: [PATCH net-next v4] selftests: net: add tests for PPP

[Guillaume Nault <gnault@redhat.com>]

On Fri, Mar 27, 2026 at 08:39:11PM +0800, Qingfang Deng wrote:
> On Fri, Mar 27, 2026 at 7:28 PM Guillaume Nault <gnault@redhat.com> wrote:
> >
> > On Thu, Mar 26, 2026 at 04:26:23PM +0800, Qingfang Deng wrote:
> > > Hi Sebastian,
> > >
> > > On Thu, Mar 26, 2026 at 4:13 PM Sebastian Andrzej Siewior
> > > <bigeasy@linutronix.de> wrote:
> > > >
> > > > On 2026-03-26 10:16:24 [+0800], Qingfang Deng wrote:
> > > > > Add ping and iperf3 tests for ppp_async.c and pppoe.c.
> > > >
> > > > Oh thank you for doing this.
> > > > I haven't look in detail but this cover the "invalid loop" cases that
> > > > ppp tries to catch?
> > >
> > > By "invalid loop", do you mean transmit recursion?
> > > https://lore.kernel.org/all/20250715150806.700536-2-bigeasy@linutronix.de/
> > >
> > > AFAIK, this can only happen with PPTP or L2TP, which were not included
> > > in this patch.
> >
> > The problem was originally reproduced using L2TP, indeed. But I guess
> > that it could also be reproduced with PPPoE by using a UDP tunnel
> > device like VXLAN (like sending a packet through a PPP interface,
> > handled by PPPoE, running on top of a VXLAN device, that routes the UDP
> > encapsulated packet back to the original PPP interface).
> 
> Yeah I've thought of that also. It can technically happen, though
> there's no practical use. For self tests purpose, I may just add the
> recursion test into PPTP or L2TP tests.

Yes, no problem.

> Regards,
> Qingfang
>