Re: [PATCH net v3] seg6: Fix unset mac header in SRv6 End.DT VRF decapsulation
From: Andrea Mayer
Date: Mon Nov 09 2026 - 19:39:34 EST
On Fri, 6 Nov 2026 14:56:07 +0800
Boyan Liu <yymhvert@xxxxxxxxx> wrote:
> Hi Andrea,
>
> It has been about two weeks since you mentioned that you were
> looking into the NFQUEUE case.
Hi Boyan,
I followed up on the NFQUEUE case on 26 Oct, in this reply to the v3:
https://lore.kernel.org/netdev/20261026003415.f7c46fa5a495144c29c03a07@xxxxxxxxxxx/
> My understanding is that NFQUEUE is the remaining open question
> for v3. Did you get a chance to check it? Should I prepare a v4
> to address that case, or is the current approach acceptable?
As I wrote there, setting the mac header at the network header with
mac_len 0 looks fine, and the reported crash is gone with v3 in my
tests. The NFQUEUE case in the Sashiko review also exists without this
patch, so it is not introduced by v3. On the SRv6 side, I think the
current approach is acceptable, and the NFQUEUE case can be handled
separately.
Ciao,
Andrea