[Intel-wired-lan] [PATCH net v4] igb: Fix XDP with PTP enabled

Jesper Dangaard Brouer brouer at redhat.com
Wed May 19 13:34:18 UTC 2021


Hi Maintainers,

What is the status on this patch?

I don't see this fix being applied on git-trees net or net-next.

[0] 20210503072800.79936-1-kurt at linutronix.de
[1] https://patchwork.kernel.org/project/netdevbpf/patch/20210503072800.79936-1-kurt@linutronix.de/
[2] https://lore.kernel.org/netdev/20210503072800.79936-1-kurt@linutronix.de/


On Tue, 4 May 2021 10:28:27 +0200
Jesper Dangaard Brouer <brouer at redhat.com> wrote:

> On Mon,  3 May 2021 09:28:00 +0200
> Kurt Kanzenbach <kurt at linutronix.de> wrote:
> 
> > When using native XDP with the igb driver, the XDP frame data doesn't point to
> > the beginning of the packet. It's off by 16 bytes. Everything works as expected
> > with XDP skb mode.
> > 
> > Actually these 16 bytes are used to store the packet timestamps. Therefore, pull
> > the timestamp before executing any XDP operations and adjust all other code
> > accordingly. The igc driver does it like that as well.
> > 
> > Tested with Intel i210 card and AF_XDP sockets.
> > 
> > Fixes: 9cbc948b5a20 ("igb: add XDP support")
> > Signed-off-by: Kurt Kanzenbach <kurt at linutronix.de>  
> 
> Thanks for fixing this!
> 
> Acked-by: Jesper Dangaard Brouer <brouer at redhat.com>
> 
> I expect that we/I will (soon) play with getting this area that is
> stored in front of the packet (the XDP data_meta area) described via
> BTF.  This way both xdp_frame and AF_XDP can get structured access (e.g.
> to the PTP timestamp in this case).
> 
> I'll be adding my notes on this project here:
>  https://github.com/xdp-project/xdp-project/blob/master/areas/tsn/
> 
> Looking forward to collaborate on this with you :-)



-- 
Best regards,
  Jesper Dangaard Brouer
  MSc.CS, Principal Kernel Engineer at Red Hat
  LinkedIn: http://www.linkedin.com/in/brouer



More information about the Intel-wired-lan mailing list