On Thu, 2 Apr 2020 08:40:23 -0700
Alexei Starovoitov <alexei.starovoitov@xxxxxxxxx> wrote:
> On Thu, Apr 2, 2020 at 2:06 AM Jesper Dangaard Brouer <brouer@xxxxxxxxxx> wrote:
> >
> > On Thu, 2 Apr 2020 09:47:03 +0900
> > Toshiaki Makita <toshiaki.makita1@xxxxxxxxx> wrote:
> >
> > > On 2020/04/02 1:15, Jesper Dangaard Brouer wrote:
> > > ...
> > > > [PATCH RFC net-next] veth: adjust hard_start offset on redirect XDP frames
> > > >
> > > > When native XDP redirect into a veth device, the frame arrives in the
> > > > xdp_frame structure. It is then processed in veth_xdp_rcv_one(),
> > > > which can run a new XDP bpf_prog on the packet. Doing so requires
> > > > converting xdp_frame to xdp_buff, but the tricky part is that
> > > > xdp_frame memory area is located in the top (data_hard_start) memory
> > > > area that xdp_buff will point into.
> > > >
> > > > The current code tried to protect the xdp_frame area, by assigning
> > > > xdp_buff.data_hard_start past this memory. This results in 32 bytes
> > > > less headroom to expand into via BPF-helper bpf_xdp_adjust_head().
> > > >
> > > > This protect step is actually not needed, because BPF-helper
> > > > bpf_xdp_adjust_head() already reserve this area, and don't allow
> > > > BPF-prog to expand into it. Thus, it is safe to point data_hard_start
> > > > directly at xdp_frame memory area.
> > > >
> > > > Cc: Toshiaki Makita <makita.toshiaki@xxxxxxxxxxxxx>
> > >
> > > FYI: This mail address is deprecated.
> > >
> > > > Fixes: 9fc8d518d9d5 ("veth: Handle xdp_frames in xdp napi ring")
> > > > Reported-by: Mao Wenan <maowenan@xxxxxxxxxx>
> > > > Signed-off-by: Jesper Dangaard Brouer <brouer@xxxxxxxxxx>
> > >
> > > FWIW,
> > >
> > > Acked-by: Toshiaki Makita <toshiaki.makita1@xxxxxxxxx>
> >
> > Thanks.
> >
> > I have updated your email and added your ack in my patchset. I will
> > submit this officially once net-next opens up again[1], as part my
> > larger patchset for introducing XDP frame_sz.
>
> It looks like bug fix to me.
> The way I read it that behavior of bpf_xdp_adjust_head() is a bit
> buggy with veth netdev,
> so why wait ?
I want to wait to ease your life as maintainer. This is part of a
larger patchset (for XDP frame_sz) and the next patch touch same code
path and thus depend on these code adjustments. If we apply them in
bpf vs bpf-next then you/we will have to handle merge conflicts. The
severity of the "fix" is really low, it only means 32 bytes less
headroom (which I doubt anyone is using).
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Principal Kernel Engineer at Red Hat
LinkedIn: http://www.linkedin.com/in/brouer
