Re: [PATCH net-next 0/2] virtio-net: re enable XDP_REDIRECT for mergeable buffer

From: "Michael S. Tsirkin" <mst@redhat.com>
To: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: netdev@vger.kernel.org, john.fastabend@gmail.com,
	linux-kernel@vger.kernel.org,
	virtualization@lists.linux-foundation.org,
	Alexei Starovoitov <alexei.starovoitov@gmail.com>
Subject: Re: [PATCH net-next 0/2] virtio-net: re enable XDP_REDIRECT for mergeable buffer
Date: Thu, 1 Mar 2018 15:40:26 +0200	[thread overview]
Message-ID: <20180301153900-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <20180301113531.7b25e2df@redhat.com>

On Thu, Mar 01, 2018 at 11:35:31AM +0100, Jesper Dangaard Brouer wrote:
> On Thu, 1 Mar 2018 17:23:37 +0800
> Jason Wang <jasowang@redhat.com> wrote:
> 
> > On 2018年03月01日 17:10, Jesper Dangaard Brouer wrote:
> > > On Thu,  1 Mar 2018 11:19:03 +0800
> > > Jason Wang <jasowang@redhat.com> wrote:
> > >  
> > >> This series tries to re-enable XDP_REDIRECT for mergeable buffer which
> > >> was removed since commit 7324f5399b06 ("virtio_net: disable
> > >> XDP_REDIRECT in receive_mergeable() case"). Main concerns are:
> > >>
> > >> - not enough tailroom was reserved which breaks cpumap  
> >
> > > To address this at a more fundamental level, I would suggest that we/you
> > > instead extend XDP to know it's buffers "frame" size/end.  (The
> > > assumption use to be, xdp_buff->data_hard_start + PAGE_SIZE, but
> > > ixgbe+virtio_net broke that assumption).
> > >
> > > It should actually be fairly easy to implement:
> > >   * Simply extend xdp_buff with a "data_hard_end" pointer.  
> > 
> > Right, and then cpumap can warn and drop packets with insufficient 
> > tailroom.
> >
> > But it should be a patch on top of this I think.
> 
> Hmmm, not really.  If we/you instead fix the issue of XDP doesn't know
> the end/size of the frame, then we don't need this mixed XDP
> generic/native code path mixing.
> 
> You could re-enable native redirect, and push the responsibility to
> cpumap for detecting this too-small frame "missing tailroom" (and avoid
> crashing...). (If we really want to support this, cpumap could fallback
> to dev_alloc_skb, and handle it gracefully).

Yea, we probably should.

However it's not nice that redirect is now gone in net.
IMHO a smaller version of patch 1/2 (without using generic code)
should go into net. tailroom tracking and fallback to dev_alloc_skb
can go into net-next.

> -- 
> Best regards,
>   Jesper Dangaard Brouer
>   MSc.CS, Principal Kernel Engineer at Red Hat
>   LinkedIn: http://www.linkedin.com/in/brouer
_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization