From: "Björn Töpel" <bjorn.topel@gmail.com>
To: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: "Karlsson, Magnus" <magnus.karlsson@intel.com>,
Magnus Karlsson <magnus.karlsson@gmail.com>,
"Duyck, Alexander H" <alexander.h.duyck@intel.com>,
Alexander Duyck <alexander.duyck@gmail.com>,
Alexei Starovoitov <ast@fb.com>,
Daniel Borkmann <daniel@iogearbox.net>,
Netdev <netdev@vger.kernel.org>,
MykytaI Iziumtsev <mykyta.iziumtsev@linaro.org>,
John Fastabend <john.fastabend@gmail.com>,
Willem de Bruijn <willemdebruijn.kernel@gmail.com>,
"Michael S. Tsirkin" <mst@redhat.com>,
michael.lundkvist@ericsson.com, "Brandeburg,
Jesse" <jesse.brandeburg@intel.com>,
"Singhai, Anjali" <anjali.singhai@intel.com>,
"Zhang, Qi Z" <qi.z.zhang@intel.com>,
Francois Ozog <francois.ozog@linaro.org>,
Ilias Apalodimas <ilias.apalodimas@linaro.org>,
Brian Brooks <brian.brook
Subject: Re: [PATCH bpf-next 10/11] i40e: implement AF_XDP zero-copy support for Tx
Date: Tue, 5 Jun 2018 15:07:52 +0200 [thread overview]
Message-ID: <CAJ+HfNiGXtbjvAV8PUkBC5OL=caTiaOZYues1VJCEvo8PsMeRA@mail.gmail.com> (raw)
In-Reply-To: <20180605144347.34351ecc@redhat.com>
Den tis 5 juni 2018 kl 14:44 skrev Jesper Dangaard Brouer <brouer@redhat.com>:
>
> On Mon, 4 Jun 2018 14:06:00 +0200
> Björn Töpel <bjorn.topel@gmail.com> wrote:
>
> > Here, ndo_xsk_async_xmit is implemented. As a shortcut, the existing
> > XDP Tx rings are used for zero-copy. This will result in other devices
> > doing XDP_REDIRECT to an AF_XDP enabled queue will have its packets
> > dropped.
>
> This behavior is problematic, because XDP Tx rings are smp_processor_id
> based, and several RX queues can (via proc smp_affinity settings) be
> assigned to the same CPU. Thus, other RX-queues (than the AF_XDP
> enabled queue) can experience packet drops. And other devices doing
> redirect through i40e, which happen to run on a CPU which XDP Tx queue
> is "hijacked" will see dropped packets.
>
> Any plans to allocate/create a dedicated TX ring per AF_XDP socket?
>
Yes -- again this was a shortcut, and must be addressed (for all the
reasons above).
Björn
> --
> Best regards,
> Jesper Dangaard Brouer
> MSc.CS, Principal Kernel Engineer at Red Hat
> LinkedIn: http://www.linkedin.com/in/brouer
next prev parent reply other threads:[~2018-06-05 13:08 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-06-04 12:05 [PATCH bpf-next 00/11] AF_XDP: introducing zero-copy support Björn Töpel
2018-06-04 12:05 ` [PATCH bpf-next 01/11] xsk: moved struct xdp_umem definition Björn Töpel
2018-06-04 12:05 ` [PATCH bpf-next 02/11] xsk: introduce xdp_umem_page Björn Töpel
2019-03-13 9:39 ` [bpf-next,02/11] " Jiri Slaby
2019-03-13 11:23 ` Björn Töpel
2018-06-04 12:05 ` [PATCH bpf-next 03/11] net: xdp: added bpf_netdev_command XDP_{QUERY,SETUP}_XSK_UMEM Björn Töpel
2018-06-04 12:05 ` [PATCH bpf-next 04/11] xdp: add MEM_TYPE_ZERO_COPY Björn Töpel
2018-06-04 12:05 ` [PATCH bpf-next 05/11] xsk: add zero-copy support for Rx Björn Töpel
2018-06-04 12:05 ` [PATCH bpf-next 06/11] net: added netdevice operation for Tx Björn Töpel
2018-06-04 12:05 ` [PATCH bpf-next 07/11] xsk: wire upp Tx zero-copy functions Björn Töpel
2018-06-04 12:05 ` [PATCH bpf-next 08/11] i40e: added queue pair disable/enable functions Björn Töpel
2018-06-04 12:05 ` [PATCH bpf-next 09/11] i40e: implement AF_XDP zero-copy support for Rx Björn Töpel
2018-06-04 20:35 ` Alexander Duyck
2018-06-07 7:40 ` Björn Töpel
2018-06-04 12:06 ` [PATCH bpf-next 10/11] i40e: implement AF_XDP zero-copy support for Tx Björn Töpel
2018-06-04 20:53 ` Alexander Duyck
2018-06-05 12:43 ` Jesper Dangaard Brouer
2018-06-05 13:07 ` Björn Töpel [this message]
2018-06-04 12:06 ` [PATCH bpf-next 11/11] samples/bpf: xdpsock: use skb Tx path for XDP_SKB Björn Töpel
2018-06-04 16:38 ` [PATCH bpf-next 00/11] AF_XDP: introducing zero-copy support Alexei Starovoitov
2018-06-04 20:29 ` [Intel-wired-lan] " Jeff Kirsher
2018-11-14 8:10 ` af_xdp zero copy ideas Michael S. Tsirkin
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CAJ+HfNiGXtbjvAV8PUkBC5OL=caTiaOZYues1VJCEvo8PsMeRA@mail.gmail.com' \
--to=bjorn.topel@gmail.com \
--cc=alexander.duyck@gmail.com \
--cc=alexander.h.duyck@intel.com \
--cc=anjali.singhai@intel.com \
--cc=ast@fb.com \
--cc=brouer@redhat.com \
--cc=daniel@iogearbox.net \
--cc=francois.ozog@linaro.org \
--cc=ilias.apalodimas@linaro.org \
--cc=jesse.brandeburg@intel.com \
--cc=john.fastabend@gmail.com \
--cc=magnus.karlsson@gmail.com \
--cc=magnus.karlsson@intel.com \
--cc=michael.lundkvist@ericsson.com \
--cc=mst@redhat.com \
--cc=mykyta.iziumtsev@linaro.org \
--cc=netdev@vger.kernel.org \
--cc=qi.z.zhang@intel.com \
--cc=willemdebruijn.kernel@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).