netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Michael S. Tsirkin" <mst@redhat.com>
To: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Cc: netdev@vger.kernel.org, "David S. Miller" <davem@davemloft.net>,
	"Eric Dumazet" <edumazet@google.com>,
	"Jakub Kicinski" <kuba@kernel.org>,
	"Paolo Abeni" <pabeni@redhat.com>,
	"Jason Wang" <jasowang@redhat.com>,
	"Björn Töpel" <bjorn@kernel.org>,
	"Magnus Karlsson" <magnus.karlsson@intel.com>,
	"Maciej Fijalkowski" <maciej.fijalkowski@intel.com>,
	"Jonathan Lemon" <jonathan.lemon@gmail.com>,
	"Alexei Starovoitov" <ast@kernel.org>,
	"Daniel Borkmann" <daniel@iogearbox.net>,
	"Jesper Dangaard Brouer" <hawk@kernel.org>,
	"John Fastabend" <john.fastabend@gmail.com>,
	"Sebastian Andrzej Siewior" <bigeasy@linutronix.de>,
	"Menglong Dong" <imagedong@tencent.com>,
	"Kuniyuki Iwashima" <kuniyu@amazon.com>,
	"Petr Machata" <petrm@nvidia.com>,
	virtualization@lists.linux-foundation.org, bpf@vger.kernel.org
Subject: Re: [PATCH 00/33] virtio-net: support AF_XDP zero copy
Date: Thu, 2 Feb 2023 06:08:30 -0500	[thread overview]
Message-ID: <20230202060757-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <20230202110058.130695-1-xuanzhuo@linux.alibaba.com>

On Thu, Feb 02, 2023 at 07:00:25PM +0800, Xuan Zhuo wrote:
> XDP socket(AF_XDP) is an excellent bypass kernel network framework. The zero
> copy feature of xsk (XDP socket) needs to be supported by the driver. The
> performance of zero copy is very good.

Great! Any numbers to share?

> mlx5 and intel ixgbe already support
> this feature, This patch set allows virtio-net to support xsk's zerocopy xmit
> feature.
> 
> Virtio-net did not support per-queue reset, so it was impossible to support XDP
> Socket Zerocopy. At present, we have completed the work of Virtio Spec and
> Kernel in Per-Queue Reset. It is time for Virtio-Net to complete the support for
> the XDP Socket Zerocopy.
> 
> Virtio-net can not increase the queue at will, so xsk shares the queue with
> kernel.
> 
> On the other hand, Virtio-Net does not support generate interrupt manually, so
> when we wakeup tx xmit, we used some tips. If the CPU run by TX NAPI last time
> is other CPUs, use IPI to wake up NAPI on the remote CPU. If it is also the
> local CPU, then we wake up sofrirqd.
> 
> Please review.
> 
> Thanks.
> 
> 
> Xuan Zhuo (33):
>   virtio_ring: virtqueue_add() support premapped
>   virtio_ring: split: virtqueue_add_split() support premapped
>   virtio_ring: packed: virtqueue_add_packed() support premapped
>   virtio_ring: introduce virtqueue_add_outbuf_premapped()
>   virtio_ring: introduce virtqueue_add_inbuf_premapped()
>   virtio_ring: introduce virtqueue_reset()
>   virtio_ring: add api virtio_dma_map() for advance dma
>   virtio_ring: introduce dma sync api for virtio
>   xsk: xsk_buff_pool add callback for dma_sync
>   xsk: support virtio DMA map
>   virtio_net: rename free_old_xmit_skbs to free_old_xmit
>   virtio_net: unify the code for recycling the xmit ptr
>   virtio_net: virtnet_poll_tx support rescheduled
>   virtio_net: independent directory
>   virtio_net: move to virtio_net.h
>   virtio_net: introduce virtnet_xdp_handler() to seprate the logic of
>     run xdp
>   virtio_net: receive_small() use virtnet_xdp_handler()
>   virtio_net: receive_merageable() use virtnet_xdp_handler()
>   virtio_net: introduce virtnet_tx_reset()
>   virtio_net: xsk: introduce virtnet_rq_bind_xsk_pool()
>   virtio_net: xsk: introduce virtnet_xsk_pool_enable()
>   virtio_net: xsk: introduce xsk disable
>   virtio_net: xsk: support xsk setup
>   virtio_net: xsk: stop disable tx napi
>   virtio_net: xsk: __free_old_xmit distinguishes xsk buffer
>   virtio_net: virtnet_sq_free_unused_buf() check xsk buffer
>   virtio_net: virtnet_rq_free_unused_buf() check xsk buffer
>   net: introduce napi_tx_raise()
>   virtio_net: xsk: tx: support tx
>   virtio_net: xsk: tx: support wakeup
>   virtio_net: xsk: tx: auto wakeup when free old xmit
>   virtio_net: xsk: rx: introduce add_recvbuf_xsk()
>   virtio_net: xsk: rx: introduce receive_xsk() to recv xsk buffer
> 
>  MAINTAINERS                                 |   2 +-
>  drivers/net/Kconfig                         |   8 +-
>  drivers/net/Makefile                        |   2 +-
>  drivers/net/virtio/Kconfig                  |  11 +
>  drivers/net/virtio/Makefile                 |   8 +
>  drivers/net/{virtio_net.c => virtio/main.c} | 564 +++++++-------------
>  drivers/net/virtio/virtio_net.h             | 317 +++++++++++
>  drivers/net/virtio/xsk.c                    | 524 ++++++++++++++++++
>  drivers/net/virtio/xsk.h                    |  33 ++
>  drivers/virtio/virtio_ring.c                | 376 +++++++++++--
>  include/linux/netdevice.h                   |   7 +
>  include/linux/virtio.h                      |  29 +
>  include/net/xsk_buff_pool.h                 |   6 +
>  net/core/dev.c                              |  11 +
>  net/xdp/xsk_buff_pool.c                     |  79 ++-
>  15 files changed, 1541 insertions(+), 436 deletions(-)
>  create mode 100644 drivers/net/virtio/Kconfig
>  create mode 100644 drivers/net/virtio/Makefile
>  rename drivers/net/{virtio_net.c => virtio/main.c} (92%)
>  create mode 100644 drivers/net/virtio/virtio_net.h
>  create mode 100644 drivers/net/virtio/xsk.c
>  create mode 100644 drivers/net/virtio/xsk.h
> 
> -- 
> 2.32.0.3.g01195cf9f


  parent reply	other threads:[~2023-02-02 11:09 UTC|newest]

Thread overview: 79+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-02-02 11:00 [PATCH 00/33] virtio-net: support AF_XDP zero copy Xuan Zhuo
2023-02-02 11:00 ` [PATCH 01/33] virtio_ring: virtqueue_add() support premapped Xuan Zhuo
2023-02-02 11:00 ` [PATCH 02/33] virtio_ring: split: virtqueue_add_split() " Xuan Zhuo
2023-02-02 11:00 ` [PATCH 03/33] virtio_ring: packed: virtqueue_add_packed() " Xuan Zhuo
2023-02-03  9:16   ` Michael S. Tsirkin
2023-02-02 11:00 ` [PATCH 04/33] virtio_ring: introduce virtqueue_add_outbuf_premapped() Xuan Zhuo
2023-02-02 11:00 ` [PATCH 05/33] virtio_ring: introduce virtqueue_add_inbuf_premapped() Xuan Zhuo
2023-02-02 11:00 ` [PATCH 06/33] virtio_ring: introduce virtqueue_reset() Xuan Zhuo
2023-02-03  9:05   ` Michael S. Tsirkin
2023-02-03  9:09     ` Xuan Zhuo
2023-02-13 12:15       ` Michael S. Tsirkin
2023-02-14  1:53         ` Xuan Zhuo
2023-02-02 11:00 ` [PATCH 07/33] virtio_ring: add api virtio_dma_map() for advance dma Xuan Zhuo
2023-02-03  9:07   ` Michael S. Tsirkin
2023-02-02 11:00 ` [PATCH 08/33] virtio_ring: introduce dma sync api for virtio Xuan Zhuo
2023-02-02 12:44   ` Magnus Karlsson
2023-02-03  9:24   ` Michael S. Tsirkin
2023-02-02 11:00 ` [PATCH 09/33] xsk: xsk_buff_pool add callback for dma_sync Xuan Zhuo
2023-02-02 12:51   ` Magnus Karlsson
2023-02-03  7:01     ` Xuan Zhuo
2023-02-02 11:00 ` [PATCH 10/33] xsk: support virtio DMA map Xuan Zhuo
2023-02-05 22:04   ` kernel test robot
2023-02-02 11:00 ` [PATCH 11/33] virtio_net: rename free_old_xmit_skbs to free_old_xmit Xuan Zhuo
2023-02-02 11:00 ` [PATCH 12/33] virtio_net: unify the code for recycling the xmit ptr Xuan Zhuo
2023-02-02 11:00 ` [PATCH 13/33] virtio_net: virtnet_poll_tx support rescheduled Xuan Zhuo
2023-02-02 11:00 ` [PATCH 14/33] virtio_net: independent directory Xuan Zhuo
2023-02-02 11:00 ` [PATCH 15/33] virtio_net: move to virtio_net.h Xuan Zhuo
2023-02-03  8:53   ` Michael S. Tsirkin
2023-02-03  9:04     ` Xuan Zhuo
2023-02-03  9:26       ` Michael S. Tsirkin
2023-02-02 11:00 ` [PATCH 16/33] virtio_net: introduce virtnet_xdp_handler() to seprate the logic of run xdp Xuan Zhuo
2023-02-03  8:55   ` Michael S. Tsirkin
2023-02-03  9:01     ` Xuan Zhuo
2023-02-02 11:00 ` [PATCH 17/33] virtio_net: receive_small() use virtnet_xdp_handler() Xuan Zhuo
2023-02-02 11:00 ` [PATCH 18/33] virtio_net: receive_merageable() " Xuan Zhuo
2023-02-02 17:16   ` Michael S. Tsirkin
2023-02-02 11:00 ` [PATCH 19/33] virtio_net: introduce virtnet_tx_reset() Xuan Zhuo
2023-02-02 17:23   ` Michael S. Tsirkin
2023-02-03  4:35     ` Xuan Zhuo
2023-02-02 11:00 ` [PATCH 20/33] virtio_net: xsk: introduce virtnet_rq_bind_xsk_pool() Xuan Zhuo
2023-02-03  8:48   ` Michael S. Tsirkin
2023-02-03  8:52     ` Xuan Zhuo
2023-02-03  9:28       ` Michael S. Tsirkin
2023-02-02 11:00 ` [PATCH 21/33] virtio_net: xsk: introduce virtnet_xsk_pool_enable() Xuan Zhuo
2023-02-02 11:00 ` [PATCH 22/33] virtio_net: xsk: introduce xsk disable Xuan Zhuo
2023-02-02 23:02   ` kernel test robot
2023-02-12  7:56   ` kernel test robot
2023-02-02 11:00 ` [PATCH 23/33] virtio_net: xsk: support xsk setup Xuan Zhuo
2023-02-02 11:00 ` [PATCH 24/33] virtio_net: xsk: stop disable tx napi Xuan Zhuo
2023-02-02 17:25   ` Michael S. Tsirkin
2023-02-03  3:24     ` Xuan Zhuo
2023-02-03  8:33       ` Michael S. Tsirkin
2023-02-03  8:49         ` Xuan Zhuo
2023-02-03  9:29           ` Michael S. Tsirkin
2023-02-02 11:00 ` [PATCH 25/33] virtio_net: xsk: __free_old_xmit distinguishes xsk buffer Xuan Zhuo
2023-02-02 11:00 ` [PATCH 26/33] virtio_net: virtnet_sq_free_unused_buf() check " Xuan Zhuo
2023-02-02 11:00 ` [PATCH 27/33] virtio_net: virtnet_rq_free_unused_buf() " Xuan Zhuo
2023-02-02 11:00 ` [PATCH 28/33] net: introduce napi_tx_raise() Xuan Zhuo
2023-02-02 11:00 ` [PATCH 29/33] virtio_net: xsk: tx: support tx Xuan Zhuo
2023-02-03  8:39   ` Maciej Fijalkowski
2023-02-03  8:55     ` Xuan Zhuo
2023-02-02 11:00 ` [PATCH 30/33] virtio_net: xsk: tx: support wakeup Xuan Zhuo
2023-02-02 11:00 ` [PATCH 31/33] virtio_net: xsk: tx: auto wakeup when free old xmit Xuan Zhuo
2023-02-02 11:00 ` [PATCH 32/33] virtio_net: xsk: rx: introduce add_recvbuf_xsk() Xuan Zhuo
2023-02-03  8:43   ` Maciej Fijalkowski
2023-02-03  8:56     ` Xuan Zhuo
2023-02-02 11:00 ` [PATCH 33/33] virtio_net: xsk: rx: introduce receive_xsk() to recv xsk buffer Xuan Zhuo
2023-02-02 11:08 ` Michael S. Tsirkin [this message]
2023-02-02 11:44   ` [PATCH 00/33] virtio-net: support AF_XDP zero copy Xuan Zhuo
2023-02-03  9:08     ` Michael S. Tsirkin
2023-02-03  9:09       ` Xuan Zhuo
2023-02-02 14:41 ` Paolo Abeni
2023-02-03  3:33   ` Xuan Zhuo
2023-02-03  8:37     ` Michael S. Tsirkin
2023-02-03  8:46       ` Maciej Fijalkowski
2023-02-03  9:09         ` Michael S. Tsirkin
2023-02-03  9:17     ` Michael S. Tsirkin
2023-02-06  2:41       ` Xuan Zhuo
2023-02-13 12:14         ` 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=20230202060757-mutt-send-email-mst@kernel.org \
    --to=mst@redhat.com \
    --cc=ast@kernel.org \
    --cc=bigeasy@linutronix.de \
    --cc=bjorn@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=hawk@kernel.org \
    --cc=imagedong@tencent.com \
    --cc=jasowang@redhat.com \
    --cc=john.fastabend@gmail.com \
    --cc=jonathan.lemon@gmail.com \
    --cc=kuba@kernel.org \
    --cc=kuniyu@amazon.com \
    --cc=maciej.fijalkowski@intel.com \
    --cc=magnus.karlsson@intel.com \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=petrm@nvidia.com \
    --cc=virtualization@lists.linux-foundation.org \
    --cc=xuanzhuo@linux.alibaba.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).